,rp „. --gp^- *r H tt^fFfwf B?
"4 ff t ^s,» , %«*<*<•*'•{
I , w i. <• H\ r- J*rt 4n
Vs-
'"s*
B*
t x* ^ 5 W
»"«. ft t«^%«
^.SWU^s^* ™?fc
T*
1 / *
% !/3 ;" S«
United States
Environmental Protection
Agency
Office of Research and
Development
Washington DC 20460
EPA/620/R-93/010
January 1993
Agroecosystem
1992 Pilot Project Plan
Environmental Monitoring and
Assessment Program
-------
-------
EPAJ620JR-93J&10
January 1993
Environmental Monitoring and Assessment Program
Agroecosystem 1992 Pilot Project Plan
(April 3, 1992)
by
Walter W. Heck, Technical Director
C. Lee Campbell, Associate Director
Alva L. Finkner
Craig M. Hayes
George R. Hess
Julie R. Meyer
Michael J. Minister
Deborah Neher
Steven L. Peck
John O. Rawlings
Charles N. Smith
Mark B. Tooley
with support from
Robert P. Breckenridge
Virginia M. Lesser
Thomas J. Moser
Gail L. Olson
This study was conducted in cooperation with
U.S. Department of Agriculture
Agricultural Research Service
Raleigh, NC 27711
U.S. Environmental Protection Agency
Office of Research and Development
Washington, D.C. 20460
Environmental Monitoring Systems Laboratory
Las Vegas, NV 89193
Atmospheric Research and Exposure Assessment Laboratory
Research Triangle Park, NC 27711
Environmental Research Laboratory
Athens, GA 30613
Environmental Research Laboratory
Corvallis, OR 97333
Printed on Recycled Paper
-------
NOTICE
Tin's research has been funded by the U.S. Environmental Protection Agency through its
Office of Research and Development (ORD) and was conducted with our research partners
under the management of the Environmental Monitoring Systems Laboratory - Las Vegas.
This work is in support of the Environmental Monitoring and Assessment Program (EMAP)
Issue. This report has been subjected to ORD's peer and administrative review and has
been approved as an EPA publication. Neither the U.S. EPA nor ORD endorses or
recommends any trade name or commercial product mentioned in this report. They are
mentioned solely for the purpose of description and/or clarification.
Proper citation of this document is:
Heck, W.W., Campbell, C.L., Finkner, A.L., Hayes, CM., Hess, G.R., Meyer, J.R., Munster,
MJ., Neher, D., Peck, S.L., Rawlings, J.O., Smith, C.N., and Tooley, M.B. 1992,
Environmental Monitoring and Assessment Program - Agroecosystem 1992 Pilot Project
Plan. EPA/620/R-93/010. U.S. Environmental Protection Agency, Washington, D.C.
- n -
-------
Table of Contents
Notice ii
List of Figures ...'..' v
List of Tables vi
Glossary of Acronyms viii
Acknowledgements ix
1. Introduction . . . . 1-1
1.1. Overview of The Agroecosystem Program 1-1
1.2. Cooperative Interaction with The National Agricultural Statistics Service
(NASS) 1-6
1.3. Cooperative Interactions with Other National Programs and Agencies . . 1-6
1.4. Projected Implementation Schedule for a National Agroecosystem
Monitoring Program 1-7
2. The 1992 Pilot Project: Rationale and Objectives 2-1
2.1. Rationale 2-1
2.2. Objectives 2-3
3. Design and Statistical Considerations 3-1
3.1. Selection of the Pilot Sample Segments for Each Plan 3-1
3.2. Evaluation of The Two Plans 3-6
3.3. Within Segment Sampling Protocols 3-10
3.4. Analysis 3-19
4. Assessment Endpoints and Indicators 4-1
4.1. Societal Values, Assessment Endpoints, and Indicators ..."..- 4-1
4.2. Selection of Assessment Endpoints for The 1992 Pilot Project 4-4
4.3. Research Activities on Candidate Indicators and Assessment
Endpoints 4-7
4.4 Current Status of the Assessment of Endpoints for
the Agroecosystem Program 4-8
5. Description of Specific Assessment Endpoints for The Pilot Project 5.1-1
5.1. Crop Productivity 5.1-1
5.2. Soil Quality: Physical and Chemical Components 5.2-1
5.3. Water Quality 5.3-1
5.4. Land Use and Cover 5.4-1
5.5. Agricultural Chemical Use . 5.5-1
6. Description of Specific Research Endpoints for The Pilot Project 6.1-1
-------
6.1. Soil Biological Health 6.1-1
6.2. Landscape Structure 6.2 - 1
6.3. Water Quality - Groundwater Monitoring, Wells and Modeling .... 6.3-1
6.4. Biological Ozone - Indicator System 6.4-1
7. Quality Assurance 7-1
7.1. Introduction 7-1
7.2: NASS Quality Assurance Procedures 7-1
7.3. Soil Quality Measurements 7-7
7.4. Water Quality Measurements 7-7
7.5. GIS Data for Albermarle-Pamlico Regions (Landscape Regions) 7-8
7.6. Additional Data . 7-8
7.7. Data Quality Objectives . 7-8
8. Logistics , 8-1
8.1. Introduction 8-1
8.2. Logistics and the NASS . 8-1
8.3. Specific Logistics Elements 8-2
8.4. Logistics for the Biological Ozone-Indicator System and
the Well Comparison Study 8-18
9. Information Management 9-1
9.1. Introduction 9-1
9.2. Information Sources and Flow 9-2
9.3. Confidentiality of Data 9-5
9.4. Data Integration and Management 9-9
9.5. Data Access 9-9
9.6. Hardware and Software Requirements 9-10
10. Resources and Implementation 10-1
10.1. Introduction 10-1
10.2. Importance of the Pilot 10-2
10.3. Tasks and Schedule for the Pilot Project 10-2
10.4. Funding and Personnel Resources 10-2
Literature Cited , L - 1
Appendix 1. Agroecosystem Resource Group Members Al - I
Appendix 2. List of N.C. Counties Sampled in the 1992 Pilot Project .. A2 - 1
Appendix 3. Expected Data Summaries from the Agroecosystem 1992
North Carolina Pilot A3 - 1
Appendix 4. Methods - Soils Analyses A4 - 1
Appendix 5. NASS Survey Questionnaires A5 - 1
Appendix 6. Example Instructions for the Enumerators A6 - 1
Appendix 7. Sample Identification for QA/QC Procedures A7 - 1
Appendix 8. Response of Two White Clover Clones to Peanut Stunt Virus
and Ozone A8 - 1
- iv -
-------
List of Figures
Figure 1-1. The Agroecosystem Implementation Schedule . 1-10
Figure 3-1. Hexagon subsamples for use in choosing the NASS Sample Segments 3-3
Figure 3-2. North Carolina Counties containing NASS Segments for 1992 Pilot 3-6
Figure 3-3. Transect sampling of field . . ". 3-15
Figure 3-4. Transect sampling of field (Bounce rules) 3-16
Figure 3-5. Hypothetical cumulative density function (cdf) for electrical conductivity
of soil in North Carolina 3-21
Figure 3-6. Examples of box-plots for the electrical conductivity of soil in three
Regions of North Carolina . 3-23
Figure 3-7. A, Kriged estimates of USLE over North Carolina (an example only); B, Display
of spacial patterns 3-24
Figure 4-1. Agroecosystem societal values that will .b addressed with a suite of indicators
to determine the status and trends in agroecosystem health 4-2
Figure 5.1-1. Some factors which influence crop productivity . . . 5.1 - 1
Figure 5.1-2. Harvested acreage of several North Carolina crops, 1990 (preliminary) 5.1 - 11
Figure 5.2-1. An example of a cumulative distribution function: electrical conductivity of soil . . . 5.2 - 33
Figure 5.2-2. General soil map of the USA (USDA, 1975) 5.2-37
Figure 5.3-1. Sampling Design for Farm Pond 5.3 - 6
Figure 5.4-1. Use of NASS Area Frame data 5.4 - 8
Figure 5.4-2. Preliminary pie chart showing the area and proportion of land in each of the eight
NASS strata for North Carolina 5.4-10
Figure 5.4-3. Use of CGIA TM data 5.4-11
Figure 5.4-4. Use of NASS JES data 5.4-13
Figure 6.2-1. Analysis of Thematic Mapper Data for Landscape Structure 6.2 - 8
Figure 6.2-2. Analysis of Aerial Photography for Landscape Descriptors 6.2 - 9
Figure 6.2-3. Comparison of Statistical Analysis of Aerial Photos to
Completely Digitized Scenes . . 6.2 - 10
Figure 6.2-4. Framework for Exploratory Analyses and Integration 6.2 - 12
Figure 8-1. Major activities for the 1992 Agroecosystem Pilot 8-4
Figure 8-2. Example of a sample-tracking postcard to be sent to the ARG by the enumerators 8-7
Figure 8-3. Logistics flow chart for the 1992 Pilot Project 8-19
Figure 9-1. Overview of the flow of data through the AIC 9-1
Figure 9-2. Row of data collected by NASS to the AIC 9-3
Figure 9-3. Flow of data from other EMAP sources and other agencies and other
agencies and institutions to the AIC and NASS data center for integration 9-5
Figure 9-4. Use of existing data to perform validity checks on data 9-7
- V -
-------
List of Tables
Table 1-1. Resource, Integration and Coordination Groups of EMAP
Table 1-2. Monitoring Activities of EMAP Resource Groups in 1992
Table 1-3. Planned implementation of Agroecosystem monitoring and
assessment across EPA regions
Table 3-1. Stratification of NASS Segments in North Carolina for 1992
Table 3-2. Crops ineligible for selection in the Agroecosystem 1992 Pilot
Table 3-3. Degrees of freedom for field sampling components of variance .
Table 4-1. Association between the Agroecosystem assessment endpoints
and societal values
Table 4-2. Association between the Agroecosystem assessment endpoints and the
indicator types '....'
Table 4-3. Vital statistics on the Assessment Endpoints for the Agroecosystem
Program •
Table 5.1-1. Principal crops eligible for selection in the Agroecosystem 1992 Pilot ....
Table 5.1-2. Conversion factors from yield to net primary productivity (NPP)
Table 5.1-3. Elements of metadata to be recorded in association with data for the
crop productivity indicators, not including ancillary data such as weather . .
Table 5.1-4. Example output table for an indicator of crop productivity
Table 5.2-1. Description of physical and chemical soil quality indicators
Table 5.2-2. Research indices of soil quality
Table 5.2-3. Requested data elements from the SCS State Soil Survey Database
Table 5.2-4. Ratings of available water capacity (AWC) by moisture regime
Table 5.2-5. Ratings of soil pH
Table 5.2-6. General ratings for exchangeable sodium percentage
Table 5.2-7. Salinity ratings based on electrical conductivity
Table 5.2-8. Ratings of hydraulic conductivity recognized by the SCS
Table 5.2-9. Sources of data for the six Universal Soil Loss Equation (USLE)
factors
Table 5.2-10. Contents of enumerator kit
Table 5.2-11. Soil analytical laboratory parameters to be measured in the 1992 Pilot . . .
Table 5.2-12. Reporting units, precision and expected concentration ranges
(December 1990)
Table 5.2-13. Private and federal laboratories contacted for chemical and
physical analysis of soils
Table 5.2-14. Data quality objectives for measurement of soil samples within
the analytical laboratory and within fields
Table 5.2-15. Metadata-for chemical and physical analysis of soils
Table 5.2-16. SCS Land Capability Classes
Table 5.2-17. Examples of using soil depth for assigning soil loss tolerance values to soils
Table 5.2-18. Examples of soil assessments
Table 5.3-1. The anticipated analysis of variance
Table 5.4-1. Steps to convert NASS area frame to ARC format
Table 5.4-2. Classification system for Albermarle-Pamlico watershed land cover data . .
. 1-5
. 1-5
. 1-9
. 3-5
3- 11
3- 17
. 4-3
. 4-5
. 4-9
5.1 -4
5.1 -6
5.1 -9
5.1 - 20
5.2-2
'5.2-4
5.2-6
5.2 - 10
5.2 - 11
5.2 - 13
5.2 - 14
5.2 - 19
5.2 - 21
5.2 - 23
5.2 - 26
5.2 - 27
5.2 - 29
5.2 - 30
5.2 - 32
5.2 - 35
5.2 - 40
5.2 - 42
5.3 - 4
5.4 - 4
5.4-5
- vi -
-------
Table 5.4-3. NASS JES land use classification 5.4-6
Table 5.4-4. Steps to acquire JES data 5.4-6
Table 6.1-1. Reporting Units, Precision and Expected Ranges for Nematode Populations
(December 1991) . 6.1-6
Table 6.1-2. Data Quality Objectives for Enumeration of Nematodes by the Enumeration
Laboratory and Within Fields (October 1991) 6.1-7
Table 6.1-3. Metadata for Biological Analysis of Soils in the 1992 North Carolina Pilot 6.1-7
Table 6.2-1. Steps to Acquire Digitized Aerial Photography 6.2 - 3
Table 6.2-2. Proposed LCG Classification System 6.2-5
Table 6.2-3. Landscape Descriptors Currently Under Consideration for use in the
Agroecosystem Program . . . 6.2 - 7
Table 8-1. Logistical issues that have been addressed by the ARG 8-1
Table 8-2. Activities in the 1992 Agroecosystem Pilot Project 8-3
Table 9-1. Examples of existing data to be used for the 1992 Pilot Project 9-6
Table 9-2. Summary of confidentiality provisions of several government agencies
with data of value to the Agroecosystem Resource Group . . . 9-8
Table 9-3. Hardware and software requirements to support the 1992 Pilot Project 9-11
Table 10-1. Tasks with schedule for conducting the Pilot Project - NC Pilot
Plans (1992-93) . . . . . 10-3
Table 10-2. 1992 Activity Chart for the ARG 10-4
Table 10-3. Program Tasks with Budget for 1992 Pilot 10-6
Table 10-4. Pilot Budget by Location/Category 10-7
Table 10-5. Personnel/Responsibilities for the Agroecosystem Pilot Project 10-8
Table 10-6. 1993 Activity Chart for the ARG 10-10
Table 10-7. Program Activities with Budget for 1993 10-12
Table 10-8. 1993 Budget by Location 10-13
Table 10-9. Personnel/Responsibilities for the 1993 Agroecosystem Program 10-14
- Vll -
-------
Glossary of Acronyms
AIC
APHIS
ARC
ARS
ASCS
ASTM
AVHRR
AWC
BRG
C
CEC
CGIA
CRP
DAT
DBAPE
DC
DLG
DQO
EC
ECD
EIC
ELISA
EMAP
EPA
ERL
ERS
ESP
FPD
GI
CIS
Hall ECD
ffl
HQ
HT
ID
IM
IMC
INEL
JES
LAI
LAN
LCG
LPT
MLRA
NAPP
NASDA
NASS
NAWQA
NC
NC-R
NC-S
Agroecosystem Information Center
Animal and Plant Health Inspection Service
Agroecosystem Resource Group
Agricultural Research Service (USDA)
Agricultural Stabilization and Conservation Service
American Society for Testing and Materials
Advanced Very High Resolution Radiometer
Available water capacity
Business Resources Group, Inc.
Carbon
Cation exchange capacity; Commission for European Communities
Center for Geographic Information and Analysis
Conservation Reserve Program
Digital audio tape
Database Analyzer and Parameter Estimator
District of Columbia
Digital line graph
Data quality objective
Electrical conductivity
Electron capture detector
EMAP Information Center
Enzyme-linked immunosorbent assay
Environmental Monitoring and Assessment Program
Environmental Protection Agency
Environmental Research Laboratory
Economic Research Service (USDA)
Exchangeable sodium percentage
Flame photometric detector
Greenness index
Geographic information system
Hall electrolytic conductivity detector
Harvest index
Headquarters
Horvitz-Thompson
Identification/identifier
Information management
Information Management Committee
Idaho National Engineering Laboratory
June Enumerative Survey
Leaf area index
Local area network
Landscape Characterization Group
Landscape pattern type
Major Land Resource Area
National Aerial Photography Program
National Association of State Departments of Agriculture
National Agricultural Statistics Service (USDA)
National Water-Quality Assessment
North Carolina; North Central
Oyresistant clone of white clover
Oj-sensitive clone of white clover
- vm -
-------
Glossary of Acronyms cont'd.
NCCGIA
NCDA
NCSU
NDVI
ME
NGO
NOAA
NPD
NPP
NRI
O,
OM
OMB
ORD
PAR
PCR
PSU
QA
QA/QC
RDBMS
RUSTIC
SAR
SAS
SCS
SE
SI
SO
SOP
SSSD
T
TD
TM
USDA
USDC
USGS
USLE
UV
WE
WRI
North Carolina Center for Geographic Information and Analysis
North Carolina Department of Agriculture
North Carolina State University
Normalized Difference Vegetation Index
Northeast
Non-governmental organization
National Oceanic and Atmospheric Administration
Nitrogen-phosphorus detector
Net primary productivity
National Resources Inventory (USDA/SCS)
Ozone
Organic matter
Office of Management and Budget
Office of-Research and Development (EPA)
Photosynthetically active radiation
Post column reaction
Primary Sampling Unit
Quality assurance
Quality assurance/quality control
Relational database management system
Risk of Unsaturated/Saturated Transport and Transformation of Chemical Concentrations
Sodium absorption ratio
Statistical Analysis System
Soil Conservation Service (USDA)
Southeast
Systeme International (version of the metric system)
South
Standard operating procedure
State Soil Survey Database
Soil erosion tolerance factor
Technical Director
Thematic Mapper
United States Department of Agriculture
United States Department of Commerce
United States Geological Survey (USDC)
Universal Soil Loss Equation
Ultraviolet
West
World Resources Institute
- ix -
-------
Acknowledgements
The members of the Agroecosystem Resource Group have interacted with many people
during the development of this Plan. Dozens of people have shared information and expertise
during various Agroecosystem workshops. Many more have responded to our written inquiries
and telephone calls. The level of interest and cooperation from other government agencies,
universities and conservation organizations is appreciated. We would like to express our
gratitude to all who have contributed to our effort.
A special note of thanks is in order for Robert Bass and Steve Manheimer and those who
work with them at USDA/NASS. They have become an integral part of the Agroecosystem
Resource Group, bringing with them years of experience in developing spatial sampling
techniques for agricultural statistics. Their insight has proven invaluable and helped us avoid
many of the pitfalls associated with a program as broad and diverse as EMAP. We would also
like to thank Ray Halley, our administrative contact at USDA/NASS, who has given much
support to NASS' involvement with the Program.
Special thanks go to Bruce Jones for his programmatic suggestions, ideas and support, and
to Ann Pitchford for her dedicated administrative support to the Program.
We would also like to thank the ex-officio members of the Agroecosystem Resource Group:
Karl Hermann, Doug Lewis and Susan SpruilL These individuals have provided valuable advice,
guidance and assistance during the development of the Plan. Robert Smith, of the Soil
Conservation Service (SCS), has also given valuable advice and is working with us to increase
the involvement of SCS with EMAP.
Finally, we would like to thank Sirena Hardy, who spent long hours at the computer
keyboard revising and compiling this manuscript; Tangela Newsome, who copied and collated
the many versions; and our librarian, Phyllis Garris, who never left us without reading material.
- x-
-------
1. Introduction
In 1992 a Pilot Project will be conducted in North Carolina by members of the Environmental
Monitoring and Assessment Program's (EMAP) Agroecosystem Resource Group (ARG). The
EMAP is an Environmental Protection Agency (EPA) initiative in which the U.S. Department
of Agriculture's (USDA) Agricultural Research Service (ARS) was asked to give technical
leadership to the Agroecosystem component Thus the Technical Director (TD) of the ARG is
with the USDA-ARS. ARS asked the USDA's National Agricultural Statistics Service (NASS)
to cooperate in the development and data collection aspects of the Pilot project These three
agencies are the principal cooperators in the Pilot, which is an important developmental step
towards the implementation of a plan for monitoring the ecological condition of agroecosystems
in the United States. This document is an implementation plan for the Pilot project and represents
the combined effort of the members of the ARG (Appendix 1). Every attempt has been made to
include pertinent information in this document; however, as plans continue to develop and
methods become refined, changes will necessarily be made.
1.1. Overview of The Agroecosystem Program
This section is intended to provide a brief overview of the Agroecosystem component of the
EMAP Program. For a more detailed description of the Program, the Environmental Monitoring
and Assessment Program (EMAP) - Agroecosystem Monitoring and Research Strategy (Heck et
al. 1991) should be consulted.
1.1.1. Establishment and Purpose
The agroecosystem monitoring program, as one component of EMAP, is a national program
administered by the EPA's Office of Research and Development (ORD) in cooperation with
several USDA agencies (Heck et al. 1991). In the past decade, environmental scientists have
identified the need for more relevant and accessible ecological data, and the EPA has been
1 - 1
-------
encouraged to adopt an ecological perspective of the environment, in which the ecosystem is the
fundamental unit of research and monitoring. In 1988, EPA, in cooperation with other agencies
and organizations, initiated EMAP to provide baseline estimates of the condition of U.S.
ecological resources and follow changes and trends that could be computed with statistical
confidence (Kutz and Linthurst 1990).
The Agroecosystem Resource is one of seven resource categories within EMAP. The
Agroecosystem Resource Group (ARG) was established in 1988 to initiate the development and
implementation of a monitoring and assessment program to determine the status and extent of
U.S. agroecosystems. Roy E. Cameron (Lockheed Engineering and Science Co.) served initially
as Acting Technical Director. In 1989, Walter W. Heck [U.S. Department of Agriculture,
Agricultural Research Service (USDA, ARS)] was named Technical Director. He has worked
with C. Lee Campbell [Department of Plant Pathology, North Carolina State University (NCSU)],
the Associate Director, in the development of an interagency, interdisciplinary group of federal,
state and private scientists (Appendix 1) which comprise the ARG. Members of the ARG
developed an initial Research Plan (Heck et al. 1989) which served as the basis for the current
Research Strategy Plan [Agroecosystem Monitoring and Research Strategy (Heck et al. 1991)].
1.1.2. Mission, Objectives, Definition, and Societal Values
The mission of the ARG is "to develop and implement a program to monitor and evaluate
the long-term status and trends of the nation's agricultural resources from an ecological
perspective through an integrated, interagency process" (Heck et al. 1991). The developmental
stages of this national program include this 1992 Pilot Project and subsequent regional pilot and
demonstration projects (see Section 10: Resources and Implementation). The pilot and
demonstration projects allow for the orderly attainment of full national implementation while
assuring essential scientific rigor.
1-2
-------
The specific objectives of the agroecosystem program parallel the overall EMAP program
objectives, focusing on agroecological resources. When fully implemented the program will meet
the following objectives:
o Estimate the distribution of agroecosystems and the status and trends in indicators of
ecological condition on a regional basis with known statistical confidence.
o Monitor indicators of pollutant exposure and habitat quality and seek associations between
anthropogenic stresses and ecological condition.
o Provide periodic statistical summaries and interpretive reports on ecological condition to
the public, to the scientific community, and to policy-makers.
For EMAP, agroecosystems are defined as land used for crops, pastures and livestock; the
adjacent uncultivated land that supports other vegetation (hedgerows, woodlots, etc.) and wildlife;
and the associated atmosphere, underlying soils, groundwater, and drainage networks (first and
second order streams, ponds, and irrigation drainage networks). This definition of agroecosystems
recognizes their complexity and emphasizes a holistic approach that considers all components of
agroecosystem landscapes.
The ARG also recognizes that certain societal values or concerns are associated with
agroecosystems. Three societal values are currently identified as highly relevant to
agroecosystems:
o Supply of agricultural commodities
o Quality of natural resources
o Conservation of biological resources
These values and concerns parallel those stated in the 1991 Research Strategy Plan (Heck et al.
1991) and have served as a focus for development of the overall strategy for agroecosystem
1-3
-------
monitoring, for the establishment of assessment endpoints, and for the selection of specific
indicators (measurements) of ecological condition of the resource. Although not specifically
mentioned, socioeconomic factors are recognized as being inherent in these societal concerns.
1J3. Relationship to Other EMAP Resource Groups and Cross-cutting Activities
EMAP comprises seven ecosystem resource groups, four integration groups and four
coordination groups (Table 1-1). Interdisciplinary and interagency groups of scientists in the
seven resource groups are responsible for the collection, analysis, and integration of data from
their ecological resource. The four integration and four coordination groups have been established
to assist the resource groups and to ensure uniform quality management, consistency, and
integration across the program.
Presently, the resource groups are in various stages of development with regard to plans for
and implementation of actual monitoring activities (Table 1-2). For example, the Estuaries
Resource Group will complete a third season of monitoring in 1992, with activities in the
Virginian and Louisianan provinces. The Forest Resource Group will continue monitoring in the
New England regions and continue pilots in the Southeastern and Western regions in 1992. The
intent, however, is that all resource groups will be ready to implement a national monitoring
program by 1997.
The ARG is in continued communication with other resource groups, particularly the
terrestrial groups, concerning cross-cutting activities such as indicators, landscape
characterization, design, statistics, logistics, QA/QC, and other areas. Discussions of joint efforts
in pilot projects have been discussed with all resource groups except Great Lakes and Estuaries.
Also, as pilot plans develop, interactions with all of the coordination and integration groups will
intensify to insure that the ARG program is consistent and compatible with other activities within
EMAP.
1-4
-------
Table 1-1. Resource, integration and coordination groups of EMAP.
Resource Groups
Agroecosystems
Arid Lands
Estuaries
Forests
Great Lakes
Surface Waters
Wetlands
Integration Groups
Air & Deposition
Integration and Assessment
Landscape Characterization
Statistics and Design
Coordination Groups
Indicators
Information Management
Logistics
Quality Assurance
Table 1-2. Monitoring activities of EMAP Resource Groups in 1992.
Resource Group
Activity
Location
Agroecosystem
Arid Lands
Estuaries
Forests
Great Lakes
Surface Waters
Wetlands
Pilot Project
Pilot Project
Pilot/Demonstration
Pilot/Demonstration
Pilot Project
Pilot/Demonstration
Pilot Project
North Carolina
Colorado
Virginian/Louisianian Provinces
NE, SE, Western Region
Great Lakes
NE
Gulf Coast
1 -5
-------
1.2. Cooperative Interaction with The National Agricultural Statistics Service (NASS)
The ARG maintains extensive, cooperative interactions with personnel of USDA/NASS, both
at the operational and administrative levels. Two members of the USDA/NASS staff serve as
regular members of the ARG and as liaison between NASS and the Agroecosystem Technical
Director (Appendix 1).
The association with NASS is an integral component of the Agroecosystem program. The
Agroecosystem Program will utilize NASS's established and well-accepted, national sampling
frame as well as NASS's long experience in performing site visits and interviews with farmers.
Over the past 30 years, NASS has developed a network of enumerators and administrators
experienced in conducting successful national surveys and monitoring activities. This nation-wide
force of trained enumerators, with its proven administrative organization, will be utilized for
much of the field assessment in the Agroecosystem Program (see following Sections). It is
important to the Program that growers throughout the U.S. are familiar with and have confidence
in NASS personnel. The NASS also has an established, well-respected program for tracking,
processing and summarizing data acquired in the field (see Section 9). The ARG is thus
developing the Agroecosystem Program to make maximum use of these aspects of NASS.
The NASS requirement of data confidentiality is established by law and is well accepted in
the agricultural community. This confidentiality requirement (Section 8.3) is essential to the
success of the ARG in working with growers in the U.S.
1.3. Cooperative Interactions with Other National Programs and Agencies
Discussions are currently in progress with three federal agencies/departments to explore
possible cooperative activities with the 1992 Agroecosystem Pilot: USDA, Economic Research
Service (USDA/ERS); U.S. Geological Survey (USGS); and the USDA, Soil Conservation
Service (USDA/SCS).
1-6
-------
The ERS initiated the USDA Area Study Program in 1991 in four study areas across the U.S.
They have identified another four study areas for 1992; one area is the Albemarle-Pamlico
drainage area of North Carolina and Virginia. The study areas used by the USDA Area Study
Program are sampled only once to aid in the development of economic models. In each study
area, ERS samples approximately 1000 sites utilizing the SCS National Resources Inventory
(NRI) area frame to identify NASS sampling units; NASS enumerators then collect the data.
Because of the intense sampling in these study areas and the similarity of ARG and Area Study
data elements, we are interested in determining if Area Study Program data can be used to
improve or enrich the data collected for the 1992 Pilot of the Agroecosystem Program. Also, ERS
is interested in determining if the Agroecosystem Program data, from our continued monitoring,
may be of benefit to ERS.
The USGS will implement the National Water-Quality Assessment (NAWQA) Program in
the 1990s. The ARG has had several discussions with USGS personnel in the North Carolina
office who will be responsible for the development and implementation of the NAWQA program
in North Carolina. The NAWQA program will monitor not only water quality within designated
watersheds, but also biological indicators of interest to the ARG including habitat quality for
wildlife. The NC office is responsible for monitoring the Albemarle-Pamlico watershed and plans
to initiate monitoring in 1993.
The SCS conducts the National Resources Inventory (NRI) every five years (from 1982). The
ARG and SCS personnel are exploring ways in which NRI data could be integrated with, or
supplement information from the Agroecosystem Program. Also, the ARG is exploring the
possibility of obtaining specific soils data from SCS for the 1992 Pilot Program.
1.4. Projected Implementation Schedule for a National Agroecosystem Monitoring Program
The ARG has developed a multiyear program to establish the national implementation of a
suite of indicators by 1997. These indicators will address the assessments endpoints (Section 4)
1-7
-------
and identified societal values associated with agroecosystems. The first stage of the program
(1990) encompassed the initial evaluation of: 1) statistical designs, 2) existing monitoring
programs (i.e., NASS, SCS, ERS), 3) assessment endpoints and associated indicators (for their
availability, validity, variability, cost) (Campbell et al. 1990), 4) 'data management and analysis
techniques, and 5) derived outputs (Meyer et al. 1990). During 1990, a national monitoring
strategy was developed (Heck et al. 1991). In the second stage of the program (1991) in-depth
examinations were conducted of several areas critical to the planning and implementation of the
1992 Pilot Project: 1) statistical design options, 2) measurements associated with specific
indicators and assessment endpoints, 3) sampling protocols, 4) cooperation with NASS, 5)
logistics, 6) total quality management, and 7) information management. Discussion and re-
examination of these areas will continue through 1992.
The 1992 Pilot Project will test aspects of the monitoring program with a limited suite of
endpoints (indicators). Experience from the 1992 Pilot will be utilized to develop a regional
demonstration of all program elements in the Southeast (1993), to implement an additional pilot
project in EPA Region Vn (1993) and to initiate cooperation in an integrated terrestrial pilot
(1993). Assuming the pilots and regional demonstrations are successful, we anticipate being ready
to implement specific components of the Program on a national basis in 1995 or 1996; funding
levels will determine the degree of regional and national implementation. The implementation
schedule for the Agroecosystem Program is shown in Table 1-3 and Figure 1-1 through 1995.
1-8
-------
Table 1-3. Planned implementation of Agroecosystem monitoring and assessment across EPA
regions.
EPA Regions Years (Funds in thousands}*7
1-NE Boston
2-NE New York
3-NE Philadelphia
4-SO Atlanta
5-NC Chicago
6-SO Dallas
7-NC Kansas City
8-NC Denver
9- WE San Francisco
10- WE Seattle
1991
(300)
..—
—
—
PI
—
—
—
—
—
1992
(800)
—
—
—
Pil
—
—
PI
—
—
—
1993
(2440)
—
—
—
De
—
—
Pil
PI
—
1994
(4060)
—
PI
—
Im
PI
PI
De
—
Pil
—
1995
(6200)
PI
Pil
PI
Im
Im
Im
Im
PI
De
PI
PI Planning (write a peer-reviewed Project Plan)
Pil Pilot Project (1 or 2 states)
De Demonstration Project
Im Implementation in EPA Region (full implementation in megaregion)
1 -9
-------
Agroccosystcm Implementation Schedule
• Planning
@ Pilot
+ Regional Demonstration
* Implementation
Figure I-I. The Agroccosystcm Implementation Schedule.
1-10
-------
2. The 1992 Pilot Project: Rationale and Objectives
This section provides an overview of the rationale for the 1992 Agroecosystem Pilot Project
and presents the specific objectives of the Project. The rationale is in keeping with the overall
program approach and rationale outlined in Section 2 of the Agroecosystem Monitoring and
Research Strategy (Heck et al. 1991).
2.1. Rationale
The Agroecosystem component of EMAP is being designed as a comprehensive monitoring
program with the intent of increasing our knowledge of the status and extent of our national
agroecological resources. It is also designed to identify associations between observed changes
in ecosystem condition and a suite of stressor/exposure indicators.
Agroecosystems are managed intensively for human welfare and activities in the crop and
non-crop components are often influenced by government programs (i.e., Conservation Reserve
and Crop Quotas) and regulations (i.e., wetlands preservation and changes in permissible pesticide
use). These intentional perturbations of agroecosystems provide a series of challenges to the
establishment of an ecological monitoring program. Although the focus of the Agroecosystem
Program is ecological, a full understanding of these intensively managed systems requires that
both ecological and more traditional agricultural information be included.
It is essential to obtain certain information on management practices for crops and livestock,
selected sociological and economic factors, and agricultural land use directly from the grower,
because of the importance of grower inputs to agroecosystems. It is also essential to obtain
specific samples, such as soil and water samples, and measurements, such as production
efficiency, that relate directly to the actual quantification of ecological condition. Thus, the Pilot
Project, and the eventual implementation of a national monitoring program, will be accomplished
through a combined survey and sampling methodology.
2- 1
-------
Pilot projects will serve to resolve a number of relevant issues prior to regional or national
implementation. These issues include the critical evaluation of indicators and assessment
endpoints, the establishment of a sampling frame and sample sizes, the evaluation of logistics and
quality control, the development of information management procedures (including provisions
for data confidentiality), and, in cooperation with NASS, the establishment of data analysis,
summarization, and reporting formats. The 1992 Pilot Project will address these issues at a
geographic scale that is large enough to provide reliable answers to specific questions concerning
the operation of the monitoring program, but is small enough to be physically and fiscally
manageable. However, not every aspect of the national monitoring plan will be evaluated. For
example, because of the status of indicator development and fiscal constraints, only a limited
suite of indicators and assessment endpoints will be evaluated in the 1992 Pilot. Also, most
components of the Pilot include research and development activities that will lead to the inclusion
of specific indicators and procedures in regional demonstrations and in the national monitoring
program.
The state of North Carolina was selected for the 1992 Pilot Project for several reasons, given
in order of importance:
1. The physiographic diversity of the state is representative of the entire Southeastern region
of the United States.
2. NASS is organized on a state-by-state basis and enumerator training is done in each state.
By staying within a single state, we only need to work with a single NASS state
organization. This simplifies the resolution of problems during the development of
logistics, design, and implementation procedures. The training of NASS enumerators will
be transferable, with only minor modifications, to each new state as states are added to
the program.
3. The core staff of the ARG is located in Raleigh. For the first pilot study, this facilitates
logistic activities.
2-2
-------
2.2. Objectives
The 1992 Pilot Project is designed to provide information that will allow the evaluation of
specific aspects of the Agroecosystem Monitoring Program. The Pilot will serve as a basis for
the development of Regional Demonstration Projects and, where needed, of additional pilot
projects to evaluate agroecological characteristics or logistic issues that are unique to specific
regions. Specifically, there are four major objectives for the Pilot Project:
1. Critically compare the relative efficiency, in terms of cost and precision, of the EMAP
Hexagon Design and the MASS Rotational Panel Design for use in a national
agroecosystem monitoring program.
2. Empirically evaluate an initial suite of indicators in order to:
o Assess the ability of an indicator to address the assessment endpoints of interest
o Establish an initial range of values for each indicator across the diverse physiographic
regions in the state
o Assess spatial variability of indicator values within and among sample units
o Identify the usefulness and sensitivity of each indicator and assessment endpoint in
determining ecological condition
o Determine the cost-effectiveness for each indicator
3. Develop and refine plans for key components of the monitoring program.
o Sampling
o Logistics
o Total quality management
o Data analysis, summarization, and reporting
o Information management
o Health indices and their interpretation.
2-3
-------
4. Develop and evaluate additional indicators that will address specific assessment endpoints.
o Soil quality - biological component
o Landscape structure
o Water quality - groundwater component
O Biomonitors of ozone impact on crops
The 1992 Pilot Project is not intended to be a full implementation of the Agroecosystem
Monitoring Program, but will provide information essential to the successful development of
regional demonstration projects. The Pilot Project represents the wise use-of resources to fully
consider issues critical for the success and implementation of the Agroecosystem Program.
2-4
-------
3. Design and Statistical Considerations
Statistical considerations for the Agroecosystem 1992 Pilot Project fall under the two topics
of sampling design and protocols, and data analysis. The basic issues associated with these topics
were discussed in the 1991 Agroecosystem Monitoring and Research Strategy (Heck et al. 1991).
The ARG has two sampling plans under consideration for the 1992 Pilot Project. Two
independent samples, one from each plan, will be used. This will provide cost and variance
information from which a comparison of the, two plans can be made. (Key information on
temporal correlations needed for a more complete comparison of the plans cannot be obtained
from a one-year pilot.) The basic sampling units in both plans are well-defined geographical areas
that will contain an unknown number of agricultural fields. A protocol for obtaining a random
sample of agricultural fields with known probabilities of inclusion is given. Some indicators
require sampling the geographical area defined by the field. A protocol is given for this within-
field sampling that will also provide information on relevant components of variance.
Data analysis will include (in addition to a simple statistical summary of the indicator
results): 1) estimation of variance components to help determine future field sampling strategies,
2) correlation analysis to understand relationships among indicators as well as spatial patterns of
the indicators, and 3) comparison of the variance and cost efficiencies of the two sampling plans.
3.1. Selection of the Pilot Sample Segments for Each Plan
Each of the two sampling plans under consideration will use the NASS Area Frame segments
as the basic sampling unit. The NASS area frame segments were defined by first stratifying the
state of North Carolina based on intensity of agriculture (See Section 3.1.2). Each stratum is
divided into Primary Sampling Units (PSUs). A random sample of PSU's are then divided into
six to eight sample segments, with segment size dependent on strata. For example, segment size
is approximately 0.1 square mile for urban strata and 1 square mile for agricultural strata.
3- 1
-------
One of the proposed Agroecosystem sampling plans is the EMAP Hexagon Plan. It uses the
centroid of selected hexagons to identify the NASS segment that will be used for indicator
sampling. The other sampling plan, called the Rotational Panel Plan, uses a subs«t of segments
from the NASS June Enumerative Survey. Both plans will be evaluated in the 1992 Pilot Project
3.1.1. The Hexagon Sampling Plan
The EMAP hexagons (40 km2) with their centroids in the state boundaries of North Carolina
were selected as the hexagon sample for the 1992 Pilot. These 203 hexagons were located on a
state map and were divided into four interpenetrating subsamples according to procedures
outlined in the EMAP Design Report (Overton et al. 1991). One of the four subsamples was
selected at random for the 1992 Pilot. Fifty-four hexagons were in the selected subsample, but
three were over bodies of water not included in the NASS water strata. These three hexagons
(numbers 8, 33 and 36 in Figure 3-1) were located in the large sounds lying between the
mainland and the barrier islands and were dropped. The 51 remaining hexagons are distributed
over 49 counties in North Carolina (Figure 3-1). The list of counties with the number of
hexagons in each county is given in Appendix 2.
The coordinates of the centroids of these 51 hexagons were forwarded to NASS for
identification of the NASS sample segments according to the following procedures.
o The primary sampling unit (PSU) that encompasses the centroid will be identified and its
ID number, i.e., stratum, substratum, county and NASS replicate will be attached. The
PSU will be assigned to a NASS technician who will divide it into segments according
to NASS' standard criteria. Special care will be taken to ensure that the assigned
technician does not know the location of the centroid within the PSU to avoid bias while
delineating the segment boundaries.
3-2
-------
3-3
A
2
S,
-------
o After segments within the PSU have been delineated, the segment containing the centroid
will be identified and included as a sample segment
o Characteristics to be described include the area of the PSU, the area of the selected
segment and, if possible, the estimated cultivated acreage and an estimate of the number
of fields within the segment.
o The boundaries of the PSU and the selected segment are to be delineated on an aerial
photo and on a county highway map for use by the ARG according to the NASS
confidentiality guidelines. Duplicates are to be prepared for NASS field staff and the
enumerators during data collection.
o Accurate time and cost records will be maintained for each step of the operations
described above.
3.1.2. The Rotational Panel Sampling Plan
The complete 1992 NASS sample for their June Enumerative Survey (JES) in North Carolina
has 321 segments stratified as shown in Table 3-1. Only four of the 100 counties in North
Carolina do not contain a sample segment.
The Rotational Panel sample for the 1992 Pilot consists of (approximately) a 20% subsample
of the JES; one replication (replication number 4) from the sub-strata that have five replications,
two (replications number 4 and 9) from the sub-strata that have 10 replications, and one
(replication number 3) from the two sub-strata that have 3 replications. Thus, 65 segments from
NASS' JES (Table 3-1) will be assigned to the Rotational Panel sample for the 1992
Agroecosystem Pilot. Reasons for selecting these particular replication numbers are:
3-4
-------
Table 3-1. Stratification of NASS segments in North Carolina for 1992.
Stratum
Sub
ID. Number Definition Strata
13
20
31
32
33
40
50
Total *
>50% Cultivated
15-50% Cultivated
>20 home/mi2 Ag-Urban
>20 home/mi2 Commercial
>20 home/mi2 Resort
<15% cultivated
50 Non-Agricultural
6
14
5
3
1
8
1
Number of
Reps Segments
5
10
10
5
3
10
3
30
140
50
15
3
80
_3
321
Pilot
Segments
6
28
10
3
1
16
J.
65
o Numbers 4 and 9 are the latest replications; they enter the sample in 1992 for the first
time. Because the segments in the hexagon will be sampled for the first time, the
comparison of the two designs will be free from any conditioning effects that might have
resulted from any previous visits.
o Replicates 4 and 9 will remain in the JES sample for five years and will be available for
re-measurement during that period.
o The latest replication (replication number 3) in Strata 33 and 50 was selected in 1991; it
presumably will remain in the sample at least through 1993.
The 65 segments selected for the 1992 Pilot fall into 55 counties and provide a reasonable
spatial representation (Figure 3-2) of the state.
3 -5
-------
Distribution of NASS Segments by County
Chosen with Rotational Panel Plan
fyq 1 segment
§%fl 2 segments
Hi 3 segments
Figure 3-2. North Carolina Counties containing NASS Segments for 1992 Pilot; there are a
total of 65 segments in 55 counties (see Appendix 2 for a list of the counties).
Maps and aerial photos will be prepared for the use of the North Carolina NASS Field Office
for each of the 321 NASS sample segments in the 1992 June Enumerative Survey. If needed,
duplicate aerial photographs of the segments selected from the 1992 pilot will be sent to the
ARG. Records of the cost of these photos and maps and the cost of their preparation will be
maintained.
3.2. Evaluation of the Two Plans
Both cost and precision will be considered in evaluating the relative efficiencies of the two
sampling plans. Because the two sample sizes differ slightly, efficiencies will be expressed in
a standardized manner such as information per unit cost.
3-6
-------
Because the Pilot is a one-year test and EMAP is a long range program, comparisons from
the Pilot will not provide information on the relative efficiency of the two sample plans for
estimation of changes or trends over time. Consideration of the relative efficiencies for the
estimation of time trends must be based on theoretical results (e.g., Lesser 1992) or simulation
rather than empirical studies (1992 Pilot).
3.2.1. Cost Comparisons
To be sure that every applicable cost is included, each step in the survey process for each
Plan will be identified and placed in its proper sequence in a flow chart Records of costs at
each of these steps will be maintained. NASS will maintain records of costs for operations they
perform such as segmentizing the hexagon sample, delineating sample segments on aerial photos,
and visiting sample sites. The ARG will maintain costs for operations they perform. The types
of activities required to prepare the sample have been noted in Section 3.1. There will be similar
field costs in training enumerators and in collecting the data. The entire field cost for the
Hexagon sample will be assigned to that Plan since it is outside the scope of the regular JES.
Costs assigned to the Rotational Plan sample will be prorated to include miscellaneous costs
associated with the conduct of the survey.
Finally, the cost of processing the data, making the appropriate population estimates for the
indicator variables and estimates of their variances will be identified. Because the Rotational
Panel sample will be a full replicate of the NASS sample in North Carolina, the estimation
procedures and variance formulae already developed by NASS will apply. The Hexagon sample
will not be stratified and will require estimation procedures that have been worked out by the
EMAP, Statistical Design Team.
There should be little or no incremental cost in preparing the Rotation Panel sample. NASS
may choose to allocate pro rata costs for the development of the JES sample and perhaps even
some costs for the development of the NASS area frame in North Carolina. These costs may have
3-7
-------
to be negotiated but whatever they are determined to be, they will be considered in the
comparison with the Hexagon Plan.
3.2.2. Estimates of Precision - The Hexagon Sample
Estimation procedures for the EMAP Hexagon sampling design are being developed by the
EMAP Statistical Design Team. The approach relies on the Horvitz-Thompson estimation
(Horvitz and Thompson 1952). The estimation procedures presented here are extracted from their
reports.
Although the actual Hexagon sample consists of NASS segments, the selection process is by
means of the centroid of the hexagons. Thus, segments are selected with probability proportional
to their area; the inclusion probability is given as
Pi = a/A, and the weight
w,= l/p,
•*.
where ps is the probability of selecting the i"1 segment within the hexagon, a; is the area (km2) of
the i& segment, A is the area of the large hexagon (~650 km2) and Wj is the sampling weight
associated with that segment
The Horvitz-Thompson (HT) formula for estimating a population total, fy , for any
attribute y, is given as
ies
where the summation is over the set of sample segments, S, in the population of interest, where
Wi is the sampling weight and y; is the value of the attribute for the sampled segment The
number of units in the population is estimated by setting y; = 1.
3-8
-------
The HT variance formula provides unbiased estimates of variance if all pairwise probabilities
are positive. Systematic sampling has a large number of zero pairwise inclusion probabilities. A
modification of the variance formula has been shown to perform satisfactorily where the pairwise
inclusion probabilities, 7^. , have been approximated (Stehman and Overton, 1987) by
2(n-l)7tv7C.
2n-n -
from which
The variance formula with this approximation is:
ies jes
3**.
It has been suggested recently that the Yates-Grundy estimator of variance might be more
appropriate. This suggestion will be investigated by the EMAP, Statistical Design Team. Also,
procedures are being developed to determine variance of the estimates empirically by means of
facsimile population bootstrap (Overton 1991).
Because the segment is not a standard size, there is little interest in estimating a mean per
sampling segment. Rather, interest will be in estimating population totals, or means for some
standardized unit such as per acre, e.g., average yield per acre. Standardization to a per unit
basis can be handled two ways. The appropriate population estimates of mean per standard unit
are ratio estimates in which both the numerator and the denominator are random variables.
Because no stratification is involved, the ratio estimate for the Hexagon sample is simply the
estimated total production of, say corn, in a given universe divided by the estimated acreage of
land planted in corn for grain in the same defined universe. The estimate of variance is an
approximation based on the Taylor Series expansion and, therefore, is biased. However, in large
3-9
-------
samples the bias is negligible. For some purposes the standardization to a per unit basis will be
done at the sampling unit level (e.g., field yield divided by field acreage). In these cases the
yield per acre will be treated as a variable in the HT estimation.
In the Hexagon Plan per se, post-stratification of the sample segments will not be used.
However, for complete comparison of the two plans the effects of post-stratification will be
investigated. Post-stratification will change the form of the inclusion probabilities.
3.23. Estimates of Precision - The Rotational Panel Sample
Estimates of population quantities on variables of interest and estimates of their variances
from Rotational Panel sample data have been worked out by NASS and can be applied directly.
Estimates will be expanded to the population by applying the appropriate weights to the sample
data in each substratum and adding up the substratum totals to provide an estimate for the region.
Similarly, the variance of the estimate will be based on standard formula for a stratified random
sample. Population estimates that are obtained as ratio estimates will require an approximation
of the variance, as discussed in 3.2.2, but the component variances will be tine appropriate
variance formula for a stratified random sample.
3.3. Within Segment Sampling Protocols
Field sampling can be divided into two parts: selecting the fields within the segments and
taking the measurements within those fields. The individual components of variance for these
two sources of variation will be explored during the Pilot.
3.3.1. Field Selection
During the June Enumerative Survey (JES), NASS enumerators will obtain land use
information on all areas of each sample segment. The location of each cultivated field in each
sample segment will be mapped on an aerial photograph and its identification number and
3-10
-------
acreage recorded. For the 1992 pilot eligible fields will be defined as the planted acreage in any
field that does not contain the crops listed in Table 3-2. All inferences made in the pilot will be
to this population. The information will be used to select a subset of fields over all segments
in the Hexagon Plan and over all segments within a given stratum for the Rotational Panel plan.
Table 3-2. Crops ineligible for selection in the Agroecosystem 1992 Pilot
Permanent Pasture
Orchards
Vine Fruits
Christmas Trees
Other woody perennial crops
Greenhouse Plants
A systematic sample with a random start will be used to select fields for inclusion in the pilot
with probability proportional to size. Fields within sampled segments will be ordered arbitrarily
first by crop and then by segments (for a given stratum for the Rotational Panel sample). All
fields that contain crops that have not been excluded from the 1992 Pilot will be included in the
ordering. The ordering by crops is to ensure that each crop is selected at least once as long as
its acreage is greater than the step size used in selecting the sample. If a field has been double-
cropped and both crops will be harvested for grain, then the crop pair will be treated as a
separate entity in the ordering of the crops. For example, in a field planted consecutively with
both wheat and soybean, each to be harvested for grain, the ordered list created for the systematic
sampling might appear:
Field #
i
Crop
wheat
wheat
i+j+m
wheat
wheat-soybean
wheat-soybean
3- 11
-------
i+j+m+l
i+j+m-f2
soybean
soybean
In double-cropped fields where only one crop was harvested for grain, the harvested crop will
be used for determining the ordering.
For the sampled segments we will define
o s as the number of sample segments
o f as the number of fields, and
o 3j as the planted acreage in the field, where the subscript indexes the fields after ordering.
The cumulative acreage of all fields in the sequence up to and including field i is
i
and the total acreage is
Preliminary analyses of field size distributions in North Carolina suggests that an average sample
size of three fields per segment will provide a reasonable representation of the major crops. This
implies that the step size, k, in the sequential sampling needs to be:
3s
where int(c) denotes largest integer less than c. This step size will be adjusted to provide the
desired sample size when the land use data from the JES are available. A random integer
between 1 and k will be chosen as the random start and then every integer m+ck, c=0,l,2, ...
until m+ck > AT, will designate a selection. Field i is selected for sampling if
A- < m+ck s A for some c.
3-12
-------
In double cropped fields where the acreage of the two crops differ, the maximum of the two
planted acreages will be used in field selection.
The fields that have been selected will be identified and marked on the aerial photographs
for use by the NASS enumerators in collecting the field data.
An alternate way to select the fields for indicator sampling is to expand the fields by their
expansion factor (the inverse of their selection probability) prior to selection. This provides a
self-weighting sample and simplifies the estimation of sample variances, means and population
totals. The ARG is currently exploring this possibility.
3.3.2. Sampling Within Fields
Soil sampling to determine soil physical and chemical properties and nematode densities will
require within-field sampling. A sample of 20 soil cores composited for each field will provide
sufficient soil for both physical and chemical analysis, and nematode density assays. Whereas
it may be desirable to have a method that would sample the entire field, field size often will
make this impractical. Consequently, the entire field will serve as the soil sampling unit only
if it is five acres or less in size. For fields larger than five acres, a pseudo-random five-acre
subregion of the field will be chosen. If a field is chosen randomly for the collection of more
than one soil sample, an independent five acre subregion will be chosen for each sampling. The
five-acre subregion will be sampled with 20 soil cores taken at equal distances along a 100-yard
transect that represents the diagonal of the five-acre subregion. The diagonal transect, as opposed
to some other method such as a grid placement, was chosen primarily because of its ease of
implementation.
NASS' procedures for locating objective yield plots will be adapted for use in locating the
sampling transect. According to these procedures, if the field is less than sixty acres the field
is divided into quarters. If the field is larger than sixty acres it is divided into ninths. The
objective yield plot is then located in only one of these subregions. The subregion that is
3 - 13
-------
selected for location to the objective yield plot is identified by the first corner of the field that
the enumerator encounters as he or she approaches the field. While this is not a random choice,
m NASS'experience they have found this procedure to be satisfactory.
The modifications to the procedure for locating the transect for the soil samples for the
Agroecosystem Pilot are as follows. If the field is five acres or less in size,, it will not be
subdivided before location of the transect; otherwise, subdivision of the field will be as described
above. A random point will be located in the subdivision based on a random number of rows
and paces along rows from the corner of the selected subsection. If rows are not present in the
field, random paces will be used. The enumerator using NASS procedures will locate the point
in the field. This point will designate the midpoint of the transect to be used for soil sampling,
and the transect will run at a 45° angle to the direction being walked by the enumerator (Figure
3-3). From this center point on the transect, the enumerator will take 10 soil cores in each
direction along the transect with the first core being 2.5 yards from the center point and each
succeeding core being an additional 5 yards away. For example, if the enumerator had come to
the selected point from due south, then the transect would run from the center point
approximately 50 yards to the northeast and 50 yards to the southwest
If the transect intersects the boundary of a field then a set of "bounce" rules will be initiated.
Upon reaching the field margin the enumerator will reflect off the boundary at an angle of 90
degrees from the direction of the transect. See Figure 3-4. This will continue for every boundary
encountered until the entire distance of the transect has been traversed.
To ensure that the sampling is not biased by the subjective placement of the core,
enumerators will mark the end of each 5-pace interval with a wooden stake, and before sampling
they will lay a marked stick along the transect at the stake. The soil sample will then be taken
at either 1.5 feet or 3 feet from the stake, depending on whether the stake is odd or even.
Appendix 6 gives additional detail on this sampling procedure.
3- 14
-------
Figure 3-3. Transect sampling of field.
3.3.3. Sources of Variation in Field Sampling
There are three principal sources of variation in field sampling: between-field variation,
within-field variation, and the variation in laboratory analyses. To obtain these components of
variation the following design will be used. Except for differing numbers of segments, the same
procedure applies to both sampling designs.
Sample segments will contain, on average, 3 sampled fields. For soil samples, every k"1 field
will be sampled twice to get an estimate of the within field variability. This will be
accomplished by choosing a second independent transect from the same field using NASS
protocols. The next most accessible corner of the field is chosen and the sampling is repeated
on a new transect defined with a new starting point Considering only the transects from these
3-15
-------
Boiince rules
Sampling starting
point
Figure 3-4. Transect sampling of field (Bounce rules).
twice-sampled fields, the soil sample from every 8th transect will be split for duplicate laboratory
analysis. To ensure enough soil for the split sampling, two cores will be drawn at each sampling
point in these particular transects.
Let N be the number of fields sampled in the pilot; this should be about N=348. Using the
ordering of the fields given previously, every k"1 field will be sampled twice with independent
transects. (Other fields will be sampled twice simply due to their size being greater than the step
size in the field sampling process). This will give 2N/k soil samples from these twice-sampled
fields. (There will be {-^)tf soil samples from the other fields.) Of the 2N/k soil
/C
samples from the twice-sampled fields, every 8th soil sample will be split for laboratory
determinations. This procedure gives N/k and 2N/ks degrees of freedom for the "samples in
fields" and "determinations in samples" mean squares. The two mean squares have equal degrees
of freedom if s=2, which seems desirable. With N=348 and s=2, the analysis of variance
(ignoring strata) is found in Table 3-3.
3-16
-------
Table 3-3. Degrees of freedom for field sampling components of variance.
Source
Among fields
Samples(fields)
Def(samples)
Total Soil Samples
df
N-l
N/k
2N/ks
„/ ks+s+2 \
1 ks )
(k,s)=(5,2)
347
70
70
487
(k,s)=(6,2)
347
58
58
464
Current plans are to use k=6 and s=2; that is, every sixth field sampled will use two
independent transects and the soil from the second of each of these transects will be split to
provide duplicate laboratory determination for soil chemical analysis and nematode assay. These
numbers may be modified by budget considerations. Additional known samples to determine
laboratory accuracy will also be included (see Appendix 7, Section 5.2.5).
3.3.4. Selection of Farm Ponds and Wells for Water Quality Sampling
Current plans are to sample farm ponds and wells for water quality analysis. The budget and
logistic constraints limit this sampling to about 50 segments. This is approximately the number
of segments associated with each of the two sampling plans (65 in the Rotational Panel Plan and
51 in the Hexagon Plan) so that the present plan is to limit the water quality sampling to only
one of the two plans. Currently, however, there is no information on the distribution of farm
ponds and wells across the state and the frequency with which they will be associated with the
randomly chosen fields. Information on the number and location of ponds and wells within each
selected segment will be obtained from the JES. Since pond and well sampling will not be done
until November (See section 5.3 and 6.3), the ARG is postponing until after the JES decisions
about which set of sample segments will be used and how ponds and wells will be selected
within the segments.
3 - 17
-------
3,3.5. Sampling within Farm Ponds and Sources of Variation
Sampling within farm ponds is still in the early stages of development. Two methods are
being discussed. The first method uses the more conventional procedure of taking samples from
a boat at several places and depths in some prescribed manner (see 5.3 for further details). While
it is recognized that sampling from a boat is preferable, and may be necessary, the logistics of
having a boat available to the enumerators and transporting the boat to all ponds makes the
procedure difficult The second method attempts to avoid the logistical problem of having a boat
available. A long pole with a water sampler attached would be used to obtain water samples at
some fixed distance (say 24 feet) from the shore at several points around the pond. Since most
farm ponds are relatively small, this composite water sample may provide a reasonable
representation of the pond. However, there are concerns that this method will produce biased
estimates. If preliminary testing shows that this procedure or some similar procedure is feasible,
the two methods, boat sampling and shore sampling, will be compared in the Pilot
Whatever method is adopted for choosing the ponds, the pond sampling plan will include a
subset of segments (approximately 25) on which the boat and shore sampling methods will be
compared and components of variance estimated. (In the remaining segments, only the shore
sampling method will be used.) The experimental design for this study will be structured as
follows (except the numbers may vary). Two ponds will be selected within each of these 25
segments and on each pond both methods of obtaining the water sample will be used. In one of
the ponds, randomly chosen, a water sample will be obtained by compositing the samples from
the two water sampling methods; in the other of the two ponds, two independent composite water
samples will be obtained by each water sampling method. This design will provide for a direct
comparison of the two water sampling methods and allow for estimation of variance between
ponds within segments and within pond sampling variability for each method.
3- 18
-------
3.4. Analysis
Analysis of the 1992 Pilot Project data will address four major topics: 1) statistical summary
of the indicator results; 2) variance estimation including assessment of the levels of precision
attained, variance component estimation, and consideration of alternative field sampling
strategies; 3) analysis of the indicators and their component variables including the correlations
among variables as well as the spatial correlations and patterns of individual indicators; and, 4)
comparison of the two sampling designs.
3.4.1. Statistical Summaries
The statistical summary of the indicator data will convey the population estimates of the
present status of the indicators for the geographical region covered by the 1992 Pilot A detailed
listing of the summary statistical data we expect to develop from the 1992 Pilot is shown in
Appendix 3.
The Pilot alone will not provide information on changes or trends. The primary purpose of
the statistical summary is to develop the methods for the annual statistical summaries. The
indicator data will be summarized by means of the estimated population cumulative density
function (cdf). The cdf for a particular indicator presents the proportion of the population (in the
Agroecosystem Pilot case, the proportion of area of cultivated land) that has values of the
indicator less than or equal to any specified value. Figure 3-5 illustrates a hypothetical cdf for
electrical conductivity of soil (mmhos/cm) in North Carolina. To the degree the estimated cdf
accurately reflects the population cdf, it conveys all the information about the distribution of the
indicator values: location, dispersion, and shape of the distribution. To facilitate interpretation
of each cdf, key quantiles of the distribution will be presented in tabular form. Construction of
the cdf must take into account the differential weights of the sampling units.
A problem arises in estimating a population cdf when the indicator variable is subject to
appreciable measurement error, as is expected for the Agroecosystem indicators. Without
3-19
-------
adjustment the empirical cdf, obtained from the indicator data, estimates an overly dispersed cdf.
Methods of disentangling the population cd from measurement error are being developed by the
EMAP Statistical Design Team.
It is anticipated that the two sampling plans will provide similar information on the
population distribution of the indicators. The estimated cdf from the two sampling plans will be
compared with nonparametric tests and, if compatible, a combined cdf will be presented. Other
methods of displaying key features of several cdfs, such as box-plots, may facilitate their
comparison and will be explored. Figure 3-6 illustrates the box-plots for three distributions. In
each case, the rectangular box encompasses the central 50% of the distribution with its top and
bottom edges denoting the 75th and the 25th percentile, respectively. The middle line is the
median and the two lines extending vertically from the box mark the 95th and 5th percentiles.
Outliers beyond these percentiles are marked with asterisks.
In addition, the statistical summary will include displays of the spatial patterns of key
indicators. The displays will be of sufficient resolution to develop contour plots or shaded maps
of the value of the indicator (See figures 3-7a and 3-7b). The precision of the kriged surface can
also be displayed.
3.4.2. Variance Estimation
Estimation of precision (variance) of population estimates is determined by the sampling
design. For each sampling design, the precision attained by the 1992 pilot survey for each of the
various population estimates will be computed as appropriate for that design. These measures
of precision will be repeated in the statistical summary and will be used to compare the variance
efficiencies of the two designs for estimation of status and to help in defining attainable data
quality objectives for measures of status for the Agroecosystem Program. For comparison of
variance efficiencies, for definition of data quality objectives, and for measures of change or
trend, assumptions on the magnitude of temporal correlations will have to be made.
3-20
-------
100
or
03
05
75
/o
-2 50
g
tr
o
Q.
o
25
2 4
Electrical conductivity (mmhos/cm)
Figure 3-5. Hypothetical cumulative density function (cdf) for electrical conductivity
of soil in North Carolina.
3 -21
-------
Components of variance will be estimated for key indicator variables. The components of
variance will be used with cost estimates to explore strategies for future allocation of sampling
effort: numbers of fields per segment, samples per field, and determinations per sample. Since
several indicators are involved, each with its own variance component structure, any sampling
strategy adopted would necessarily be a compromise. Excessively large variance components
may suggest problems in the definition of specific indicators.
3.43. Analysis of Correlation Structure
Analysis of the correlation structure of the indicators will be addressing four basic questions:
o What are the relationships among the indicators and what implications do these
relationships have with respect to defining the set of indicators to be used?
o Do the principal components of the indicators provide any meaningful suggestions about
the definition of health of the agroecosystem?
o What is the nature of the spatial correlation structure of the indicators?
o Is there also potential to use double sampling techniques to enhance the Agroecosystem
information with the correlated information collected on other variables from the full JES
sample of 16,000 units a year?
Principal component analysis and biplots will be used to investigate the relationships among
the indicators. Correlations reveal pakwise linear association of the indicators. Principal
component analyses and the biplot reveal multivariate associations; groups of indicators that tend
to behave similarly within sets. Similar behavior of several indicators many indicate
redundancies in the definition of indicators or may be suggesting a definition for one dimension
of health of the ecosystem. Different groups may be addressing different dimensions of health.
3-22
-------
5 -
4 -
3 -
2 -
0 -
T
"T
I
I
I
_L
1
1
I
I
_L
T
_L_
1
Region 1 - Mountains
Region 2 - Piedmont
Region 3 - Coastal Plain
Figure 3-6. Examples of box-plots for the electrical conductivity of soil in three Regions of
North Carolina.
3-23
-------
*1«-7*
S77JT
3
ta.»t
17.2
4D7.J
7B7.S
utr.r
Long
C2l3V9 •
see. -
400. •
see. -
see. -
lee. -
fl/ * I * I 3
"O "^n1 °n *n ° DD
•• x LJ xc O «U o ..
QUQ * • IT
D°« X x n X * D X ! D n D D
0° OaovX
X
X
*e. see. 400. eoe.
X
f*ia nD +DY+i**
°D° ° X 4X 4 t 4
n 0 V X 4 f
OCX 4.
D x x xx \
X * +
X 4
4
usl*
4 < .651
X < 1.336
Q < 2.507
X >=2.507
0 - Mtsstnv
eea. ieae. 1200.
Standardized Scale
Figure 3-7. A, Kriged estimates of USLE over North Carolina (an example only); B, Display
of spatial patterns.
3-24
-------
Interpretation of the principal component analysis will require close collaboration with the
scientists.
The nature of the spatial correlation structure will be investigated by fitting variogram models
that depend on the spatial relationships of the observations. This will require knowledge of the
site locations and must be done in conformity with the confidentiality requirements of NASS.
The variogram information will be used in constructing spatial displays of regional patterns. The
spatial correlation structure and some method of spatial interpolation may also be used as one
possible way of masking the true locations of the sample points (maintaining confidentiality)
while retaining data that have properties similar to the original observations.
Correlations between the agroecosystem indicator values and variables measured by NASS
in the full JliS sample will be explored to determine if any of the correlations are of sufficient
magnitude to warrant use of double sampling techniques.
3.4.4. Comparison of the two Sampling Plans
Both sampling designs, the Hexagon Plan and the Rotational Panel Plan, sample exactly the
same reference population and both are proper probability samples. Consequently, the estimates
of population total or means obtained from the two designs are estimating the same population
quantities; any differences would be due to sampling error. The primary differences in the two
designs are the use of stratification and the rotation of segments out of sample after five years.
Both stratification and the systematic spatial coverage may affect precision of all estimates. The
precision of the estimates of status for the two plans will be adjusted for differences in sample
size before being compared. The direct effect of stratification on precision will also be
determined by applying post-stratification to the Hexagon sample.
The rotation of the sample segments will affect precision estimates of change or trend; The
magnitude of the effect will depend on the magnitude of the temporal correlations. The temporal
3-25
-------
correlations cannot be estimated from a single year study so that any comparisons of precision
of change or trend estimation will necessarily require assumptions about the temporal correlation.
The costs per observation will differ considerably between the two plans due to the closer
coordination of the Rotational Panel Plan with the ongoing JES. Costs will be determined for
each sampling plan and combined with the measures of precision to obtain information per unit
cost for each indicator population estimate.
This empirical comparison of the two designs will be limited because only one realization
of the sampling process will be available and because variances in the limited region covered by
the 1992 Pilot may not adequately reflect variances for other regions. Consequently, simulation
will also be used for a more thorough comparison of the designs and these results will be
compared with the theoretical work of Lesser (1992).
3-26
-------
4. Assessment Endpoints and Indicators
The goal of the Agroecosystem program is to monitor and assess the long-term status and
trends in the health of the nation's agricultural resources from an ecological perspective. Because
agroecosystem health cannot be measured simply and directly, a number of assessment endpoints
and associated indicators have been proposed that, when monitored, will describe collectively the
overall condition of agroecosystems.
4.1. Societal Values, Assessment Endpoints, and Indicators
For the purposes of EM AP, the ARG has identified three societal values that are of primary
importance in determining agroecosystem condition. These societal values are: 1) supply of
agricultural commodities, 2) quality of natural resources, and 3) conservation of biological
resources (Figure 4-1). Supply of agricultural commodities addresses the ability of an
agroecosystem to provide adequate crop and livestock yield and quality over the long term.
Quality of natural resources is the freedom of natural resources from harmful levels of
substances such as trace metals, pesticides, fertilizers, pathogens, salts and pollutants in one or
more components of the agroecosystem. These are present usually as a result of human
activities, may be persistent and mobile in the environment, have potential to bioaccumulate in
the food chain, or have potential short- or long-term adverse effects on biota, including humans.
Conservation of biological resources reflects the desire to maintain the ecological soundness
of crop and non-crop components of the agricultural landscape as habitat for plant, animal and
microbe species.
Assessment endpoints are quantitative or quantifiable expressions of the environmental value
being considered in the analysis (Suter 1990). Seventeen assessment endpoints have been
identified for possible use in the Agroecosystem monitoring program. These assessment
endpoints and their relationship to the three societal values discussed above are shown in Table
4-1. These assessment endpoints address the agricultural and ecological aspects of agroecosystems
4- 1
-------
Supply of
Agricultural
Commodities
Conservation
of Biological
Resources
Status & Trends
in Agroecosystem
Health
Quality of
Natural
Resources
Figure 4-1. Agroecosystem societal values that will be addressed with a suite of
indicators to determine the status and trends in agroecosystem health.
and have been selected through a process of consultation with experts and extensive discussions
within the ARG over a two-year period (Heck et al. 1991). Although the list is comprehensive,
it can be changed. Also, because of fiscal and logistical limitations, it may not be possible to
retain, all of the assessment endpoints within the eventual regional and national monitoring
program.
Indicators (measurements) are characteristics of the environment that, when measured,
quantify the magnitude of stress, habitat characteristics, degree of exposure to stressors, or degree
of ecological response to an exposure (Hunsaker and Carpenter 1990). Indicators serve as the
basis for quantification of the assessment endpoints. For example, water-holding capacity, amount
of erosion, and indices of soil biological activity are indicators that serve to quantify the
assessment endpoint of soil quality.
4-2
-------
Table 4-1. Association between the Agroecosystem assessment endpoints and societal values.
Assessment Endpoint
Crop Productivity
Soil Quality: Physical/Chemical
Water Quality: Ponds
and Existing Wells
Land Use/Land Cover
Agrichemical Use
Soil Biological Health (Nematode
indices)
Landscape Structure
Groundwater/Well Comparisons
Biological Ozone Indicator
(Clones of white clover)
Socioeconomic Health
Pest Density
Foliar Symptoms
Beneficial Insects
Genetic Diversity
Habitat Quality
Wildlife Populations
Livestock Productivity
Nonpoint Source Loading
Water Quantity (irrigation)
Other Biomonitor Species
Supply of
Agricultural
Commodities
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Quality
of Natural
Resources -
X
X
X
X
X
X
X
X
X
X
X
Conservation
of Biological
Resources
X
X
X
X
X
X
X
X
X
X
X
X
X
X
- Air, soil, and water, including transport of contaminants into, within, and out of agroecosystems.
4-3
-------
Four types of indicators are defined for EMAP. The relationships of these indicator types to
the assessment endpoints are shown in Table 4-2. The four indicator categories are:
I. Response indicator: a biological/ecological characteristic measured to provide evidence
of the condition of a resource at the organism, population, community, or ecosystem level
of organization.
2. Exposure indicator: an environmental characteristic measured to provide evidence of the
occurrence or magnitude of contact with a physical, chemical or biological stressor.
3. Habitat indicator: a physical, chemical or biological attribute measured to characterize
the conditions necessary to support an organism, population, community or ecosystem in
the absence of stressors.
4. Stressor indicator: a characteristic measured to quantify a natural process, an
environmental hazard or a management action that results in changes in exposure or
habitat
4.2. Selection of Assessment Endpoints for The 1992 Pilot Project
One of the objectives of the 1992 Pilot Project is to evaluate empirically an initial suite of
measurements or indicators which will address several of the selected endpoints for monitoring
the ecological condition of agroecosystems (Section 2.2).
4.2 J. Rationale for Selection of Pilot Endpoints and Indicators
Because the 1992 Pilot Project has several primary objectives (Section 2.2), a balance was
required between the selection of assessment endpoints and other aspects of the project. Also,
endpoints needed to be selected based upon the information derived from the Pilot which would
aid in judging the suitability of specific measurements or indicators and upon the likelihood of
4-4
-------
Table 4-2. Association between the Agroecosystem assessment endpoints and the indicator types.-
Assessment Endpoint
Crop Productivity
Soil Quality: Physical/Chemical
Water Quality: Ponds and Existing Wells
Land Use/Land Cover
Agrichemical Use
Soil Biological Health (Nematode indices)
Landscape Structure
Groundwater/Well Comparisons
Biological Ozone Indicator
(Qones of white clover)
Socioeconomic Health
Pest Density
Foliar Symptoms
Beneficial Insects
Genetic Diversity
Habitat Quality
Wildlife Populations
Livestock Productivity
Nonpoint Source Loading
Water Quantity (irrigation)
Other Biomonitor Species
Indicator Types
Response
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Exposure
X
X
X
X
X
X
X
X
X
X
X
X
Habitat
X
X
X
X
X
X
X
X
X
X
Stressor
X
X
X
X
X
X
X
X
X
- See definitions on page 4-4.
4-5
-------
success in implementing associated indicators. This likelihood of success was judged on the basis
of 1) the ability of NASS enumerators to collect the required survey data and samples, 2) the
availability of analytical and assay procedures that fit within the quality and fiscal standards of
the ARC, and 3) the ability of the ARG to utilize and interpret the data obtained.
The first criterion was essential because one element of the pilot project is to establish and
refine the working relationship between the ARG and NASS. It is the current intent of the ARG
that NASS enumerators will serve as the primary grower contact and as the primary field
personnel for acquiring specific samples (e.g., soil, water, etc.). In the Pilot Project it will be
essential to establish this as a viable and realistic approach.
The second criterion reflects the desire of the ARG to produce the best quality product (a
pilot assessment of the condition of agroecosystems in a limited geographic area) within the
constraints of available budget Budgets often dictate what is and is not possible. The challenge
to the ARG has been to assemble a suite of endpoints and indicators that is scientifically credible
and informative within such fiscal constraints.
The third criterion acknowledges the difficulty of interpreting monitoring data in an integrated
assessment that is intended to assess system health. Thus, the approach emphasizes key, critical
areas in the ecological assessment of agroecosystem condition. Also, from the perspective of the
design and statistics components of the pilot, indicators need to be selected that have a relatively
clear, known interpretation so that variability within and among sample units could be analyzed
and placed in the proper context
4.2.2. Assessment Endpoints Selected for the 1992 Pilot Project
Based upon the three criteria identified in section 4.2.1, five assessment endpoints were
selected for initial implementation in the 1992 Pilot Project:
o Crop Productivity
4-6
-------
o Soil Quality
o Water Quality
o Land Use
o Agricultural Chemical Use
All three societal values are addressed by this group of assessment endpoints. The specific
societal values addressed by these endpoints are identified in Table 4-1.
The selected assessment endpoints will be quantified primarily via response, exposure and
habitat indicators or measurement endpoints (Table 4-2). The specific, candidate indicators and
measurements to be obtained during the pilot are identified in Section 5. The measurements
needed to quantify these endpoints are generally well known. However, critical decisions must
still be made concerning the specific measurements and techniques of data analysis related to the
endpoints that will be appropriate for EMAP.
The assessment endpoints selected have both agricultural and ecological interpretations. The
apparent emphasis on agricultural characteristics reflects the availability of a greater number of
agricultural system attributes that can be readily characterized, because long-term monitoring of
agricultural attributes has been carried out within agricultural systems. The ecological applications
of the endpoints are also appropriate when the total agroecosystem is considered. As agroecology
develops as a discipline and as the ARG continues to make progress in indicator development
(see Sections 4.3 and 6), the program emphasis will obtain the desired agroecological approach
to monitoring the status of agroecosystems.
4.3. Research Activities on Candidate Indicators and Assessment Endpoints
Because of the desire of the ARG to have an ecological focus, four specific research projects
have been selected for inclusion in the 1992 Pilot Project. These projects (see Sections 6.1 - 6.4)
4-7
-------
include research on: 1) a response indicator of the biological health of soils based upon the
prevalence and frequency of occurrence of specific trophic groups of soil-inhabiting nematodes
(Section 6.1); 2) a series of currently available and new habitat indicators (a landscape ecology
perspective) to characterize the structure and quality of agricultural landscapes (Section 6.2); 3)
a direct comparison of the quality of water available for irrigation and consumption in existing
and newly-established wells in agroecosystems with a series of exposure indicators (section 6.3);
and, 4) a biomonitor of the impact of ozone on crop production systems (section 6.4). If these
research projects confirm the suitability of one or more of the candidate indicators for
agroecosystem monitoring, the indicators or measurement endpoints will be included at some
level in the 1993 demonstration and pilot projects.
4.4. Current Status of the Assessment Endpoints for the Agroecosystem Program
Table 4-3 presents a summary of the current status of the assessment endpoints identified for
use and development by the ARG. Expected source measurement/indicator data and sample
design for obtaining 1992 Pilot data is indicated, the index period for obtaining measurements
is shown, the parties responsible for collecting/handling/summarizing the data are listed, and-the
stage of development for each assessment endpoint is shown. The definitions of the
developmental stages are shown as footnote 2 in the Table.
The flow of the major activities for the 1992 Pilot with emphasis on the assessment endpoints
and collection by survey or sampling is shown in Section 8 (Rgure 8-1).
4-8
-------
Table 4-3. Vital statistics on the Assessment Endpoints for the Agroecosystem Program.
Assessment Endpoint
Crop Productivity
•Soil Quality: Physical/Chemical
Water Quality: Ponds and Existing
Wells
Land Use / Land Cover
Agrichemical Use
Soil Biological Health
(Nematode indices)
Landscape Structure
Groundwater/Well Comparisons
Biological Ozone Indicator
(Clones of white clover)
Socioeconomic Health
Pest Density
Foliar Symptoms
Beneficial Insects
Genetic Diversity
Habitat Quality
Wildlife Populations
Livestock Productivity
Nonpoint Source Loading
Water Quantity (irrigation)
Other Biomonitor Species
Source of
Data
Survey
Sampling
Sampling
Survey-
Survey
Sampling
Remote?
Sampling
Sampling
Survey
Sampling
Sampling
Sampling
Survey
Sample
Design from
which data
wil come
Boti>?
Both
Hexagon or
Rotational
AD of JES
Both
Hexagon or
Rotational
Off-frame
Off-frame
Off-frame
Index
Period^
Fall
Fall
Fall
May-June
Fall
Fall
Several
Summer
Spring/
Summer
Responsible
Party
ARG/NASS
ARG/NASS
ARG with
Athens-ERL
ARG/NASS
ARG/NASS
ARG/NASS
ARG
Athens-ERL
ARS
cooperators
Stage of
Develop-
ment
1
1
1
1
1
2
3
3
3
4
4
4
4
4
5
5
5
5
5
5
- Period during which data is taken (Note: survey data may actually represent earlier events)
- l=developmental 2=research 3=off-frame research 4=onder consideration 5=proposed
Numbers 1,2,3 will be included in the 1992 Pilot
2 Both=segments from both the hexagon design and the NASS rotational panel
H Will also make use of the NASS strata for North Carolina (developed in 1978)
4-9
-------
-------
5. Description of Specific Assessment Endpoints for The Pilot Project
5.1. Crop Productivity
5.1.1. Introduction
When people are concerned about agriculture, crop production is often the focus of their
concern — Will There Be Enough Food? was the title of the 1981 Yearbook of Agriculture
(USDA 1981). In addition to its crucial importance to human society, the crop plant also
provides food for soil microbes, plant-eating insects, and other organisms. Crop productivity is
thus an important ecological parameter and an important response indicator of agroecosystem
condition.
Figure 5.1-1 illustrates
some of the elements
which affect crop
productivity. Many more
anthropogenic factors
influence crop plants than
influence plants in less
highly managed systems.
In industrial agriculture,
the marketable yield of
crop plants is being
optimized through
management (tillage, Figure 5.1-1. Some factors which influence crop productivity.
planting date, fertilizer,
irrigation, etc.), with a view toward economic profit. Mitchell (1984) points out that in
monocropping, the economic-agricultural system determines which crop varieties are planted and
Natural Factors
[ Temperature |
| Precipitation |
""H SoHQuaWy I
| SoUr Radiation |
Disease / Pests |
| Catastrophes I
Management Inputs
I Crop Variety |
1 Fertilizers I
Irrigation |
1 TiBage j
| Pesticides F^
[ Crop Rotation 1
Crop Productivity ^B
\ Air Pollutants |
Soil Erosion 1
[Soil Compaction I
Saltnization 1™"^
JHerb. Carryover |
| Global Change j"
Anthropogenic Stressors
J Commodity PricM 1
j Input Costs 1
j Commodity Pgm«. j
IconcaiVn Re*acvt |
| Pesticide Regs. |
"] Erosion Control |
Soclo-Economlc Factors
5.1- 1
-------
what yields arc obtained. Crop productivity could be measured in either economic or ecological
terms; it is the latter which are of interest in EMAP. An indicator of productivity should be
responsive to environmental stresses such as air pollution, climate change, soil degradation and
water contamination.
Crop productivity as an assessment endpoint has four facets: total production in a region,
yield (production per unit land area), yield as a biological response indicator adjusted for inputs,
and production efficiency (production per unit input). Quantifying either of the last two requires
a knowledge of inputs as well as yield, but the two perspectives are subtly different. To use
yield as a biological response variable, one must adjust for those factors that contributed to yield
but are considered extraneous to ecosystem health. These may include natural inputs (e.g.
rainfall), human-produced inputs (e.g. pesticides), or both. Production efficiency would quantify
agroecosystem status by comparing production achieved to resources expended, whether those
resources contributed directly to yield or not Again the overlap of ecological and socio-
economic issues in agroecosystems is apparent. The proposed emphasis for the crop productivity
indicators for the 1992 Agroecosystems Pilot is on the third of the four facets of productivity:
a biological response that points toward agroecosystem health. Production efficiency will also
be considered, especially during the assessment phase.
At a workshop in January 1992 in Athens, Georgia, the ARG endorsed three types of
indicators of crop productivity: standardized yield, net primary productivity, and measures
derived from remote sensing. These are briefly introduced below.
Crop yields have, of course, been survej'ed and reported for decades, but yield alone is not
a sufficient indicator of "health". If one field produces a higher yield than another because of
additional fertilizer, is that first field therefore "healthier"? Before deciding this it is necessary
to account for the effects of management inputs and perhaps for the influence of weather. During
the Pilot, several such standardized indicators will be tested, including output/input indices and
adjusted yields.
5.1-2
-------
Adjusted or not, crop yield reflects only part of a plant community's productivity. A fuller
measure of productivity can be calculated by dividing yield by the harvest index (HI) of that
particular crop, to give the total aboveground dry matter. The HI is usually defined as dry matter
yield divided by total aboveground dry matter (Tivy 1990, Donald and Hamblin 1976). Given
crop yield, HI, and the root-shoot ratio at harvest, an estimate of total dry matter production (i.e.,
net primary productivity, NPP) can be calculated.
NPP = (harveste& dry matter) ,- roots *
(harvest index)shoot
Alternatively, factors to convert directly from yield to NPP can be constructed from experimental
data (Sharp et al. 1976, Klopatek 1978) (See Sections 5.1.3.3 and 5.1.7.3).
Remote sensing is discussed briefly in Section 5.1.9.2, "Possible Alternative Measures of
Productivity." The ARG lacks the resources to launch a major effort on this indicator at this
time.
A secondary step in the development of indicators of productivity will be to standardize and
combine data from different crops. Yields vary among species, even when expressed in common
units. The simplest method of combining such data would be to express the yield (or other
indicator value) from a sample field as a fraction of the average yield of that crop. The average
would be calculated over a reference region and time period. This would partially adjust for the
local effects of soils, management practices, and climate while allowing trends to be followed
for different crops and locations. A slightly more sophisticated method that adjusts variances as
well as means is described in Section 5.1.7.1. Reference means and variances are readily
available for yield data, but other productivity indicators may require several years of sampling
to establish a baseline.
5.1-3
-------
5.12. Data to be Collected by NASS
NASS will gather essential information about both inputs and outputs. They will ask for the
production from the harvested area of each sample field (Section 3.3) as well as the units in
which these arc expressed. For inputs, the Agroecosystem 1992 Pilot Questionnaire will contain
questions about timing and amounts of fertilizer, lime, and pesticide applications; about the tillage
system; and about irrigation (Appendix 5). For the 1992 Pilot, sample fields will be drawn from
the population of planted cropland in North Carolina, to include those crops listed in Table 5.1-1.
Crop lands excluded are shown in Table 3-2.
The NASS enumerators will also be collecting soil samples. Data from soil analyses may
or may not be used in the process of standardizing yields.
Table 5.1-1. Principal crops eligible for selection in the Agroecosystem 1992 Pilot
Barley
Corn
Hay
Irish Potato
Oat
Peanut
Rye
Sorghum
Soybean
Strawberry
Sweet potato
Tobacco
Upland Cotton
Vegetables
Winter Wheat
5.1.3. Essential Complementary Data
5.1.3.1. Weather Data
Some productivity indicators should be adjusted for year-to-year weather fluctuations. This
will most likely require the use of weather data and some sort of crop growth model.
5.1 - 4
-------
Geographically referenced weather data will probably be obtained from the National Oceanic and
Atmospheric Administration (NOAA) personnel stationed at the EPA laboratory in Research
Triangle Park, NC. An initial list of needed data was discussed with Sharon LeDuc, Ellen
Cooler, and Brian Eder of (NOAA) in late November 1991. We anticipate that daily precipitation
totals and daily high and low temperatures will be needed. Some measure of insolation or
photosynthetically active radiation (PAR) on a daily basis and a database containing drought
index values, should also be obtained. A final list of weather data needs will be compiled
following consultation with several crop growth modelers and after selection of models to be
used.
5.1.3.2. Production Practices Not Queried
Certain values needed for calculations will not be asked on the Survey Questionnaire or will
not be known by some farmers, so the values must be obtained from other sources. For example,
industry standards for moisture content of the major U.S. crops will be obtained from NASS.
Crop models may also require parameters such as plant density (number of plants per unit area).
Typical values of such factors will be obtained from the literature or from the Cooperative
Extension Service personnel.
5.1.3.3. Conversion Factors
The most difficult complementary data to obtain will be the conversion factors for
standardizing inputs and outputs. Two issues must be addressed: 1) conversion factor
availability for a given crop or input and 2) variability of the conversion factor. If energy
productivity is calculated, conversions of input values to energy units will come from the
literature (e.g., Southwell and Rothwell 1977, Fluck and Baird 1980, Stout 1990). Factors for
calculating NPP present several challenges. The HI and root-shoot ratios can be used to calculate
primary productivity from harvested yield, but these ratios depend on variables such as fertility,
crop variety, water status, and ozone stress (Donald and Hamblin 1976, Pettersson 1987, Temple
1990). For the Pilot, conversion factors used by Sharp et al. (1976) for eleven crops will be the
5.1-5
-------
starting point (Table 5.1-2). These factors allow conversion from yield to productivity. Standard
moisture contents will be obtained from NASS, rather than assuming 12% for every crop, as
Sharp et al. has done. Using these conversion factors means accepting that they have not been
validated, as well as assuming that they have not changed in 20'years. It would be impractical
for the ARG to try to derive conversion factors empirically, so further research into the literature
is planned to find conversion factors and their variability. The search initially may be restricted
to six to eight major crops. There are many unanswered questions: Are there different
conversion factors for different locations or soil types? How is below-ground productivity
included? Can conversion factors derived at one location be applied to a larger region? Can
crop growth models be used to verify conversion factors?
Table 5.1-2. Conversion factors from yield to net primary productivity (NPP).^
Crop
Wheat
Soybean
Com
Oats: winter
spring
Irish and sweet potatoes
Cotton (lint yield)
Tobacco
Peanuts
Hay
Conversion Factor
3.69
4.52
2.62
5.30
5.22
2.47
2.08
2.03
2.00
1.30
Source: Sharp et al. 1976. Yields are first expressed in tons/ha,
then multiplied by the given factors, and then adjusted to 0%
moisture to give NPP in tons/ha. Values for wheat, corn, oats, and
tobacco had been calculated from North Carolina data.
5.1-6
-------
5.1.3.4. Reference Yield Values
For purposes of calculating normalized yield, county average yields for the reference period
of 1980-1989 will be used (see Section 5.1.7.1). These have been obtained from NASS for the
major crops in North Carolina.
5.1.3.5. Soil Type
Some models may require more detailed information about the soil than the data which the
ARG will be obtaining from each field. Procedures are being developed to determine the soil
series at each sample site. This will allow access to information from the State Soil Survey
Database (SSSD).
5.1.4. Logistics
The field-level data needed for the crop productivity indicators will be taken from tfo
Agroecosystem Survey Questionnaire to be administered by NASS in the fall of 1992 (Appendix
5). Logistics for obtaining soil samples are described in Section 5.2. As mentioned above, some
soils data will be derived from the SSSD, which the ARG is obtaining. Complementary data
from the literature and other sources will be obtained by the ARG. Weather and climate
databases will be obtained through the ARG information manager, the EMAP Information
Management Committee, and possibly NOAA cooperators.
5./.5. Quality Assurance
Quality Assurance (QA) procedures for data collected by NASS are discussed in Section 7.
For weather data, QA procedures will be discussed with the supplier of the database. It is
anticipated that the complementary data will present several QA problems. For example, the
factors from Sharp et al. (1976) do not carry associated estimates of variability. Procedures will
be investigated for developing QA standards for conversion factors. QA for indicators derived
5.1 - 7
-------
using crop growth models will also require development. In particular, Data Quality Objectives
(DQOs) will not be estimated until after the Pilot is completed.
5.1.6. Metadata Requirements
Because data from so many sources will be needed, metadata requirements will be .extensive.
They will fall into different groups, depending on their level of applicability (Table 5.1-3). The
QA/QC procedures will be part of the metadata.
5.1.6.1. Data keys will be needed to identify (ID) the sample: date (year), frame (hexagon or
MASS), PSU/segment ID and sample (field) number. Although the association of the sample
number with a particular field must be kept confidential, some geographic information will be
needed. In particular, the name of the county will be needed so that county averages can be used
to normalize yields. It may also be necessary to indicate the region to which the field belongs,
if summary statistics are calculated for subregions of the state, although it may be possible to use
the PSU identifier for assigning fields to regions.
5.1.6.2. Certain metadata items will be the same for each record in the entire database. These
include the descriptions of each variable. In many cases the description will be the question
which was asked in the questionnaire. The description is to include the units in which the
quantity is expressed, for example "acres" for the area of a field or area under irrigation, "dollars
per gallon" for the price of fuel, and "acre-inches" for the amount of irrigation water. Currently,
NASS data are taken in U.S. units. Some quantities, such as moisture content and fertilizer
analysis, will be dimensionless ratios and should be expressed as decimals (not percentages). The
description of fertilizer analysis should indicate the chemical form in which the analysis is
expressed, e.g. P vs. P2O5. A few conversion factors will apply to the entire database, for
example the conversion from acres to hectares. All final summary statistics will be expressed
in metric units or as dimensionless ratios.
5.1-8
-------
Table 5.1-3. Elements of metadata to be recorded in association with data for the crop
productivity indicators, not including ancillary data such as weather.
1. Data keys
o Year
o Frame (hexagon or NASS)
o PSU/segment ID
o Sample (field) number
2. Elements which apply to every record in the database
o Description of variable (including the form in which chemical species are
expressed, such as P vs. P2O3)
o Units (may be dimensionless)
o Coding tables for pesticide product codes, etc.
3. Elements which are associated with a particular crop, land use or input
o Units for crop yield (bu., tons, etc.)
o Conversion factors from yield to NPP, including units
o Conversion factors from inputs to energy (if used)
o Source of conversion factors
o Variability of conversion factors
o Common names for pesticides
4. Elements associated with individual records (these elements vary among or within
sampled fields)
o Units for fertilizer, manure, and pesticides
o Units, source and base period for county averages used to standardize yields
Another category of metadata which will be the same for the entire database will be the
translation tables for those variables that are recorded by code numbers (unless NASS converts
them to text before shipping to the ARG): crop and land use codes; fertilizer timing and
application method; pesticide product code, timing, application method, and applicator, type of
manure; tillage system; erosion control methods; irrigation system; and source of irrigation water.
If yes and no responses are stored as 1's and O's, this needs to be documented.
5.1.6.3. Some of the metadata will be associated with a particular crop, land use, or input. For
example, yields are expressed in different units for different crops, and different crops have
different factors for converting yield to NPP. The descriptions of the conversion factor variables
5.1-9
-------
will need to reveal the quantity to which they apply and what units are being changed to what
other units. The origin of the conversion factors should be recorded, including the source from
which they were obtained, an estimate of their variability, and the mathematical derivation of
composite conversion factors. If production inputs are to be standardized, the conversion factors
(e.g. for expressing fuel in energy units, manure as nutrient equivalent, etc.) will likewise need
extensive documentation.
5.1.6.4. A fourth group of metadata needs to be associated with individual records in the
dataset, because there may be differences among.or within fields. This group includes the units
for fertilizer, manure, and pesticides. Care must be taken that the correct conversion factors are
applied to these quantities, because units will vary. Another group of data at this level will be
the units, source and years of the county averages used for standardizing yields. These need not
be stored with the individual record, but must be indexed by the particular county and crop.
5.1,7. Data Analysis and Integration
Figure 5.1-2 shows the 1990 harvested acreage for 28 North Carolina crops. Except for
apples, peaches and blueberries, any of these crops would be eligible to be sampled in the fall
of 1992. However, many occupy such small areas that they will be missed. Sample fields will
be drawn according to the protocol outlined in Section 3. Except for NPP, indicator values will
be calculated separately for each crop, so that shifts in the productivity of one crop do not mask
shifts in the productivity of others. Keeping the indicators separate is also a way of recognizing
the important differences among the requirements and adaptation of different crop plants.
It may be impossible to calculate some or all of the productivity indicators for certain crops.
For example, the sampling scheme might draw too few rye fields for a reliable estimate, or
existing crop growth models may be inadequate for calculating adjusted yields of sweetpotatoes.
It is not known how often these problems will occur, but it is unlikely that indicators will be
reported for all of the crops found on the sample fields (Table 5.1-1).
5.1 - 10
-------
Harvested Acreage of N.C. Crops, 1 99O
Soybaana for baana
Corn for grain
Whaat for grain
Preliminary
i i i i | i i , i | i 1 , , 1
kBHBHa^HHa^HHB sso.ooo
Tobacco (all) ••HBIaHl 2S4.2OO
Cotton, upland
Paanuta
Corn for •ll»0o
Oata for grain
Sor0hum for grain
Barloy for grain
Sorghum for allaga
Ryo for grain
All hay
Alfalfa hay
Swaatpotatoaa
Irlah potatoaa (all)
Cucumbar* (prooaaa)
Watarmalona
Qraan pappara
Snap baana
Cucumbara (fraah)
Cabbaga
Swaat corn
Strawbarriaa
Tomatoaa
Apptoa
Paachaa
Bluabarriaa
•••• 200.000
•HH 1B4.OOO
IHBS-OOO
• 40,000
|4O.OOO
| 3O.OOO
| 2O.OOO
| 15.OOO
^••••i^B 470.000
|30.000
| 34.0OO
| 17.6OO
| 24.600
O.5OO
7.OOO
6.700
6.100
9.40O
3.2OO
2.1OO
1.400
1 15.000
4.2OO
2.900
1 * • • 1 ....1. . . , 1
o 500,000 1,000,000 1,500,000
Acres harvested
Figure 5.1-2. Harvested acreage of several North Carolina crops, 1990 (preliminary). Source: 1991 North Carolina
Agricultural Statistics.
Efforts will focus on calculating indicators separately for each crop, but attempts will also
be made to combine data from different crops into some sort of overall index. It may be possible
to weight and then directly combine individual values if they are on a common scale (as NPP
will be). If indicators such as adjusted yield are not on a common scale, the combined index
may need to wait until the ARG has established baseline indicator values for each crop. These
baselines can then be used for standardization and for detecting trends. "Normalized yield", as
described below, will be a test of this sort of calculation.
The five types of indicators proposed here range from the simple and straightforward to the
complex and uncertain, but potentially more useful. Simple yield is a key building block of the
5.1 - 11
-------
other indicators. Normalized yield is an easy way to try to combine data across crops. The net
primary productivity indicator will use data from all crops to make a broad ecological statement
Output/input relationships and adjusted yield are expected to unrnask inherent productivity
differences hidden by management or other variables.
5.1.7.1. Simple yield
Straightforward yield figures are routinely reported by agencies such as NASS. Nevertheless,
there are four reasons for the ARG to report simple agronomic yields:
o The values obtained can be compared to the estimates which NASS gets from a much
larger sample.
o Simple yields form the foundation for the adjusted yields. It will be interesting to see
the spatial distributions and cdf s of such yields before adjustment
o A simple assessment can be done by plotting yields over time and comparing them to
changes in inputs over time. This may not be the most powerful use of the data, though,
since inputs and output will be known on a fieldrby-field basis.
o The method of aggregating over sub-regions within the state can be tested with simple
yield. Data from NASS are not aggregated this way; the approach may be unique to
EMAP.
5.1.7.2. Normalized yield
Normalized yield QO will be calculated for each field by using that field's yield per acre
(Y = production from field/area harvested), the county average for the arbitrary reference period
1980-1989 (Yre£), and the standard deviation of that average yield (s). Similar to a standard
normal variate, the calculation will be
5.1 - 12
-------
V_V
The number 5 is added arbitrarily so that the distribution of Y' will have a mean of 5. Because
the standard deviation of Y' is 1 and its mean is 5, negative values will be conveniently rare.
The advantage of this method of standardization is that both means and variances of different
crops are put on a similar scale. For simplicity, s will be calculated from temporal (year-to-year)
variation in the county means. Values of Y' will be averaged over the fields within each segment
and weighted according to the inclusion probabilities of the field being sampled. Segment means
will be used to calculate regional means, quartiles, etc., with weighting and stratification .as
appropriate (see Section 3).
5.1.7.3. Net Primary Productivity
Net primary productivity (NPP) is the net accumulation of plant biomass per unit area per
unit time. It is a useful ecological indicator because it allows comparisons among different types
of ecosystems. An estimate of seasonal NPP will be calculated from the Pilot yield data. The
yield of each crop will be expressed in kg/ha, and then converted from economic yield to dry
matter production, using conversion factors like those from Sharp et al. 1976 (Table 5.1-2), along
with standard moisture contents. A sample calculation is given in Section 5.1.7.6, though the
method of aggregation and the method of handling double crops have yet to be determined. Net
primary productivity will not be reported for individual crops, since it is simply a multiple of the
yield. Also, NPP will not be compared across crops, because different species would be expected
to have different NPPs. Instead, NPP will be aggregated over all sampled crops within a region.
The method of integrating these values over each region of interest will need to account for the
area occupied by each crop. It would seem that regional productivity will be a function of both
the productivity of individual fields and the patterns of land use (Sharp et al. 1976; Turner 1987).
5.1 - 13
-------
5.1.7.4. Output/Input Relationships
One way of measuring plant health using ARG data is to look at the response of yield or
NPP to various inputs, either singly or collectively. This might involve single-factor or multi-
factor productivity indices. Another possibility would be to use Agroecosystem data to determine
the coefficients relating yield to inputs (see for example Lin et al. 1991).
Within the concept of a multi-factor index, one approach to aggregating the input data would
be to put them on a common energy scale. Ideally, this should be done using process analysis
and on the basis of energy resource depletion (Southwell and Rothwell 1977) or some similar
philosophy. Various types of energy output/input ratios have been used in agriculture (Fluck and
Baird 1980); however, the validity of the energy ratio (energy output per energy input) has been
questioned, and energy productivity (e.g., kg of production per unit of input energy) has been
suggested as a better measure (Fluck 1979).
5.1.7.5. Adjusted Yield
Another way to develop an indicator of crop health in agroecosystems would be to estimate
what the yield on each field would have been if a standard set of inputs had been used. These
adjusted values can then be aggregated, mapped, or treated in other ways. Such adjustments
would come from existing research findings on the response of yield to inputs. A similar method
would be to build an indicator from a difference or ratio between a field's yield and the yield
predicted by a statistical or process model. Much work needs to be done in this area. The two
critical steps are (1) deciding which inputs, natural and anthropogenic, should be accounted for
and (2) finding the means to make those adjustments. Whether indicators should be adjusted for
fluctuations in the weather is a major question. Such an adjustment may stabilize the variability
inherent in yield data, but if all weather variations were accounted for, it might be more difficult
to detect changes caused by global climate change (except through shifts in land use).
5.1 - 14
-------
Rather than using the predicted yield from a model, an adjustment could be made using an
estimate of the optimum yield for a given crop on a given soil series. Such values are published
by the USD A Soil Conservation Service for the six or eight major crops in each soil survey, but
these are not updated very often, and the updates occur at staggered intervals (USDA Soil
Conservation Service 1970, 1983). If this approach is to be useful, a set of estimates would be
needed that were made within a few years of each other and covering the entire state (eventually
the country).
5.1.7.6. Sample Calculations
Following are examples of how two of the indicators would be calculated, given the
following hypothetical yields for 1992.
Hypothetical corn yield in sample field: 80 bu/A @ 12.5% moisture
Hypothetical soybean yield in sample field: 26 bu/A @ 8.5% moisture
Sample calculation: normalized yield (Section 5.1.7.2)
Assume mean yields and standard deviations in County A for the reference period 1980-1989:
Corn: mean = 71 bu/A, std. dev. = 20 bu/A
Soybean: mean = 25 bu/A, std. dev. = 7 bu/A
Normalized yields
V— V
Corn: (80 - 71)/20 + 5 =" 5.45
Soybean: (26-25)/7 + 5 = 5.14
These will be calculated for each field.
5.1 - 15
-------
Sample calculation: net primary productivity (Section 5.1.7.3)
NPP = yield*(unit conversion to tons/ha)*(NPP conversion factor)*(l,-fraction moisture)
o Unit conversion factors from standard bushel weights for corn (561b) and soybeans (601b)
o NPP conversion factors from Sharp et al. 1976, see Table 5.1-2
Corn
(80^X0.0628^/^X2.62^X1.00-0.125)
A la A yitltf
=11.5
tons
ha
Soybean
A
ha A yield
=7.23
tons
ha
In practice, NPP might be calculated by starting with the production of each crop over the
entire region, which can be converted to net primary production, summed over crops, and divided
by the total harvested area to give the regional NPP estimate.
5.1 - 16
-------
5.1.8. Interpretation of Indicators
The above indicators will serve as biological response variables that show whether
productivity is rising, falling, or remaining the same over time. The goal of the ARG is to
measure the status of productivity indicators, along with other agroecosystem indicators, to obtain
a picture of the condition of U.S. agricultural resources with regard to sustaining both the supply
of agricultural commodities and the ecological integrity of the system. Because of the great
differences among crop species, most indicators will be interpreted separately for each crop
before a composite index for all crops is calculated. The exception will be NPP, which will only
be reported as a composite.
Assigning value judgements to an indicator (good status vs. bad status) can only be done in
reference to some criterion. Unfortunately, such criteria for crop productivity are hard to come
by, with the possible exception of optimum yields published in soil surveys or perhaps yield
contests or the outputs of crop models. Thus, the main use of productivity indicators will be for
following trends. It may be difficult to detect trends, and innovative ways must be developed
to distinguish the effects of changes in the natural resource base from other effects. Complicating
factors include shifts in production decisions caused by price shifts and changes in government
programs (W.E. Foster, NCSU, personal communication). The easiest way to try to interpret the
data will be to look at the graph of simple yields or NPP over time alongside the graphs of
various inputs .such as fertilizer and land use. The more complex indicators will then-be
examined. These will be designed to be less sensitive to extraneous factors.
A secondary type of assessment will be to look for associations among indicators and for
associations between indicators and the forces that may drive them. Spatial maps of the crop
indicators can be used to overlay the maps of other agroecosystem indicators and other data. For
example, maps for yield could be used to overlay maps of soil quality or insolation. This
technique can serve the ARG's primary goal (i.e., using multiple indicators to get a picture of
agroecosystem condition in various regions) and it might also be used to generate hypotheses for
why certain indicators are responding as they are. Similarly, trends in each response indicator
5.1 - 17
-------
over time can be compared with trends in other response indicators and trends in stressor
indicators (including inputs). This may preserve some of the information that is lost when
indicators are adjusted for other factors.
Particular care is needed when interpreting normalized yield (Section 5.1.7.2). Because each
value is relative to a county average, most spatial (county-to-county) variability will have been
removed. The normalized yield will show whether a particular region is producing above, below,
or at about the same level as it did during the reference period. Because yields vary widely from
year to year, it will take some time to determine if such differences are true trends. The
calculation of normalized yield does nothing to account for changes in technology, climate.,,or
cropping pattern, so such factors will be reflected in the trends that are found. Despite these
concerns, this indicator will be a test of one method of combining data across crops. Historic
yield data allow this to be done in the Pilot, while standardization of other indicators may need
to wait until baselines have been established. ,
An output table (such as Table 5.1-4) will be generated for each indicator, as will a graph
of the cumulative distribution function (example not shown), and a map of indicator values. For
examples of output tables, cdf's, and other ways of presenting data, see Section 3 and Meyer et
al. 1990. Because NPP will include data from all crops within a summary region, cdfs will not
be meaningful. Instead, population estimates of the mean and variance for each region will be
reported.
5.1.9. Research Goals and Applications
5.1.9.1. Continuing Research on the Indicators of Crop Productivity
As mentioned above, extensive research is required before adjusted yield, output/input
indices, and NPP become useful indicators. Statistical and process models must be examined.
More yield-to-NPP conversion factors must be found for different crops (possibly for cultivajs
within crops), and the variability of those conversion factors must be determined. Improvements
5.1 - 18
-------
in breeding can alter the harvest index (i.e., Turner 1987) so these may need to be updated over
time. The inclusion of an estimate of root production is desirable, and the conversion factors
from Sharp et al. (1976) attempt to do this, but root production is likely to be underestimated
because of sampling error (Mitchell 1984) and because root exudates and "slough-off might not
be included (Coleman et al. 1976). Also, stresses on the crop (e.g., drought, air pollution) will
affect the root-shoot ratio. If energy is to be used as a common currency for inputs, then
conversion factors must be found. The energy equivalents of different production inputs will
change as industries try to be more energy-efficient; however, the factors reported by Southwell
and Rothwell (1977) should still be valid (Terry Rothwell, A+E Engineering, personal
communication). As new technologies are adopted, especially new pesticides, their energy
content must be determined. Trying to keep up with such changes may be prohibitively time-
consuming.
One limitation of the Pilot statistical design is the small sample size. Ways of using a larger
NASS sample for indicator calculations will be investigated. Of course when associations with
other indicators are tested, only the values from the Pilot sample can be used. It is not yet
known what the summary regions will be, other than the entire state of North Carolina. It is also
undecided what region should serve as the basis for means and standard deviations used in
normalizing the various indicators. Counties will serve this purpose for the normalized yield
indicator in the Pilot, but there may never be enough data to do county-by-county standardization
of other indicators. Therefore, larger regions should be tested for normalizing yield in the Pilot
Regions should be chosen so as to reduce the variability of the normalized indicator.
The applicability of measures of yield and crop productivity indicators for the future must
be addressed. When the Agroecosystem Program expands to include pastures, orchards and
livestock, and adjoining lands, what changes will be needed?
The suitability of the Survey Questionnaire will be evaluated after the pilot (i.e., should
seeding rate be asked). Can some questions be dropped? It is important that enough data be
gathered to allow calculation of a wide range of indices according to the developing interests of
the ARG and the interests of future researchers.
5.1 - 19
-------
Table 5.1-4. Example output table for an indicator of crop productivity
Average Normalized Yield in Regions of North Carolina ^
Coastal Plain Piedmont Mountains
mean 1st quartile median 3rd q'tile N mean 1st quartile median 3rd q'tile N mean 1st quartile median 3rd q'tile N
Soybeans
Com (grain)
Wheat (grain)
Tobacco (all)
Cotton
Peanuts
Corn (silage)
Oats (grain)
Sorghum (grain)
Barley (grain)
Sorghum (silage)
Rye (grain)
Hay (all)
Swcetpotatocs
Irish potatoes
Composite Index
For All Crops
& Regions shown are for illustration only. Actual summary regions have yet to be determined.
& It may not be possible to calculate all indicators for all crops. See Section 5.1.7.
Note: Pastures, idle land, and woody perennials will not be included in the 1992 pilot.
5.1 - 20
-------
5.1.9.2. Possible Alternative Measures of Productivity
Some of the above indicators may be unreliable because they depend on ancillary information
of unknown quality. It would be helpful to have a more straightforward indicator of productivity.
The use of remotely sensed data must be considered, along with some method of ground truthing,
possibly using micrometeorological techniques for calculating carbon flux (Tim Ball, Desert
Research Institute, personal communication). Again, this could be a common indicator across
terrestrial ecosystems. Note that weed productivity would be included in such a measurement,
but not in the currently proposed indicators. The following paragraphs summarize the report of
the workgroup which discussed remote sensing at the January 1992 Agroecosystems crop
productivity workshop.
Indicators computed from remotely sensed data could integrate plant productivity at a
landscape scale and would complement indicators of crop productivity currently planned at the
field level. Remotely sensed vegetation indices could provide a measure of plant productivity
at the full agroecosystem scale, including both crop and noncrop plants. Productivity of idle
land, Conservation Reserve Program (CRP) land, and adjacent noncrop areas, all considered part
of the agroecosystem, could be measured.
Indicators that can be derived from remotely sensed data include (1) vegetation indices such
as the Normalized Difference Vegetation Index (NDVT) or "greenness index", (2) actual
transpiration, (3) CQ2 flux and (4) leaf area index (LAI) (Wiegand et al. 1991, Box et al. 1989).
The group decided to focus on the possibility of data from the Advanced Very High Resolution
Radiometer (AVHRR) for the following reasons: (1) frequent collection (twice daily), (2)
historical record of over ten years, (3) inexpensive purchase cost and (4) known to address net
primary productivity (Box et al. 1989). The EMAP Integration and Assessment team is in the
process of obtaining 1990 and 1991 AVHRR data for all of the U.S.
The greenness index can be calculated from AVHRR data and is responsive to many kinds
of plant stress. In an associative, diagnostic study, the indicator could be used in combination
5.1-21
-------
with other data such as land use, ozone, drought, and disease epidemics, that arc readily available
at the appropriate scale. Ancillary data useful for associative studies are also available from
remote sensing. These include weather variables, soil temperature and moisture, water stress
index and solar radiation.
A vegetation index could be explored in a pilot project by obtaining the AVHRR pixels for
the NASS PSUs used in the 1992 Agroecosystem Pilot AVHRR pixels are about 1.1 km2. The
greenness index is best calculated with three to five pixels for each location (about 5x5 km
resolution). About 35 AVHRR pixels would be needed to cover a 6-8 mi2 PSU. The PSU, rather
than the segment, would probably be the most appropriate scale for this indicator, but this needs
to be explored further. The use of Thematic Mapper (TM) data, rather than AVHRR data, could
also be evaluated. After the PSUs were identified, the AVHRR (and/or TM) data for the past
ten years would be obtained and the greenness index (GI) for each PSU calculated for each year.
This value could be used as a baseline for evaluating the GI for 1992. The index would not be
normalized with respect to management (e.g. fertilizer inputs). These management factors would
be considered a part of the variability.
5.1 - 22
-------
5.2. Soil Quality: Physical and Chemical Components
5.2.7. Introduction
About 13,100 kinds of soils have been recognized in the U.S.; more than twice as many kinds
of map unit deliniations exist when slope, erosion, rocky, and stony phases are considered
(McCracken et al. 1985). Soils can be thought of as "archives" of the long-term interactions
among the major soil-forming factors of climate, parent material, plant material and topography,
and are organized and structured natural entities in their own right.
Soils function as sinks and sources of biogeochemical elements, as filters for pollutants, and
as an environment for growth and development of plants and other biological communities. They
are liable to change, gradually or abruptly and partly irreversibly, due to human use. Soil
structure is especially sensitive to human activities (Kay 1989). The main activities affecting
soils in agroecosystems are vehicular traffic, tillage, use of agricultural chemicals, waste disposal
and land use. In response to the perceived need to protect and conserve agricultural soils from
degradative processes, specific practices such as conservation tillage, residue management, crop
rotation, careful selection of crops for specific soils, and use of organic amendments are now
widely implemented on U.S. cropland. The long-term goal of soil quality monitoring and
assessment in agroecosystems is to provide a regional assessment of the cumulative soil response
to these conservation efforts.
The focus of soil quality assessment in agroecosystems will be on the presence, extent and
change in those soil properties that are (1) important to the functioning of the soil system, (2)
known to be affected by agricultural land management, and (3) can be adequately measured in
one sampling period at a regional scale. The physical and chemical indicators of soil quality to
be measured in the pilot are defined in Table 5.2-1. Biological indicators would also be valuable
indicators of soil response to management; research on nematode trophic groups as a biological
indicator of soil quality is discussed in Section 6.1.
5.2- 1
-------
Table 5.2-1. Description of physical and chemical soil quality indicators.
SUMMARY STATISTIC
Single measures
DESCRIPTION
Organic carbon
Clay content
Available water capacity
Porosity
SoilpH
Base saturation
Exchangeable acidity
(humid regions)
Exchangeable sodium percentage
(arid regions)
Electrical conductivity
Extractable aluminum
(humid regions)
Mercury
Bulk density (intact core)1'
Hydraulic conductivity (intact core)f
Quantity of organic carbon in first 20 cm of soil
(plow layer)
% clay in plow layer
Water retention between -33 and -1500 kPa
Water retension at -5 kPa and -10 kPa
Measure of soil acidity and nutrient availability
Extent to which the soil exchange capacity is
occupied by base cations
Extent to which the exchange capacity is
occupied by H and Al
Extent to which the exchange capacity is
occupied by Na
Measure of salt concentration in soil water
Quantity of aluminum in the plow layer .
Quantity of mercury in the plow layer
Mass of dry soil per unit volume
Rate at which soils transmits water while
saturated
- will not be included in initial pilot but will be included in subsequent pilots as sampling
protocol for intact cores is completed
5.2-2
-------
The main short-term objective in the assessment of soil quality is to determine the range and
frequency distribution (in proportion of land area) of individual indicators and to begin evaluation
of how well the chosen measurements and derived indices will reflect changing conditions.
Because standards of soil quality will vary with climate and soil, determination the rate of change
of soil quality will be an important long-term objective. A second long-term objective is to
combine indicator measurements into quantitative indices so that general statements about soil
quality on a regional basis can be made. Several possible indices include structure, tilth, fertility,
contamination, and productivity (Table 5.2-2). Thirdly, soil quality information will be combined
with other pilot data into a picture of over-all agroecosystem health. A fourth long-term
objective is to integrate information on the health of agricultural soils in the U.S. with
information on soils in forests and arid lands to provide an overall picture of soil quality across
terrestrial ecosystems.
5,2 J. Data Sources
Data for soil quality assessment in the pilot will come from soil samples taken by National
Agricultural Statistics Service (NASS) enumerators. Data will also be obained from the Soil
Conservation Service (SCS) State Soil Survey Database (SSSD) and Natural Resources Inventory
(NRI).
The State Soil Survey Database (SSSD) databases are currently being compiled at a state
level by linking the information from the Soil Interpretations Record Data Base (commonly
known as the Soils-5 database), with the specific map unit identified in the county-level soil
surveys (compiled in the Soils-6 database). The SSSD is, therefore, a more refined and accurate
source of information about a specific soil than the Soils-5 database because the information is
linked to a specific geographic location (SCS National Soil Survey Lab, Lincoln, NE, personal
communication, 1991).
5.2-3
-------
Table 5.2-2. Research indices of soil quality.
Assessment or index
Measurements
Conlamination/roxins
Anthropogenic
Nonanthropogenic
acidification
salinization
alkalinization
Soil structure
(tilth, porosity)
Soil fertility
Leaching Potential/
Adsorption Potential/
Run-off Potential
(SCS ratings)
Sensitivity to
degradation from
intensive agriculture
Erosion
Productivity
Lead/cadmium/mercury
and other trace metal contaminants.
PH
Exchangeable sodium percentage
Base saturation
Exchangeable acidity
Extractable Al
Electrical conductivity
Organic carbon
Bulk density
Available water capacity
Porosity
Organic carbon
% clay
Base saturation
Extractable P
Organic matter
pH
Exchangeable acidity
Slope ,
Infiltration (Hydrologic group)
Horizon depth
Organic matter .
K factor
Texture
Drainage
Erosion rate
Erodibility index (R*K*L*S/T)
Soil depth
Rooting depth
Depth to water table
Restrictive soil layers
Landscape position (hillslope)
taxonomic order or suborder
'Highly Erodible Land' rating-water
'Highly Erodible Land' rating-wind
Erosion rate (USLE)
Erosion index (USLE/T)
Soil properties such as bulk density, OM and pH
5.2 - 4
-------
Selected data elements from the SSSD to be used in the pilot arc listed in Table 5.2-3. Useful
data elements from this database include grouping variables, such as taxonomic classification,
Major Land Resource Area, and soil depth, to be used in statistical analyses. The database also
contains data on soil physical properties, such as bulk density, which will not be measured
directly in the first pilot until sampling protocols for intact soil cores are completed. The SSSD
data will be linked to Agroecosystem data by identifying the map unit of the sample point on
NASS aerial photos and the appropriate SCS county soil map, and requesting the selected data
elements for each mapping unit from the state SGS office. Possible collaboration, in the office
and/or in the field, with SCS personnel is being explored to help identify the soil mapping unit.
SSSD data can also be used to determine expected ranges in each state of many soil
properties, including pH, bulk density, available water capacity, organic matter, permeability and
clay. The Agroecosystem Resource Group (ARG) is in the process of acquiring data from the
North Carolina Soil Conservation Service Office for this purpose.
5.2.3. Indicators
The initial set of physical and chemical indicators of soil quality to be measured in the Pilot
are described in some detail below. These indicators were chosen for evaluation because they
are known to be important to the functioning of the soil system, are affected by anthropogenic
stresses, and are likely to be measureable in a single sampling period on a regional basis. Many
are key variables in soil productivity models. Methodologies for sample collection and laboratory
analyses are described in Section 5.2.4.
Approaches to data analysis and application are discussed in Section 5.2.7 and 5.2.8.
Generally, ranges and within- and among-site variance will be determined for each measurement
to help refine sampling design and to determine what magnitude of change could likely be
measured over time at a regional scale. Secondly, the data will be used to begin an evaluation
of how well the indicators, and derived indices, would truly reflect good, poor, or changing
conditions. Although identified ranges for indicators and benchmark references of soil quality
5.2-5
-------
Table 5.2-3. Requested data elements from the SCS State Soil Survey Database
Data element
Definition
MLRA
survey area ID
map unit ID
map unit symbol
map unit name
class code
soil layer
soil layer
soil layer
available water capacity
available water capacity
bulk density
bulk density
cation exchange capacity
cation exchange capacity
clay
clay
organic matter
organic matter
permeability
permeability
pH
pH
K factor
T factor
SCSLCC
SCS LCC
slope
slope
hydrologic group
drainage
prime farmland
depth
code for Major Land Resource Area
code for state+FIPS (state soil survey area)
stssaid+musym: uniquely identifies a mapunit within a state
map unit symbol
map unit name
code for taxonomic classification of the soil
identifies the original layers on the Soils-5 record
depth to the lower boundary of the soil layer (inches)
depth to the upper boundary of the soil layer (inches)
maximum value for the range of awe (inches/in)
minimum value for the range in awe (inches/in)
maximum value for the range in moist bulk density (g/cm3)
.minimum value for the range in moist bulk density (g/cm3)
maximum value for the range in CEC
minimum value for the range in CEC
maximum value for the clay content (% in less than 2 mm fraction)
minimum value for the clay content (% in less than 2 mm fraction)
maximum value for the range in OM (% by weight)
minimum value for the range in OM (% by weight)
maximum value for the range in permeability (inches/hour)
minimum value for the range in permeability (inches/hour)
maximum value for the range in pH
minimum value for the range in pH
credibility factor; can be used in USLE (tons/acre)
soil loss tolerance factor, can be used to interpret USLE (tons/acre)
SCS Land Capability Class rating (nonirrigated)
SCS Land Capability Class- subclass rating
maximum value for the range of slope within a mapunit (%)
minimum value for the range of slope within a mapunit (%)
the SCS hydrologic group
code identifying the natural drainage condition/frq-i-duration when
saturation-free
SCS prime farmland classification
depth to water table
depth to bedrock
5.2-6
-------
are generally lacking, soil quality standards and ratings are of great interest to SCS and other soil
scientists, who are making good progress with respect to specific soil uses or functions (see Table
5.2-17) and it is likely that this will continue to be an active area of research that can be applied
in the EMAP program. Evaluation of soil quality indicators is made even more complex by the
fact that what is a good or poor indicator range or value will vary with climate, soil and
management scenarios. General, baseline reference points with which to group soils and
sampling sites for indicator evaluation are needed, and different approaches to this will be
explored in this and subsequent pilots (see Section 5.2.7).
5.2.3.1. Organic carbon
The organic matter content of surface soils range from 0.1% in mineral soils to nearly 100%
in organic soils (Schnitzer 1982). Organic matter is considered important for the long-term
physical, chemical and biological functioning of soils; it stabilizes soil structure, increases the
cation exchange capacity and water-holding capacity of sandy soils, and supplies nutrients for
plants and microorganisms.
Carbon is the main element present in soil organic matter, comprising from 48 to 58% of
total weight (Nelson and Sommers 1982). Organic carbon (C) will be used initially as a measure
of soil organic matter because 1) soil organic matter is difficult to estimate quantitatively (Nelson
and Sommers 1982) and 2) different organic matter fractions considered important to nutrient
cycling, structure and biological activity in soils require different extraction and analysis
procedures (Schnitzer 1982, Stevenson and Elliot 1985).
A large amount of data exists on changes in organic C when forests and grasslands were
converted into agricultural land. Mann (1986) confirmed several previous reports that the greatest
rates of change occurred in most soils in the first 20 years after conversion. Soils very low in
C tended to gain small amounts after cultivation; soils high in C lost at least 20% in the top 30
cm during cultivation. After the initial rapid loss of C after land conversion, rate of C loss in
cultivated soils tends to slow and approach a new equilibrium (Mann 1986). Loss of organic
5.2-7
-------
matter is increased by tillage and affected by management practices such as choice of crops,
stubble mulching, fallowing and use of organic amendments. Organic C is lost due to soil
erosion, often accompanied by a loss in nutrients, deterioration of soil structure and diminished
soil workability (Pierce et al. 1991, Frye et al. 1982). Depletion of soil organic matter and
erosion are spirally cyclic because a decrease in organic matter increases the susceptibility of a
soil to erosion (Pierce et al. 1991). Changes in land management, such as the increasing
implementation of no-till practices, may affect rates of organic C loss (Coleman et al. 1990).
Burke et al. (1989) developed predictive models of organic C loss in U.S. grasslands using
climate, soil texture, landscape position and management practices as driving variables. These
models help to identify areas which are most vulnerable to organic C loss. An "ideal" or
"healthy" standard of organic C in soils does not exist because it depends on soil-forming
processes of each soil. The goal of the Agroeeosystem Program is to provide a broad-scale,
long-term picture of organic C in agricultural soils. A decline in organic C would be interpreted
as a warning of decline in soil quality.
5.2.3.2 Clay content
Clay content is the weight percentage of the particle size class smaller than 0.002 mm
diameter that is present in the < 2 mm soil fraction. Clay may have thousands of times more
surface area per gram than silt or sand and is, therefore, the most chemically and physically
active part of the mineral soil (USDA, SCS 1983).
Under conditions of accelerated erosion, the subsurface soil layers are increasingly
incorporated into the plow layer (Indorante et al. 1991, Frye et al. 1982, Stone et al. 1985, Pierce
et al. 1991). This is due to selective removal of fine particles during the erosion process, and
to mixing of subsoil into the surface layer. The implications of changing the surface soil texture
on crop productivity can be significant. The kind and amount of clay affects available water
capacity, permeability, credibility and workability (Frye et al. 1982, Lai 1987, Pierce et al.
1991).
5.2 - 8
-------
As in the case of organic C, an "ideal" or "healthy" standard of clay content in soil does not
exist. The indicator is intended to provide a broad-scale, long-term picture of clay content in the
top 20 cm of agricultural soils. An increase in clay content would be interpreted as an indicator
of soil loss and a warning of decline in soil quality.
5.2.3.3. Available water capacity (AWC)
Available water capacity is the capacity of a soil to hold water available to plants; it is
usually expressed in inches of water per inch of soil depth. AWC is commonly measured with
a pressure plate apparatus as the amount of water held by the soil at tensions between field
capacity and wilting point (-33 and -1500 kPa); and is mainly determined by the pore size
distribution of the soil.
Large quantities of water are needed to supply the evapotranspiration requirements of growing
plants. Except in the areas of abundant and timely rainfall, most of it must come from the soil.
Thus, the amount of water a soil can hold available for plant use is an important property
(USDA, SCS 1983). Available water capacity is one of the soil properties most affected by
erosion and management practices (Pierce et al. 1991, USDA.SCS 1981, Frye et al. 1982, Larson
et al. 1985). One reason for this is because the silt fraction is the major factor that governs pore
size distribution in a soil, which in turn affects AWC (USDA, SCS 1983). Because the silt
component is very susceptible to erosion, accelerated erosion leads to reduced AWC.
Classes of AWC are not standardized throughout the country because of the different effects
of AWC on plant production in different moisture regimes. In areas of the country where
moisture is seldom in short supply, AWC has a minimum effect on plant production so the
classes are based on a relatively thin root zone. In dryer areas, however, production of plants
is highly dependent on AWC and classes are based on deeper depth (USDA, SCS 1983) (Table
5.2-3). The AWC classes listed in Table 5.2-4 will be used initially to interpret AWC values and
rate soils based on AWC. A decrease in AWC based on these classes would be considered a
warning of decline in soil quality.
5.2 - 9
-------
Table 5.2-4. Ratings of available water capacity (AWC) by moisture regime -
AWC
class
very low
low
moderate
high
very
high
Aquic,
Perudk
(wet)
in/40 in soil
<2
2-3
3-4
>4
Udic, Ustic
(moderately
wet)
in/60 in soil
<3
3-6
6-9
9-12
>„
Aridic,
Xeric(dry)
in/60 in soil
<2.5
2.5-5
5-7.5
7.5-10
»0
" From: USDA, Soil Conservation Service 1983
Aquic soils are water-saturated long enough for reducing conditions to exist; in perudic regimes,
precipitation exceeds evapotranspiration for every month of the year, in Udic regimes, sofls are
not dry as long as 90 cumulative days per year; in Ustic regimes soils are dry for more than 90
consecutive days per year; soils in Aridic regimes are never dry for more than 90 consecutive
days and are dry for more than one-half the time when not frozen; in xeric moisture regimes soils
are dry >45 consecutive days in the summer and wet >45 consecutive days in the winter (Buol
et al. 1980).
In the Pilot, AWC will be measured in the top or surface 20 cm of soil. The range in AWC
values for the lower horizons of the soil type will be obtained from the State Soil Survey
Database. Later, if direct measurements of AWC in lower horizons seem important to obtain,
we can explore the possibility of taking deeper soil cores.
5.2.3.4. SoilpH
Soil pH is an indicator of possible chemical constraints to the growth of roots and other
biological communities. Chemical constraints usually associated with pH include the presence
of inhibitory compounds (e.g., Al, salts), or a nutrient deficiency (e.g., P fixation), (Pierce et al.
1991). As soil weathering and leaching processes progress, base cations are removed from soil
5.2 - 10
-------
and the pH declines. The amount of rainfall, rate of percolation, and evaporation leave a definite
impression on pH and on the morphology of the soil profile. The pH is higher in soils of arid
regions than in humid regions, higher in younger soils than older soils, lower on flat topography
than on steep slopes. Agriculture accelerates the process of soil acidification on many soils when
soil liming is not practiced. For this reason, agricultural soils are often amended with liming
compounds such as calcium carbonate.
Classes of soil pH used by the SCS are listed in Table 5.2-5. These ratings could be used
to give a more qualitative interpretation of pH values. An increase in land in highly acid or
highly alkaline classes would be interpreted as a warning of decline in soil quality. The use of
liming amendments will be monitored as an indicator of the need to neutralize acidification. As
the Program is implemented in western states, alkalinization processes would become more
important
Table 5.2-5. Ratings of soil pH U
Class
ultra acid
extremely acid
very strongly acid
strongly acid
moderately acid
slightly acid
neutral
mildly alkaline
moderately
alkaline
strongly alkaline
very strongly
alkaline
pH value
<3.5
3.5-4.4
4.5-5.0
5.1-5.5
5.6-6.0
6.1-6.5
6.6-7.3
7.4-7.8
7.9-8.4
8.5-9.0
>9.0
- From: USDA, Soil Conservation Service 1983
5.2- 11
-------
5.2.3.5. Base saturation
Base saturation is a measure of the proportion of base cations on the cation exchange sites
of a soil. It is the most common measure of soil fertility with regard to available nutrients for
plants and microorganisms, and would be an important parameter in a soil fertility index.
The data would be used to provide a broad-scale, long-term picture of base saturation in the
top 20 cm of agricultural soils. An increase in the proportion of land area with a decrease in
base saturation would be interpreted as an indicator of decline in soil quality (Ewel et al. 1991).
The use of liming amendments would be monitored as an indicator of the need to increase base
saturation.
5.2.3.6. Exchangeable acidity
Exchangeable acidity is a measure of the proportion of hydrogen and aluminum on the cation
exchange sites of a soil. Exchangeable acidity is an indicator of possible chemical constraints
to the growth of roots and other biological communities, including the presence of inhibitory
compounds (e.g., aluminum, manganese) or a nutrient deficiency (e.g., P fixation) (Pierce et al.
1991). Interpreted together with pH, exchangeable acidity is a good measure of sioil acidity. An
increase in acid saturation or a soil acidity index would be considered an indicator of decline in
soil quality (Ewel et al. 1991). The use of liming amendments would be monitored as an
indicator of the need to neutralize acidification.
5.2.3.7. Exchangeable sodium percentage (arid soils).
The presence of large quantities of sodium in fine-textured soils is undesirable because of its
degradative effect on soil structure (Tisdale and Nelson 1975). Irrigation is known to increase
sodium content of soils because evaporation of saline water deposits salts in fields. A measure
of sodium in soils is likely to become most important as the Program is implemented in western
states with arid soils.
5.2 - 12
-------
Plant growth in alkaline soil, especially for irrigated agriculture, depends critically on the
exchangeable sodium percentage (ESP) of the soil. ESP is the proportion of sodium of the total
exchangeable cations in the soil. Soil with greater than about 15% ESP deflocculates readily and
is difficult to make or keep permeable. Soil with ESP in the range of 7.5-15% needs careful
management, especially under irrigation. Where the ESP is less than 7.5% the soil is not
appreciably affected by sodium. Therefore, the critical ratings in ESP are 7.5 and 15% (Webster
and Oliver 1990, Russell 1973) (Table 5.2-6). An increase in the proportion of soils in the
>7.5% class would be interpreted as an indicator of decline in soil quality. Other salinity
measures, such as the sodium absorption ratio (SAR) will be evaluated in future pilots.
Table 5.2-6. General ratings for exchangeable sodium percentage -
Rating
Exchangeable sodium
percentage
Not affected
Affected
Seriously affected
<7.5
7.5-15
U From: Russell 1973, Webster and Oliver 1990
5.2.3.8. Electrical conductivity (salinity)
Salinity is the concentration of dissolved salts in water. High concentrations of neutral salts
such as sodium chloride and sodium sulfate may interfere with the absorption of water by plants
through the development of a higher osmotic pressure in the soil solution than in the plant cells.
Salts may also interfere with the exchange capacity of nutrient ions, thereby resulting in nutrient
deficiencies in plants (USDA, SCS 1983).
5.2 - 13
-------
The electrical conductivity of a saturated extract is the standard measure of salinity. The
standard international unit of measure of electrical conductivity is decisiemins per meter (dS/m)
corrected to a temperature of 2.5 C. A value > 4 dS/m is considered a saline soil (Table 5.2-7).
Therefore, an increase in the proportion of soils in a region with > 4 dS/m would be interpreted
as a warning of decline in soil quality.
Table 5.2-7. Salinity ratings based on electrical conductivity -
Classes
Electrical
conductivity
(dS/m)
not saline <2
very slightly saline 2-4
slightly saline 4-8
moderately saline 8-16
highly saline >16
i' From: USDA Soil Conservation Service 1983
1 dS/nn = 1 mmhos/cm
5.2.3.9. Extractable aluminum (humid soils)
Extractable aluminum is a measure of trivalent aluminum ions (Al3*) on the exchange sites
of a soil. Aluminum is the main source of exchangeable acidity in soils and is responsible for
the detrimental biological effects of soil acidification (Veitoh 1902). The Al3* ion is the main
species present at soil pH values of <5.0 and is the species most toxic to plants and,soil
microorganisms. Microbial processes known to be affected by exchangeable Al include
symbiotic and nonsymbiotic nitrogen fixation (Cooper et al. 1985, Rosswall et al. 1985,
Alexander 1985, Katznelson 1940), decomposition (Mutatkar and Pritchett 1966), and growth of
soil fungi (Ko and Hora 1972). The concentrations of aluminum known to be toxic
5.2 - 14
-------
or nontoxic to plants and soil microorganisms are available in the literature and could be
compiled into a rating scale for interpretation of Al values.
5.2.3.10. Trace metals
Municipal sludge and industrial or urban waste water are commonly applied to agricultural
soils as an organic amendment and a waste control strategy (Korentajer 1991). Atmospheric
deposition also contributes to the presence of contaminants in soil. Nearly all the earth's surfaces
have received atmospheric deposits of lead released from burning fossD fuels (Brams 1977, Page
and Ganje 1970). Soils in some areas have received lead, cadmium and/or copper pesticide
sprays containing these metals.
Although an active microflora will degrade most potentially harmful contaminants, the safety
and desirability of waste application has been controversial because of the potential of gradual
contamination of soils with toxic and persistent contaminants, such as trace metals, and the
potential ecological effects. Municipal sludge (sewage sludge) applied to soils nearly always
contain lead, cadmium, chromium, copper, nickel and zinc (Baker et al. 1979, Baker and Chesnin
1975).
Soil contamination with trace metals poses a direct risk of toxicity to plants, soil organisms
and microbial functioning (Brookes and McGrath 1985). Vesicular-arbuscular mycorrhizal fungi,
important plant symbionts in agrpecosystems, are usually sensitive to high levels of trace metals
(Tyler et al. 1989). Some contaminants are taken up by plant material and pose a risk of
accumulation in grazing livestock and in humans.
In the 1992 pilot, mercury (Hg) levels in agricultural soils will be measured as an initial
indicator of trace metal concentrations. Mercury contamination of the environment is a serious
problem. Terrestrial ecosystems receive continuous fallout of Hg estimated at 100,000 tons
annually (Lindsay 1979), from fossil fuels, Hg-consuming industry and natural evaporative losses
from soils and rocks.
5.2 - 15
-------
The natural mercury content of soils is depends on the nature of the parent material, pH,
drainage and organic matter content (Stewart and Bettany 1982)-. Mercury occurs as a mineral
at shallow depths. It has a high vapor pressure, is very volatile, and has the ability to form many
organic and inorganic compounds and complexes (Lindsay 1979). Background levels are
generally less than 100 pg/g soil (Stewart and Bettany 1982). In soils that have developed on
shale or sedimentary deposits, the Hg content can range from 1 to 50 pg/g soil (Warren et al
1966) to 250 pg/g (Jonasson and Boyle 1971). Atmospheric fallout, application of municipal
sludge, and seed fungicide treatments can result in elevated levels of Hg in the surface horizon
of agricultural soils several orders of magnitude greater than background levels.
Mercury levels in the surface 20 cm of soil will be measured. The number of fields which
have received applications of municipal sludge will also be determined. It is likely that there is
enough available data in the scientific literature to determine potentially toxic levels of Hg for
plants, animals, and soil organisms that could be compiled and used for the interpretation of Hg
data. Additional trace metals may be included in further pilots.
5.2.3.11. Bulk density
Bulk density is an indicator of how well plant roots are able to extend into the soil (USDA,
SCS 1983). Bulk density is expressed as soil weight per volume dry soil and generally ranges
from about 1.0 to about 2.0 g/cm3 in agricultural soils; Because bulk density is defined as the
volume,of both solids and pores, soils that are loose and porous will have low weights per
volume (bulk density) and those that are compact will have higher bulk densities (Brady 1974).
Soils that contain organic matter and have good aggregation have low bulk densities.
Bulk density is used as a parameter most closely related to mechanical impedance of root
growth in models that relate soil properties to soil productivity (Kiniry et al. 1983, Pierce et al.
1983). Crop rotation and soil management of a given soil affects the bulk density, especially of
the surface layers. Accelerated erosion and intensive cultivation increases bulk density; adding
5.2 - 16
-------
crop residues, manure or planting cover crops tend to lower it (Frye et al. 1982, Brady 1974,
Groenevelt et al. 1984).
Nonlimiting, critical and root-limiting bulk densities are generally known, and vary with the
texture class of the soil (USDA, SCS 1975, Pierce et al. 1983). Bulk density as an indicator
would be used to provide a broad-scale, long-term picture of bulk, density in the top 20 cm of
agricultural soils, and perhaps in lower soil horizons as the Program develops. An increase in
the proportion of soils reaching critical bulk density values within their texture class would be
interpreted as an indication of decline in soil quality. Bulk density would be an important
component in a soil structure index.
Bulk density measurements are most accurate when taken in intact cores. However, because
the procedure to obtain intact cores in the context of a large survey has not been developed, this
indicator will not be measured directly in the 1992 Pilot Bulk density data from the SSSD will
be used in the 1992 Pilot for initial exploration of this indicator.
5.2.3.12. Soil porosity
The pore space of a soil is that portion occupied by air and water. Continuous cropping,
particularly of soils originally high in organic matter, often results in a reduction of pore space.
The reduction is usually associated with a decrease in organic matter content and a consequent
lowering of granulation and soil structure (Brady 1974).
Both macro- and micropore spaces occur in soils. Although there is no sharp line of
demarcation, macropores characteristically are those which allow the ready movement of air and
percolating water. Air movement is generally impeded in micropores and water movement is
restricted to capillary movement Thus, in a sandy soil, in spite of the low total porosity, the
movement of air and water is rapid because of the dominance of the macrospaces. In
fine-textured soils, dominated by micropores, the total pore space is large but the micropores are
usually filled with water. Aeration can frequently be inadequate, especially in the subsoil (the
5.2-17
-------
soil below the plow layer), for satisfactory root development and desirable microbial activity.
Therefore, the size of the individual pore spaces rather than their combined volume is the
important consideration (Brady 1974). Soil porosity is the main response variable used in a
conceptual model of changes in soil structure under different cropping systems (Gibbs and Reid
1988).
Two approaches will be taken to measure and interpret soil porosity. The first is similar to
the work of Thomasson (1978), who used the relative proportion of macropores (>60um) and
mesopores (0.2pm to 60pm) to define four classes of soil structure. Pores greater than about 60
pm (termed air capacity) could be measured with a pressure plate at -5 kPa. Mesopores (termed
available water) can be measured as the volume of water held between about -33 and 1500 kPa
(same measurement as available water capacity). The best soil class has a macroporosity >= 15%
and a mesoporosity of 20-35%. The worst class has a macroporosity >5% and a mesoporosity
of <35% (Kay 1989). A second approach will measure the percent of soil volume occupied by
pores of approximately 30 and 60pm (measured at -10 and -5 kPa with a pressure plate,
respectively), which are important for water drainage and the survival, growth and movement of
soil microflora and fauna (Duniway 1979). An increase in the proportion of soils in the lower
classes of Thomasson's rating scale, and/or at the lower levels of porosity critical to microbial
functioning, would be interpreted as an indication of decline in soil quality.
Porosity can be measured with a pycnometer on intact soil cores or with a pressure plate
apparatus on nonintact cores. Because the procedure to obtain intact cores in the context of a
large survey has not been developed, pressure plate measurements will be used initially.
5.2.3.13. Hydraulic conductivity (permeability)
Permeability or hydraulic conductivity is the quality of the soil that enables water or air to
move through it and is determined by pore geometry. Hydraulic conductivity is especially
important in drainage, water erosion and leaching potential of a soil. It is a main variable used
5.2 - 18
-------
in the algorithm developed by Goss and Wauchope (1990) to calculate soil leaching potential of
pesticides.
Hydraulic conductivity is a measure of the rate at which soil transmits water while saturated.
Classes of hydraulic conductivity used by the SCS are listed in Table 5.2-8. These ratings could
be used to give more of an qualitative interpretation to hydraulic conductivity values. An
increase in the proportion of soils in the lower classes of the SCS rating scale would be an
indication of decline in soil quality.
Hydraulic conductivity measurements are most accurate when taken on intact cores. The
x
procedure to obtain intact cores in the context of a large survey has not been developed.
Therefore, this indicator will not be directly measured in the 1992 Pilot. Hydraulic conductivity
data from the SSSD will be used for initial exploration of this indicator.
Table 5.2-8. Ratings of hydraulic conductivity recognized by the SCS.-
Hydraulic
class
very low
low
mod low
moderate
high
very high
*f USDA,
pm/s
<0.01
0.01-0.1
0.1-1
1-10
10-100
>100
SCS 1983
5.2 - 19
-------
5.2.3.14. Erosion (water)
The Universal Soil Loss Equation (USLE) will be used as an estimate of erosion due to
water. The USLE is a model of soil erosion developed in the 1950's from many years of field
experimentation throughout the U.S. (Wischmeier and Smith 1978). The equation is designed
to predict long-term losses of soil through sheet and rill erosion from specific land areas under
specified cropping and management and is widely used by the SCS and conservation planners
to determine appropriate soil management strategies. Although termed the soil loss equation, the
USLE is actually an estimate of soil movement or displacement within a field, rather than an
estimate of actual soil loss from the field.
The equation,
A=R*K*LS*C*P
groups six major factors whose site-specific values can be expressed numerically. The equation
parameters are defined by Wischmeier and Smith (1978):
A Soil loss (displacement or movement within the field)
R The rainfall and runoff factor, is the number of rainfall erosion index units plus a
factor for runoff from snowmelt or applied water where such runoff is significant
K The soil credibility factor, is the soil loss rate per erosion index unit for a specified
soil as measured on a unit plot, which is defined as a 72.6 ft length of uniform
9-percent slope continuously in clean-tilled fallow. The soil properties that influence
soil credibility are infiltration rate, permeability, total water capacity, and those
properties that resist dispersion, splashing, abrasion and transportation forces of rainfall
and runoff. The Soil Conservation Service has estimated K for most agricultural soils.
LS The slope-length factor, is the ratio of soil loss from the field slope length to that from
a 72.6 ft. length under identical conditions; the slope-steepness factor, is the ratio of
soil loss from the field slope gradient to that from a 9-percent slope under otherwise
identical condition.
5.2 - 20
-------
C The cover and management factor, is the ratio of soil loss from an area with specified
cover and management to that from an identical area in tilled continuous fallow.
P The support practice factor, is the ratio of soil loss with a support practice like
contouring, stripcropping, or terracing to that with straight-row farming up and down
the slope.
Data for the six USLE factors are obtained from field measurements (LS factor), from grower
interviews (C and P factors), and from the State Soil Survey Database (K and R factors) (Table
5.2-9).
Table 5.2-9. Sources of data for the six Universal Soil Loss Equation (USLE) factors.
USLE
parameter
Data source
R Published maps (U.S. or state, e.g. Wischmeier and Smith, 1978 or
USDA/SCS 1990), State Soil Survey Database or STATSGO
K State Soil Survey Database, State SCS USLE Technical Guides (e.g. -
USDA/SCS 1990) or STATSGO
L Field - procedure needs development
S Clinometer measurement - needs development
C Calculated by EMAP staff from technical guides (e.g. Wischmeier and
Smith, 1978 or USDA/SCS 1990) and site-specific data on crop type and
tillage practices (data questionnaire) - estimate may be rough - procedure
needs development
P Calculated by EMAP staff from technical guides (e.g. Wischmeier and
Smith, 1978 or USDA/SCS 1990) and site-specific data on crop type,
slope, and tillage practices (data questionnaire) - estimate may be rough -
procedure needs development
5.2 - 21
-------
The soil erosion tolerance factor (T) is also available from the SSSD and is used in the
interpretation of the USLE values (USDA 1989). The T factor is defined as the maximum rate
of annual soil erosion that will permit crop productivity to be sustained economically and
indefinitely (USDA 1975). There are five classes of T factors, ranging from two tons per hectare
per year for shallow or otherwise fragile soils to eleven tons per hectare per year for deep soils
that are least sensitive to damage by erosion (see Table 5.2-17).
The goal of soil erosion estimates as an indicator is to provide a regional, long-term picture
of soil erosion due to water. Spatial and temporal patterns in soil erosion could be evaluated
with respect to other Agroecosystem indicators (land use, crop productivity, agrichemical use and
soil chemical, physical and biological measurements). For the initial exploration of this indicator,
the data from the SCS National Resource Inventory (NRI) on soil erosion will be used (USDA
1989). The SCS has national soil erosion data from 1982 and 1987 that allow analysis at a
substate (multi-county) level. The NRI for 1992 is in progress. If it is determined that NRI data
are not adequate for the desired assessments, procedures for measuring USLE factors from points
within sampling fields would be developed. First, the procedure for extrapolation of the
point-based S and L measurements to a field basis will be addressed, and protocols for NASS
enumerators or other field staff developed. Then methods and algorithms for automating the
computation of the C and P factors from data in the NASS survey questionnaire would be
derived.
5.2.4. Logistics
Each NASS enumerator will sample approximately 10-15 segments and receive a kit
containing the items listed in Table 5.2-10 at the NASS training session. Within the enumerator
kit will be a soil sampler/probe set. In the probe set, three tips will be available for the core tube
for sampling soil under a range of conditions. The regular (2 notches), mini (1 notch), and super
(4 notches) duty tips are for sampling moist, dry, and stony soils, respectively. Extra parts will
be available at 1509 Varsity Drive, Raleigh, NC 27606 (Agroecosystem Program headquarters)
and can be shipped by overnight express delivery upon demand. Several phone numbers, where
5.2 - 22
-------
Table 5.2-10. Contents of the enumerator kit
Item
Purpose
Manual describing sampling methods
Indication of whether that field will be
sampled in duplicate or not
3-foot hinged ruler
stakes (20 red, 10 yellow)
36-inch Oakfield probe set
[contains 12" handle, 212" extension rods, 12" tube
which extracts a sample 8" long x 13/16" diameter,
3/4" tips for moist, dry, and stony soil, a footstep
for dry or compacted soils, and a fiberboard case]
Extra 12" core tube
14-qt. plastic bucket with handle
2 500-ml plastic beakers
4x2x 12" plastic bags
3 qt. plastic bags
Pre-labeled paper-wire tags
-write sample number on the label twice, one above
the other
2 polystyrene ice chests
Pre-printed mailing labels
Postage-paid container for mailing
Postage-paid, insulated container for mailing
Roll of strapping tape with cutting edge
Postcards
internal check of laboratory variability
measure 45° angles for transect
marking transect and location of soil cores along
the transect
collect 20 20-cm deep cores
in case the core in the kit becomes twisted
collect and homogenize soil cores
measure volume of soil for .nematode and
chemical/physical analysis; 2 are included in the
kit to allow 1 extra
store 500 ml soil sample for nematode
enumeration at the moisture content of the field
mail samples back to preparation lab.
labeling nematode samples appropriately for
enumeration laboratory
lightweight, insulated container for storage of
samples in an environment to prevent lethal
temperatures
tracking samples from field to analysis
laboratories
transporting samples for chemical/physical
analysis directly to analysis laboratory
transporting samples for enumeration of nematodes
packaging samples for mail
tracking of samples that are mailed
5.2 - 23
-------
someone could be reached at all times, will be included in the enumerator's manual for use in
the event of equipment loss or breakage.
Sample collection. For each field, the enumerator will be given the following information
printed on their survey form: the sample number(s), whether or not a second composite sample
must be collected in that field, and the number of paces along and into the field to determine the
midpoint of the sampling transect. Two labels will be provided for composite samples that will
be divided into duplicate samples at the preparation laboratory. All labels will be printed in
cooperation with the North Carolina Agricultural Statistics Division in Raleigh (an office of
NASS). The sampling design was constructed to include measures of within-field variability (a
second composite sample collected for every sixth field sampled) and within-sample or laboratory
variability (duplicates are taken from the second composite sample from every twelfth field) (see
Appendix 7). The enumerator will collect soil cores according to the sampling design described
in Section 3.3.2. Example instructions for the NASS enumerators are listed in. Appendix 6.
Twenty cores (2-cm diameter) of soil are necessary to provide enough soil (1256 cm3) for
the required analyses. Total soil volume of each composite sample must exceed 500 plus 550
cm3, the respective volumes required for chemical/physical analysis and nematode enumeration
(Section 5.6.4). The volume designated for nematode enumeration contains 50 cm3 for
calculation of the volume:weight ratio described in Section 6.1.4. When a field is selected for
a duplicate sample (Appendix 7), 40 cores per transect will be required to collect enough soil for
all laboratory determinations.
Within each field, one core will be taken at each of 20 locations, except for duplicate samples
where two cores will be taken at each of 20 locations, equally spaced along a 100 yard diagonal
transect (Section 3). For each core, the soil tube will be pushed straight down into the soil,
without twisting, to the depth that fills the entire length of the tube (20 cm). "The tube will be
pulled up and the soil core placed into a plastic bucket If the core is unsatisfactory, another core
will be taken in the same location within 15 cm. When all 20 cores have been deposited into
the bucket the enumerators will be instructed to mix the soil thoroughly by hand, breaking up
5.2 - 24
-------
soil clumps gently. Any roeks larger than 2 cm in diameter will be discarded, but all surface
organic matter should be kept as part of the soil sample. When appropriate, soil for nematode
enumeration (Section 6.1) will first be removed. Pre-labeled mailing containers will then be
filled with soil for the chemical and physical analyses, and stored in an insulated container (ice
chest). Samples will be mailed the same day they are collected or first thing the next day
through Federal Express (1-800-238-5355 for pick-up). Sample(s) will be mailed to the
preparation laboratory (Attn: Charles Harper, Box 7616, North Carolina State University, Raleigh,
NC 27695) in the pre-addressed, postage-paid container. Postage will be paid using a Federal
Government account through the Air Resources Research Consortium at North Carolina State
University.
Laboratory analyses-physical and chemical. ARG personnel will air-dry, homogenize, and
grind the samples in the preparation laboratory according to specifications listed in Appendix 4.
Then all samples will be mailed to the analysis laboratory in batches of approximately 40
samples. Within each box will be a list of the enclosed samples.
The analysis laboratory will analyze the soil samples for the specified chemical and physical
parameters using the prescribed procedures (Table 5.2-11). Detailed laboratory procedures are
described in Appendix 4. Reporting units and precision are listed in Table 5.2-12.
Future activities. The relationship between slope and fertility must be quantified and
evaluated. The evaluation could result in future division of fields by slope region, with separate
composite samples taken from each slope region. A decision to divide fields into subregions has
the disadvantage that soil samples may be collected on a unit smaller than a whole field (a 5-acre
area), which is the unit size for most other indicators.
Indicators of soil compactness are important because distribution and size of pore spaces is
important for root growth, distribution of soil microbes, and earthworm populations; the activity
of microbes and earthworms improve soil fertility and porosity, respectively. Potential indicators
include bulk density, pore size distribution, or surrogate measures of compaction such as
5.2-25
-------
Tabte 5.2-11. Soil analytical laboratory parameters to be measured in the 1992 Pilot
Parameter
Description of Parameter
%WATER Air-dry soil moisture determined gravimetrically and expressed as a percentage on an oven-dry
weight basis; mineral soils are dried at 105 C, organic soils at 60 C
SAND Sand is the portion of the sample with particle diameter between 0.05 mm and 2.0 mm; it is
measured using a hydrometer method
SILT Silt is the portion of the sample with particle diameter between 0.002 mm and 0.05 mm; it is
measured as [100 minus (SAND + CLAY)]
CLAY Clay is the portion of the sample with particle diameter less than 0.002 mm; it is measured using
a hydrometer method
EC Electrical conductivity determined in a deionized water extract using a 1:1 mineral soil to solution
ratio or 1:4 organic soil to solution ratio; it is measured with an electrical conductivity meter
PH_H20 pH determined in a deionized water extract using a 1:1 mineral soil to solution ratio or 1:4 organic
soil to solution ratio; it is measured with a pH meter and combination electrode.
XCA Exchangeable calcium determined in a buffered (pH 7.0) Mehlich 111 extract using direct current
plasma.
XMG Exchangeable magnesium determined in buffered (pH 7.0) Mehlich III extract using direct current
plasma.
XK Exchangeable potassium determined in buffered (pH 7.0) Mehlich III extract using direct current
plasma.
XNA Exchangeable sodium determined in buffered (pH 7.0) Mehlich III extract using direct current
plasma.
XAL Exchangeable aluminum determined in buffered (pH 7.0) Mehlich III extract using direct current
plasma.
CEC Cation exchange capacity will be calculated as the concentration (meq/1 OOg) of the exchangeable
cations plus acidity.
ACIDITY Total exchangeable acidity is a measure of the exchangeable acidic cations on the soil cation
exchange complex. It will be determined in an unbuffered (pH 8.2) barium chloride triethanolamine
solution using a 1:30 soil to solution ratio and a back titration procedure
BASE Percent base saturation; may be calculated as the sum of exchangeable Ca, Mg, K and Na divided
by CEC
MIN N Mineralizable nitrogen is a good predictor of soil nitrogen availability due to biological activity; an
~~ incubation technique for determination of anaerobic nitrogen as ammonium nitrogen is preferred.
P Extractable phosphorous determined by a Bray II extractant using direct current plasma.
5.2 - 26
-------
Tabte 5.2-11. (cont'd)
Parameter
Description of Parameter
ORG C
Easily oxidizable humus determined as loss by combustion at 350 C.
Hg Total mercury analyzed as a cold vapor using atomic absorption spectrometry.
kPa Soil moisture determined on nonintact cores at -33 and -1500 kPa (-0.3 and -15 bars) and -10 kPa
and -5 kPa (-100 and -50 mbars) soil matric potential using a pressure plate apparatus. The first
measurements are those of permanent wilting capacity and field capacity for calculation of water
available for plant extraction. The latter two tensions are those required to drain soil water from
pores of size important for microbial survival and movement (i.e., approximately 30 and 60 urn
diameter), respectively.
Table 5.2-12. Reporting units, precision and expected concentration ranges (December 1990)
Parameter Reporting units!' Reporting precision^ Expected range (median)^
%WATER
SAND
SILT
CLAY
EC
PH_H20
XCA
XMG
XK
XNA
XAL
CEC
ACIDITY
BASE
MIN N
P
ORG C
Hg
wt%
wt%
wt%
wt%
dS/m
pH units
meq/10Og
meq/100g
meq/100g
meq/100g
meq/100g
meq/100g
meo/IOOg
%
mg N/100g
mg P/kg
wt%
mg Hg/kg .
1.0
1.0
1.0
1.0
1.00
1.00
1.00
1.00
1.00
1.000
1.0
1.00
1.00
1.00
1.0
1.0
1.0
1.0
15.7-88.1 (70.2)
3.4-56,9 (23.9)
1.0-20.0(5.1)
0.14-0.38 (0.23)
4.6-6.6 (5.7)
1.07-9.72(3.14)
0.28-3.02(1.04)
0.13-0.67 (0.35)
0.03-0.10(0.04)
0.01-0.98 (0.52)
2.5-21.9 (6.2)
1.5-45 (24) (%)
47.0-94.0 (76.0)
13-195(76.5)
0.7-1 9.4 (2.1 )i'
kPa
vol%
1.0
All values expressed on an oven-dry soil weight basis.
Number of decimal places that each unit should be determined for
Expected concentration ranges in reporting units for soil samples, based on the 1 st, 95th, and (50th) percentiles
of data collected from previous surveys.
Estimated from organic matter determinations.
5.2 - 27
-------
hydraulic conductivity or water infiltration rate. These measurements require either intact cores
or complicated protocols. They were not included in the 1992 Pilot Project because the logistical
concerns had not been resolved, but they will be studied for inclusion in a 1993 pilot
5.23. Quality Assurance I Quality Control
Samples. Each sample will be enclosed in a pre-labeled container, with a unique sample
number from 1 to 447 (Appendix 7). The code number will not reveal the actual location of the
field where the sample was collected. The containers will not contain contaminants that would
bias or interfere with detection of chemical parameters and will be provided by or purchased
from the analysis laboratory. The date the sample was collected, mailed, and received by the
preparation laboratory will be recorded on the mailing container using permanent ink. A pre-
addressed postcard will be mailed by the NASS enumerator to the ARG information manager
(1509 Varsity Drive, Raleigh, NC 27606) for each sample at the same time the sample is mailed
to analysis and preparatory laboratories to facilitate tracking of samples.
The analysis laboratory will be provided with a list of soil samples in each container shipped.
As each sample is received, the date of receipt will be recorded by laboratory personnel in a log
that later will be returned to ARG personnel.
Laboratory analyses. Five private laboratories and one federal laboratory (Table 5.2-13)
were compared for analysis methods and costs for a specified list of desired soil analyses,
analysis procedures (Table 5.2-11) and QA/QC requirements (Table 5.2-12). An official bidding
process will be conducted through USDA-ARS based on the desired procedures and cost. These
results will provide justification for selection of a laboratory for analysis of chemical and physical
parameters of the soil samples.
A legal contract or interagency agreement will be written with the contract laboratory to
address the following topics: analysis precision and method of determining precision, cost, and
time of completion of analyses. Laboratory accuracy will be determined by including one known
5.2 - 28
-------
Table 5.2-13. Private and federal laboratories contacted for chemical and physical analysis of
soils.
HUFFMAN LABORATORIES, INC.
4630 Indiana
Golden, CO 80403
Contact: Suzanne J. Zeller, Technical Services Coordinator
303/278-4455 . '
WEYERHAEUSER ANALYTICAL AND TESTING SERVICES
32901-32 Drive, S.
Federal Way, WA 98003
Contact: Ron Isaacson
206/924-6149
MICRO-MACRO INTERNATIONAL (MMI)
183 Paradise Blvd., Suite 108
Athens, GA 30607
Contact: J. Benton Jones, Jr.
404/548-4557
AGRICO RESEARCH LABORATORY
P.O. Drawer 639, 1087 Jamison Road N. W.
Washington Court House, OH 43160
Contact: Scot Anderson
800/321-1562
BROOKSIDE FARMS LABORATORY
308 South Main St.
New Knoxville, OH 45871
Contact: Mark Flock
419/753-2448
SCS NATIONAL SOIL SURVEY LABORATORY
Federal Building, Rm 152
100 Centennial Mall North
Lincoln, NE 68508-3866
5.2 - 29
-------
sample for every 40 samples submitted to the analysis laboratory (see Appendix 7). Laboratory
discrepancy can be statistically removed from estimates of within-field variability to permit
greater accuracy of variance estimates.
The data from chemical/physical analyses will be sent as an ASCII file on diskette to the
ARG information manager. The ARG information manager will perform validation tests on the
data to determine whether the values for each parameter fit within the expected range (Table 5.2-.
12) and precision objectives established for within-laboratory analysis and within- field variability
(Table 5.2-14). The outlier samples will be resubmitted to the laboratory for a second analysis.
Table 5.2-14. Data quality objectives for measurement of soil samples within the analytical laboratory and within
fields (October 1991, Wake and Johnson Counties, NC)
Precision objectives -
Parameter
SAND
SILT
CLAY
EC
PHJH20
XCA
XMG
XK
XNA
XAL
CEC
ACIDITY
BASE
MiN N
P
ORG C^
Hg
kPa
Reporting
units
wt%
wt%
wt%
dS/m
pH units
meq/100g
meq/100g
meq/100g
meq/100g
meq/100g
meq/100g
meq/100g
%
mg N/100g
mg P/kg
wt%
mg Hg/kg
vol%
Laboratory
SD %CV
12.83
6.60
9.90
0.083
0.363
1.246
0.395
0.173
0.027
105.46
1.842
9.50
9.50
46.82
0.999
19.0
38.6
63.9
32.1
6.4
40.7
40.2
56.2
50.6
13.8
29.2
37.3
12.8
47.4
44.7
Field
SD
15.04
7.40
11.72
0.106
0.585
1.647
0.501
0.195
0.024
139.76
2.356
13.97
13.97
63.7
1.065
%CV
23.0
41.7
69.3
38.5
10.3
50.8
49.2
60.8
45.7
18.1
35.2
53.7
18.9
64.8
45.0
- For the field samples, objective is 2X analytical samples.
- Estimated as % organic matter
% CV-standard deviation x 100
mean
5.2 - 30
-------
All samples must pass laboratory precision tests. Submitted samples will be archived at the
analysis laboratory until laboratory personnel are notified that all analyses passed the precision
tests. After all data have been collected, validated and transformed (as needed), the ARG
information manager will work with NASS personnel to integrate the soils data into the larger
Agroecosystem Pilot dataset at the North Carolina Agricultural Statistics Division (Raleigh).
5.2.6. Metadata requirements
In addition to the analysis data, metadata will be recorded to permit future interpretation of
the database. Metadata will include methods of analysis, reporting units, whether data are
integers or characters, name of analysis laboratory, and comments recorded during sampling or
processing procedures (Table 5.2-15).
5.2.7. Data Analysis
A major objective of the pilot study is to determine the range of values and the within- and
among-site variance for each indicator. Pilot data will be supplemented with that from literature
searches and from the State Soil Survey Database (SSSD). The ranges are needed for data
editing programs as part of the quality assurance procedures. Ranges and estimates of variance
are also needed, in combination with data from the literature, to determine what magnitude of
change in indicator values is likely to occur and if this magnitude could be measured at the
regional scale.
The main statistical presentation of data in the Pilot and in the implemented program will be
cumulative distributions of indicator values in a region and interpretation of the values as the
proportion or amount of land in a region that has values of concern with regard to soil quality.
For example, the proportion of land with electrical conductivity values > 4 mmhos may indicate
the proportion of land affected by salinization (Figure 5.2-1). The initial focus will be on
cropland only; this focus will be expanded in the future to include other soils in agroecosystems
(e.g., idle land, land adjacent to cropped fields, and land in the Conservation Reserve and other
set-aside programs).
5.2 - 31
-------
Table 5.2-15. Metadata for chemical and physical analysis of soils.
Variable
%WATER
SAND
SILT
CLAY
EC
PHJH20
XCA
XMG
XK
XNA
XAL
CEC
ACIDITY
BASE
,viIN_N
P
ORG_C
HG
kPa
Type
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Integer
Unit*'
wt%
wt%
wt%
wt%
dS/m
pH units
meq/lOOg
meq/lOOg
meq/lOOg
meq/lOOg
meq/lOOg
meq/lOOg
meq/lOOg
%
mg N/lOOg
ppm
wt%
mg/kg
vol%
Anal. Method
gravimetric
hydrometer
hydrometer
hydrometer
1:1 soil: sol
1:1 soihsol
Mehlich m
Mehlich m
Mehlich EQ
Mehlich HI
Mehlich m
calculated
BaCl 1:30
Ca+Mg+K+Na/CEC
KMn7O4
Bray H
Combustion
Color vapor
pressure pit
Lab Comments
AGRICO3'
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
AGRICO
- All values expressed on an oven-dry soil weight basis.
- Used as an example laboratory.
Pilot data will be used to begin an evaluation of how well the indicators and derived indices
truly reflect good, poor, or changing conditions. Although identified ranges for indicators and
benchmark references of soil quality are generally lacking, soil ratings based on specific soil uses,
properties or functions are available (see sections on individual indicators and Table 5.2-18)
These soil ratings can be explored for application to regional soil quality monitoring. Because
5.2 - 32
-------
100
CO
CD
CO
.55
•4—
O
o
o
CL
o
Electrical conductivity (mmhos/cm)
Figure 5.2-1. An example of a cumulative distribution function: electrical conductivity of soil.
5.2 - 33
-------
indicator values will vary with climate, soil and management scenarios, general baseline reference
points with which to group soils and sampling sites for indicator evaluation arc needed. One
approach may be to use the SCS Land Capability Classification (Table 5.2-16) as a reference
standard (USDA 1961). One could assign each data point from the pilot into one of the 5 classes
and to determine if the indicator values reflect better or poorer soils as defined by this soil rating
scheme. In subsequent pilots, specific reference sites might be sampled, such as Class I (very
good) and Class IV (very poor) soils or soils known to be poorly managed and degraded.
Determination of the rate of change of soil quality (change in the proportion of land area
with specific ranges in indicator values) is an important long-term objective. Because the
program is designed to give regional estimates of each indicator and standards of soil quality will
vary with climate and soil, some grouping of the data will probably be necessary (see Section
5.2.7.1) (Webster and Oliver, 1990).
Using soil quality data to form a larger picture of agroecosystem condition is a long-term
goal of the Program. Some aspects might be explored with Pilot data, perhaps supplemented by
NRI and SSSD data, where needed One type of assessment would be to explore spatial patterns
of soil "stresses" and soil quality indicators. For example, patterns of land use or of soil erosion
could be compared with those of soil structure indicators (i.e., AWC, porosity, clay content,
organic Q on a regional scale. Trends in soil indicators might be compared with trends in
overall implementation of soil conservation practices. However, it should be emphasized that
ascribing a cause to observed indicator values or trends (e.g., soil erosion effect on soil structure)
is not a goal of the regional monitoring and assessment component (called Tier 2 in EMAP).
Rather, associations among Tier 2 data are meant to be an initial look at broad spatial or temporal
relationships. If broad assocations are observed, more extensive sampling and/or research would
be initiated to determine if a cause and effect relationship exists.
5.2 - 34
-------
Table 5.2-16. SCS Land Capability Glasses
Class
Description
Class I
Class II
Class III
Class IV
Class V
Class VI
Class VH
Class VIII
Soils have few limitation that restrict their use
Soils have moderate limitations that restrict the choice of plants or that
require moderate conservation practices
Soils have severe limitations that reduce the choice of plants or that require
special conservation practices or both.
Soil have very severe limitations that reduce the choice of plants or that
require very careful management
Soils are not likely to erode but have other limitations, impractical to
remove, that limit their use
Soils have severe limitations that make them generally unsuitable for
cultivation
Soil have very severe limitations that make them unsuitable for cultivation
Soil and miscellaneous area have limitations that nearly preclude their use
for commercial crop production
Capability subclasses are soil groups within one class and reflect major limitations such as risk
of erosion, water in or on the soil surface that interferes with plant growth or cultivation, shallow,
stony or droughty soils or a very cold or very dry climate.
5.2.7.1 Soil spatial variability and statistical approaches
The spatial and temporal variability of many soil properties is large and may make real
changes in soil quality difficult to detect. Because EMAP is designed to provide regional
estimates of indicator values, some aggregation will likely be necessary to minimize the broad
inherent differences among agricultural soils. Several methods used to group soils according to
taxonomic classes or soil properties would be appropriate for EMAP data. These include:
5.2 - 35
-------
O Derived geographic classifications such as the Land Resource Regions and Major Land
Resource Areas (USDA, SCS 1981)
o Taxonomic order or suborders (Figure 5.2-2)
o General landscape position or slope (Stone et al. 1985, Larson et al. 1983)
o Soil depth (Larson et al. 1983, 1985)
These aggregation groups are described briefly below. The 1992 Pilot was not designed to
address how many samples are needed for different aggregation approaches, but the range of data
values across the three physiographic regions of North Carolina may allow some initial
exploration of these approaches. Data not collected directly that would be required to group each
sample into the suggested aggregations are available in the SSSD.
Derived geographic classifications. Two of the most important technical groups and their
derived geographic classifications are: Land Resource Regions and Major Land Resource Areas
(MLRA) (USDA, SCS 1981). A MRLA can be treated .as an agroecological zone with a
relatively homogeneous pattern of soils, climate, water resources and land use (McCracken et al.
1985). Examples of the use of MLRA's in soil quality assessment include Larson et al, (1983)
who estimated soil erosion rates and changes in productivity hi two MLRA's broken down by
slope class, and Turner et al. (1986) who aggregated data by MLRA's in an assessment of soil
characteristics that indicate sensitivity to acidic deposition. Land Resource Regions are larger
aggregations of MLRA's.
Taxonomic groups. The soil classification system used by the National Cooperative Soil
Survey (SCS county-based soil surveys) is based on properties related to soil development and
allows the placement of soil series into broader groups for progressively more general
interpretations: soil families, subgroups, great groups, suborders and orders (USDA 1975). The
general soil map of the USA identifies 27 suborders of soils that have been delineated in 61 areas
(Figure 5.2-2). The most appropriate grouping of soils for statistical analysis is likely to be at
the level of soil orders, suborders, or great groups (Larson et al. 1985). Ten soil orders are
recognized. The differences among orders reflect the dominant soil-forming processes and the
5.2 - 36
-------
SOIL TAXONOMY
I
-------
Figure 5.2-2 (cont'd). Soil orders and suborders in the U.S. (USDA.SCS 1981; Hall et al. 1985).
Order and Suborder
Alfisols
Aqualfs
Boralfs
Udalfs
Ustalfs
Xeralfs
Aridisols
Argids
Orthids
Entisols
Aquents
Orthents
Psamments
Histosols
Inceptisols
Andepts
Aquepts
Ochrepts
Umbrepts
Mollisols
Aquolls
Borolls
Udolls
Ustolls
Xerolls
Spodosols
Aquods
Orthods
Ultisols
Aquults
Jumults
Udults
Vertisols
Uderts
Usterts
Areas with little soil
Map symbol
Al
A2
A3
A4
A5
Dl
D2
El
E2
E2
H
11
12
13
14
Ml
M2
M3
M4
M5
SI
S2
Ul
U2
U3
VI
V2
X
Land Area (%)
13.4
1.0
3.0
5.9
2.6
0.9
11.5
8.6
2.9
7.9
0.2
5.2
2.2
0.5
18.2
1.9
11.4
4.3
0.7
24.6
1.3
4.9
4.7
8.8
4.8
5.1
0.7
4.4
12.9
1.1
0.8
10.0
1.0
0.4
0.6
4.5
5.2 - 38
-------
degree of soil formation. Each order is divided into suborders primarily on the basis of
properties that influence soil genesis and are important to plant growth or properties that reflect
the most important variables within the orders. Each suborder is divided into great groups on
the basis of close similarities in kind, arrangement, and degree of horizon development; soil
moisture and temperature regimes; and base status. The range of values for each indicator would
be much smaller, and trends more likely detectable, if soils were aggregated at some taxonomic
levels during data analysis.
Topographic position. Recent studies have shown that spatial variation in soil properties
is controlled mainly by topographic position (Stone et al. 1985; Daniels et al. 1987, Ovalles and
Collins 1986, Pierce et al. 1991). Lower slope soils are nearly always more fertile (and less
susceptible to change or degradation) than ridgetop or upper slope soils. Compositing or bulking
soil samples within and across sample fields with topographic variability will reflect primarily
the properties and changes in the more fertile bottomland soils. Because topographic position
and erosion are not mutually exclusive, and are confounded mainly by water relations, Stone et
al. (1985) and others conclude that much published data dealing with the effects of erosion on
plants and soil are confounded by the effect of topographic position.
Changes in soil quality due to erosion and management practices will likely be undetectable
in some topographic positions such as bottomland soils, while changes hi ridgetop soils may be
of substantial importance. Therefore, stratified within-field sampling according to slope and/or
interpretation of indicators within slope classes (Larson et al. 1983) will likely be necessary at
some time in the development of the program. This will require exact protocols for the
enumerator on when and how to divide fields for sampling and/or how to determine the slope
at the sample point. Field sampling methods for the 1992 Pilot that minimize within-field spatial
variability will be chosen (see Section 3 and 5.2.4) but will not include slope considerations at
this time. Field sampling will be reevaluated after the 1992 pilot for subsequent pilots.
Soil depth. Degradation of irreplaceable soil attributes is much more serious on some soils
than on others when compared at the same erosion rates (Larson et al. 1983, Hall et al. 1985).
5.2 - 39
-------
For example, a deep alluvial soil is much less vulnerable to degradation by erosion, in the short
term, than is a shallow, weathered soil, or soils with biologically unfavorable subsoils. For this
reason, soil depth is used as the basis for determining SCS soil loss tolerance CO values (Table
5.2-17). Soils could be post-stratified according to horizon depth when interpreting soil quality
indicators; the depth of each soil mapping unit is available in the SSSD.
Table 5.2-17. Examples of using soil depth for assigning soil loss tolerance values to soils. ?
Soil depth
cm
<25
25-51
51-102
102-152
>152
Renewable soil -
t/ha
2.2
4.5
6.7
9.0
11.2
Nonrenewable soil -
t/ha
2.2
2.2
4.5
6.7
11.2
From: USDA-SCS 1983; Hall et al. 1985
Soils that have a favorable substratum and can be renewed by tillage, fertilizer, organic matter
and other management practices
Soils that have an unfavorable substratum, such as rock, and that cannot be renewed
economically
5.2.5. Research Goals and Applications
Long-term assessment of soil quality has become a high priority for agroecologists
(McCracken et al. 1985, Shirley 1991, Pierce et al. 1991, Haberern 1991). The World Resources
Institute listed soil condition and extent of degradation as high priority environmental information
needed for decisionmakers (WRI 1991). The earliest soil assessments for agroecosystems
attempted to develop numerical ratings of soil productivity and were motivated by the need to
compare different soils for purposes of land use planning and tax assessments. These ratings
5.2 - 40
-------
were based primarily on crop yield (Huddleston 1984). Several newer soil productivity models
are based on soil properties such as bulk density and texture, often with the goal of predicting
the effect of accelerated soil erosion on long-term crop yields (Williams et al. 1984, Pierce et al.
1983, Kiniry et al. 1983, Huddleston 1982). The Soil Conservation Service is currently
developing a new Soil Rating for Plant Growth, which is also based on soil properties (Ray
Sinclair, SCS, Lincoln, NE, personal communication 1992). Because soil structure is central to
the functioning of soils and is susceptible to long-term damage from intensive agriculture,
attention is also being given to conceptual models that characterize soil structure and the rate of
change due to agricultural land management (Kay 1989, Gibbs and Reid 1988, Thomasson 1978).
Table 5.2-18 lists several examples of published work on soil assessments that will be useful
for identification of ranges for indicator values (e.g., USDA-SCS 1983) and in the development
of indices of soil quality (e.g. Lai 1991, Singh et al. 1992, Pierce et al. 1983, Kiniry et al. 1983,
Huddleston 1982, Thomasson 1978). The basis for establishing rating scales to interpret
indicator values should ultimately include not only the capacity of the soil to sustain crop
production, but should also allow for interpretation of how changes in soil indicators affect soil
organisms, nutrient cycling, soil resiliency, vulnerability to erosion and thresholds of irreversible
change. For example, soil porosity values, and changes over time, could be interpreted in the
context of microbial ecology as well as adequate aeration for root growth.
Many of the assessments listed in Table 5.2-18 combine and query GIS databases on a
regional or national scale (Burke et al. 1989, Turner et al. 1986, Nielsen and Lee 1987, Bliss and
Reybold 1989). Examples of soil assessments conducted on a regional scale include soils or land
area likely to be sensitive to intensive agricultural use (Federoff 1987, Yassoglou 1987), sensitive
to acid deposition (Turner et al. 1986) or susceptible to organic matter loss (Burke et al. 1989).
Goss (1991) develped a rating scheme of the soil leaching potential of agricultural chemicals that
has been applied to a national assessment of groundwater vulnerability (Nielsen and Lee 1987).
This scheme (Goss 1991) combines chemical and physical information on soils and on pesticides
and can be used as a management tool to "match" appropriate types and rates of agricultural
chemicals to soils, in an effort to keep runoff and residues out of water systems. This is one of
the .many soil quality assessment questions that could be addressed using pilot data.
5.2-41
-------
Table 5.2-18. Examples of soil assessments.
Productivity indices
Berger et al. 1952
Stone .1978
Kiniry et al. 1983
Larson et al. 1983
Pierce et al. 1983
Gersmehl and Brown 1986
Huddleston 1984
Huddleston 1982
Erosion Productivity Impact Calculator
Tilth index
Changes in soil structure due to cropping systems
Williams et al. 1984
Singh et al. 1992
Kay 1989
Gibbs and Reid 1988,
Thomasson 1978
Extent of erosion and land degradation
Soil leaching potential/groundwater vulnerability
USDA/SCS RCA Appraisal 1989
Goss 1991
Nielsen and Lee 1987
Sensitivity of soil to acidification from acid deposition
Land use effects on soil organic matter dynamics
Organic matter dynamics
Sustainability index: production per unit soil loss or per unit
decline in soil properties
Sensitivity of soil to degradation
Soil ratings for specific uses
Global change
Turner et al. 1986
Cole et al. 1989
Burke et al. 1989
Lai 1991
Federoff 1987
Yassoglou 1987
USDA/SCS 1983
Bliss 1990
Sombreck 1990
5.2 - 42
-------
5.3. Water Quality
5.3.1. Introduction
Agroecosystems are often irrigated and provide a source of drinking water for many
Americans. In one sense the agroecosystem is highly stressed from both the use of agricultural
chemicals (fertilizer nitrates/phosphate, and pesticides) and the mechanical operations and
landscape manipulations associated with food and fiber production. Alternatively, it is the very
use of agricultural chemicals and land management that permit farmers to deliver a dependable
and plentiful supply of crops for food, fiber and fuel. Agriculture appears to be the largest
source of non-point source pollutant loadings to streams and lakes in the United States, and its
sediment burden remains a major factor in aquatic habitat degradation. The use of agricultural
chemicals and land management practices affect agroecosystem productivity and stress ecological
health, including habitat quality, size and diversity of wildlife communities, aquatic populations,
and soil biota. The use of agricultural chemicals impacts other connecting ecosystems by exports
from the agroecosystem to lakes and streams, wetlands and estuaries, and by leaching to
groundwater supplies.
Irrigation is often used during the cropping season to supplement natural rainfall, particularly
during drought periods. Irrigation water is obtained from many sources, including farm ponds,
lakes, streams, and wells. In North Carolina, irrigation water is obtained primarily from farm
ponds that are recharged from wells (personal communication with Ron Snead, Agricultural
Engineering Department, North Carolina State University). Chemical applications usually occur
during early spring planting of crops and throughout the crop season. Applications of chemicals
are made again during the planting of winter cover crops of grain such as winter wheat, rye, oats,
and barley. Usually growers will make decisions concerning usage of chemicals, such as
herbicides and certain pesticides, prior to or at the time of planting. The original objective of
this water quality initiative was to sample farm ponds and wells used for irrigation purposes.
However, irrigation practices are scattered throughout the state and many segments may not have
irrigated fields. Thus, the principle focus was changed to sample farms ponds and wells
5.3-1
-------
regardless of their use for irrigation purpose. Sampled ponds or wells used for irrigation will be
so noted.
The principle objective of the water quality monitoring initiative is to assess the quality of
water in farm ponds and wells on a statewide basis (North Carolina Pilot).
5.3.2. Sampling Design and Sample Collection
This effort will involve monitoring and sampling across the entire state of North Carolina.
The statewide effort will provide information on water quality in a descriptive sense (e.g. detect
or non-detect) for farm ponds and groundwater (existing on-farm wells). Farm ponds and wells
will be identified in each sampled segment during the June Enumerative Survey (JES) by NASS
enumerators. Information on chemical use at each site is critically needed from the JES to
determine sampling and analysis requirements.
Water samples will be collected from farm ponds and wells from either the Hexagon Design
(51 segments) or from the NASS Rotational Panel Design (65 segments). Sample collection will
be consistent with strategies planned by the ARC. Sampling at each site will lie conducted by
NASS enumerators in general accordance with guidelines provided in the EPA Region IV SOP
Manual (LT.S. EPA 1991b). Chemical analyses will be conducted by EPA's Environmental
Research Laboratory, Athens, Georgia.
These water samples will be analyzed for specific chemicals such as atrazine, carbofuran,
aldicarb, and other selected pesticides (applied to crops such as tobacco, peanuts, corn, and
cotton) and nitrates. Testing for pesticide metabolites and sampling of sediment from some farm
ponds may be included if resources permit. All agricultural chemicals selected for monitoring
will be widely used for crop production in North Carolina. It is not anticipated that any
extensive effort will be devoted to determining spatial variability characteristics within farm
ponds at this stage, although some limited activity and literature research may be started.
5.3-2
-------
Two sampling approaches will be used for sampling farm ponds: a "boat" sampling method
and a "bank" sampling method.
The "Boat" method will utilize a boat to move to three locations on the pond where two
samples will be collected at different depths. After compositing the six samples, a sample
for analysis will be taken from the composite.
The "Bank" method will utilize a sampling device on a long pole (about 16 feet) which will
be extended over the pond while the enumerator stands on the bank. Six samples will be
collected from points around the pond; these will be composited, and then a sample for
analysis will be taken from the composite.
Either the Hexagon or Rotational Panel sampling frame will be used to select the monitored
segments. The actual number will depend on the frame used and the number of segments
containing ponds that can be sampled. Of these, half will be selected randomly and used to
examine only one pond using the "Bank" technique. In each of the other segments, two ponds
will be selected. In each of these ponds, both methods ("bank" and "boat") will be used. In each
of the segments where two ponds are utilized, one pond will be selected randomly and replicate
samples will be collected by both methods. This design involves a total of 75 ponds and 175
samples, and it provides at least 24 degrees of freedom for each variance component of interest
(Table 5.3-1).
Farm pond sampling by boat will require a small jon boat (12 ft.) or canoe and a plumb line
or, preferably, a fathometer (depth finder). From a logistical viewpoint and the remote location
of some ponds, it could be very difficult for NASS samplers to utilize a boat for sample
collection. Therefore a technique for obtaining a representative sample a short distance from the
bank is under development. A telescoping pole sampler which will take a 1-liter sample at a
depth of 1-foot approximately 15 feet from the bank is being constructed. This prototype will
be tested on area ponds to perfect the design. For the pilot study, a comparison of the boat
collection and the pole sampler bank collection techniques will be conducted.
5.3 - 3
-------
Table 5.3-1. The anticipated analysis of variance.
SOURCE
df
Variance component
Segments 49
Bank-Only Seg vs
Bank & Boat Seg 1
Bank-Only Segments 24
Bank & Boat Segments 24
Ponds (B&B_Segments) 25
Method y 1
Method x B&B_Segment 24
Method x Ponds (B&B_Segs) 25
Residual 50
Obs (Bank_Method) 25
Obs (Boat_Method) 25
Total 174
o2
a2
y The error term for testing Methods is MxS.
The recommended protocol for sampling ponds from the boat, involves composite sampling
as follows:
Three sampling sites within each pond should be selected in accordance with the
diagrams in Figure 5.3-1. At each sampling site, samples will be taken in a vertical
profile at 1 foot below the surface and at half the depth using a designated sampling
device. These samples will be composited for the two depths as well as for the three
locations. Wherever possible the boat should first be positioned at the approximate
deepest point of the pond (for impounded ponds, this usually will be behind the dam
about one fourth the distance of the pond, and, for natural ponds, this usually will be
5.3 - 4
-------
at the center of the pond). The depth at this point can be determined by using a plumb
line (or a depth meter). Each sample will be transferred into a suitably large glass
container (plastic containers are not acceptable) which will serve as the compositing
vessel. The same procedure will be followed at the remaining locations using the same
compositing vessel. There should be at least two gallons of water in the vessel after
all sites have been visited. The composited sample will be mixed well, and three 1-qt
subsamples will be poured off. The samples will be placed on ice immediately for
transport to the analytical laboratory or holding facility.
The recommended protocol for "bank" sampling of ponds is being developed.
For well sampling, the following protocol is to be followed:
Locate a faucet at a nearby wellhead. The well first must be purged by opening the
cold-water faucet to remove stored water, usually requiring at least five well volumes
of water before a sample is collected. The volume of water to purge depends on the
storage/pressure tank volume. A complete exchange of the volume of water in the
tank is required to collect a representative sample of ground water. About 30 minutes
is a reasonable time estimate if the faucet is located behind the storage tank, or about
five minutes if the faucet is located between the storage tank and pump motor and/or
plumbing entering the well. During the purging process, measurements of temperature,
conductivity and pH can be made to determine if the stored water is removed from the
system. When the measurement parameters stabilize or when the designated purging
time has elapsed, a sample can be collected directly into a one-quart glass bottle and
placed on ice immediately for transport to the analytical laboratory.
5.33. Essential Complementary Data
Information on chemical use on each segment is critically needed from the NASS JES to
determine sampling and analysis requirements. A farm pond (identify if used for irrigation
purposes) and well will be located in each segment for sampling purposes.
5.3-5
-------
W777
DAM
Vertical profile sampling
Constructed Impoundment Natural Impoundment
Figure 5.3-1. Sampling Design for Farm Pond.
5.3 - 6
-------
5.3.4. Logistics
Sample collection and transport to analytical laboratory:
All samples will be collected by NASS, stored on ice immediately, and shipped in insulated
containers to Athens-ERL for residue analysis. Samples will be- shipped by fastest possible
means the same day as collected If unforeseen events make same day shipment impossible, the
samples will be stored under refrigeration at 34-40°F (2-4°C) until shipment Sampling must be
scheduled so that samples will not be stored by the collector over the weekend Samples will
be stored at 2-4°C at the laboratory until analysis.
5.3.5. Quality Assurance
Sample Collection:
All water samples must be properly (i.e. according to protocol) collected in 1-qt amber glass
bottles (Athens will supply sampling containers). Fortified samples will be held under identical
storage conditions as field-collected samples and analyzed at regular intervals to assess storage
stability. Ten percent of field samples will be analyzed in duplicate. Outliers will be analyzed
in triplicate, if possible.
Prior to the collection of field samples, duplicate spiked samples will be run at several
concentrations to determine method accuracy and precision and to establish lower limits of
detection. During the analysis period, fortified recoveries will be analyzed as dictated by the
situation, but not less than one set per month. Spiking levels and range will be determined by
that time.
One reagent blank will be run each time samples are extracted (sample set). Standard
instrument calibration curves will be prepared at least once each instrument operating day.
Individual laboratory log books and instrument log books will be kept current and reviewed by
5.3-7
-------
the project officer on a regular basis. Analytical standards will be obtained from the EPA
repository at Research Triangle Park, NC or check-analyzed against an EPA standard if obtained
from another source.
Data Quality Objectives (DQO) will be established prior to the generation of sample data.
Approved EPA methodology will be utilized whenever possible and standard operating
procedures (SOP) referenced or written as needed. Quality control activities are a key component
for assuring high quality data. To minimize systematic bias attributable to laboratory techniques
and to ensure objectivity in measurements, samples will be analyzed in random order. Such
randomization of samples helps ensure that observed trends are actually due to field responses.
Laboratory Analyses:
All analytical support for this effort will be conducted at EPA's Environmental Research
Laboratory, Athens, Georgia: Analysis of pesticides will require analytical sensitivities in the low
parts per billion range in extractions from both water and sediment The analysis requires
production-line efficiency for large numbers of samples with multiple extractions. Depending
on the sample type and the test compounds, samples will be extracted using solid phase, liquid-
liquid, ultrasonic, or Soxhlet extraction techniques. Also, depending on test compounds, the
analyses of the extracted residue will be conducted by gas chromatography utilizing electron
capture (ECD), flame photometric (FPD), nitrogen-phosphorus (NPD), or Hall electrolytic
conductivity (Hall ECD) detection systems or high pressure liquid chromatography utilizing post
column reaction (PCR) and ultraviolet (UV) detection systems.
Depending upon available resources, residue analysis at Athens-ERL may include atrazine,
carbofuran, aldicarb, other selected pesticides, metabolites and nitrate.
5.3 - 8
-------
5.3.6. Metadata Requirements
Metadata will include methods of analysis, reporting units, data formats, and pertinent
comments by samplers or laboratory personnel.
5.3.7. Data Analysis and Integration
The result of this effort will be a database with generally the same design characteristics as
other parameters being examined within the Pilot Standard statistical techniques will be applied
to summarize the data in meeting Program objectives.
5.3.8. Research Goals and Application
Assess quality of irrigation water supplied by farm ponds and wells on a statewide basis in
the North Carolina Pilot.
5.3-9
-------
-------
5.4. Land Use and Cover
5.4.1. Introduction
A concept central to the field of landscape ecology is that the changing spatial structure of
a landscape affects the flow of energy, materials, and organisms across its components.
Agricultural landscapes, largely as a result of human activity, are characterized by spatial and
temporal patchiness on many scales. There are annual changes in agricultural land use, as well
as decades-long processes as land cycles between agriculture and other uses. One must consider,
for example, the influence of a hedgerow on an adjacent field, as well as the cumulative effects
of agricultural systems on the habitat of far-ranging species.
Changes in land use patterns, which may represent significant ecological change in their own
right, may also foreshadow further ecological change in agricultural landscapes. For example:
o An increase in chemical-intensive crops might affect water quality in surrounding areas.
o Removal of hedgerows and shelterbelts may lead to increased soil erosion.
o Changes in the amount and spatial structure of non-cropped land areas in the landscape
may affect populations of plants and animals which utilize those areas.
Land use changes may also reflect changing ecological conditions. For example:
o Global climatic changes may bring about major shifts in cropping regions or cropping
patterns within regions.
o Degradation of soil or water quality may lead to the abandonment of cropped land.
The ARG has defined two closely related assessment endpoints to address these issues.
o Land use and cover: an accounting of the amount of land in various land use and cover
categories. The'remainder of this section focuses on this assessment endpoint.
5.4- 1
-------
o Landscape structure: a more comprehensive analysis of the spatial structure of the various
components of agricultural landscapes. This research effort is described in section 6.2.
For EMAP, the agricultural landscape is comprised of several broad categories of land use,
including the land area 1) in cropland, by crop; 2) in permanent pasture, set-aside programs, or
fallow; 3) in use for managed animal production; 4) in farm ponds or other water; 5) in non-
cropped areas, such as hedgerows, woodlots and grassed waterways; and 6) devoted to buildings
and paved areas.
For the 1992 Pilot, land use will be monitored at multiple scales using:
o NASS area frame materials (broad spatial and temporal scale for all land).
o Thematic Mapper data (medium spatial and temporal scale for all land).
O Survey data collected by NASS (fine spatial and medium temporal scale for cropped
land).
o Interpretation of aerial photographs (fine spatial and medium temporal scale for non-
cropped land; see section 6.2 Landscape Structure).
5.43. Data Acquisition
Area Frame Material
The NASS area frame (Cotter and Nealon 1987) provides complete coverage of the
conterminous United States and Hawaii. Sampling frames are developed by state and are
currently updated every 15-20 years. The components of the area frame are summarized below.
Detailed information, including the strata used for the North Carolina frame, may be found in the
Design and Statistical Considerations section (Section 3) of this document, and in Cotter and
Nealon (1987).
o Strata: A state's land area is stratified according to intensity of cultivation. Stratification
is performed by county.
5.4-2
-------
o PSU: Strata are further subdivided into primary sampling units (PSU). The size of the
PSU's varies by stratum, but is 15-20 square kilometers for most agricultural strata. A
random sample of PSU's is drawn to represent each stratum. PSU boundaries are
digitized by NASS.
o Segment: All selected PSU's are further subdivided into segments of approximately 2.6
square kilometers each. One segment is selected at random from each selected PSU. The
resulting set of segments comprise the NASS sample.
The NASS stratification of land area provides a framework in which to analyze long-term
changes in land use patterns over large geographic areas. For the 1992 pilot, a procedure for
creating a geographic information system (GIS) coverage based on the NASS strata will be
developed and tested using the North Carolina area frame. The current frame for North Carolina
was developed in 1978 and will serve as a baseline against which future frame changes in North
Carolina will be measured. Strata maps for the state can be created by combining county strata
maps using GIS techniques. Because the strata represent very broad categories of land use
intensity (e.g., 15-50% agriculture), these maps will directly reflect only large changes in
agricultural land use intensity. Table 5.4-1 summarizes the steps needed to create an ARC
coverage of the NASS area frame.
Thematic Mapper Data
The State of North Carolina has developed ARC coverages of land use and cover for the
Albemarle-Pamlico watershed, which covers a large portion of northeastern North Carolina and
southeastern Virginia (Khorram et al. 1991). The coverages are based on Thematic Mapper data
collected during the winter of 1987-88. The classification, performed at North Carolina State
University as part of the Albemarle-Pamlico Estuarine Study, used a hierarchical classification
system as shown in Table 5.4-2.
5.4-3
-------
Table 5.4-1. Steps to convert NASS area frame to ARC format
Step
1) For each county, NASS Area Frame Division registers PSU map to
latitude / longitude. QA checks applied by NASS.
2) For each county, NASS Area Frame Division converts PSU map from
internal to DLG format.
3) DLG files shipped to ARG with paper map and PSU area listing for
each county.
4) DLG maps converted by ARG to ARC format and strata map for each
county produced. If a county does not convert cleanly, the county map is
plotted on paper and returned to NASS for clarification.
5) ARC county coverages edge-matched by ARG to provide seamless PSU
map for the state. QA check of boundaries with other coverages of state
and county borders.
6) Dissolve PSU boundaries between like strata to provide a seamless strata
map for the state.
7) Provide complete documentation of procedure used to create coverages.
Also document all GIS files created according to ASTM standards.
Target
Completion
Already
complete
8/30/91
10/30/91
12/15/91 -
5/15/92
5/15/92
6/15/92
Actual
Completion
-
9/30/91
files: 9/30
areas: 10/30
1/15/92
These data have been purchased by the EMAP-Landscape Characterization Group at the
request of the ARG. The ARG and the Landscape Characterization Group will cooperate in the
use and analysis of these data. For the 1992 Pilot, the ARG will use these data to summarize
land cover, at the Level 1 classification, for the portions of the Albemarle-Pamlico watershed
within North Carolina. These data will also be used for research in landscape structure indicators
for large geographic areas (see Section 6.2).
5.4-4
-------
Table 5.4-2: Classification system for Albemarle-Pamlico watershed land
cover data.
Level 1
Urban or Built-Up
Agriculture / Grassland
Forest Land
Shrub / Scrub
Water
Wetland
Barren Land
Other
Level 2
Low density
Medium density
High density
Agriculture/grass fields
Disturbed land
Hardwood
Pine
Mixed Pine / Hardwood
Low Density Vegetation
Water
Bottomland Hardwood
Riverine Swamp
Evergreen Hardwood / Conifer
Atlantic White Cedar
Low Pocosin
Low Marsh
High Marsh
Sand
Undetermined
June Enumerative Survey (JES) Data
Land use data for all selected sample segments are collected annually by NASS during the
June Enumerative Survey (JES). The entire land area of the segment is classified into one of the
categories shown in Table 5.4-3. As described in the Design and Statistical Considerations
section (Section 3) of this document and in Cotter and Nealon (1987), these values are expanded
to give land use estimates within each stratum and for the entire state. It is anticipated that JES
data for North Carolina will be available from NASS in July 1992. The entire North Carolina
JES sample will be utilized to calculate land cover estimates. Although NASS typically
5.4-5
-------
maintains JES data in SAS datasets, the Table 5.4-3. NASS JES land use classification.
precise form and manner in which these
data will be received and analyzed by the
ARG is still to be determined. Each JES
record must be identified by county and
PSU number.
These data provide extensive
information about the land used for
agricultural production and very little information about, other components of the landscape.
Consequently, these data will be used primarily to analyze changes in land used for agricultural
production. Steps to acquire JES data are shown in Table 5.4-4.
Land Use Classification
Cropland, by crop
Permanent pasture
Pastured cropland
Idle cropland
Occupied farmstead or dwelling
Other (woods, waste, roads, ditches, etc.)
Table 5.4-4. Steps to acquire JES data.
Step
1) ARG develops survey instrument with NASS.
2) NASS obtains OMB approval.
3) JES data collected by NASS enumerators.
4) JES data released to ARG.
Target
Completion
9/GO/91
3/30/92
6/15/92
7/15/92
Actual
Completion
9/30/91
5.43. Logistics and Quality Assurance
No special field sampling is required. Some QA aspects are discussed under data acquisition.
Standard NASS QA procedures will be used during administration of the JES.
5.4 - 6
-------
QA procedures used by NASS for area frame development are documented in Cotter and
Nealon (1987).
Procedures used in developing the Albemarle-Pamlico database are documented in Khorram
et al. (1991).
These documents are available at the ARG headquarters in Raleigh, NC.
5.4.4. Metadata Requirements
GIS Coverages
All GIS coverages will be documented in accordance with ASTM Draft Proposed
Specifications for Meta-Data Support in Geographic Information Systems (August 1991), which
has been adopted as a standard by the GIS Team of the EMAP Information Management Task
Group. The manner in which these data will be stored has not been determined.
JES Data
For each data element, at least:
Name
Brief description
Data type (integer, real, character)
Measurement type (categorical, nominal, ordinal, interval, ratio)
Definition of categories (for categorical and nominal data)
Units (for ordinal, interval and ratio data)
Data collection method
Error information
5.4-7
-------
5.4.5. Data Analysis and Integration
Figures 5.4-1, 5.4-3 and 5.4-4 show the flow of data from collection through analysis to
development of the final reporting product.
Data <
Source
Keys
Elements
Analyses
Products
-------
Source of data: NASS area frame
Summary statistic for segment: not applicable
Sampling method: entire state covered; not sampled
Variance structure: base map accuracy, digitizing, CIS conversions
Trend to be detected: long-term (15-20 years) changes in land use
Base period: 1978, year of current area frame for North Carolina
Nominal and Subnominal: not appropriate
Note: Figure 5.4-2 is a preliminary pie chart showing the area and proportion of land in each
of the eight NASS strata for North Carolina.
Figure 5.4-3 shows how Thematic Mapper (TM) land cover data will be used to develop
indicators of overall land cover. For the 1992 Pilot, only the North Carolina portion of the
Albemarle-Pamlico watershed will be analyzed. Indicators of Overall Land Cover and Overall
Land Cover Diversity will be calculated. The GIS coverage of the NASS area frame will be used
to stratify the TM data.
Indicator 2) Overall Land Cover: report estimated area and area! proportion of land for each
Level 1 land cover category (Table 5.4-2) for each stratum and for the North Carolina portion
of the Albemarle-Pamlico watershed.
Source of data: TM data & NASS area frame
Summary statistic for stratum: hectares of land in each TM category
Sampling method: entire stratum covered; not sampled
Variance structure: measurement, digitizing, classification, overlay
Trend to be detected: changes in land cover
Base period: 1987-88, date of TM data acquisition
Nominal and Subnominal: not appropriate
5.4-9
-------
to
1 CO
I
I
U
M.
CO:
CO.
.3
o
•a
S
a
§,
1
I
I
1
1
1
I
: en
5.4 - 10
-------
Data <
Source
Keys
Elements
Analyses
Products
-------
Calculations:
Simpson's Index
D = EPl2
Shannon-Wiener Index
H= -
Shannon-Wiener Eveness Measure
_ H _ H
log (number of categories)
Trend to be detected: changes in overall land cover diversity
Base period: 1987-88, date of data acquisition
Nominal and Subnominal: unknown
Figure 5.4-4 summarizes the development of several indicators using JES data. These data
will be used to calculate indicators of Production Land Use and Production Land Use Diversity.
Production Land includes all but the other (wood, waste, etc.) land use categories on the JES.
The values of these indicators will be reported and tracked over time. The entire North
Carolina JES sample will be utilized to calculate these indicators.
Indicator 4) Production Land Use: report estimated total area and areal proportion of
production land for each JES land use category (Table 5.4-3) for each stratum and for the entire
state.
Source of data: JES
Summary statistic for segment: acres of land in each JES category
Sampling method: see Design and Statistical Considerations, Section 3
Variance structure: see Design and Statistical Considerations, Section 3
Trend to be detected: annual changes in production land use ......
Base period: 1991
Nominal and Subnominal: not appropriate
Indicator 5) Production Land Use Diversity: use Production Land Use Proportions to calculate
a crop diversity index for each stratum and for the entire state.
5.4 - 12
-------
Data <
Analyses
Products <
Source
Keys
Elements
-
June Enumcrative Survey I
LandUseDaU •
T^-L
PSU# I
Cropland, by crop
Permanent pasture
Pastured cropland
Idle cropland
Farmstead
Other
f
Area & proportion of 8
production land use • ; _
by stratum and total • 1
_^
Table I
Pie Chart or Bar Graph 8
Plot over time •
^J
CIS Coverage of Agricultural •
Land Use Strata 1
Geo-referenced 8
County •
PSU# •
1 Stratum •
1
|
^_ Diversity Indices •
^"" by stratum and total •
^^|
| " - -
Report value 1
Plot over time I
Figure 5.4-4. Use of NASS JES data.
Source of data: JES
Summary statistics for segment: proportion, pit of land in each JES category
Sampling method: see Design and Statistical Considerations, Section 3
Variance structure: see Design and Statistical Considerations, Section 3
Calculations: same formulae as Overall Land Cover Diversity
Trend to be detected: annual changes in production land use diversity
Base period: 1991
. Nominal and Subnominal: unknown
5.4 - 13
-------
5.4.6. Research Goals and Applications
Error Structure
The error structure of the land use indicators must be understood and quantified in order to
determine the magnitude of land use changes which may be detected by these approaches.
The error structure of production land use data collected during the NA.SS JES is known.
It has been documented by NASS (Cotter and Nealon 1987) and is described in the Design and
Statistical Considerations section (Section 3) of this document.
Quality assurance procedures and error rates for the Albemarle-Pamlico land cover data are
described in Khorram et al. (1991). This document contains error matrices for classification of
the satellite data based on "ground truthing" of a sample of one acre sites within the study region.
"Ground truthing" for this study was carried out using 1:58,000 scale National High Altitude
Photography images. The error matrices provide an estimate of the accuracy of land cover
classification.
The NASS area frame for North Carolina was developed using 1/2 inch : 1 mile scale county
highway maps as base maps. Accuracy assessment of this material may be difficult. Errors in
the frame arise from many sources, including errors in the base map, errors in the digitization
process, and errors in registration. More work is required in determining how to quantify the
error associated with the area frame. Consultation with members of the NASS Area Frame
Section, the EMAP GIS Team, and the EMAP Landscape Characterization Group will be
necessary. Recent changes at NASS, including increased automation of frame development,
should simplify accuracy assessment of new area frames.
5.4 - 14
-------
Indicator Correlation
One goal of the pilot program is to determine if selected indicators are highly correlated and
possibly redundant. For example, land use cover may be correlated with surface water quality.
The monitoring program might be streamlined by eliminating redundant indicators. The data
from the 1992 pilot will be analyzed to determine if any of the land use and cover indicators are
correlated with any of the other indicators. Tier III research might be required to further study
any unexpected correlations among indicators.
5.4 - 15
-------
-------
5.5. Agricultural Chemical Use
5.5./. Introduction
Agricultural chemical use is a quantitative measure of rates and spatial and temporal
distributions of chemicals applied to agroecosystems.
Objectives:
o Determine actual use of pesticides and fertilizers
o Use as a surrogate measure for pest density and pest spectrum
o Use in risk analysis of potential ecological impacts of agrichemical use
5.5.2. Data to be collected by NASS (See Appendix 5)
o Type, rate and frequency of fertilizer use
o Type, rate and frequency of pesticide (insecticide, fungicide, nematicide) use
o Type, rate and frequency of herbicide use
o Crop treated
o Number of acres treated
o Mode of application
o Time of application
5.5.3. Essential Complementary Data
o Costs of chemical inputs
o Chemical grouping (type of compound) for each chemical (other grouping properties may
include persistence, toxicity, chemical formulation, and mode of action)
5.5 - 1
-------
o Spectrum of plants and pests against which the herbicides and pesticides are effective; for
which crops and pests they are registered in each state
o Reason grower applied a specific compound
5.5.4. Logistics
o See NASS survey logistics
5.5.5. Quality Assurance
o See NASS survey logistics
5.5.6. Metadata Requirements
o Trade name of compound
o Formulation
o Manufacturer of compound
5.5.7. Data Analysis and Integration
1. Classify pesticides into ecologically meaningful groups such as persistence, toxicity, chemical
formulation, mode of action, and spectrum of pests affected
o Classes need to be identified.
o Data management strategies need to be worked out to classify the many different
individual compounds that will be present in the raw data.
5.5-2
-------
2. Frequency distribution of fertilizer and pesticide use (proportion of acres treated with a
certain class of pesticide and fertilizer)
3. Spatial distribution of pesticide and fertilizer use
5.5.8, Research Goals and Applications
1. Nontarget effects on soils and biological communities
Data currently collected by the National Agricultural Statistics Service (USDA 1991) can also
be used for this assessment
Examples:
o What proportion of herbicides used are highly degradable?
o What proportion, are highly persistent?
o What proportion of insecticides used are organophosphates?
o What proportion are pyrethroids?
2. Nontarget effects on water resources
o Relate soil leaching and runoff potential to the leaching and runoff potential of the
specific pesticide (Goss 1991). Use to calculate relative overall potential for leaching and
runoff.
o Present as frequency of land area with high, medium or low potential for pesticide
leaching or runoff.
o Because specific chemicals will be related to the specific soil of the treated area, EMAP
data, which is taken at the same sample point, would be the best data for this assessment
5.5-3
-------
3. Surrogate measure for incidence of specific weeds or pests
o Many pesticides are registered and targeted for management of a specific weed, insect or
pathogen. The amount of specific pesticides applied may, therefore, serve to indicate
which weed/pest problem either was or were expected to be a problem in a region during
a given growing season. Data currently collected by the National Agricultural Statistics
Service (USDA 1991) can also be used for this assessment.
5.5-4
-------
6. Description of Specific Research Endpoints for the Pilot Project
6.1. Soil Biological Health
6.1.1. Goals and Approach
Free-living nematodes comprise up to 90% of the total nematodes in agricultural soils
(Stinner and Crossley 1982) and are a group of soil fauna that have promise for use as an
indicator of pollution exposure and the restoration capacity of soil ecosystems (Schouten et al.
1990). Nematodes have the following attributes that make them useful as ecological indicators
(Freckman 1988).
o Nematodes are small with short generation times, allowing them to respond quickly to
changes in food supply; they are ubiquitous, even in polluted or disturbed areas; they are
frequently the last animals to die.
o Nematodes have the ability to survive desiccation and revive with moisture.
o Populations are relatively stable with soil, thus any change is viewed as the result of an
environmental perturbation.
o Perturbation of nematode populations usually reflect a change of trophic structure.
o Trophic, or functional, groups can be separated easily, primarily by anterior structures
associated with various modes of feeding (Yeates and Coleman 1982, Freckman 1988).
Therefore, species identification is not necessary and the cost associated with identification
is relatively small.
6.1- 1
-------
o Abundance and size of nematodes makes sampling easier and less costly than for other
microflora and fauna.
Functional groups of nematodes are present in three positions of food webs in soil. Plant-
parasitic nematodes are herbivores, feeding on plant roots and are, therefore, consumers of
primary production. Bacterivores and fungivores consume bacteria and fungi (including
mycorrhizae), respectively, and are, thus, involved directly with decomposition and nitrogen
mineralization (Parmelee and Alston 1986; Seastedt et al. 1988; Sohlenius et al. 1988; Moore and
de Ruiter 1991). Omnivores add "connectedness" to the food web (Cpleman et al. 1983) by
feeding on more than one food source, including bacteria, flagellates and amoeba. Predaceous
nematodes feed upon all the other functional groups of nematodes (Moore and de Ruiter 1991).
6.1.2. Data to be Collected
Populations of nematodes in soil will be quantified by five trophic (functional) groups: 1)
plant parasites, 2) bacterivores (microbivores), 3) fungivores, 4) omnivores, arid 5) predators
(Yeates 1971). Numbers of nematodes in each trophic group will be counted in 500 cm31 soil
(Section 5.2.4) and transformed to numbers per kg dry soil to standardize values among soils
with different soil moistures (Section 6,1.4). ;
6.13. Essential Complementary Data
Various soil characteristics influence populations of nematodes. Soil parameters measured
will include organic carbon, exchangeable calcium, exchangeable sodium, pH, electrical
conductivity, soil texture, and gravimetric soil moisture (see Section 5.2). Iri addition, data
concerning 1) application of nematicides, by tradename and formulation, within the past 2, 2-4,
or 4-12 months; 2) crop(s) planted; 3) cropping history; and 4) tillage practices will be obtained
from the NASS Questionnaire (see Appendix 5). The NASS Questionnaire will also include
questions regarding applications of herbicides and pesticides that may be used to interpret
observed community patterns of nematodes (Section 6.1.7).
6.1 - 2
-------
6.1.4. Logistics
Sample collection. Only soil sampled from the Rotational Plan Design will be analyzed for
nematode populations. An autumn sampling period is proposed, following cultivation of crops
harvested in the fall. Populations of bacterivorous and fungivorous nematodes are favored at this
time because 1) crop residues are incorporated into soil by cultivation, and 2) temperatures are
favorable (15-20 C) (Stinner and Crossley 1982). Samples should not be collected from saturated
soils, otherwise anaerobic conditions would develop in the plastic bags during storage and
transport. Anaerobic conditions could decrease the estimates of nematode populations.
Although there are few quantitative studies describing the spatial patterns of bacterial-feeding
nematodes (McSorley et al. 1985), populations are probably aggregated around plant roots and
organic debris in a manner similar to plant-parasitic nematodes. Therefore, ridges, furrows, and
plant rows should be sampled with equal probability within a field. Because nematode
populations are aggregated spatially, soil samples will be collected using a systematic design
described in Section 3 and Appendix 6. Except for fields that are chosen for two composite
samples, 20 cores (2-cm diameter), taken to 20-cm depth, will be collected along a diagonal
transect described in Section 3.3.2, across a five-acre area, chosen at random, and pooled as one
composite sample for estimation of field populations. After all cores have been collected in a
bucket and gently (excessive pressure or abrasion will damage or kill nematodes) homogenized,
a 550-cm3 (500- ml beaker filled to the edge) subsample will be transferred toa4x2x 12 inch
plastic bag. The bag will be closed with a pre-labeled wire tag with the appropriate identification
code and stored in an insulated container or at temperatures < 30 C (Barker 1985b) until mailed,
to avoid temperatures that may affect estimates of nematode populations. All equipment
necessary for collection of the samples will be included in the enumerator kit (Table 5.2-9 in
Section 5.2).
Samples will be mailed using Federal Express (call 1-800-238-5355 for pickup) either the day
of sampling or the following morning to the enumeration laboratory (ATTN: Kitty Kershaw or
Ken Barker, 840 Method Rd Unit II, Raleigh, NC 27606). Prior to mailing, the soil sample
6.1-3
-------
should be placed in a padded (with bubble wrap) envelope, which is pre-addressed and postage-
paid to the enumeration laboratory. Methods for transport of samples to the enumeration
laboratory were tested in the December 1991 nematode survey in, North Carolina. There were
no significant differences in nematode populations when mailed or carried to the analysis
laboratory. There was also no significant effect of mailing an ice pack with the soil sample.
Samples should be mailed between Monday and Thursday so they arrive in the enumeration
laboratory on a weekday. Otherwise,, the laboratory should be notified (919-515-3330) so that
samples can be placed in appropriate environmental conditions immediately upon arrival, rather
than be stored in the post office or postal truck over a weekend. As samples are received by the
enumeration laboratory, the date of receipt;; will be logged on the list of identification codes and
sent to the laboratory before sampling is. started. Samples will be stored at 15 C and processed
within 14 days of receipL
Laboratory analyses. Nematodes will be extracted from 500 cm3 soil using a semiautomatic
elutriator followed by sucrose centrifugation (Barker 1985a). Elutriation was chosen as the
extraction method because this process, allows for the extraction of both live and dead nematodes,
which permits use of samples that may have been mishandled before reaching the enumeration
laboratory. A dissection microscope will be used as an aid to identify and enumerate nematodes
in soil by trophic group; compound-light microscopy will be used to confirm uncertain
identifications. The remaining 50 cm3 soil will be weighed, both moist and oven-dry (90 C for
48 hr), to determine the dry weight per cm3 soil; Numbers of nematodes in each trophic group
will be standardized as numbers per g or kg and per m2 (assuming 20-cm core depth and 2 cm
diameter) to permit meaningful comparisons with other methods and reports. Statistical analyses
will be conducted on non-transformed population data. A log (x+1) transformation will be used
if required to normalize the variance. , -
6J-4
-------
6.13. Quality Assurance
V
Samples. Samples will be shipped in pre-labeled containers with unique sample numbers and
logged on an inventory sheet as received by the enumeration laboratory as described in Section
5.2.5. Duplicate samples will be submitted to the enumeration laboratory for determination of
within laboratory and within field variability (Appendix 7). The variability will be compared to
expected ranges, standard deviations and % coefficients of variation established from preliminary
surveys conducted in 1990 and 1991 (Tables 6.1-1, 6.1-2). It is impossible to submit known
blanks with the field samples because of complex inoculation, handling and storage procedures
involved in handling biological organisms.
Laboratory analyses. Because the indicator encompasses many genera and trophic groups,
it is not possible or realistic to determine the extraction efficiency for each individual species.
Extraction efficiencies for an elutriator can range from 30-60% depending on the type of soil,
the screen, and the amount of organic matter. Alternatively, an average extraction efficiency for
plant-parasitic nematodes will be reported, to provide an efficiency of the extraction method for
the laboratory chosen for enumeration services (Section 6.1.8).
Nematodes from 10% of the submitted samples will be preserved in formalin-aceto-alcohol
(FAA) solution (90 ml of 50% ethanol, 5 ml of glacial acetic acid, and 5 ml of 37%
formaldehyde) and stored at room temperature (Daykin and Hussey 1985). The preserved
samples will be kept for one year and utilized if information about genera within trophic groups
is needed.
The data from the nematode enumeration laboratory will be sent as a hardcopy to the ARG
information manager who will arrange to have the data key-punched at the North Carolina
Agricultural Statistics Division (Raleigh) of NASS. After the data are entered into the computer,.
they will be combined with the other soils data described in Section 5.2.5. Necessary
transformations will be performed by the ARG information manager before the data are
6.1 - 5
-------
Table 6.1-1. Reporting Units, Precision and Expected Ranges for Nematode Populations (December
1991)
Parameter Reporting units'
PLPAR" noAg
BACT* noAg
FUNG' noAg
OMNI8 noAg
PRED* noAg
WT_VOL g/cc
Reporting precision*
1.0
1.0
T.O
1.0
1.0
t.a
Expected range (median)*
0 - 3697 (500.5)
53-1884(483.5)
0-541 (102.5)
0 - 57 (0)
0 - 208 (40)
0.33-1.06(0.90)
* Number of significant decimal places
b Expected concentration ranges in reporting units for soil samples, based on the 1 st, 95th, and (50th) percentiles of
data collected from the December 1991 survey; n - 122 for nematodes and n » 80 for WT_VOL
e All values expressed on an oven-dry soil weight basis.
* Plant-parasitic nematodes
• Bacterivorous nematodes
1 Fungh/orous nematode&
9 Omnivorous nematodes
h Predaceous nematodes
integrated into the larger NASS data set at the North Carolina Agricultural Statistics Division
(Raleigh).
6.1.6. Metadata Requirements
In addition to data used for analysis, metadata, will be recorded to permit future interpretation
of the database. Metadata will include methods of analysis, reporting units, whether data are
integers or characters, name of analytical laboratories, and comments recorded during sampling
or processing procedures (Table 6.1-3).
6.1.7. Data Analysis and Integration
Several indices will be computed for the nematode community in each soil sample. The
fields will be compared using cluster analysis (Hodda 1986) to test the indices for their ability
6.1 - 6
-------
Table 6.1 -2. Data Quality Objectives for Enumeration of Nematodes by the Enumeration Laboratory and Within Fields
(October 1991)
Precision objectives
Parameter
PLPAR'
BACT*
FUNGC
OMNI"
PRED*
PLPAR
BACT
FUNG
OMNI
PRED
Reporting
units
noJkg soil
noTkg soil
noAg soil
noTkg soil
noTkg soil
In (no./kg soil)'
In (no./kg soil)
In (no./kg soil)
In (no./kg soil)
In (no./kg soil)
Laboratory
SD %CV
565.7
670.2
120.9
28.29
59.33
1.001
0.702
1.860
1.849
1.623
82.1
75.4
85.6
139.7
81.6
16.3
10.7
44.8
103.5
44.7
Field
SD
953.5
477.9
134.6
33.7
64.7
1.50
0.723
1.546
1.804
1.660
%CV
130.8
69.2
101.2
185.2
101.4
25.6
11.5
36.5
117.3
48.7
WT VOL
g/cc
* Plant-parasitic nematodes
b Bacterivorous nematodes
c Fungivorous nematodes
" Omnivorous nematodes
* Predaceous nematodes
' Statistical analyses are run on In (x+1); x = nematode population
NOTE: For the field samples, the DQO is 2X that of analytical samples.
% CVsstandard deviation x 100
mean
Table 6.1-3. Metadata for Biological Analysis of Soils in the 1992 North Carolina Pilot
Variable Type Unit*
Anal. Method
Lab
Comments
PLPAR"
BACT0
FUNGd
OMNI"
PRED'
Integer
Integer
Integer
Integer
Integer
no ./kg
no 7kg
no ./kg
no ./kg
noAg
elutr/sucrs cent
elutr/sucrs cent
elutr/sucrs cent
elutr/sucrs cent
elutr/sucrs cent
BARKER
BARKER
BARKER
BARKER
BARKER
WT_VOL Integer g/cc
dry wt.
BARKER
a All values expressed on an oven-dry soil weight basis.
b Plant-parasitic nematodes
c Bacterivorous nematodes
d Fungivorous nematodes
* Omnivorous nematodes
' Predaceous nematodes
6.1-7
-------
to measure relative ecological health or stability of the soil. Indices that will be compared
include:
o Bacterivores + fungivores (as proportions of total)
o Omnivores + predators (presence-absence categories)
o Omnivores + predators / omnivores + predators + plant parasites (numbers or proportions
give the same value); index ranges from 0-1 (ecosystem stability potential/production
decrease)
o Omnivores alone
o Dorylaimidae (family of omnivores)
o Predators alone
o Shannon index of diversity (Ludwig and Reynolds 1988)
o Simpson index of diversity (Platt et al. 1984; Ludwig and Reynolds 1988).
The Simpson index has the advantage that, unlike the Shannon index, it does not give
disproportionate weight to rare species (Ludwig and Reynolds 1988).
The variance structure will be characterized into laboratory measurement error and within
field variation using cumulative density functions (cdf) and box-plots. The cdfs are useful when
the cumulative extent of some resource is less than, or equal to, a specified pereentile of the data.
Box-plots can also be used to display the distribution of the data. They are especially useful in
allowing comparisons of several distributions across time and space (Section 3A1). Variograms
will be used to inspect the covariance structure in the spatially distributed data (Section 3.4.2).
Interpretation of indices. Agricultural fields are characterized by an abundance of
bacterivorous and plant-parasitic nematodes and a low frequency of omnivores and predators
(Wasilewska 1979). High numbers of plant-parasitic nematodes are detrimental tc crop growth
and yield. Applications of nematicides initially decrease populations of plant-parasitic nematodes,
although populations may increase dramatically later. An abundance of bacterivores, considered
together with fungivores, is considered "healthy" (Freckman 1988). High numbers of bacterivores
and fungivores infer rapid decomposition rates (especially when Rhabditis spp. are abundant), and
6.1-8
-------
may be associated with low organic matter and with either low or high populations of bacteria
or fungi. Microbial populations may be decreased by nematode feeding or increased by the
feeding activity and feces of nematodes (Wasilewska 1979). Bacterivorcs are highly resistant to
chlorine, fungicides, nematicides (Wasilewska 1979) and herbicides (Dmowska and Kozlowska
1983), and they increase in abundance with cultivation (Wasilewska 1979).
High numbers of bacterivores are generally associated with low numbers of omnivores (< 5%
total nematodes) and predators (< 2% of total nematodes) under conditions favoring growth of
microflora (i.e. high soil humus content, high organic and mineral fertilization) (Wasilewska
1979). Together, omnivores and predators may serve as a soil bioindicator (Corny 1976) because
they are sensitive to anthropogenic disturbances, including cultivation (Wasilewska 1979).
Omnivores and predators have longer life cycles than bacterivores or fungivores and are found
in higher percentages in soils with perennial crops than in soils with annual crops (Wasilewska
1979, Bostrom and Sohlenius 1986). Omnivores do not depend on one kind of food; therefore,
they represent more stable conditions and more diverse biocenoses. Theoretically, small increases
in heterotroph (omnivore) biomass help re-establish system equilibrium and counteract
perturbation (O'Neill 1976). The presence of predators lengthens food-chains resulting in greater
stability of the soil ecosystem, and their numbers increase when conditions are stable
(Wasilewska 1979).
6.1.8. Further Research and Eventual Applications
Investigation of laboratories to enumerate nematode communities by trophic group will
continue. Presently, only two investigators in the United States, are known to be qualified to
enumerate nematodes by trophic group and in the use of the specified extraction method used
[i.e. Kenneth Barker (North Carolina State University) and Diana Freckman (University of
California at Riverside)]. Extraction by elutriation and Baermann funnels was compared for a
subset of the samples collected during a survey field study conducted in December 1991.
6.1 - 9
-------
A field study to compare sampling designs, within-field variability, and within-laboratory
variability of nematode communities was conducted in October 1991. The results from this study
provided the data quality objectives for the pilot study (Table 6.1-2).
The indices developed to describe nematode community structure will be compared for the
two surveys conducted across North Carolina in December 1990 and 1991. Using a probability
sampling frame, three annual crops (corn, soybeans and wheat) were sampled'in 1990. In 1991T
nematode communities were compared in an annual crop (soybeans), a short-term perennial (>
3-yr alfalfa) and a long-term perennial (pasture for > 10 yr). A variety of growth forms were
chosen to provide a broad range of index values. In 1991, nematode community patterns were
also compared to microbial biomass in soil; total and active bacteria and total and active fungi
were enumerated by Elaine Ingham at Oregon State University (Corvallis, OR). Microbial
biomass data are being evaluated in reference to nematode populations to evaluate the nematode
community indicator's ability to reflect the health of the decomposer foodweb in soil.
Plant-parasitic nematodes were enumerated to genus in both the 1990 and 1991 survey
studies. A diversity index such as Shannon or Simpson (Ludwig and Reynolds 1988) will also
be applied to that data to determine if, within a single trophic group, it might prove to be an
appropriate indicator.
6.1 - 10
-------
6.2. Landscape Structure
6.2.1. Introduction
The rationale for analyzing landscape structure as part of a monitoring program for
agroecosystems is described in Agroecosystem Monitoring and Research Strategy (Heck et al.
1991), and briefly in Section 5.4 (Land Use and Cover) of this document In short, the spatial
structure of the landscape affects the flow of energy and materials, and the movement of
organisms, among its components.
The indicators detailed in the Land Use and Cover Section provide information about the
amount and proportion of various land use and cover classes. These analyses may describe
landscapes A, B and C as 65% agriculture and 35% forest However, in landscape A forested
lands may be in three large parcels; in landscape B the forested land may be one large parcel;
in landscape C it may be in dozens of small, disjoint woodlots. The ecology of these three
landscapes is likely to be quite different. Indicators developed for landscape structure are
intended to provide quantitative measures of these and other ecologically relevant differences in
the spatial structure of landscapes. Since these indicators will describe the spatial structure of
the landscape, the terminology landscape descriptors has been adopted.
6.2.2. Research Objectives
Many landscape descriptors with ecological relevance have been proposed in the literature
(e.g., Turner 1989; Turner and Gardner 1991). However, most work has been theoretical with
no attempt to synthesize these measures into the framework of a national monitoring program.
The overall objective of the Landscape Structure Research Project is to develop a multi-scale,
quantitative, and ecologically relevant description of agricultural landscape structure using an
appropriate combination of landscape descriptors.
6.2- 1
-------
Initial research on landscape descriptors will be conducted as part of the 1992 Pilot Project
of the Agroecosystem Program. The research will focus on the North Carolina portion of the
Albemarle-Pamlico watershed, which covers a large part of the most northern and northeastern
portions of North Carolina, as well as part of southern Virginia. This region has been chosen
because 1) the 1992 Pilot is being conducted in North Carolina, 2) a Thematic Mapper (TM)
based land use and cdver GIS database exists for the watershed, and 3) GIS land use and cover
datasets are not available for any other large region of North Carolina. Limited resources and
the agroecosystem mission of the ARG will further restrict analysis to land in the agricultural
strata of the NASS Area Frame.
Specific objectives for the Landscape Structure Research Project are highlighted.
o Select and calculate appropriate landscape descriptors, using TM-based data, for each
of the following NASS land use strata within the North Carolina, portion of the
Albemarle-Pamlico watershed: <15% agriculture, 15-50% agriculture and >50%
agriculture,
o Within these three agriculture strata, obtain aerial photography for the NASS PSUs
within the Albermarle-Pamlico watershed; the EMAP-Landscape Characterization Group
(LCG) (or another contractor) will digitize the scenes according to Level 1 of the
Albemarle-Pamlico classification system (Table 5.4-2).
o With the LCG, determine the best type and source of imagery.
o Select and calculate appropriate landscape descriptors for these scenes.
o Analyze variance structure and attempt to determine an optimum number of PSUs
to be analyzed.
o Develop and carry out a program (jointly with the LCG) to study techniques for
analyzing aerial photography without digitizing the entire scene. It is anticipated that
such an approach would be less costly than digitizing the entire scene, and ultimately
would allow a larger sample to be analyzed. The techniques will be judged on their
6.2-2
-------
ability to 1) accurately reflect the land use characteristics of the scene and 2) accurately
reflect the structural attributes of the landscape.
6.2.3. Data Acquisition
This research will utilize the NASS Area Frame coverages and TM data acquired for the
analyses described in Section 5.4 of this document Steps to obtain these data are detailed in that
Section.
In addition, the acquisition and digitization of aerial photography will be required (Table
6.2-1). Two sources of photography are currently under consideration.
Table 6.2-1. Steps to Acquire Digitized Aerial Photography.
Step
1) ARG identifies areas (PSUs) for which photos are required.
2) NASS, EMAP-LC, or another contractor obtains
3) EMAP-LC (or another contractor) classifies and
Level 1. ARC coverages will be produced.
appropriate imagery.
digitizes at EMAP-LC
4) ARC coverages shipped to ARG.
Target
Completion
6/1/92
9/1/92
12/30/92
1/30/93
Actual
Completion
National Aerial Photography Program (NAPP)
The 1:40,000 scale black-and-white stereo imagery is available from NAPP for North
Carolina for the years 1989 and 1981. The LCG has established procedures for obtaining and
digitizing these photographs.
6.2-3
-------
USDA Agricultural Stabilization and Conservation Service (ASCS) Slides
The ASCS annually obtains almost complete coverage of North Carolina between May and,
August. Only large forested areas, such as the Great Smoky Mountains National Park, are
excluded. The imagery available is low-altitude (approximately 5200 feet), true-color, 35mm
slides. Each slide covers an area of 640-1000 acres, depending upon exact flight altitude.
Because ASCS is required to retain these photographs, an historical record is available for North
Carolina since 1984. The LCG currently has neither a mechanism for obtaining these images nor
the facilities for analyzing them. The ARG personnel have contacted the North Carolina office
of the ASCS and have an agreement in principle to obtain copies of these slides. Details could
be worked out if this imagery is to be utilized.
6.2.4. Essential Complementary Data
NASS County Road Maps
In order to identify the photography required, county road maps showing the location of
the sample PSUs will be required. These will be supplied by the North Carolina NASS office.
Land Cover Classification System
A classification system is required for the interpretation of aerial photography. Ideally,
data from different scales (e.g. satellite and aerial photography) should be interpreted according
to the same hierarchical classification system. Broad-scale data may be classified using the upper
levels of the hierarchy (ie. Level 1); fine scale data may be classified in more detail using lower
levels of the hierarchy (ie. Level 2).
The proposed classification system for the LCG is summarized in Table 6.2-2. The
classification system used for the Albemarle-Pamlico study is summarized in Table 5.4-2 (Section
6.2-4
-------
Table 6.2-2. Proposed LCG Classification System.
5.4). Differences between them are
minor at classification Level 1, most
notably the combination of the
agriculture and grassland classes in
the Albemarle-Pamlico system.
This is a reasonable combination of
classes for the classification of TM-
based data.
For the 1992 Pilot, Level 1 of
the LCG classification system will
be used for interpreting aerial
photographs. Using this system
allows a distinction to be made
between agriculture and grassland in
the PSUs. Separation of agricultural
fields from grassland is desirable for
comparison with analyses based on
data from the June Enumerative
Survey collected within the PSU.
The two classes may be combined
for analyses requiring consistency
with TM-based data.
Level 1
Urban / Built-up Lands
Agriculture
Grassland
Forest
Shrubland
Water
Wetland
Barren Lands
Snow / Ice / Glacier
Other
Level 2
High Intensity (Urban Center)
Low Intensity (Suburban)
Cropland
Orchard /Vineyard
Permanent
Man-controlled
Pasture
Evergreen
Deciduous
Mixed
Evergreen
Deciduous
Mixed
Marine
Estuarine
Fresh
Estuarine emergent
Estuarine woody
Palustrine emergent
Palustrine woody
Unconsolidated shore
Permanent
Disturbed / Transitional
Glacier
Snow /Ice
Indeterminable
6.2.5. Logistics and Quality Assurance
No field sampling is required.
6.2-5
-------
Logistics and QA for the NASS Area Frame and Albemarle-Pamlico data are detailed in
Section 5.4.
The LCG has well-established procedures for obtaining, classifying, and digitizing aerial
photography from NAPP. If another contractor is chosen they will be required to follow the
LCG's QA procedures. If the ASCS photography is required, appropriate logistics and QA
procedures will be developed.
6.2.6. Metadata Requirements
Metadata requirements for GIS coverages are described in Section 5.4 of this document
6.2.7. Data Analysis and Integration
Proposed Landscape Descriptors
A suite of landscape descriptors is proposed for monitoring agricultural landscapes. These
measures, drawn from the literature, describe various aspects of landscape structure which are
likely to affect ecological processes. Table 6.2-3 summarizes the landscape descriptors currently
under consideration. Formulae and algorithms for calculating these descriptors are described in
Appendix A8.3 of the Agroecosystem Monitoring and Research Strategy (Heck et al. 1991).
Thematic Mapper Data
Figure 6.2-1 summarizes the analysis of TM data. Land areas within the North Carolina
portion of the Albemarle-Pamlico watershed will be stratified using the digitized NASS area
frame, and the suite of landscape descriptors calculated for each stratum. This analysis will
provide broad-scale structural information for each stratum. Preliminary analysis of the
Albemarle-Pamlico dataset indicates that stratification at the landscape level will be critical to
the interpretation of these data. For example, as a result of fairly heavy tree cover in suburban
6.2-6
-------
Table 6.2-3. Landscape Descriptors Currently Under Consideration for use in the Agroecosystem Program.
Measure
Fractal analyses
Nearest neighbor analysis
Contagion index
Dissection index
Amount of edge of land cover A
adjacent to land cover B (%)
Describes
Broad-scale pattern; spatial
complexity
Fragmentation; dumpiness;
"connectedness"
Fragmentation; dumpiness
Patch edge-to-area relationship
Frequency with which Land cover
A is directly adjacent to land
cover B
Affects (examples)
Ability of organisms to utilize
habitat patches
Movement of organisms;
spread of disturbances
Movement of organisms;
spread of disturbances
Types of organisms which may
utilize patches
Border movement processes
(such as flow of sediment into
surface waters)
Raleigh, the area is classified largely as forest. Although accurate from the perspective of a
satellite, the ecological characteristics of a tree covered area are very different from those of, for
example, the Great Smoky Mountains National Park. Including both of these tree-covered areas
in a calculation of the total area of forest resources in North Carolina would be very misleading.
Landscape level stratification, such as the NASS Area Frame, draws a clear distinction between
these areas, and calculation of tree-covered areas within each stratum is much more meaningful.
Exactly how each descriptor will be calculated using the available data is one of the
unknowns to be determined by this study. Most of the examples in the literature focus on
calculating the values for simple geometric shapes (squares are very popular). The methods may
not be directly applicable to the irregular polygons of the NASS area frame. Further, many of
the measures are sensitive to perimeter length. The artificial edge created by using the NASS
stratum boundaries as an overlay must be taken into account
6.2-7
-------
Data <
Source
Keys
Elements
CG1A Thematic Mapper
Land Use Data
CIS Coverage of Agricultural
LandUieStnl*
Geo-rcfereoced
i
GcO-fCICTCDCBQ
Analyses
Products
Select
<15% agriculture
15-50* agriculture
>50* igricakme
Table
Plot over time
J
Figure 6.2-1. Analysis of Thematic Mapper Data for Landscape Structure.
Aerial Photo-Interpretation and Analysis
Figure 6.2-2 summarizes the analysis of digitized aerial photography. Because extensive
interpretation of aerial photographs is too expensive, a sampling approach will be developed for
land within the North Carolina portion of the Albemarle-PamUco watershed. For the Pilot, the
PSUs chosen for the Pilot (in the three agricultural strata) will serve as the sample. This will
allow landscape measures of the PSUs to be correlated with other indicator measurements
obtained during the Pilot (see discussion of overall integration below). For each PSU, the suite
of landscape descriptors described above will be calculated. This will provide a statistical
measure of fine-scale structure for each stratum. The use of other landscape descriptors (e.g.
hedgerow length and spacing, number of hedgerow connections) may be explored as resources
permit.
6.2-8
-------
Data <
Source
Keys
Elements
Analyses
Products 50% agriculture
Land cover, by stratum and total
Landscape descriptors, by stratum
I
Table
Plot over time
J
Figure 6.2-2. Analysis of Aerial Photography for Landscape Descriptors.
Sampling A nalysis of A erial Photographs
Because aerial photo-interpretation is costly, a study will be conducted in cooperation with
the LCG to test various within-photograph sampling techniques for analyzing aerial photography.
These techniques include 1) applying a grid to the photograph and classifying only the cover
under each grid point and 2) classifying several transects across the imagery. These analyses will
be carried out using data from the Agroecosystem Pilot as well as data from the LCG Ten
Hexagon 1990 Pilot Study. These studies" may also be carried out on simulated landscapes.
Figure 6.2-3 summarizes an approach to this analysis.
6.2-9
-------
Data <
Keys
Elements
t Digitized Aeral
Geo-referenced
i
Analyses
Products <
Aeriil Photographs
h
Results of Statistical
Analyses of Photos
I
Does statistical analyses reflect
land cover aod landscape strucnve.
by stratum and overall?
I
Journal Article
Recommendatkias for 1993 Danonstratioa
Figure 6.2-3. Comparison of Statistical Analysis of Aerial Photos to Completely Digitized Scenes.
6.2.8. Further Research and Eventual Applications
Determination of Error Structure
"Although the use of remote sensing data for spatial databases is increasing rapidly, our
understanding of associated data processing errors, especially for integrating multiple data sets,
lags far behind" (Lunetta et al. 1991). A typical procedure of developing a CIS database from
remotely-sensed data includes several phases: data acquisition, data processing, data analysis,
data conversion, error assessment and final product presentation. Error may be introduced in all
phases, and propagated and transformed from one phase to the next Another level of sampling
error is introduced when selecting a sample of PSUs for interpretation. These Individual errors
6.2 - 10
-------
must be quantified, and the way in which they propagate and combine understood, in order to
place confidence intervals around landscape descriptors based on GIS data.
This issue of error detection and classification looms large in the future development of
EMAP, because many of the proposed products are to be developed using GIS overlay
techniques. The Landscape Characterization Group, the GIS team and the Integration and
Analysis Group are discussing this issue. The ARG will continue to work closely with these
groups, with our ARG statisticians, and with the Statistics and Design Team of EMAP to address
this issue.
Integration Across Scales
Combining information obtained from TM data with information derived from aerial
photography will provide the desired multi-scale perspective. This is an area of current research
from which EMAP efforts stand.to benefit The ability to make predictions at one scale using
data collected at another scale is desirable but difficult. One approach, which will likely be more
profitable, is to develop procedures for using relatively inexpensive satellite data to guide the
allocation of resources for the solution of more expensive procedures such as aerial photo-
interpretation. Using NASS stratification to select these areas may be a viable approach.
Overall Integration
Other indicator data will be collected from fields and non-cropped areas within the PSUs
subject to landscape structure analysis. During the 1992 Pilot Project, the focus will be on
indicators of crop productivity, soil quality, and nematode populations. Exploratory analysis of
the relationships between the values of the indicators and the various landscape descriptors may
be carried out as shown in Figure 6.2-4. More detailed research will likely be required to
confirm any hypotheses developed from these analyses.
6.2- 11
-------
Landscape Pattern Types
The LCG is developing a Landscape Pattern Type (LPT) classification which could be used
to stratify land areas based on certain landscape structural attributes. A comparison of landscape
descriptors calculated using the LPT stratification to results from the NASS stratification would
be instructive and should be attempted.
Agricultural Landscape
Farm Reid
NASS PSD
J
Watershed / Region
J
Nation / Continent
J
Data Analysis
Reid-sampled Indicators J
Land use and landscape descriptors
based on NASS PSUs
Landscape descriptors based on
NASS agricultural stratification
and GIS databases
Watershed / Regional analysis of
correlations between landscape patterns
and field-sampled indicators
National / Continental Analysis
I
Figure 6.2-4. Framework for Exploratory Analyses and Integration.
6.2 - 12
-------
6.3. Water Quality - Groundwater Monitoring, Wells and Modeling
6.3.1. Introduction
Monitoring conducted using existing on-farm wells may be subject to built-in bias and may
be of questionable sample quality due to a number of factors (e.g., well construction and
materials, location, and type of use). One way to address this question is to compare data
derived from existing wells with similar data obtained using "research wells" drilled and sampled
under controlled conditions by EPA cooperators from the Environmental Research Laboratory
in Athens, Georgia (Athens-ERL staff).
Supplementary research comparing sampling results from existing wells to those from
research wells will be -conducted at the same time as the Agroecosystem Pilot Project for
groundwater model testing. Funding will be through the Groundwater Matrix Management
Program at Athens-ERL. This work is expected to complement the Agroecosystem activities and
may have an impact on the manner in which future monitoring is implemented.
Mathematical models have become useful tools for predicting movement of chemicals from
agricultural sites to surrounding environmental media. For EMAP, models may be useful for
estimating export loads to other ecosystems, so that monitoring requirements can be kept to a
minimum. A model testing and applications question, therefore, is whether existing groundwater
agricultural chemical threat models can be used reliably on large spatial scales as an alternative
approach to detailed monitoring for the non-point source loading indicators of the Agroecosystem
Program. The DBAPE system (Imhoff et al. 1990) and RUSTIC model (Dean et al. 1989) will
be used for this planned research effort.
Two objectives have been identified for this well comparison study.
o Assess the advantages and disadvantages of the use of existing on-farm wells versus
"research wells" for monitoring organic pesticides and nitrates in groundwater
6.3-1
-------
o Conduct preliminary field testing of an agricultural chemical groundwater model such as
DBAPE/RUSTIC on both a statewide scale (North Carolina) and a county or EMAP
hexagon level
6.3.2. Sampling Design and Collection by NASS and EPA, Athens-ERL
This project is research-directed and covers a smaller area of the state than the primary study
(Section 5.3). Water samples will be collected and analyzed at the Athens-ERL. This study will
compare the relative value of using research wells as opposed to existing wells for detecting
contaminants in groundwater. This emphasis will help determine whether existing wells can be
used to provide unbiased samples of groundwater quality. Results will also be used to test an
appropriate model for its predictive capability.
An area in the coastal region of North Carolina will be identified for use in this more
intensive groundwater sampling project The size of the area will be several square miles and
will correspond to some geographic unit compatible with a relatively uniform modelling scenario,
possibly to an EMAP hexagon, or to a NASS segment or PSU. A moderate number of research
monitoring wells will be installed randomly throughout the area for the purpose of obtaining
reliable groundwater samples. A similar or larger number of existing wells will be identified
within the same area for the purpose of obtaining groundwater samples. The data derived from
these samples will be used for comparison of the two types of wells. Standard statistical methods
will be applied. It is anticipated that samples will be obtained from each well on three or four
occasions throughout the year.
Detailed groundwater model predictions will be developed for the sampling area. Model
performance testing will be conducted using the methodology of Parrish and Smith (1990) on
both the existing-well and research-well data.
6.3-2
-------
6.3.3. Essential Complementary Data
Additional site data for parameterizing models will be required. These data will be obtained
from sources such as SCS. Data will include, but not be limited to, the following: soil series,
specific horizon depths, texture, areal distribution, depth to water table, horizon thicknesses,
hydraulic conductivity (by horizon), soil water retention, chemical degradation rates, retardation
(bulk density, field capacity, partition coefficient), meteorology data and chemical application
rate. Much of this will be derived from existing databases using DBAPE (Imhoff et al. 1990).
6.3.4. Logistics
All samples are to be collected by the Athens-ERL staff, transported to the analytical
laboratory, stored on ice and kept frozen until analyses are complete. Standard protocols will be
followed for sampling wells.
Location of the intensive research study area in the North Carolina coastal plain will be
coordinated with local representatives to identify sites suitable for research-well installations and
existing-well sampling.
6.3.5. Quality Assurance
Sample Collection
All water samples must be properly collected (i.e., according to protocol) in one-quart amber
glass bottles (Athens-ERL will supply sampling containers). Fortified samples will be held under
identical storage conditions to assess storage stability. Ten percent of field samples will be
analyzed in duplicate. Outliers will be analyzed in triplicate, if possible.
Prior to the collection of field samples, duplicate spiked samples will be run at several
concentrations to determine method accuracy and precision and to establish lower limits of
6.3-3
-------
detection. During the analysis period, fortified recoveries will be analyzed as dictated by the
situation, but not less often than one set per month. Spiking levels and range will be determined
at that time.
One reagent blank will be run each time samples are extracted (sample set). Standard
instrument calibration curves will be prepared at least once each instrument operating day.
Individual laboratory log books and instrument log books will be kept current and reviewed by
the project officer on a regular basis. Analytical standards will be obtained from the EPA
repository at Research Triangle Park, N.C. or check-analyzed against an EPA standard if obtained
from another source.
Data Quality Objectives (DQO) will be established prior to the generation of sample data.
Approved EPA methodology will be utilized whenever possible and standard operating
procedures (SOP) referenced or written as needed. Quality-control activities are a key component
for assuring high-quality data. Samples will be analyzed in random order, to minimize systematic
bias attributable to laboratory procedural techniques and to ensure objectivity in measurements.
Such randomization of samples helps ensure that observed trends are actually due to field
responses.
Laboratory Analyses
Analytical support for this research will be at EPA's Environmental Research Laboratory,
Athens, Georgia. Analysis of pesticides will require analytical sensitivities in the low parts-per-
billion range in extractions from both water and sediment The analysis requires production-line
efficiency for large numbers of samples with multiple extractions. Depending on the sample type
and the test compounds, samples will be extracted using solid phase, liquid-liquid, ultrasonic, or
Soxhlet extraction techniques. Also, depending on the test compounds, the analyses of the
extracted residue will be conducted by gas chromatography using electron capture (ECD), flame
photometric (FPD), nitrogen-phosphorus (NPD), or Hall electrolytic conductivity (Hall ECD)
6.3-4
-------
detection systems or high pressure liquid chromatography utilizing post column reaction (PCRS)
and ultraviolet (UV) detection systems.
Depending upon available resources, residue analyses at Athens-ERL may include atrazine,
carbofuran, aldicarb, other selected pesticides, metabolites and nitrate.
6.3.6. Metadata Requirements
Metadata will include methods of analysis, reporting units, data formats and pertinent
comments by samplers or laboratory personnel.
6.3.7. Data Analysis and Integration
Standard techniques will be used for statistical data analysis and modelling activities.
6.3-5
-------
-------
6.4. Biological Ozone-Indicator System
6.4.1. Introduction
Tropospheric ozone (03) causes more damage to plants than all other air pollutants
combined. Thus, there is increasing interest in developing ways to monitor its effects on
ecosystem health. A useful tool in such an effort would be a plant system that produces
measurable responses to ambient levels of O3 and can be calibrated to estimate O3-induced losses
to important plant species; such a system might be able to estimate the biological responsive O3
concentration. Optimally, such a system would account for potential effects of climatic variables
on plant growth and vigor per-se and on the magnitude of its response to O3. A system that can
predict changes on a short term basis would be preferred to one that has a single seasonal
endpoint
A plant system that utilizes the relative response to O3 of two clones of white clover
Trifoliwn repens L. has undergone preliminary field testing and has the potential to meet t' .;
criteria given above. The clover system can separate the effects of climate per-se from O3
response because it utilizes the differences in response between an O3-sensitive clone (NC-S) and
an O3-resistant clone (NC-R). Results from three years of field tests at Raleigh, North Carolina
indicate that both clones show similar response to climatic conditions and that clonal differences
in growth, foliar injury and foliar chlorophyll content were due to differences in sensitivity to O3.
The goal of this project is to use the differences in O3 sensitivity of NC-S and NC-R to
estimate biologically active O3 doses and ultimately, to estimate the impact of O3 on agricultural
ecosystems. Clonal differences in foliar injury symptoms (chlorosis and necrosis), foliar
chlorophyll content, and biomass production will be used as estimators in 1992.
6.4- 1
-------
6.4.2. Background
The NC-S and NC-R clones were survivors of a two-year field study (Heagle et al. 1989,
Rebbeck et al. 1988) in which a mixture of white clover (Regal) and tall fescue, Fesctuca
arundinacea Schreb. (Kentucky 31), was exposed to six levels of O3 in open-top field chambers
from April to October over two seasons. Ozone caused significant decreases in white clover
growth with a simultaneous increase in fescue growth, probably due to decreased competition
from clover. After two seasons of exposure, there was a decline in the number of live clover
plants in the high O3 treatments, while clover was still thriving in the low O3 treatment. Cuttings
from clover plants that survived the two-year field experiment were propagated vegetatively. One
clone (NC-R) that survived exposure to the high O3 treatment (seasonal 12 hour per day mean
of 89 ppb) was subsequently shown to be highly resistant to O3. The other clone (NC-S) was
selected from a charcoal-filtered-air plot (seasonal mean of 26 ppb) and was shown to be highly
sensitive to O3. Methods used in the selection and development of these clones have been
published (Heagle et al. 1991).
Three seasons of field testing (1989-1991) have shown that foliar injury, foliar chlorophyll
content and seasonal biomass production of both clones are directly related to the O3
concentration and that NC-S is always the more sensitive of the two. Ambient levels of O3 in
Raleigh routinely injure leaves and decrease growth of NC-S but not NC-R. At higher O3 levels,
both clones show response to O3 but NC-S is always much more sensitive.
6.43. General Approach
Virus-free plants (Heagle et al. 1991) of the two white clover clones will be propagated
vegetatively in a charcoal-filtered-air environment. Sixteen field sites will be selected to provide
a range of meteorological conditions, to include ozone. The sites chosen will be within reasonable
proximity to an O3 and meteorology monitoring station or will have one at the site. Plants will
be transported to field sites in early May and transplanted into large pots (30 cm diameter)
containing 15 liters of a uniform potting medium. Plants will be watered as needed to prevent
6.4 - 2
-------
moisture stress and will be fertilized regularly. At 28-day intervals over a 112-day period, leaves
will be sampled to estimate foliar injury and to measure chlorophyll. Clover forage will be cut,
dried, and weighed. Foliar injury of NC-S and NC-R and the NC-S/NC-R ratios for chlorophyll
and forage biomass will be used to estimate the O3 concentrations for individual 28-day periods
and for the entire 112 days. Relationships between climate, O3 concentrations, and the relative
response of NC-S and NC-R will be defined.
Most of the detailed methods to be used in this pilot project were published with results
of a field study performed in 1989 (Heagle et al. 1992); Appendix 8. Changes or additions to
the published protocols are provided with the outline of the procedures given here.
6.4.4. Cultural Methods
Virus-free plants of NC-S and NC-R will be maintained in the Southeastern Plant
Environment Laboratory at North Carolina State University at Raleigh, NC. Periodic enzyme-
linked immunosorbent assay (ELISA) (Heagle et al. 1991) tests will be performed to insure the
virus-free status of the plants. Rooted cuttings from this stock will be used in all field tests.
During the second week in March, stem cuttings containing from four to five nodes each were
placed in small (10 cm diameter) pots containing Metro-Mix (Metro Mix is a commercial mixture
of peat, perlite, and vermiculite with nutrients). Two weeks later, plants were inoculated with
Bradyrhizobium to promote nodulation and nitrogen fixation; each pot was fertilized with 150
ml of a solution containing 2 g of soluble fertilizer [5-11-26 (N-P-K)].
Plants will be moved to the field sites during the second or third week in April and
transplanted to 30 cm diameter (15-liter capacity) pots containing a mixture of 2 parts sandy loam
top soil, 1 part coarse washed sand, and 1 part Metro Mix. Plants will be watered to prevent
moisture stress and will be fertilized at two-week intervals with one liter per pot of the fertilizer
solution described above. Insects, will be controlled with applications of Talstar (an artificial
pyrethrin) at two-week intervals starting immediately after the first cutting. Water requirements
of the plants will vary widely, depending on the amount of foliage present and weather
6.4-3
-------
conditions. For example, daily irrigation will probably be required to prevent wilting during the
week before harvests under normal summer conditions in North Carolina, while no irrigation will
be needed on the day or two after harvests under most conditions.
The experimental exposure period will begin once the plant canopy covers the soil surface
(no soil visible from above) on more than 80 % of the pots at a given site. .At this time, plants
will be cut at a height of 7 cm (pre-study harvest). We anticipate that the pre-study harvests will
occur during mid May.
6A3. Monitoring Design
The monitoring design at each site will be two replicates of three pots each per clone.
Each replicate will consist of three randomly positioned pots of each clone in a rectangular
arrangement. The replicates will be spatially separated from each other. Plants will be sampled
on an individual pot basis on four dates at 28-day intervals after the pre-study harvest (after 28,
56, 84, and 112 days). The data from each of the three pots per clone in each replicate will be
pooled for statistical analyses. The data will be analyzed for each 28-day harvest and for various
combinations of the four 28-day periods.
The sixteen-site monitoring design for the 1992 Pilot will focus on four widely spaced sites
in the Eastern half of North Carolina. Four research stations of the NC Agricultural Research
Service will be the primary monitoring locations (Raleigh, Rocky Mount, Plymouth and
Whiteville). Four locations (on the station or on farms in the near vicinity of the station) will
be identified for each research station. Each research station has a meteorological monitoring
station. Two ozone monitors will be located at each research station. Preliminary review of
ozone data suggests the 12-hr/day mean ozone concentrations averaged over the 28-day growth
period will differ at several of the four primary locations. Monthly meteorological data may also
differ, but we have not seen data to support this comment.
6.4-4
-------
6.4.6. Logistics and Quality Assurance
The ARG members will be responsible for the operation of this indicator system. The
basic handling of plant culture has been covered (Section 6.4.3, 6.4.4 and Appendix 8). A State
Extension .Specialist has been contacted and is taking the lead to contact county extension
personnel to cooperate in the program. The county personnel will identify the farmers most
likely to cooperate and will visit the farmers with our ARG representative(s). Once an
understanding is reached, each monitoring site will be set up with the requisite number of pots
containing each clover clone. Provisions will be made for semi-automatic watering of the pots.
The participants (farmers or research station personnel) will be trained in the care and handling
of the monitoring station. Each site will be visited by ARG personnel every two weeks. The
first visit after the site set-up will be to bring the plants, set them up and review care with the
operator. The second visit (2 weeks) will be for site inspection and special care for the plants
(fertilizing and preventive pesticide spray). On the third visit (mid-May) the plants will be cut
to the 7 cm height and the pre-study harvest will be completed. A revisit of system care will be
done and all details completed for the four month (28 days each) monitoring design. Henceforth
at 2, 6, 10, and 14 weeks each site will be visited for routine care (fertilizing and pesticide) and
a check of plants. At weeks 4, 8, 12 and 16 the routine care and check will be done as well as
the collection of all data. Logistics will be developed so that the 4-, 8-, 12- and 16-week visits
are each accomplished in a two-day time frame.
Quality assurance (QA) is built into each step of the process. Initial site operator training
will be done by a single ARG member. The basic areas of QA include: 1) care of mother plants,
culture of cuttings and transplanting; 2) operator training and care of plants in the field; and 3)
sampling and analytical procedures.
Details for logistics and QA will be developed as the program develops during the Pilot
phase. A detailed protocol will be ready for either the 1993 demonstration or the 1993 pilot
6.4-5
-------
6.4.7. Metadata
Metadata will include all the details associated with the design and collection of data from
the 16 study sites. Additionally, it will include information associated with the ozone and
meteorological data. Protocol for handling these data will be developed during the Pilot study.
6.4.8. Data Analysis
Iniurv Estimates - Injury estimates involve the potential for subjectivity amd bias, but this
is a rapid procedure that, with practice ("calibration"), can achieve close agreement (within + or
- 5%) between different estimators for given leaves. Estimates of foliar injury will be made as
the total percentage chlorosis and necrosis (in 5% increments from 0-100%) on each of 5
adjacent trifoliolate leaves per stem on one randomly selected plant of each clone per replicate
(2 plants per clone per site). The physiological maturity of leaves measured will be standardized
by using the youngest "fully expanded" leaf as the first leaf on each stem. Injury estimates will
be made and recorded separately for individual trifoliolates at each stem position.
Chlorophyll Measurements - Leaves used for the injury estimates will be used for the
chlorophyll analyses. The five leaves from each of the plants will be placed in approximately
70 ml of ethyl alcohol in a brown glass bottle (150 ml capacity) and placed in the dark. After
3 days, the volume of alcohol for each container will be increased to 100 ml and chlorophylls
a and b will be measured spectrophotometrically. Dry weights of each 5-leaf sample will be used
to convert the chlorophyll values from micrograms per liter of solution to micrograms per gram
of dry leaf sample.
Biomass Measurements - Above-ground biomass (forage) production will be measured by
cutting the plants at a height of 7 cm above the soil surface. Stolons growing outside of the 30
cm pot diameter will also be cut. The cut forage (leaves, petioles, and/or flowers and stolons)
from each pot will be placed in paper bags, dried in an oven, and weighed.
6.4 - 6
-------
7. Quality Assurance
7.1. Introduction
Decision makers, the public, and other users of EMAP data must have a high degree of
confidence in the data and statistics generated by the Program. The purpose of quality assurance
is to ensure that the data will yield sound and unbiased conclusions related to the principal
questions being addressed. Quality assurance (QA) for the Agroecosystem Program is being
developed to assure the reliability of measurements. It is recognized that the development of a
QA plan is an iterative process. Thus, we expect to learn much in the Pilot that will enhance the
QA plans. In the Quality Assurance Project Plan (to be completed before full Program
implementation) several key components of QA, including data quality objectives, standard
operating procedures, QA project plans, audits, QA annual reports, and work plans will be
developed. The general philosophy on QA for the ARG is clearly developed in the Research
Strategy document (Heck et al. 1991). QA information will be incorporated into the metadata
associated with Pilot data.
7.2. NASS Quality Assurance Procedures
Because the Agroecosystem Program is being developed as a cooperative effort between the
USDA/ARS, the EPA and USDA/NASS, the ARG will take advantage of QA procedures already
employed by NASS. NASS views quality control as the process of eliminating as many survey
errors as possible. To limit errors, every survey process must be associated with some type of
quality control procedure. The ARG intends to use all of NASS's established quality control
procedures in each survey process. The major survey processes in the Agroecosystem Pilot
Project amenable to quality control considerations include:
7- 1
-------
Area sampling frame
Construction
Maintenance
Sampling
Survey specifications
Questionnaire design
Preparation of manuals
Interviewer's
Supervising and Editing
Survey software
Training schools
Survey management
Questionnaire handling/processing
Manual data review and coding
Data edit and review
Summarization
Post-survey evaluations
Survey research
The ARG will work with NASS personnel to identify sources/writeups on QA to cover those
areas above that will be incorporated into the Agroecosystem Program.
7.2.1. Area Frame Development
General procedures for selection of Pilot segments according to either the rotational panel
or hexagon scheme are presented in Section 3.1. Prior to drawing segments, however, the area
frame must have been constructed. This activity is handled by NASS, so the QA work rests with
them, as will the QA to cover the selection of segments. Some of the quality assurance methods
in frame development are documented in Area Frame Design for Agricultural Surveys (Cotter
and Nealon 1987). These procedures are important for ensuring that no land area is double-
counted or unintentionally omitted, and that strata are correctly identified. QA is also an issue
in sampling from the area frame (e.g., in marking off sample segments, Cotter and Nealon 1987).
The development of PSUs around EMAP hexagon centroids is a new activity for NASS. The
GIS lead within the ARG is verifying these PSUs to assure that they contain the appropriate
hexagon centroid.
7.2.2. Conversion of NASS Area Frame to ARC/INFO Format
The current NASS strata, developed in 1978, will be used in the development of a land use
indicator (Section 5.4). As a part of this process, the Area Frame is being converted to
ARC/INFO format. This process requires special boundary checking and careful tracking of error
7-2
-------
sources. The latter is an area of current investigation. Further information about these issues can
be found in Table 5.4-1 and in Section 5.4.6.
7.2.3. Survey Data: Collection, Processing and Output
The QA procedures utilized by NASS will be identified and developed in this section. We
do not plan to include all the detail here, but we will reference sources and have copies available
for review. The ARG will be responsible for data summarization and will develop the QA for
these procedures.
Survey data for the ARG will come from two surveys administered by NASS: the June
Enumerative Survey (JES) and a special Agroecosystem questionnaire. The JES is NASS's
annual effort to collect land use data. NASS will add eight questions to the JES for the ARG
1992 Pilot: three regarding farm ponds, three on wells and two on land use (irrigated acres and
idle cropland in government programs). The Agroecosystem 1992 Pilot Study Questionnaire (i.e.,
"Fall Survey") requests information on yields and management practices; it is reproduced in
Appendix 5, as are the eight additional JES questions.
NASS has procedures for both controlling and assessing the quality of the data collected by
their surveys. Unless otherwise noted, these procedures will apply to both of the surveys
described above. The process starts with the survey specifications. The actual Pilot
Questionnaire, along with the questions added to the JES, have been developed in cooperation
with the ARG. Close contact between NASS and the ARG is one way in which quality is being
assured.
Enumerator training is the next important part of quality control. For the JES, a national
workshop will be held in April 1992, with the state "school" to follow in May 1992. The Fall
Survey, which will actually start in November, will run concurrently with the sampling of soil
and water. NASS and the ARG will cooperate in planning and running the training session for
7-3
-------
these activities. NASS enumerators will be taking soil and water samples, as well as
interviewing farmers.
Once field interviews begin, the supervisory enumerators are responsible for assuring that
data are taken correctly. They accompany new enumerators on their first day of interviewing and
meet with experienced enumerators after the first few interviews of each survey. If there are any
problems, the supervisor either instructs the enumerator individually or holds a re-training
meeting if needed.
For both surveys, approximately two interviews from each enumerator's workload will be
checked by telephone follow-up. Questions from a worksheet will be asked, to verify that the
interviewer did contact the farmer, that a particular crop was grown, etc. Such worksheets will
be printed in the Supervising and Editing manuals (see below). For the JES, the supervisory
enumerator does an on-the-ground check of a couple of random farmers from each enumerators
workload, to be sure that field boundaries were drawn correctly. The responsibilities of
supervisory enumerators are given in the NASDA Supervisory Enumerator Handbook
(USDA/NASS, 1990). NASDA is the National Association of State Departments of Agriculture.
Two manuals will be prepared for each survey: an Interviewer's Manual and a Supervising
and Editing Manual. These are done annually for the JES (e.g., USDA/NASS 1991a) and will
be developed separately for the Agroecosystem Fall Survey. They will be published in May 1992
for the JES and in October 1992 for the Fall Survey. For the latter, the instructions for
interviews will be part of an Enumerator's Manual that will also cover soil and water sampling.
Survey data are subject to a three-stage editing process. First, the supervisory enumerator
checks the data for reasonableness and sends the questionnaire back to the enumerator if there
are problems to be resolved. Once received by NASS, questionnaires are edited by a statistician,
who returns unsatisfactory forms to the supervisory enumerator. After these two manual edits,
data are entered into the computer, where another detailed edit is performed. The computer
verifies that responses are appropriate to questions, runs tests for internal consistency and checks
7-4
-------
that data items are within the expected ranges for North Carolina. Problems discovered at this
level are brought to the attention of the statistician.
The statistician scores the quality of the enumerator's work on a scale of 1-5 for each
interview. These scores are reported back to the supervisory enumerator for both surveys. For
the Fall Survey only, enumerators will rank the quality of the interview (scale to be developed).
A post-survey analysis is done to calculate response rates, account for costs, and similar items.
This is described in Section 8, Logistics.
7.2.4. Field Samples
Several steps in the field sampling process have quality assurance as one of their functions.
These include enumerator training, duplicate sampling, sample shipping and sample tracking.
Several of these are described below.
7.2.4.1. Soil Sampling
One of the first steps in assuring the quality of soil samples will be the training of
enumerators. This will be done cooperatively by NASS and the ARG before the enumerators
begin the Fall Survey and sampling. There are no plans to double-check on enumerator
performance by re-sampling any fields, although ARG members will accompany enumerators on
a few of the sampling trips and will be available to answer questions and troubleshoot problems.
Three sources of variation will be distinguished in soil data: variation among fields, variation
within fields, and variation of laboratory analyses. These will be sorted out by sampling second
transects in certain fields and by splitting certain of those samples in two (for separate analyses).
These procedures are detailed in Section 3.3.3. The current plan is that the extra transect will
be sampled in every sixth field, and half of those second-transect samples will be split.
Appendix 7 is a detailed table showing the number of regular, repeat, and duplicate (split)
samples, as well as "knowns". This and other QA for soil quality measures is addressed in
7-5
-------
Sections 5.2.5 and 6.1.5. These sections are primarily concerned with what happens to the
sample once it is taken from the field. Proper procedures for mixing, shipping and tracking the
samples (Sections 5.2.4 and 5.2.5) will ensure that they arrive quickly and in good condition at
the processing and analysis laboratories.
The soil series of each sample field will be determined by comparing the NASS aerial
photograph (on which the field has been outlined) with the most recent soil survey photography.
To satisfy confidentiality requirements, this will need to be done at the Raleigh NASS office.
The SCS may be asked to provide assistance by visiting some fields to verify those soil types.
7.2.4.2. Water Sampling
Selection of farm ponds and wells to be sampled will follow the procedures given in Sections
3.3.4 and 5.3.2. Data quality will depend on adherence to sampling procedures given in Section
5.3.2. These include instructions for the location and depth at which pond samples must be
taken. Also specified are the type of container (glass only) and methods for compositing the
water samples and then subsampling. Six samples are to be drawn to form each composite
sample, from which a sample will be taken for analysis. When sampling from wells, a critical
step is to properly purge the system before taking the sample (Section 5.3.2). Prompt chilling
and shipping of water samples is also essential. Enumerator training will be important for both
quality assurance and for safety. Members of the Athens-ERL will assist in pre-testing of the
methods and in training enumerators, although they will not take any of the actual samples from
the area frame segments.
One of the issues unique to the pond water quality indicators is the effect of the sampling
method on data quality. If a simple and reasonable way of taking samples from the bank of the
pond can be devised, it will compared to the standard method of using a boat (Sections 3.3.5 and
5.3.2). There would be a logistical advantage if boats are not needed, but samples taken from
the bank may be biased. To answer this and other questions about data precision, some ponds
will be sampled with both the "boat" and "bank" methods. From a subset of those ponds, a
7-6
-------
second, independent composite sample will be taken using each method (Section 3.3.5), for study
of within-pond sample variation.
7.3. Soil Quality Measurements
The quality of the sampled soil will be evaluated using both physical/chemical and biological
measurements. Our current approach to QA in these areas is found in Sections 5.2.5 and 6.1.5.
These include procedures for tracking and archiving samples. Such precautions should help
reduce the number of missing data points and also will provide information such as the length
of time that each sample spent in transit between the field and the laboratories. There are also
procedures for checking laboratory precision and, to a limited degree, accuracy. One in six field
samples will be split (to test within-lab variability), and one in 40 soil samples will be a
"known". The expected batch size is 40 samples. The source of "knowns" is still to be
determined. Such samples will serve as a type of accuracy check on the laboratory analyses.
Also, data from the State Soil Survey Database (SSSD) will be used to develop range checks for
soil quality measurements from the various soil types. It is not feasible to submit check samples
with known nematode numbers (Section 6.1.5).
Quality assurance procedures that are to be followed by the laboratories doing soil analysis
or nematode counts will be specified in the contract or interagency agreement, whichever is
appropriate. The EMAP Quality Assurance Program Plan requires that a QA Review form
(QAR-C, Revision 1, 1981) must accompany all procurement/order requests over $25,000 (U.S.
EPA 199la). Proposed methods for soil analyses are given in Appendix 4. We will work to
improve QA documentation during the Pilot Project One purpose of the Pilot is to identify and
improve both the QA and logistics plans.
7.4. Water Quality Measurements
QA procedures currently found in Sections 5.3.5 and 6.3.5 will be followed during the Pilot
and revised following the Pilot experiences. They include the use of fortified samples, duplicate
7-7
-------
analysis of ten percent of samples, and triplicate analysis (if possible) of outliers. Also covered
are procedures for reagent blanks, instrument calibration curves, log books, analytical standards,
and the random ordering of samples. Data quality objectives are to be determined before the data
are produced (Sections 5.3.5 and 6.3.5). The laboratory which will be analyzing the water
samples has extensive experience in quality assurance.
7.5. GIS Data for Albemarle-Pamlico Regions (Landscape Measures)
QA for GIS data for the Albemarle-Pamlico area will be obtained from Khorram et al.
(1991), as mentioned in Section 5.4.3. One possible way of checking the quality of landscape
indices will be to test their robustness to shifts in boundaries.
7.6. Additional Data
Additional data needing QA includes weather data, conversion factors (such as NPP
conversion factors and moisture contents, see Section 5.1.3), and other acquired data. We expect
this to be a difficult section to complete because we are dealing with data over which we have
no control. QA for aerial photography from NAPP will follow procedures developed by the
EMAP-LCG. QA procedures will need development if ASCS photography is used (Section
6.2.5).
7.7. Data Quality Objectives
The process of developing data quality objectives (DQOs) for the Pilot indicators is only just
beginning. Variances generated from Pilot data will be used, along with assumptions about
temporal correlations, to help set achievable DQOs (Section 3.3.7). A few measurement quality
objectives have been proposed for the soil physical and chemical properties (Tables 5.2-12 and
5.2-14) and for nematode counts (Tables 6.1-1 and 6.1-2), based on preliminary samples taken
5n 1990 and 1991. A guidance document for DQOs in EMAP is now in draft form.
7-8
-------
8. Logistics
8.1. Introduction
Implementation of the Agroecosystem Pilot Project
has required detailed logistics planning, including
coordination and oversight of all support and data
collection activities. Although not complete, all activities
listed in Table 8-1 have been considered by the ARG.
The logistics activities are closely tied to both QA and
information management for the Pilot Project. Logistics
considerations for each indicator are included in each
indicator subsection (Sections 5 and 6 of this document).
A schematic of the logistics for the Pilot is given in
Figure 8-3 at the end of this section. A Logistics
Notebook will be maintained by the ARG with details on
logistics for the Pilot.
Table 8-1. Logistical issues that have been
addressed by the ARG.
Staffing
Design of Survey Questionnaires
Communications
Training
Safety
Sampling Schedule
Site Access and Reconnaissance
Procurement and Inventory Control
Field Operations
Laboratory Operations
Waste Disposal
Information Management
Quality Assurance
Cost Tracking
Review of Logistics
8.2. Logistics and the NASS
A major goal of the Pilot Project is to determine whether the NASS enumerators can collect
all field data required for the indicators being tested in the Pilot. The enumerators, operating
within the NASS organization, will use procedures selected and developed jointly by the ARG
and NASS. From the standpoint of logistics, working with NASS has several benefits. Based
on the integrity and reliability of their personnel, NASS has developed a relationship over time
with the agricultural community which will greatly facilitate the collection of data. Additionally,
NASS has a fully developed infrastructure for the collection of agricultural data, including well-
developed logistical procedures and strict quality controls. Use of this infrastructure greatly
reduces the resources that would be needed for the ARG to develop similar procedures. The
8- 1
-------
ARG is using the Pilot to define more completely the interactions between NASS and the ARG
and to further develop and refine logistics procedures for the 1993 demonstration/pilot projects.
Some NASS procedures are documented in non-published sources such as ithe Interviewer's
Manual. Copies of these documents will be obtained by the ARG and will be available for
review at the ARG headquarters in Raleigh, North Carolina.
8.3 Specific Logistics Elements
Any EMAP logistics planning needs to consider fifteen elements (Baker and Merritt 1991).
These elements are discussed below as they apply to the four Pilot activities: the June
Enumerative Survey (JES), the Fall Survey, soil sampling and water sampling. Sampling will
take place during the Fall Survey.
8.3.1. Overview
Table 8-2 lists the major activities involved in developing the 1992 Agroecosystem Pilot and
identifies the responsible party for each activity. The flow of the major activities planned for the
1992 Pilot is diagrammed in Figure 8-1. This figure shows sampling and survey activities, on-
and off-frame activities with general locations, general data flow and responsible parties.
8.33. Staffing
The ARG will maintain a scientific and statistical staff for the analysis and synthesis of the
information collected. Appendix 1 lists the names and addresses of the Agroecosystem Resource
Group members. The ARG consists of a group of eight scientists located in Raleigh, North
Carolina and a number of other individuals at locations such as Athens, Georgia; Idaho Falls,
Idaho; Corvalh's, Oregon; and Las Vegas, Nevada. Pilot activities will be coordinated from
Raleigh where the Technical Director, Associate Technical Director and most of the indicator
leads are stationed. Raleigh is also where the North Carolina state office of NASS is located.
Responsibility for the development of indicators and indices of agroecosystem health will reside
8-2
-------
Table 8-2. Activities in the 1992 Agroecosystem Pilot Project
ACTIVITY
Statistical design
Selection of segments
Developing indicators
Obtaining ancillary data
Testing sampling procedures
Developing survey questionnaires
Writing manuals:
- Enumerator's/Interviewer's
- Supervising & Editing
Training enumerators:
- June Enumerative Survey (JES)
- December Survey and sampling
Equipment procurement - Survey
Equipment procurement - Sampling
Conducting the JES
Selection of fields, ponds and wells
Clover biomonitor
December S urvey
Initial summarization of survey data
Survey Administration Analysis
(includes cost estimates)
Sampling:
- Soil
- Pond water
- Well water
Comparison of existing vs. research wells
Soil processing
Soil analysis
Water analysis
Compiling of data
Indicator calculations and analysis
Sample Statistical Summary and
Interpretive Report
RESPONSIBLE PARTY
ARC Statistical Team
ARG Statistical Team
Indicator leads
ARG IM
NASS/ARG/Athens-ERL
NASS/ARG
NASS/ARG
NASS/ARG/Athens-ERL
NASS
ARG
NASS Enumerators
ARG Statistical Team
ARG/Extension Service
NASS Enumerators
NASS
NASS
NASS Enumerators
Athens-ERL
ARG
Contract Laboratory
Athens-ERL
ARG IM
ARG
ARG
8-3
-------
I
tt
>%
U2
O
8
I
<
CS
I
4)
00
S
8,-4
-------
with the ARG staff, which includes NASS personnel. Expertise in water analysis will be
provided by the Athens EPA Environmental Research Laboratory (ERL).
Interagency agreements are in place between EPA and both ARS and NASS. Memoranda
of Understanding are being developed between EMAP and both ARS and the Soil Conservation
Service (SCS) to cover overall cooperation for the Pilot and beyond. The work of administering
questionnaires and collecting soil and water samples will be done by NASS enumerators, hired
on a part-time basis through the National Association of State Departments of Agriculture
(NASDA). Most enumerators are local farmers, members of farm families, retired rural residents,
or other persons with an interest in agriculture. They are located throughout each state, which
expedites data collection and minimizes travel time and expenses. NASS will be responsible for
hiring and supervising enumerators and will handle payroll and other administrative functions.
For the Agroecosystem Pilot there will be a slight expansion of the standard JES -- segments
will be sampled on the EMAP hexagon design that would not otherwise have been visited [eight
questions (7a, 10,51,51a, 51b, 52, 52a, 52b) have been added to Section D of the questionnaire].
In the fall, survey and sampling activities will require enumerators capable of the physically
demanding work of sampling soil and water. It is most likely that individual enumerators will
conduct field and well sampling, and that teams of two persons will handle pond sampling.
Enumerators will operate according to NASS guidelines, including those regarding confidentiality,
as well as procedures outlined in the manuals to be developed by NASS and the ARG. It is
expected that at most 15 to 20 enumerators will be needed to collect the Pilot data.
One of the ARG members is also a professor in the Department of Plant Pathology at North
Carolina State University. Permanent and hourly employees of his program, along with members
of the ARG, will do the work of drying and grinding soil to be sent out for analysis. This effort
is referred to as the "soil preparation laboratory."
8-5
-------
Soil physical/chemical analyses and nematode counts will be done by contract laboratories.
They are responsible for their own staffing. The Athens-ERL has the staff for analyzing water
samples for chemical composition.
8.3.3. Design of Survey Questionnaires
Two questionnaires will be used for data collection during the Pilot The first will be the
June Enumerative Survey (JES), an annual NASS activity to determine land use, livestock
numbers, and crop stocks. The ARG will obtain land use/ land cover data from the JES (Section
5.4). In cooperation with NASS, the ARG has developed eight supplementary questions for the
JES. Two of the questions specifically ask about land uses (irrigated acreage and idle cropland
in government programs) not normally included in the questionnaire. Three other questions will
help determine the presence and use of farm ponds. The last three will determine the presence
and use of wells. These questions will be included in the entire North Carolina JES in 1992.
A portion of the JES (land use page plus pond and well questions) will be conducted on the Pilot
segments which were selected by the Hexagon Design (Appendix 5).
The ARG worked closely with NASS to design the Agroecosystem 1992 Pilot Study
Questionnaire (i.e., Fall Survey), found in Appendix 5. This will provide data on crop yields,
cropping sequence, fertilizer, pesticides, irrigation, tillage system and other management practices.
These questions are geared toward the selected fields within the selected segments (see Section
3.3). There will be 65 segments sampled on the Rotational Panel Plan and 51 on the Hexagon
Plan. These 116 segments are located in 83 of North Carolina's 100 counties.
8.3.4 Communications
During surveys, communication will involve established NASS procedures for maintaining
contact between the supervisory enumerator and the enumerators in the field. This will include
reporting of work hours and mileage, progress of surveys, and similar information. For sample
collection, additional lines of communication must be developed for sample tracking, general
problems and emergencies.
8-6
-------
Date mailed
Sample number
collected on /
mailed to the nematode
the above date.
Enumerators:
Date received
was
_J92 in
County and
laboratory on
uniy ) — —
Sample and equipment tracking. Each
soil sample will have a unique identification
code. As an enumerator sends each sample,
he or she will also mail a postcard (Figure 8-
2) indicating that the sample has been shipped
and the date on which it was collected and
mailed (Section 5.2.5). Room for special
explanatory notes (e.g., "sample very wet due
to recent rains") may also be provided. These
postcards will be sent to the ARC Information Fi8ure 8'2- E^pte of a sample-tracking postcard to be
sent to the ARG by the enumerators.
Manager (1509 Varsity Dr., Raleigh, NC
27606). Some soil samples will go to the soil preparation laboratory at North Carolina State
University, while others will go to the nematode enumeration laboratory. Logs of samples
received will be kept at both locations. The possibility of using bar codes and computer software
for sample tracking is being investigated.
The ARG will also record shipments of samples from the soil preparation laboratory
(Raleigh, NC) to the contract soil analysis lab (to be determined). The analysis laboratory will
keep a log of incoming samples sent to them and will report the samples they have received on
a weekly basis. Similarly, weekly contact with the nematode enumeration laboratory will allow
the ARG to keep track of the samples that have arrived there.
A scheme for tracking water samples will be developed in cooperation with the EPA-ERL
at Athens, Georgia.
Equipment must also be tracked. Each piece of equipment (or each set of pieces, e.g., for
soil probe accessories) will be marked "Property of the Federal Government" and enumerator kits
will be numbered. Enumerators will sign for their equipment when received and again when
returned.
8-7
-------
Emergencies and lesser problems. Emergency communication in the field will be handled
according to NASS protocol. The first priority is the health and safety of the enumerators; the
communication policy should reflect this, providing for contact with police or sheriff, fire
department, and ambulance services.
There are non-emergency problems which may arise in the field. Issues such as inability
to locate a site or collect a sample will be handled through NASS. If an enumerator needs to
replace lost or damaged equipment, he or she will be able to contact the supervisory enumerator,
who will contact the ARG (Section 5.2.4).
8.3.5. Training
NASS enumerators are part-time employees with a wide range of educational backgrounds.
Prior to participating in any data collection efforts, they undergo an intensive training program
in sample and data collection methods. Training of enumerators will be a joint responsibility of
the ARG and NASS. It will be documented that successful training has been completed.
NASS will be responsible for enumerator training for survey questionnaires. A three-day
training school will be held in Asheville, NC in mid-May for the North Carolina June
Enumerative Survey (JES). NASS will have this year's Interviewer's Manual ready by that time.
Training includes background information and a question-by-question review of the Survey
instrument. Group and one-on-one practice exercises are conducted to strengthen the
enumerators' knowledge of the questions.
A one-day training session will be held in Raleigh one week prior to enumerators going to
the field in the fall. Earlier there will have been a practice training session with several
enumerators, to identify weaknesses and make necessary changes in the training or procedures.
NASS will be responsible for training enumerators for the collection of post-harvest survey data;
the ARG, including members of the Athens-ERL, will train enumerators in soil and water
sampling techniques. These techniques will be taught in the classroom and field. An
8-8
-------
enumerator's manual for the fall collection of Agroecosystem data is being developed by NASS
with input from the ARG. A first draft will be prepared by mid-August, with the completed
version to be ready in October. It will include information on the background and objectives of
the Pilot Project and will define specific interview and sampling procedures for the fall survey
and sampling period.
8.3.6. Safety
The safety of NASS enumerators must be a prime concern and must be considered in
planning, in writing manuals and in giving training sessions. Conduct of surveys should not
present any hazards that are unfamiliar to the enumerator, but soil, well, and especially pond
sampling will require comprehensive planning. The safety issues are much less intense than for
other EMAP resource groups which often sample in remote locations. Enumerators live in the
same area that they work, so they should be aware of where such hazards as poisonous snakes,
fire ants, and tick-borne diseases (e.g., Lyme, Rocky Mountain Spotted Fever) may be
encountered. Common sense should guide decisions about dressing for the weather and working
under extreme conditions (e.g., high temperatures, lightning storms).
Two areas will require special training: proper use of equipment and water safety. Trying
to drive a soil probe into hard ground by repeatedly stomping on it can cause knee injury. This
and other cautions need to be mentioned in the training session. Water safety will be covered
for those enumerators who will be sampling ponds. Life jackets will be required of all those
sampling ponds, even if the enumerator is working from the bank. Those using boats will be
instructed in their proper care and use. Finally, all enumerators will wear blaze orange vests (or
orange life jackets) while taking soil or water samples because the sampling period coincides
with the deer hunting season in North Carolina.
Three other areas need to be discussed with NASS before a decision is made to include them
in the safety plan and training. (1) According to the Guidelines for Preparing Logistics Plans,
"First aid and CPR training are required for all personnel, especially those who will be working
8-9
-------
in remote locations" (Baker and Menitt 1991, Section 2, Page 11 of 20). The guidelines also say
that the safety plan is supposed to designate the American Red Cross First Aid textbook as a
guide for first aid and CPR. Whether these are binding directives and how they should be
applied within the Agroecosystem Program must be investigated. (2) During NASS Objective
Yield Surveys, enumerators work in fields of standing crop; therefore, pesticide safety is a part
of their training (USDA-NASS 1991b and 1991c). It will be discussed with NASS whether this
is warranted for the type and timing of the sampling that will be done for the ARG. (3) A policy
may need to be developed for the unlikely event that an enumerator runs across a marijuana field,
moonshine still, poaching, or other illegal activity.
The soil analysis and nematode enumeration laboratories will be contractors, responsible for
their own safety plans. The soil preparation laboratory is an activity run through the university
research project of one of the ARG members. Safety will be the responsibility of that scientist
and his technician. Chemical analysis of water samples will be done at the U.S. EPA
Environmental Research Laboratory, Athens, Georgia. That laboratory also has its own safety
plan.
8.3.7, Sampling Schedule
Data will be collected by NASS enumerators for the ARG during the June Enumerative
Survey and during survey and field visits in the fall. The period of field activity for the JES is
mid-May to mid-June. Administering the Agroecosystem questionnaire and taking soil and water
samples will be done from November through early December. NASS will be responsible for
the development of detailed sampling schedules within each survey or sampling period.
8.3.8. Site Access and Reconnaissance
NASS has an excellent record with the agricultural community at the national, state, and
local levels. Obtaining permission for site access is rarely a problem. During the JES,
enumerators locate and interview all farm owners or operators in sampled segments. From
8- 10
-------
special questions added to the JES, the ARG will know whether farm ponds or wells are present
on the segment and what they are used for. During the Fall Survey, enumerators will solicit
permission to collect soil and water samples.
Physical access to fields and wells should not be difficult, but thick brush or muddy banks
may hinder access to certain ponds. Therefore, the pond to be sampled should be visited
immediately after permission is obtained, to determine if such problems exist and to make
necessary preparations. Enumerators should not cut brush or otherwise disturb the site.
Enumerators will keep notes on any problems they encounter, such as impossible access.
8.3.9. Procurement and Inventory Control ,
The ARG will provide equipment and supplies for the collection of soil and water samples
and for the transportation of samples to contract analytical laboratories. NASS will provide all
survey instruments and supplies (e.g., aerial photos) associated specifically with the
questionnaires. A list of the equipment to be found in the soil sampling kits is found in Section
5.2.4. At least 12 such kits will be needed for the Pilot. Four soil probe sets are already on
hand, but they are not yet marked "Property of " or numbered. Types of mailing
containers will be chosen after a soil analysis laboratory is selected and after consultation with
NCDA about rules for shipping soil from quarantined areas. Soil samples sent to the nematode
laboratory will be enclosed in a padded envelope, lined with bubble-wrap, to minimize the impact
of sample handling on nematode viability. Pre-printed labels for each sample will be produced
by NASS in consultation with the ARG.
Probes and other soil sampling equipment for the Pilot will need little storage space and will
be kept at or near the ARG headquarters in Raleigh. Enumerators will sign out equipment and
receive supplies at the training session. Equipment will be returned at the debriefing following
the Pilot. Enumerator kits for water sampling will be more complex, especially since boats are
to be used. Procurement and inventory of this equipment will be developed in conjunction with
the Athens-ERL.
8-11
-------
8,3.10. Field Operations
NASS will be responsible for all data and sample collection activities during both the June
Enumerative Survey and the Fall Survey. The ARG will be responsible for all field activities
involved in the development of new indicators during the initial stages of testing: (See logistics
subsections in Sections 5 and 6.)
Surveys. Survey logistics will be the responsibility of NASS, which has years of experience
in this area. NASS uses specialized computer software to track the progress of their surveys.
Details on how the enumerators are to conduct the surveys will be found in the manuals which
will be developed for both the JES and the Fall survey. For an example, see the Interviewer's
Manual for the 1991 JES (USDA-NASS 1991b). The period of field activity for the JES is mid-
May to mid-June. The Fall Survey and associated sampling will take place from November to
early December. Questions on the Fall Survey will be directed toward individual fields. NASS
and the ARG Statistics Group will select the fields, using the scheme described in Section 3.3.1,
and will outline them on aerial photographs that will be given to the enumerator.
Soil sampling. Soil and water sampling will be done during the same period as the Fall
Survey. To save time and travel, soil samples will be taken right after the questionnaire is
completed and permission is obtained. Water may or may not be sampled during the same visit.
Soil sampling will be done as outlined in Sections 3.3.2, 5.2.4, 6.1.4 and Appendix 6. A very
brief description is presented here.
The field will be sampled with 20 soil cores taken at equal distances along a 100-yard
transect. Enumerators will have the aerial photographs, sampling equipment (including probes,
bags, labels and mailers) and the number of paces (printed on a label on the survey form) needed
to locate the transect midpoint. This midpoint will be determined using a modification of the
NASS method for locating objective yield plots. The enumerator will take the assigned number
of paces along and into the field, and will orient the transect at a 45° angle to his or her path of
entry into the field. Samples will be taken at five-yard intervals along the transect. The 20 cores
8-12
-------
will then be composited and mixed. Soil clumps are to be broken apart gently. Samples for soil
analysis (500 cm3) will be drawn from the composited sample, labeled, and packaged. The
enumerator will ship the samples that same day or early the next day via Federal Express (call
1-800-238-5355 for pick-up) to the Soil Preparation Laboratory at N.C. State University. Postage
will be paid using a Federal Government account through the Air Resources Research Consortium
at N.C. State University. Delivery on weekdays and weekends should be requested on the
mailing label. Subsamples for nematode determination (550 cm3) will be drawn from the
composited samples from fields on only one of the two sampling designs (Rotational Panel Plan).
These subsamples must be kept in an insulated box (ice chest) away from extreme heat or cold
until they can be shipped to the enumeration laboratory. They should be sent between Monday
and Thursday so that they reach the laboratory on a weekday. If this is not possible, the
enumerator will notify the laboratory so that the sample can be handled properly when it does
arrive. After shipping any sample, the enumerator will complete and mail the sample-tracking
postcard to the ARG Information Manager (see Figure 8-2). Procedures for labeling, handling
and tracking samples are still being developed by the ARG with input from NASS.
A preliminary test of soil sampling procedures showed that samples shipped via Federal
Express on Monday-Thursday from various points in North Carolina reached the destination
laboratories in Raleigh, NC and Corvallis, OR in one or sometimes two days. During the study,
the time needed to actually sample a diagonal transect (starting at the end, not the midpoint)
averaged 22 minutes.
Water sampling. Methods for sampling, handling and transporting water samples are not
yet finalized (Section 5.3). It is also not yet known which analytes are to be measured and which
segments will be selected for pond and well sampling. JES data will be the basis for selecting
ponds and wells within segments (Section 3.3.4). This may be done before the enumerator goes
out, or the enumerator may have to make the selection in the field.
There are two pond sampling methods under discussion (see Sections 3.3.5 and 5.3.2). The
standard method requires a boat. A logistically easier "bank" method will be tested, and
8-13
-------
compared to the "boat" method. The "bank" method will involve the use of a long pole (-16
feet) that allows an enumerator on the bank to lower a sampler into the pond. Multiple water
samples from each pond will form a composite sample from which analysis samples will be
taken. Water samples from wells will be easier to take, but must only be taken after an
appropriate purge (Section 5.3.2).
Once the list of analytes has been chosen, detailed water sample handling procedures will
be developed. Samples will need to be shipped in amber glass containers within insulated
cartons. Also, instant cold packs will be used to keep the samples near 0°C (Section 5.3.4). The
EPA Region IV SOP manual (U.S. EPA 1991b) will be the guide for water sample handling
(Section 5.3.2). Samples will be shipped overnight express (Federal Express) directly to the
Athens-ERL. Methods of identifying and tracking samples have yet to be developed. Procedures
for return of equipment must also be discussed.
8,3.11. Laboratory Operations
Procedures to be used at the soil preparation laboratory (N.C. State University) are found
in Section 5.2.4 and Appendix 4. As samples are received, they will be logged in and spread in
metal pans to air dry. Once dry, soil will be ground in a hammer mill and mixed thoroughly.
Subsamples will be sent to the soil analysis laboratory, and the remainder of each sample will
be archived in case re-analysis is needed. Two subsamples will be taken from some of the
samples, as a check of the analysis laboratory's precision.
The analysis laboratory has not yet been selected, but will be asked to test the soil using the
procedures listed in Table 5.2-11. Detailed laboratory procedures are described, in Appendix 4.
Log books for sample tracking will also be required.
Information on procedures for nematode enumeration are given in Section 6.1.4. The
nematode enumeration laboratory will log in samples and store them at 15°C. Samples will be
processed within 14 days of receipt From each sample, 50 cm3 will be used for gravimetric
8-14
-------
determination of soil density and moisture content. The other 500 cm3 will be used for nematode
extraction by elutriation. Nematodes will then be identified to trophic group. The laboratory of
Dr. K.R. Barker, NCSU, will handle nematode extraction and enumeration for the Pilot.
Water samples will be analyzed for contaminants at the Athens-ERL. The techniques to be
used will depend on what chemical species are of interest (see Sections 5.3.5 and 6.3.5).
8.3.12. Waste Disposal
During sampling. There are few hazards associated with the soil or water samples, but some
precautions need to, be taken. Soil should be cleaned from probes, buckets, and other items
before the next field is visited. This will reduce the chance of spreading weeds, nematodes, and
other soil-borne pests. It is especially critical in areas under quarantine for witchweed (Striga
asiatica Lour.) or imported fire ant (Solenopsis invicta Buren). NASS has a compliance
agreement with federal and state authorities responsible for quarantines. This may need to be
expanded to include soil samples. Soil from all counties will be treated and packaged as if it
came from quarantined areas.
Certain lakes in Wake County and a few isolated bodies of water in 11 other counties
contain hydrilla (Hydrilla verticillata (L.S.) Royle), a noxious aquatic weed (Gene Cross, NCDA,
personal communication). After taking pond water samples in these counties, enumerators should
remove all weeds that may have stuck to the sampling equipment, including the boat.
At the laboratories. The first disposal issue for the laboratories is that they not discard any
sample too soon. The soil preparation and analysis laboratories will be required to archive
samples until data have been determined to meet quality objectives (Section 5.2.5 and Appendix
4). Some extracted nematodes will be preserved and stored, in case more detailed identification
is needed (Section 6.1.5).
8-15
-------
All laboratories handling soil samples from the Pilot will be required to follow specified
procedures before disposing of soil from areas under quarantine for witch weed or imported fire
ant. These procedures specify the temperature and duration to heat the soil (or soil screenings)
to kill the pests. The soil preparation laboratory, run by the ARG, will sign a compliance
agreement with the Animal and Plant Health Inspection Service (APHIS) and the North Carolina
Department of Agriculture (NCDA) guaranteeing that the procedures will be followed. The
nematode and soil analysis laboratories will be chosen from among a list of laboratories that
follow acceptable protocols.
Witchweed and fire ant quarantine counties are located in southeastern North Carolina,
though many of these counties are only partially under quarantine. For simplicity, the contract
laboratories will handle all samples as if they came,from quarantined areas.
Waste disposal of any other hazardous materials, such as reagents used to test soil, at
analytical and enumeration laboratories is to be^ done ma.responsible way according to the
methods generally used by those laboratories.
8.3.13. Information Management
Data collected during the Pilot, as well as massive amounts of existing data (e.g., Natural
Resources Inventory, Agricultural Census, State Soil Survey Database, weather data, GIS data)
will need to be managed by the ARG Information Manager, through the Agroecosystem
Information Center (AIQ.. Cooperation with NASS and the logistics of managing data transfer
between NASS and the AIC are being developed. Pilot Survey data will go first to NASS; while
soil and nematode analysis data will come first to the ARG. A critical concern is to ensure the
confidentiality of data from individual farms. These issues, along with hardware and software
requirements, are covered in Section, 9. Some of the logistics for the acquisition of the data to
be used for land use/ land cover and landscape descriptors are covered in Sections 5.4, 6.2.3 and
6.2.5. These include data from satellites and possibly aerial photography.
8-16
-------
An additional information/logistics issue is how to report back to the farm operators. They
should be given the results of surveys and analyses. One question is how much of an
interpretation should be put on the results. Giving raw numbers may be meaningless, so some
interpretation is needed. Should it go so far as to recommend remedial measures for problem
situations, or refer the farmer to sources of such recommendations? Even negative results can
present a problem. A farmer might assume that his well or pond is certified free of
contamination just because there were no detectable levels of the contaminants which were of
interest for the Pilot (observation of Dr. W. Payne, Athens-ERL).
8.3.14. Quality Assurance .
A streamlined logistical operation helps to ensure that data are collected properly and that
good records are kept. Also, a number of QA procedures for the Pilot will have to be
incorporated into the logistics area. Some of these are already part of NASS operations (eg., the
work of supervisory enumerators and call-backs to verify questionnaires). Others are sampling
operations that will be added for QA purposes. For example, a second composite soil sample
will be taken from every sixth field. Half of those second composite samples will contain 40 soil
cores, two from each point on the transect. This larger sample has enough soil that it can be split
at the preparation laboratory and sent as duplicate samples to the analysis laboratory. Section
7 gives further details on Quality Assurance for the 1992 Pilot.
8.3.15. Cost Tracking
One of the objectives of the Pilot is to compare the efficiency (cost and precision) of the
NASS Rotational Panel and the EMAP Hexagon Design (Section 3.2). To be sure that every
applicable cost is included, each step in the survey process for each Plan will be identified and
placed in its proper sequence in a flow chart. Information on costs will be recorded by NASS
for every step in the Pilot: frame development, sample selection, preparation of maps and aerial
photographs, and conduct of the sampling and surveys. Some of these are discussed in Sections
3.1.1, 3.1.2 and 3.2.1. It is standard practice for NASS enumerators to report their time and
8-17
-------
mileage during surveys, but the cost of office work by NASS will need to be included.
Allowance will be made for the fact that certain activities (e.g., soil sampling for nematodes,
water sampling) will be done only on one design or the other. Costs of conducting each NASS
survey are reported in the "Survey Administration Analysis" (see below). The difficult part will
be estimating costs (per track and per segment) attributable to the EMAP Hexagon Design.
8.3.16. Review of Logistics
During the Pilot Project several reviews of the logistic plan and procedures will be
conducted. Members of the ARG, NASS, and the EMAP Technical Coordinator for Logistics
will participate in these reviews. The purpose of the reviews will be to identify areas of missing
information associated with the monitoring program and procedures for incorporating this
information; also, to re-examine all phases of the logistics plan.
After the Pilot, enumerators will be debriefed to determine strengths and weaknesses in the
logistics. Enumerators will return sampling equipment at this time. A post-Pilot meeting will
also be held with NASS administrators. A Logistics End-of-Season Summary Report will
document problems and propose solutions.
Within 12 months of each survey conducted by NASS, a report called a Survey
Administration Analysis is produced which contains specific information about the survey (i.e.,
response rates, cost accounting). Estimated completion dates are November 1992 (for the
previous JES) and August 1993 (for the Pilot Fall survey).
8.4. Logistics for the Biological Ozone-Indicator System and the Well Comparison Study
Logistics for two aspect of the Pilot have not been documented above: the use of ozone-
sensitive and ozone-resistant clones of white clover (Trifolium repens L.) as biomonitors and the
comparison study of existing wells with research wells. These projects will be conducted in a
slightly different way from the rest of the Pilot The ozone biomonitor system will be tested in
8-18
-------
conjunction with the Cooperative Extension Service, and plants will be deployed in the North
Carolina coastal plain and piedmont. This design is still being developed. Logistical details may
be found in Section 6.4.
The well study will compare the chemistry of groundwater samples taken from existing wells
with the chemistry of water from specially drilled research wells. The goal is to determine if
there is a bias in samples taken from existing wells. This study will be performed by members
of the Athens-ERL in a selected area of the coastal region of North Carolina. A full description
may be found in Section 6.3, with logistics details in Sections 6.3.2 and 6.3.4.
NASS Enumerators
Equipment:
Acquisition, Storage, Inventory
Recruited
Provided Equipment
I
Identify Fields
t
Sample Soil
Mix, Package, Split
X
Send Postcard Soil to
for Tracking Nematode Lab
Log In
f
Extract Nematodes
To Field
Return Equipment
•Debrief Enumerators -
Discussion with
NASS Administrators
"Soil to
Soil Prep. Lab
Log In
Samples
x
Identify Ponds
— Sample Water
Composite and
Subsample
Package with
Cold Pack
t
Send to Athens-ERL
Discar
Dry, Grind, Package
4Z^ ^*- *
'd Soil Count Nematodes Send to
Analytical Lab
Archive Analyze
Archive Analyze
Survey
NASS HQ
Data to ARG
Figure 8-3. Logistics flow chart for the 1992 Pilot Project.
8 - 19
-------
-------
9. Information Management
9.1. Introduction
The Agroecosystem 1992 Pilot Project will require that data be obtained, stored, manipulated,
integrated and analyzed. These new and existing data will come from many sources, including
joint ARG-NASS data collection efforts and from other EMAP Resource Groups, other
government agencies, cooperating non-governmental organizations (NGOs), and academic
institutions (Figure 9-1). The information collected, together with existing data, must be
integrated in such a way as to make meaningful analysis possible. The focal point of this
integration will be the Agroecosystem Information Center (AIQ. The AIC will be developed
during the Pilot to provide computer equipment, data storage, data processing, software
development and data communications facilities to the ARG.
Other
Existing j
Data
Agroecosystem
Data Collected
With NASS
Agroecosystem
Information
Center
Other
EMAP
Data
Agroecosystem Resource Group
PRODUCTS
o Reports
0 Data
Figure 9-1. Overview of the flow of data through the AIC
9-1
-------
A major emphasis of the Agroecosystem Pilot Project is the development of a close working
relationship with USDA-NASS (Section 1.2). As discussed in Section 3.2, NASS uses an area
frame to gather data on crop acreage, cost of production, farm expenditures, crop yield, specialty
crops, livestock production, chemical usage, irrigation, water quality and other items of interest
to the agricultural community. Statistics are compiled and reported annually from the June
Enumerative Survey (JES) using some 16,000 Primary Sampling Units nationwide. A primary
objective in the Pilot will be developing and fine tuning the logistics and cooperation required
for moving and integrating data from several sources into a data management system that is
available to the ARG. Group members will perform statistical, modeling, geographical and other
types of analyses, using the data.
Confidentiality of data, and consequently data security, are particularly critical issues to the
ARG/NASS relationship. Meeting the program objectives requires that data be collected from
individual farmers and corporations. Because these NASS data, at some level of summarization,
are then to be available outside the confines of NASS facilities, there must be a policy and
mechanism which continues to protect the privacy of the individual respondents. As a part of
the Pilot, the ARG will work closely with NASS to establish methods and procedures -for
maintaining strict confidentiality and security of all microdata (i.e., data which can be associated
with individual growers and operations). .
9.2. Information Sources and Flow
Information that will be used for analyses and reporting in the pilot will originate from two
general sources: those data actively collected at the farm field sites and those existing data (both
current and historical) that have been collected by other agencies.
Data for the 1992 Pilot will be collected under an Interagency Agreement developed with
USDA-NASS under which NASS enumerators collect all of the agricultural field level
information. This information will consist of both survey data and physical samples for
laboratory analysis (Figure 9-2). The enumerators will operate within the NASS organization,
9-2
-------
using procedures selected and developed jointly by the ARG and NASS. The survey data will
be entered, verified, validated, and stored on NASS computers. The soil laboratory data will be
sent as an ASCII file to the ARG Information Manager. The ARG Information Manager will
perform validation tests on the data. Once validation is complete, the laboratory data will be sent
to NASS for integration with the survey data. Only aggregated or summarized data will be
transferred to the AIC within the constraints of NASS confidentiality agreements.
Survey Design 1
,
DataC
by ]>
haBBBB
Cor
Sen
NAS
/
ollection | NASS
T , nr • ^M. Data
JASS I ^
^^•^J Center
'"I
Data Integration 1
and Analysis at L^
ifidentialty NASS by ARG |
:ening and ^^••^••••J
Agroecosystem 1
Information 1
Center 1
Field Sampling Prodcedures |
.
Sample Collection 1
by NASS §
\
i
1 Lab Analyses 1
Figure 9-2. Flow of data collected by NASS to the AIC
From the standpoint of information management, working with NASS is important for a
number of reasons.
o Over time, NASS has developed a relationship with the agricultural community which
will greatly facilitate the collection of data.
9-3
-------
o NASS provides the assurance of data confidentiality to individual farm operators. This
allows the organization to collect data which farmers might otherwise be reluctant to
supply for fear of legal or regulatory action.
o NASS has a fully developed infrastructure for the collection, recording, summarization,
analysis and publication of agricultural data, including strict quality controls. Use of this
infrastructure greatly reduces the expenditure of resources on the development of
duplicate logistics and QA procedures.
o NASS has developed the computer resources to organize, analyze, and quickly report on
large volumes of data. Use of these resources may reduce the overall need for data
processing within the ARG.
In addition to field data, a broad array of existing data will be required for the 1992 Pilot
Project (Figure 9-3 and Table 9-1). The ARG is committed to the use of existing data whenever
possible, assuming the scope and quality of the data are sufficient for our needs,. Although there
may be some effort required to transform existing data to conform with EMAP standards, this
effort is usually substantially less than that required to collect new data. The existing data we
anticipate using, fall into two major categories and have a variety of uses. These are:
o Physical and biological parameters: used to provide complementary data, verify values
of collected data, provide a basis for the implementation of summarized data (validation),
and for indicator research.
o Geographic based data: used for boundary establishment, provide spatial distributions,
perform geographic visualization, and for indicator research.
The physical and biological data, although distinct from geographic data, are frequently
V
associated with one another in the same database. This is often referred to as geo-referenced
data. For example, meteorological data, which are critical to the development and interpretation
of any crop yield or productivity measures for an area, are associated with a collection point.
The use of existing data also permits the analysis of historical trends. In this way, it may be
feasible to validate and correlate measurements associated with specific endpolnts by predicting
9-4
-------
present conditions using historical data. Existing data will be used to develop expected values
for performing verification and validation of both survey and sample data (Figure 9-4). The
ARG will import data as needed and appropriate from other EMAP efforts as well as other
agencies and organizations to support pre- and post-Pilot activities.
EMAP
Landscape
(GIS) Data
Other EMAP
Resource Group
Data
Other Data
at EMAP
Information Center
Agroecosystem
Information
Center
NASS
Data
Center
t
Other Existing Data
Figure 9-3. Flow of data from other EMAP sources and other agencies and institutions to the AIC and
NASS data center for integration
9.3. Confidentiality of Data
In order to protect the rights of individual respondents, legal confidentiality provisions (Table
9-2) apply to all data collected by NASS. The NASS cannot release microdata; data are currently
available, in most cases, at the county level. The United States Department of Commerce's
(USDC) Census of Agriculture is also legally subject to confidentiality provisions, and the USDA
Soil Conservation Service's (SCS) National Resource Inventory (NRI) follows confidentiality
9-5
-------
Table 9-1. Examples of existing data to be used for the 1992 Pilot Project.
Description
Weather and Climate Data
State Soil Survey Database
(N.C. derivation of SOILS-5)
National Resources Inventory
Herbicide Use Database
Ag. Land Use and Cover Data
Census of Agriculture
Soil Ratings for Pesticide Loss
Major Land Resource Areas
Albermarle-Pamlico Watershed Coverages"
Aerial Photography
Source
NOAA
SCS
SCS
Resources for the Future
NASS
USDC
BRG/Data & SCS
SCS
NCCGIA
ASCS
restrictions. Table 9-2. summarizes some of these policies. The rationale behind such assurances
is clear. Confidentiality laws protect individual respondents from prosecution which might
otherwise result from their participation in a data collection effort. Without such assurances,
respondents may be hesitant to comply with any survey or data collection efforts, either voluntary
or legally required. Also, without these assurances, respondents may be more likely to falsify
information on surveys. Violation of this confidence would result in loss of NASS credibility
with survey respondents and seriously hamper future data collection efforts. Hence, NASS is very-
serious about maintaining data confidentiality.
The current view of the. ARG with respect to these confidentiality provisions is positive. In
a review of confidentiality in EMAP, Franson (1990) writes that the EPA is presently unable to
issue a blanket statement of confidentiality for EMAP data. Requests for EMAP data from the
9-6
-------
Raw Data
Verified Data
Validated Data
Check Values for Reasonableness
eg.: Range Check
Duplicate Entries
Format Check
Inappropriate Codes
Internal Consistency
Check Data for Reasonableness
eg: Comparison with Historical Data
Statistical Analyses
Figure 9-4. Use of existing data to perform validity checks on data
regulatory arm of EPA, or from other agencies, corporations, and individuals, must be reviewed
on a case-by-case basis. Farmers are unlikely to provide data without a confidentiality agreement
In fact, the question of whether farmers would cooperate with any team identifying themselves
with the EPA, even with promises of confidentiality, should not be lightly dismissed.
Although the microdata from agencies employing confidentiality provisions are not available,
there are solutions which allow the ARC to make use of the data collected by NASS.
Aggregated data: Whereas many agencies will not release their microdata, they will all release
data aggregated at various levels (Table 9-2). The goal is to aggregate the data in such a way that
individuals cannot be identified. Obviously, for the Agroecosystem Program, the lower the level
of aggregation, the better (i.e., county-level data are better than state-level data.)
9-7
-------
Table 9-2. Summary of confidentiality provisions of several government agencies with data of value to the
Agroecosystem Resource Group.
Organization
USDA National Agricultural
Statistics Service
USDC Census of Agriculture
USDA Soil Conservation
Service (National Resource
Inventory)
US Environmental Protection
Agency
Policy
Public Law 99-198.
No release of data with identity of individual
respondent.
US Code Title 13.
No release of data with identity of individual
respondent.
No release of exact location (Primary Sampling
Unit) at which data are collected.
Freedom of Information Act requests handled
on a case-by-case basis.
Lowest Level of
Aggregation Normally
Available
County
Zip Code (5 digit)
County
Varies
Analysis requests: The owning agency may accept requests for tabulation and analysis from
another agency. The analysis would be performed by the owning agency's personnel, the results
returned to the requesting agency. Release of any confidential material would be strictly avoided.
Deputization: It is possible, at least with some government agencies (including NASS), for an
individual to be deputized by that agency. Deputization requires completion of a non-disclosure
agreement A deputized individual is permitted to access the data for the purpose of performing
analyses. Typically, the analysis would have to be performed at the owning agency's facilities.
Only aggregated results may be removed, and are subject to confidentiality screening.
All of these possibilities will be explored by the ARG with NASS, and any other agencies
with such provisions, during the Pilot program. Currently, the view of the ARG is to treat this
as a true Pilot Project; data will not be released, in any form, to anyone outside of the EMAP
until it has been summarized into a publishable format. Regardless of the confidentiality
provisions, it is our belief that because of the preliminary nature of these data, they would not
be of use to others, except in a final publication format.
9-8
-------
9.4. Data Integration and Management
The integration of EMAP data with data subject to confidentiality provisions presents a
unique challenge that can be resolved only through close interagency cooperation. The NASS
data are used for economic forecasts which have the potential for affecting the livelihood of
many people. They are closely guarded and access is severely restricted and carefully monitored.
No microdata may be removed from NASS facilities, and computers containing microdata may
not be connected to foreign networks. During the pilot program NASS will allow the ARG to
install a workstation at the NASS facilities in Raleigh, NC. ARG members will be able to
examine and analyze NASS microdata using the workstation. Integration of data from other
sources may be accomplished by loading that data onto the ARG workstation at the NASS
facilities and performing the required analyses on that workstation (Figure 9-3). Although this
mode of operation will suffice during the pilot program, other approaches will be explored for
future pilots, demonstrations and implementations.
Because of the data confidentiality and security requirements discussed previously, the task
of data management becomes paramount In order to coordinate and facilitate the movement,
integration and selection of collected and ancillary data, a full-featured relational database
management system (RDBMS) must be employed. This becomes especially critical when ARG
members require different "views" of the data so that different analyses can be performed on
various subsets. Carefully constructed data dictionaries are essential to maintaining flexible
access to all of the data. Another important concept to be established and tested during the pilot
is that of maintaining metadata associated with the collected data. These metadata will provide
different characteristics of the data (i.e., collection methods and units), which will furnish
invaluable information when the data are evaluated in the future.
9.5. Data Access
Providing ARG members access to the pilot and ancillary data in a convenient and organized
manner will prove to be the foremost challenge of the AIC. Individual ARG members are
9-9
-------
currently located in several cities throughout the United States (Appendix 1). Because of the
requirements for data storage locations (Figures 9-2 and 9-3), the logistics of locating, accessing
and transporting that data to the individual investigator will require a carefully planned and
designed information system.
Although we anticipate no release of pilot data outside of EMAP, eventually aggregated data
from demonstration projects and implementation will be made available to outside groups and
agencies. This will necessitate identifying what is available, where it is located, its characteristics
(metadata) and how to obtain it Development of this reference, the data catalog, will be a
component of the Pilot Project Plans call for working cooperatively with the Information
Management Committee (JMQ of EMAP, which is attempting to standardize the process for
cataloging EMAP data.
9.6. Hardware and Software Requirements
In order to establish and further develop the AIC in support of the Pilot, additional hardware
and software procurements are anticipated. Listed in Table 9-3 are the items of major
significance that the ARG will purchase to use in the 1992 Pilot Not listed in the Table are the
smaller items (Le., software and computing supplies). Although a great deal of thinking has gone
into planning the AIC, unexpected situations can arise that change the data processing
requirements of the group. One such situation is the need of the ARG for ancillary data. The
size of the datasets and computing resources required for their transformation and integration are
not completely known at this point. As indicated in Table 9-3 a substantial upgrade of both disk
and tape capacity is planned in response to increased data storage requirements.
Two more workstations will be needed for the pilot One of these will be located at the
NASS data center office for analyzing and aggregating confidential data. Additional personal
computers will be purchased for staff that currently are lacking them. Because; of the travel that
will be involved with the Pilot, a notebook computer for the ARG will prove a valuable resource
for documentation and communication.
9-10
-------
Several software requirements are critical to the success of the Pilot A major upgrade of
ARC/INFO is expected in early 1992. This upgrade will provide a point-and-click interface as
well as other changes to improve ease of use. An additional copy of ARC/INFO and SAS will
be required for the workstation at NASS. In order to allow adequate time for software
development and testing, a RDBMS must be procured soon. At this time, EPA has not awarded
the contract for the RDBMS.
Planning is currently underway for the installation of a local area network (LAN) for the
ARG facility in Raleigh. The LAN will provide more convenient access to, and faster movement
of, the data associated with the Pilot Preliminary plans call for attaching the LAN to the Internet
via the North Carolina State University campus fiber optic backbone. This connection will
permit the Raleigh location to exchange data and information with ARG members at other sites
and with the EPA laboratories. This link will be invaluable to the ARG during summarization
and assessment activities of the Pilot With the Internet connection established, ARG members
all over the country will have interactive access to the AIC. In the future, when the EMAP
Information Center (EIC) will become the repository for ancillary data and resource group data,
the link will permit interactive uploading and downloading of data from other resource groups
and cooperating agencies.
Table 9-3. Hardware and software requirements to support the 1992 Pilot Project
Item
PC and Workstation
Requirements
Peripherals
Major Software
Requirements
Telecommunications
Requirements
ARG (Varsity Drive, Raleigh)
Development workstation
80386 PC (2)
Notebook PC
1.2 Gb Disk Drive (2)
4mm DAT Tape Drive
CD/ROM Drive
RDBMS
ARC/INFO upgrade
ARC/INFO GRID module
ARC/VIEW module
LAN for ARG
Connection to Internet
ARG/NASS HQ (Raleigh)
GIS workstation
80386 PC
Plotter
ARC/INFO
SAS
9- 11
-------
-------
10. Resources and Implementation
10.1. Introduction
The Agroecosystem Resource Group (ARG) has developed a five year program strategy
(Heck et al. 1991) for implementation of a suite of indicators for monitoring agroecosystem status
and trends. This five-year period (1991-1995) includes time to test concepts relating to design,
indicators, data analysis, QA, logistics and information management at the pilot and
demonstration program stages. A primary emphasis is the development of close working
relations between personnel from NASS and the ARG, so that issues relating to design, QA,
sampling, logistics, information management, and data analysis can be identified and addressed.
The first stage of the program (1990) encompassed the evaluation of: 1) statistical designs,
2) existing monitoring programs (i.e., NASS, SCS, ERS), 3) assessment endpoints and associated
indicators (availability, validity, variability, cost), 4) data management and analysis techniques
and 5) derived outputs. During 1990, a conceptual national monitoring plan was also develope .
The second stage involves the design and execution of pilot and demonstration projects prior to
implementation. The 1992 Pilot Project in North Carolina, will test all aspects of the monitoring
program for a selected suite of indicators. Results will be utilized to develop a regional
demonstration of all program elements in the Southeast (SE) and a Pilot in EPA Region 7 for
1993. The pilot and demonstration projects will address specific concerns of the different
geographic areas of the country.
This Section addresses specific budgetary and personnel resources, and tasks planned for the
1992 Pilot. The Section does not address details as to how the various tasks and plans for the
Pilot will be accomplished, since the details are contained in earlier sections of this Plan.
Timelines are not included in this Plan for the Pilot but are being developed by ARG members
for inclusion in the 1993 Demonstration/Pilot Plans.
10- 1
-------
This section briefly addresses activities and budgetary and personnel resources required to
proceed with a full Southeast (eight state) Demonstration and a full Region 7 (four state) Pilot
in 1993.
10.2. Importance of the Pilot
It is essential for this Pilot Project to be considered a research Pilot with sufficient flexibility
to try a number of innovative approaches to all facets of the Pilot. This is important, if we are
to continually improve the various components of the monitoring approach in preparation for
implementation on a regional/national basis. The Pilot, as planned, will permit a critical
evaluation of the monitoring design, individual indicators, data analysis and integration, logistics,
QA and information management in preparation for the 1993 Program elements (Heck et al.
1991). The Program, as designed, allows for the orderly establishment and rigorous evaluation
of preliminary protocols and for the full utilization of existing data bases and networks.
10.3. Tasks and Schedule for the Pilot Project
The principle tasks associated with the Pilot Project are listed in Table 10-1 with a schedule
for completion of the tasks. The ARG expects to follow the time schedule closely to assure a
successful Pilot and permit a complete development of plans for 1993. An activity chart (Table
10-2) is shown that addresses all aspects of the planned ARG activities for 1992. These are
shown without timelines.
10.4. Funding and Personnel Resources and Products
These are shown for all activities associated with the ARG in 1992.
10-2
-------
Table 10-1. Tasks with schedule for conducting the Pilot Project - NC Pilot Plan (1992-93)
Tasks
1. Supply procedural manual for enumerator training for Fall survey
and sampling
2. Use suite of indicators developed in this Plan.
3. Assure that logistic, QA and information management strategies
are in place.
4. Participate in the NASS Enumerator Training Schools: a) May -
procedures for the JES, b) October - procedures for Fall Survey
and sampling.
5. Obtain all necessary equipment and materials for the pilot
6. Sample 1 16 NASS segments using NASS personnel, logistics, QA
and data management protocol.
7. Work with NASS on data management and data analysis.
8. Send soil and water samples to contract laboratories for analysis
and data return.
9. Compare the two design approaches through sampling of units in
the field; cost, variance, biases; determine covariance structure to
refine DQOs.
10. Data analysis: provide statistical summaries, compare cumulative
distribution functions, explore spatial distribution patterns, and
examine statistical properties.
11. Develop data summary to derive initial indices to classify
agroecosystems as "healthy" or "unhealthy".
Schedule
(1992)
July
June-Dec
Mar-Sept
May
Oct
Apr-Oct
June
Nov-Dec
(NASS Survey
Dates)
June-Dec
Mar 1993
(NASS time
periods)
Nov-Jan 1993
Dec-Mar 1993
May 1993
July 1993
10-3
-------
Table 10-2. 1992 Activity Chart for the ARC.
1. Primary Voces oT the North Carolina Pilot - The NASS enumerators will be trained to understand
the questionnaire and sampling techniques required to obtain data for the NC pilot; a detailed
enumerator's manual will be provided. Data will be collected by the enumerators and data quality
and logistics will follow NASS guidelines. Soil and water samples will be sent to appropriate
laboratories following standard procedures. Data will be managed initially by NASS and then
processed by the Agro Resource Group for data analysis and summarization. Design options and
indicators will be evaluated and a statistical summary will be prepared. Most of this effort is
directed at the 5 basic indicators identified for the Pilot
a. Prepare Enumerator's Manual
c. Acquire Materials for Sampling Soil and Water
e. NASS, Other Responsibilities
g. Sample Preparation and Analysis (water)
i. Evaluation of Design Options
k. Acquire Equipment/Software for Information Mgmt
m. Evaluate/Update QA/QC
o. Prepare First Annual Statistical Summary
b. Train NASS Enumerators
d. Data/Sample Collection -
NASS Enumerators
f. Sample Prep. & Analysis (soils)
h. Management and Analysis of Data
j. Evaluation of Five Indicators
1. Evaluate/Update Information Mgmt
n. Evaluate/Update Logistics
2. Acquire Found Data and Test Compatibility with Data from the Pilot - Search other data sets to
see what data is present and may be of value for the Agroecosystem. Test ways to determine whether
the data is compatible with agio data or can be used in some way to aid in interpretive reports.
Data sets of interest include atmospheric, terrestrial and water inputs (SCS, ERS, EPA, etc.).
a. Water Quality
b. Terrestrial and Atmospheric (i.e., weather, ozone, soils, pesticides, etc.)
3. Evaluation of Research Indicators Tested as Part of the Pilot - The Agroecosystem Resource
Group will continue to test indicators designed to monitor additional components of the agro-
ecosystem resource. Work will continue to develop the nematode as an indicator of the biological
"health" of the soil system. Additional effort will be put into identifying other specific
measurements for water quality. A major effort will continue in the development of habitat
indicators that will monitor the vitality of lands adjacent to agricultural fields. The
development of habitat indicators will be continued in close cooperation with other Resource
Groups and with other agencies, such as USGS and SCS. These indicators will be field tested in
NC during the Pilot.
a. Nematodes - Soil Biological Health
c. Water Quality - Irrigation, Farm Ponds, Wells
b. Habitat - Extent and Quality
d. Clover - Ozone Biomonitor
4. Activities Supportive of the Agroecosystem Resource - The development of an integrated pilot, in
conjunction with the other terrestrial resource groups, will continue with the expectation-that the
pilot will be undertaken in 1993. Additional work will be done with the Integration and Assessment
team in preparation for an example of an integrated assessment Additional efforts will be made to work
with the Regions.
a. Linkages/Integration of Resource Groups
c. Regional Interests
b. Integration and Assessment
5. Develop 1993 Plans - The Agroecosystem Resource Group is planning for a Demonstration project in the
South East and a Pilot project in Region VII for 1993. Detailed plans, revised questionnaires, revised
enumerator manuals, and revisions to all cross-cutting activities are needed for these two field studies.
10-4
-------
10.4.1. Funding
The budget by tasks is shown in Table 10-3 and the budget by Location/Category is shown
in Table 10-4. Although funding is not at the level requested for an in-depth pilot, it provides
sufficient support for a well designed monitoring program to address the issues highlighted in this
Plan. It required the use of fewer sampling segments (116) as opposed to our recommended
(200) and does not give full funding to several of our ARG members. Direct EMAP support for
the Program is shown (Tables 10-3 and 10-4) as well as support coming from other cooperators
in the Program.
10.4.2. Personnel
Personnel associated with the ARG are shown in the Organization Structure (Appendix 1).
This translates to the number and full time equivalents shown in Table 10-5. Because of the
small percentage of time we were able to budget for several of the contact people, we will not
receive as much dedicated effort from this group. We expect to have more of their time in 1993.
The enumerator's time, covered by NASS, is not shown in the table.
10.4.3. Program Products (Outputs)
In addition to this Pilot Plan we have several other outputs planned in 1992. These are listed
below with a tide, brief description, due date and comments.
Title
o Agroecosystem 1992 Pilot
Project Plan
o Monitoring the Conditions of
Agroecosystems
o Comparison of Periodic Survey
Designs Employing Multistage
Sampling
o Sustainable Agriculture
o Enumerators Manual
o Report on Indicator Testing -
Soil Nematodes
Brief Description Due Date
Pilot Study Plan 4/6/92
Overview Document 4/92
Comparison of the 7/92
Hexagon and Rotational
Panel Designs
Symp. Proceedings 6/92
Instructions/Training for 10/92
Enumerators
Analysis of Nematode 10/92
Data
Comments
Submitted
4/3/92
Submitted
3/92
Completed
Completed
In Process
In Process
10-5
-------
Table 10-3. Program Tasks with Budget for 1992 Pilot
Primary
Task Activities Funding (Thousands)
(Number) I/ EMAP Other2/
• Conduct North Carolina Pilot la-g
• Manage and analyze data from pilot lh,o
* Evaluate design and sampling options li
• Evaluate pilot indicators (integration) Ij
• Evaluate and update data management protocol lk,l
• Evaluate and update QA/QCprotocolsAogistics lm,n
• Collect and analyze data from existing data bases - 2
Determine applicability to the Agro database
• Evaluate research indicators tested in the Pilot 3
• Activities supportive of the ARG 4
• Develop plans for 1993 Regional Demonstration 5
and Pilot
Totals
Total Pilot Funds
$ 275
125
30
50
45
20
30
85
33.6
50
$ 743.6
$ 1,237.6
$ 183
70
10
61
10
10
20
90
15
25
$ 494
I/ Corresponds to activity number in Table 10-2
2/ Other funds from EPA laboratories, ARS, NCSU and NASS
10 - 6
-------
Table 10-4. Pilot Budget by Location/Category
Location
Funding Thousands
EMAP
Otherl/
1. Locations
a) Athens (ERL)
b) Corvallis (ERL)
c) Idaho (INEL)
d) Las Vegas (ERL)
e) RTP (ERL)
2. NASS (DC/NC)
3. USDA/ARS/NCSU
a) Personnel
b) Travel
c) Supplies/Service
d) Advisory Com.
e) Equipment
f) Sample Costs
g) Utilities/Space
h) Athens
i) Indirect (12.7% waived)3-/
Totals
Total for Pilot Funds
10
10
10
200
513.6
322
29
38.6
10
34
30
10
40
(65.3)
743.6
$1,237.6
180
5
5
113
20
171
141
2
3
25
(+65.3)
494
I/ Other funds from EPA laboratories, NASS, ARS and NCSU
2/ Direct support will come from the USDA budget ($40,000)
1J If the waived indirect costs are included, funds from other sources is $559,300
and direct EMAP funds are $678,300.
10-7
-------
Table 10-5. Personnel/Responsibilities for the Agroecosystem Pilot Project!/
Position
Technical Director
Associate Director
Professional Staff
Biological
Statisticians
Research Assoc.
(PI. Path/Biomath)
Statistician
Information Man.
QA/Log.
Technicians
Support Staff
Number
1
1
7
4
2
2
1
-
4
5
Total 27
FTEs
1.0
0.3
2.5
2.3
2.0
1.0
1.0
0.2
3.2
1.5
15.0
Organization
USDA/ARS
NCSU
EPA/Contract
Labs/ARS
NCSU/NASS/Athens
NCSU
NCSU
NCSU
NCSU
NCSU/Athens/ARS
NCSU
If The Table does not include the time of the enumerators covered by NASS
10-8
-------
10.5. Activities, Funding and Personnel Needs for 1993 Demonstration and Pilot Projects
The Agroecosystem Program is planning two primary programs in 1993. The information
detailed in this section is based on the level of funding shown in the tables for the planned
activities. In the first program, costs are based on a broad coverage Demonstration Project in the
S.E. to include eight states (Delaware, Virginia, North Carolina, South Carolina, Georgia,
Alabama, and Mississippi). Costs are based on obtaining data from 100 sampling units
(segments) per state for a total of 800 segments. This is a broad coverage of the S.E. and
permits addressing results on both political and ecological regions of the S.E. The second
program is a large Pilot to include the four states (Iowa, Missouri, Kansas and Nebraska) of EPA
Region 7. Costs are based on obtaining data from 100 sampling units (segments) per state for
a total of 400 segments. This may be sufficient to address results on ecological regions as well
as political regions.
The ARG believes it can accomplish the above programs primarily because of the
infrastructure that NASS has developed across the country. This permits us to utilize the
information developed in the 1992 Pilot across all 12 states with some revision relating to
differences in levels of agriculture in several of the states. This will be a major challenge for
the ARG but one we can accomplish, if detailed planning can start early and new staff can be
added fairly quickly. We are proposing this program to the EMAP Steering Committee in the
spring of 1992.
The list of planned activities is detailed in Table 10-6. Budget by activity is shown in Table
10-7 and by category/location in Table 10-8. Personnel needs are developed in Table 10-9.
Detailed planning for 1993 is an iterative process that has already begun in the current
document Revisions of this document will form the basis for both the Demonstration and Pilot
Programs planned for 1993.
10-9
-------
Table 10-6. 1993 Activity Chart for the ARC.
1. Primary Activities for the S.E. Regional Demonstration - The NASS enumerators will be trained within the
selected SB states to understand the questionnaire and sampling techniques required to obtain data for the
SE Demonstration project Data will be collected by the enumerators and data quality and logistics will follow
NASS guidelines. Soil and water samples will be sent to appropriate laboratories following standard procedures.
Data will be managed initially by NASS and then processed by the Agro Resource Group for data analysis and
summarization. Design options and indicators will be evaluated and a statistical summary will be prepared.
a. Review/Prepare Enumerator's Manuals
c. Acquire Materials for Sampling Soil and Water
e. NASS, Other Responsibilities
g. Sample Preparation and Analysis (water)
i. Evaluation of Selected Indicators
k. Prepare Annual Statistical Summary for Demonstration
b. Train NASS Enumerators
d. Data Collection - NASS Enumerators
f. Sample Preparation and Analysis (soils)
h. Management and Analysis of Data
j. Acquire Equipment/Software for Info. Mgmt.
2. Primary Activities for the Region \H Pilot - The same basic tasks will be required for the Region VH pilot
on a single-state basis as was required for the SE Demonstration. This is expected to be a single State effort
and will not require multiple training sessions or working with multiple groups of NASS personnel. A separate
statistical summary will be prepared for this pilot
a. Review/Prepare Enumerator's Manuals
c. Acquire Materials for Sampling Soil and Water
e. NASS, Other Responsibilities
g. Sample Preparation and Analysis (water)
i. Evaluation of Selected Indicators
b. Train NASS Enumerators
d. Data Collection - NASS Enumerators
f. Sample Preparation and Analysis (soils)
h. Management and Analysis of E>ata
j. Prepare Annual Statistical Summary for Pilot
3. Cross Cutting Activities Supportive of Both the Demonstration and Pilot Projects - These activities
include information management, QA/QC, and logistics. Special effort with go into these activities to assure
thai all
three are compatible with other EMAP activities. A sample integration report will be prepared as part of this
overall activity,
a. Evaluate/Update Information Management
b. Evaluate/Update QA/QC
c. Evaluate/Update Logistics
d. Prepare a Sample Integration Report
e. Explore Ways to Integrate "Found" Data into Data from Pilot and Demonstration Projects
f. Continue Staff Development ^ •-...•'
4. Evaluation of Research Indicators Tested in Either the Demonstration or Pilot Project - The
Agroecosystem Resource Group will continue to evaluate additional indicators and insert them into the
monitoring designs on a limited basis. Wo* on water quality and habitat will be continued in conjunction with
other groups. Preliminary work will be initiated with several socio-economic indicators that have gone through
some level of testing. A literature review combined with one or two workshops will be undertaken to establish
possible indicators for use with farm animals. These indicators will be considered preliminary but might see
limited field testing in 1994.
a. Nematodes - Soil Biological Health
b. Habitat - Extent and Quality
c. Water Quality - Irrigation, Farm Ponds, Wells .
d. Clover - Ozone Biomonitor
e. Farm Animals
f. Socib-Economic
10- 10
-------
4. Evaluation of Research Indicators Tested in Either the Demonstration or Pilot Project - The
Agroecosystem Resource Group will continue to evaluate additional indicators and insert them into the monitoring
designs on a limited basis. Work on water quality and habitat will be continued in conjunction with other groups.
Preliminary work will be initialed with several socio-economic indicators that have gone through some level of
testing. A literature review combined with one or two workshops will be undertaken to establish possible indicators
for use with farm animals. These indicators will be considered preliminary but might see limited field testing in
1994.
a. Nematodes - Soil Biological Health
b. Habitat - Extent and Quality
c. Water Quality - Irrigation, Farm Ponds, Wells
d. Clover - Ozone Biomonitor
e. Farm Animals
f. Socio-Economic
5. Participate in An Integrated Pilot for The Terrestrial Ecosystems - The Technical Director of
Agroecosystem will work closely with the Technical Directors of Arid Lands and Forest Lands in the initiation of
an integrated pilot, probably in Colorado. Planning for this pilot was initiated in 1992. A primary purpose of this
integrated pilot is to test concepts and the importance of integrated pilots.
6. Activities Supportive of the Agroecosystem Resource - We will continue to identify additional areas in which
the Agroecosystem can form linkages both within and outside of EMAP. The sample integration report will be
completed and other ways to interact with the Integration and Assessment team will be explored. Further exploration
of Regional interest and ways to work more effectively with personnel in the regions will be undertaken.
a. Linkages/Resource Group Integration
b. Integration and Assessment
c. Regional Interests
7. Develop 1994 Plans for the Agroecosystem Resource - The Agroecosystem Resource Group is planning for
three major activities in 1994. This includes a pilot in Region DC, a Demonstration in Region VII, and
Implementation in Region IV (the SE). This will require final development of a suite of indicators that will become
core indicators for the agroecosystem program. This core group will be used for implementation in Region IV. In
the other two regions additional indicators will be tested in addition to the core group. The questionnaires and
enumerator manuals will be revised to reflect results from the 1992 and 199J activities.
10- 11
-------
Table 10-7. Program Activities with Budget for 1993
Activity i
1. Primary Activities for the S.E. Regional
Demonstration (8 states)
2. Primary Activities for the Region VH Pilot (4 states)
3. Cross Cutting Activities Supportive ;of Both the
Demonstration and Pilot Projects
4. Evaluation of Research Indicators Tested ;in Either
the Demonstration or Pilot Project
5. Participate in An Integrated Pilot for The Terrestrial ;
Ecosystems ;
6. Activities Supportive of the Agroecosystem Resource
7. Develop 1 994 Plans for the Agroecosystem Resource
Total '
funding
(Thousands)
EM&P
$1,200
600
140
285
50
50
15
$2,400
10-12
-------
Table 10-8. 1993 Budget by Location
Location
Funding
(Thousands)
EMAP
1. Locations
a) Athens (ERL)
b) Corvallis (ERL)
c) Idaho (INEL)
d) Las Vegas (ERL)
2. USDA/NASS (DC/States)
3. USDA/SCS (DC/States)
4. USDA/ARS/NCSU
a) Personnel - current
b) Personnel - new
c) Travel
d) Supplies/Service
e) Advisory Com.
f) Equipment
g) Sample Costs, soil
h) Utilities/Space
i) Indirect (12.7%)1/
Totals
300
250
75
90
25
65
130
25
140
130
75
75
20
900
100
1,100
$ 2,400
We will ask for waiver of indirect costs; if approved, funds will be used to
increase operations budget, which is low.
10-13
-------
Table 10-9. Personnel/Responsibilities for the 1993 Agroecosystem Program
Position
Technical Director
Associate Director
Professional Staff
Biological
Statisticians
Research Assoc.
(PL Path/Biomath)
Statistician
Information Man.
QAyLog.
Technicians
Support Staff
SubTotals
Program Totals Staff 40
FTEs 28.3
Current
No. FTEs
1
1
7
4
2
2
1
-
4
5
27
1.0
0.3
2.5
2.3
2.0
1.0
1.0
0.2
2.8
1.5
14.6
New
No. FTEs
0.2
3 3.0
1 1.0
2 2.0
1 1.0
1 1.0
4 4.0
1 1.5
13 13.7
Organization
USDA/ARS
NCSU
EPA/Contract
Labs/ARS
NCSU/NASS
NCSU
NCSU
NCSU
NCSU
NCSU
NCSU
10-14
-------
LITERATURE CITED
Alexander, M. (1985) Ecological constraints on nitrogen fixation in agricultural ecosystems
Adv. Microbial Ecol. 8:163-183.
American Society for Testing and Materials. (1991) Draft Proposed Specification for Meta-Data
Support in Geographical Information Systems. August 1991. ASTM Committee D.18.01.05.
Baker, D.E. and Chesnin, L. (1975) Chemical monitoring of soils for environmental quality and
animal and human health. Adv. Agon. 27:305-374.
Baker, J.R. and Merritt,-G.D. (1991) Environmental Monitoring and Assessment Program:
Guidelines for Preparing Logistics Plans. May 1991. U.S. Environmental Protection Agency.
Office of Research and Development Environmental Monitoring Systems Laboratory Las Vegas
Nevada. EPA/600/4-91/001.
/
Baker, D.E., Amacher, M.C. and Leach, R.M. (1979) Sewage sludge as a source of cadmium
in soil-plant-animal systems. Environ. Health Perspectives 28:45-49.
Barker, K.R. (1985a) Nematode extraction and bioassays. In: An Advanced Treatise on
Meloidosvne Vol. II: Methodology, ed. K.R. Barker, C.C. Carter and J.N. Sasser. North Carolina
State University Graphics, Raleigh, pp. 19-35.
Barker, K.R. (1985b) Sampling nematode communities. In: An Advanced Treatise on
Meloidosvne Vol. II: Methodology, ed. K.R. Barker, C.C. Carter and J.N. Sasser. North Carolina
State University Graphics, Raleigh, pp. 2-17.
Berger, K.C., Hole, F.D. and Beardsley, J.M. (1952) A soil productivity score card. Soil Sci
Soc. Am. Proc. 16:307-309.
Bliss, N.B. (1990) A hierarchy of soil databases for calibrating models of global climate
change. In: Soils and the Greenhouse Effect, ed. A.F. Bouwman. John Wiley and Sons, Ltd. pp.
529-533.
Bliss, N.B. and Reybold, W.U. (1989) Small-scale digital soil maps for interpreting natural
resources. Journal of Soil and Water Conervation (Jan.-Feb.):30-34.
Bostrom, S. and Sohlenius, B. (1986) Short-term dynamics of nematode communities in arable
soil: Influence of a perennial and an annual cropping system. Pedobiologia 29:345-357.
Box, E.G., Holben, B.N. and Kalb, V. (1989) Accuracy of the AVHRR vegetation index as a
predictor of biomass primary productivity and net CO2 flux. Vegetatio 80:71-89.
L- 1
-------
Brady, N.C. (1974) The Nature and Properties of Soils. Eighth Edition. MacMiUan Publishing
Co. New York.
Brams, E. (1977) Relationships of soil cadmium and lead to the environment peripheral to
industrial cities. In: Proceedings, International Seminar of Soil Environment and Fertility
Management in Intensive Agriculture (SEFMIA), Japan 1977. Soc. of the Science of Soil and
Manure, Tokyo, pp. 369-382. (Cited in Methods of Soil Analysis, Part 2. Agronomy Monograph
No. 9, Second edition. 1982. ASA-SSSA. p. 362. Madison, WI.)
Brookes, P.C. and McGrath, S.P. (1985) Effects of heavy metal accumulation in field soils
treated with sewage sludge on soil microbial processes and soil fertility. In: Microbial
Communities in Soil, ed. V. Jensen, A. Kjoller and L.H. Sorensen. Elsevier Applied Science
Publ., New York. pp. 327-343.
Buol, S.W., Hole, F.D., and McCracken, R.J. 1980. Soil Genesis and Classification. Second
Edition. The Iowa State University Press, Ames.
Burke, I.C., Yonker, CM., Parton, W.J., Cole, C.V., Flach, K. and Schimel, D.S. (1989)
Texture, climate and cultivation effects on soil organic matter content in U.S. grassland soils.
Soil Set. Soc. Am. J. 53:800-805.
Campbell, C.L., Heck, W.W., Moser, T.J., Breckenridge, R.P., Hess.G.R. and Meyer, J.R. (1990)
Indicator Fact Sheets for Agroecosystems. In: Environmental Monitoring and Assessment
Program: Ecological Indicators, ed. C.T. Hunsaker and D.E. Carpenter. EPA/600/3-90/060.
U.S.EPA, Office of Research and Development, Washington, DC. pp. F-l to F-23.
Cole, C.V., Stewart, J.W.B., Ojima, D.S., Parton, WJ. and Schimel, D.S. (1989) Modelling land
use effects of soil organic matter dynamics in the North American Great Plains. In: Ecology of
Arable Land., ed. M. Clarholm and L. Bergstrom, Kluwer Academic Publishers, pp. 89-98.
Coleman, D.C., Andrews, R., Ellis, J.E. and Singh, J.S. (1976) Energy flow and partitioning in
selected man-managed and natural ecosystems. Agroecosystems 3:45-54.
Coleman, D.C., Cole, C.V. and Elliot, E.T. (1990) Decomposition, organic matter turnover, and
nutrient dynamics in agroecosystems. In: Agroecology, ed. R. Lowrance, B.J. Stinner, and G.J.
House. John Wiley and Sons, New York. pp. 83-104.
Coleman, D.C., Reid, C.P.P. and Cole, CV. (1983) Biological strategies of nutrient cycling in
soil systems. Adv. in Ecol. Res. 13:1-55.
Cooper, J.E., Wood, M. and Bjourson, A.J. (1985) Influence of acidity and aluminum on the
nodulation of Marsh Trefoil (Lotus pedunculatus) and white clover (Trifolium repens) by
Rhizobia. In: Microbial Communities in Soil, ed. V. Jensen, A. Kjqller and L.H. Sorensen.
Elsevier Applied Science Publishers, New York.
L-2
-------
Cotter, J. and Nealon, J. (1987) Area Frame Design for Agricultural Surveys. U.S. Department
of Agriculture, National Agricultural Statistics Service, Research and Applications Division, Area
Frame Section. Washington, DC.
Daniels, R.B., Gilliam, J.W., Cassel, D.K. and Nelson, L.A. (1987) Quantifying the effect of
soil erosion on present soil productivity. J^ Soil Water Conserv. (May-Jun.): 183-187.
Daykin, M.E. and Hussey, R.S. (1985) Staining and histopathological techniques in nematology.
In: An Advanced Treatise on Meloidosvne Vol. II: Methodology, ed. K.R. Barker, C.C. Carter,
and J.N. Sasser. North Carolina State University Graphics, Raleigh, pp. 39-48.
Dean, J.D., Huyakorn, P.S., Donigian, A.S., Jr., Voos, K.A., Schanz, R.W., Meeks, Y.J. and
Carsel, R.F. (1989) Risk of UnsaturatedlSaturated Transport and Transformation of Chemical
Concentrations (RUSTIC), I. EPA/600/3-89/048a, U.S. EPA, Athens, Georgia.
Dmowska, E. and Kozlowska, J. (1983) Effects of some herbicides on soil nematodes under
laboratory conditions. Zeszytz Problemowe Postepow Nauk Rolniczych 278:130-131.
Donald, CM. and Hamblin, J. (1976) The biological yield and harvest index of cereals as
agronomic and plant breeding criteria. Adv. Agron. 28:361-405.
Duniway, J.M. (1979) Water relations of water molds. Ann. Rev. Phytopathol. 17:431-460.
Ewel, J.J., Mazzarino, M.J. and Berish, C.W. (1991) Tropical soil fertlity changes under
monocultures and successional communities of different structure. Ecol. Applications 1:289-303.
Federoff, N. (1987) The production potential of soils: Part I. Sensitivity of principal soil types
to the intensive agriculture of north-western Europe. In: Scientific Basis for Soil Protection in
the European Community, ed. H. Barth and P. L'Hermite. Elsevier Applied Science, New York.
pp. 65-85.
Fluck, R.C. and Baird, C.D. (1980) Agricultural Energetics. AVI Publishing Co. Inc.,
Westport, Connecticut
Fluck, R.C. (1979) Energy productivity: A measure of energy utilization in agriculture systems.
Agric. Systems 4:29-37.
Franson, S.E. (1990) Data confidentiality in the Environmental Monitoring and Assessment
Program (EMAP): Issues and Recommendations. U.S. EPA Report 600/X-90/129. U.S. EPA,
Las Vegas, NV.
Freckman, D.W. (1988) Bacterivorous nematodes and organic-matter decomposition. Agri.,
Ecosys. and Environ. 24:195-217.
L-3
-------
Frye, W.W., Ebelhar, S.A., Murdock, L.W. and Blevins, R.L. (1982) Soil erosion effects on
properties and productivity of two Kentucky soils. Soil Sci. Soc. Am. J. 46:1051-1055.
Gersmehl, P.J. and Brown, D.A. (1986) Regional differences in the validity of the concept of
innate soil productivity. Ann. Assoc. of Am. Geographers 76:480-492.
Gibbs, RJ. and Reid, J.B. (1988) A conceptual model of changes in soil structure under
different cropping systems. Adv. in Soil Sci. 8:123-149.
Corny, M. (1976) The problem of bioindication in soils of degrade biotopes. Wiad. ekol.
22:224-230 (cited by Wasilewska 1979).
Goss, D. (1991) Screening procedure for soils and pesticides relative to potential water quality
impacts. In: Using Computer Simulation Models in Pesticide Registration Decision Making.
Weed Sci. Soc. of Am., Louisville, KY.
Goss, D. and Wauchope, R.D. (1990) The SCS/ARS/CES Pesticide Properties Database II:
Using it with soils data in a screening procedure. In: Pesticides in the Next Decade: The
Challenges Ahead., ed. D.L. Weigmann. pp. 471-493.
Groenevelt, P.H., Kay, B.D. and Grant, CD. (1984) Physical assessment of a soil with respect
to rooting potential. Geoderma 34:101-114.
Haberern, J. (1991) A soil health index. /. Soil and Water Conser. 47:6.
Hall, G.F., Logan, T.J. and Young, K.K. (1985) Criteria for determining tolerable erosion rates.
In: Soil Erosion and Crop Productivity, ed. R.F. Follett and B.A. Stewart. ASA-CSSA-SSSA,
Madision, WI. pp. 173-188.
Heagle, A.S., McLaughlin, M.R., Miller, I.E., and Joyner, R.E. (1992) Response of two white
clover clones to peanut stunt virus and ozone. Phytopathology 82:254-258.
Heagle, A.S., McLaughlin, M.R., Miller, J.E., Joyner, R.E., and Spruill, S.E. (1991) Adaptation
of a white clover population to chronic ozone stress. New Phytol. 119:61-68.
Heagle, A.S., Rebbeck, J., Shafer, S.R., and Heck, W.W. (1989) Effects of long-term ozone
exposure and soil moisture deficit on growth of a ladino clover-tall fescue pasture.
Phytopathology 79:128-136.
Heck, W.W., Campbell, C.L., Moser.TJ., Rawlings, J.O., Hess, G.R., Breckenridge, R.B., Smith,
C.N., Byers, G.E. and Finkner, A.L. (1989) EMAP-Agroecosystem Research Plan. Informational
Report.
L-4
-------
Heck, W.W., Campbell, C.L., Breckenridge, R.P., Byers, G.E., Finkner, A.L., Hess, G.R., Meyer,
J.R., Moser, T.J., Peck, S.L., Rawlings, J.O., and Smith, C.N. (1991) Environmental Monitoring
and Assessment Program, Agroecosystem Monitoring and Research Strategy. U.S. EPA, Office
of Research and Development, Environmental Monitoring Systems Laboratory, Las Vegas NV
EPA 600/4-91/013.
Hodda, M. (1986) Cluster analysis of marine nematodes. Nematologica 32:419-438.
Horvitz, D.G. and Thompson, DJ. (1952) A generalization of sampling without replacement
from a finite universe. J. Am. Statist. Assoc. 47:663-685.
Huddleston, J.H. (1984) Development and use of soil productivity ratings in the United States
Geoderma 32:297-317.
Huddleston, J.H. (1982) Agricultural Productivity Ratings for Soils of the Willamette Valley.
Oregon State University Extension Service, Corvallis, OR.
Hunsaker, C.T. and Carpenter, D.E. eds. (1990) Environmental Monitoring and Assessment
Program: Ecological Indicators. EPA/600/3-90/060. U.S. EPA, Office of Research and
Development, Washington, DC.
Imhoff, J.C., Carsel, R.F., Kittle, J.L. and Hummel, P.R. (1990) Data Base Analyzer and
Parameter Estimator (DBAPE) Interactive Computer Program. User's Manual
EPA/600/3-89-083, U.S. EPA, Athens, Georgia.
Indorante, S.J., Hammer, R.D. and Maatta, J.M. (1991) Horizontal and vertical distribution of
soil properties on native and eroded central Missouri deep loess soils. Soil and Water Conserva.
Soc. Annual Meeting, p. 21. Abstract.
Jonasson, I.R. and Boyle, R.W. (1971) Geochemistry of mercury. In: Mercury in Man's
Environment. Proc. Royal Soc. Canada Symp., Ottawa, Ontario, Canada. 15-16 Feb 1971.
Geological Survey of Canada, Department of Energy, Mines and Resources, Ottawa, Canada pp
5-21. VV'
Kaiznelson, H. (1940) Survival of Azotobacter in soil. Soil Sci. 49:21-35.
Kay, B.D. (1989) Rates of change of soil structure. Adv. in Soil Sci. 12:1-52.
Khorram, S., Cheshire, H., Siderelis, K. and Nagy, Z. (1991) Mapping and CIS Implementation
of Land Use and Land Cover Categories for the Albemarle-Pamlico Drainage Basin. North
Carolina State University Computer Graphics Center and North Carolina Department of
Environment, Health and Natural Resources, Center for Geographic Information and Analysis.
Draft Copy date July 1991.
L-5
-------
Kiniry, L.N., Scrivner, C.L. and Keener, M.E. (1983) A Soil Productivity Index Based Upon
Predicted Water Depletion and Root Growth. University of Missouri-Columbia Agricultural
Experiment Station Bulletin 1051.
Klopatek, J.M. (1978) The Energetics of Land Use in Three Oklahoma Counties! Norman: The
University of Oklahoma Graduate College. Dissertation.
Ko, W.H. and Hora, F.K. (1972) Identification of an Al ion as a soil fungitoxin. Soil Sci
113:42-45.
Korentajer, L. (1991) A review of the agricultural use of sewage sludge: benefits and potential
hazards. Water SA 17:189-196.
Kutz, F.W. and Linthurst, R.A. (1990) A systems-level approach to environmental assessment.
ToxicoL and Environ. Chem. 28:105-114.
Lai, R. (1991) Soil structure and sustainability. J. Sust. Agric. 1:67-92.
Lai, R. (1987) Effects of soil erosion on crop production. CRC Critical Rev. in Plant Sci
5:303-367.
Larson, W.E., Fenton, T.E., Skidmore, E.L. and Benbrook, CM. (1985) Effects of soil erosion
on soil properties as related to crop productivity and classification. In: Soil Erosion and Crop
Productivity., ed. R.F. FoUett and B.A. Stewart. ASA-CSSA-SSSA, Madision, WL pp. 189-211:
Larson, W.E., Pierce, F.J. and Dowdy, R.H. (1983) The threat of soil erosion to long-term crop
production. Science 219:458-465.
Lesser, V.M. (1992) A comparison of Periodic Survey Designs Employing Multi-State
Sampling. Chapel Hill: The University of North Carolina. Dissertation.
Lin, B-H, Hansen, L., Daberkow, S. and Dreitzer, M. (1991) Substitutability of crop rotations
for agrichemicals: preliminary results. In: Agricultural Resources: Inputs Situation and Outlook
Report, AR-24. October 1991. USDA, Economic Research Service, Resources and Technology
Division. Washington, DC. pp. 24-29.
Lindsay, W.L. (1979) Mercury. In: Chemical Equilibria in Soils. John Wiley and Sons, New
York. pp. 344-363.
Ludwig, J.A. and Reynolds, J.F. (1988) Statistical Ecology: A Primer on Methods and
Computing. John Wiley & Sons, New York. pp. 85-103.
L-6
-------
Lunetta, R.S., Congalton, R.G., Fenstermaker, L.K., Jensen, J.R., McGwire, K.C. and finney,
L.R. (1991) Remote sensing and geographic information system data integration: error sources
and research issues. Photogrammetric Eng. and Remote Sensing 57:677-687.
Mann, L.K. (1986) Changes in soil carbon storage after cultivation. Soil Sci. 142:279-288.
McCracken, R.J., Lee, J.S., Arnold, R.W. and McCormack, D.E. (1985) An Appraisal of Soil
Resources in the USA. In: Soil Erosion and Crop Productivity., ed. R.F. Follett and B.A.
Stewart. ASA-CSSA-SSSA, Madison, WI. pp. 49-66.
McSorley, R., Dankers, W.H., Parrado, J.L. and Reynolds, J.S. (1985) Spatial distribution of
the nematode community on perrine marl soils. Nematropica 15:77-92.
Meyer, J.R., Peck, S.L., Rawlings, J.O. and Campbell, C.L. (1990) Annual Statistical Summary:
Agroecosystems. A Hypothetical Example. EPA/600/X-90/208. U.S. EPA, Environmental
Monitoring Systems Laboratory, Las Vegas, NV.
Mitchell, R. (1984) The ecological basis for comparative primary production. In Agricultural
Ecosystems: Unifying Concepts, ed. R. Lowrance, B.R. Stinner, and G.I. House. Wiley-
Interscience, John Wiley & Sons, New York. pp. 13-53.
Moore, J.C. and de Ruiter, P.C. (1991) Temporal and spatial heterogeneity of trophic
interactions within below-ground foodwebs. Agric., Ecosys. Environ. 34:371-397.
Mutatkar, V.K. and Pritchett, W.L. (1966) Influence of added aluminum on carbon dioxide
production in tropical soils. Soil Sci. Soc. Am Proc. 30:343-346.
Nelson, D.W. and Sommers, L.E. (1982) Total carbon, organic carbon, and organic matter.
Methods of Soil Analysis, Part 2. Agronomy Monograph No. 9, 2nd Edition. ASA-SSSA
Madison, WI.
Nielsen, E.G. and Lee, L.K. (1987) Magnitude and Costs of Groundwater Contamination from
Agricultural Chemicals: A National Perspective. USDA Economic Research Service Report
AGES870318. USDA/ERS, Washington, DC.
North Carolina Agricultural Statistics Division. (1991) North Carolina Agricultural Statistics:
1991. North Carolina Department of Agriculture, Raleigh.
O'Neill, R.V. (1976) Ecosystem persistence and heterotrophic regulation. Ecology 57:1244-
1253.
Ovalles, F.A. and Collins, M.E. (1986) Soil-landscape relationships and soil variability in North
Central Florida. Soil Sci. Soc. Am J. 50:401-408.
L-7
-------
Overton, S.O. (1991) Probability Sampling and Population Inference in Monitoring Programs.
(unpublished manuscript)
Overton, S.O., White, D. and Stevens, D.L., Jr. (1991) Design Report for EMA1P Environmental
Monitoring and Assessment Program. U.S. EPA/600/xxx/xxx. (draft)
Page, A.L. and Ganje, T.J. (1970) Accumulation of lead in soils for regions of high and low
motor vehicle traffic density. Environ. Sci. Technol. 4:140-142.
Parmelee, R.W. and Alston, D.G. (1986) Nematode trophic structure in conventional and no-
tillage agroecosystems. /. Nematology 18:403-407.
Parrish, R.S. and Smith, C.N. (1990) A method for testing whether model predictions fall within
a prescribed factor of true values, with an application to pesticide leaching. Ecological Modelling
51:59-72.
Pettersson, R. (1987) Primary Production in Arable Crops: Above-Ground Growth Dynamics,
Net Production and Nitrogen Uptake. Report 31. Uppsala: Swedish University of Agricultural
Sciences, Department of Ecology and Environmental Research.
Pierce, FJ. and Lai, R. (1991) Monitoring soil erosion's impact on crop productivity. In:
Research Methods in Soil Erosion. Second Edition, (in press).
Pierce, FJ., Larson, W.E., Dowdy, R.H. and Graham, W.A.P. (1983) Productivity of soils:
Assessing long-term changes due to erosion. / Soil Water Conserv. Soc. (Jan.-Feb.):39-44.
Platt, H.M., Shaw, K.M. and Lambshead, PJ.D. (1984) Nematode species abundance patterns
and their use in the detection of environmental perturbations. Hydrobiologia 118:59-66.
Rebbeck, J., Blum, U. and Heagle, A.S. (1988) Effects of ozone on the regrowth and energy
reserves of a ladino clover-tall fescue pasture. /. Applied Ecology 25:659-681.
Rosswall, T., Schniirer, J. and Soderlund, S. (1985) Interactions of acidity, aluminum ions and
microorganisms. In: Microbial Communities in Soil., ed. V. Jensen, A. Kjoller and L.H.
Sorensen. Elsevier Applied Science Publishers, New York. pp. 395-410.
Russell, E.W. (1973) Soil Conditions and Plant Growth. Longman Group Ltd., London.
Schnitzer, M. (1982) Organic matter characterization. Methods of Soil Analysis, Part 2.
Agronomy Monograph No. 9, 2nd Edition. ASA-SSSA, Madison, WI.
Schouten, A.J., van Breemen, E.M. and Slob, W. (1990) Estimating nematode abundance in
natural sandy forest soil; a comparison of two soil samplers. Pedobiologia 34:239-246.
L-8
-------
Seastedt, T.R., James, S.W. and Todd, T.C. (1988) Interactions among soil invertebrates,
microbes and plant growth in the tallgrass prairie. Agric., Ecosys. and Environ. 24:219-228.
Shannon, C.E. and Weaver, W. (1949) The Mathematical Theory of Communication. Univ.
Illinois Press, Urbana.
Sharp, D.D., Lieth, H., Noggle, G.R. and Gross, H.D. (1976) Agricultural and Forest Primary
Productivity in North Carolina 1972-1973. Tech. Bui. No. 241, North Carolina Agricultural
Experiment Station, Raleigh, NC.
Shirley, C. (1991) Experts to issue "Soil Report Card". The New Farm. (Sep.-Oct.):5-6.
Simpson, E.H. (1949) Measurement of diversity. Nature 163:688.
Singh, K.K., Colvin, T.S., Erbach, D.C. and Mughal, A.Q. (1992) Tilth Index: An approach to
quantifying soil tilth. Trans. ASAE (in press).
Sohlenius, B., Bostrom, S. and Sandor, A. (1988) Carbon and nitrogen budgets of nematodes
in arable soil. Biol. Pert. Soils 6:1-8.
Sombreck, W.G. (1990) Geographic quantification of soils and changes in their properties. In:
Soils and the Greenhouse Effect, ed. A.F. Bouwman. John Wiley and Sons, Ltd. pp. 225-244.
Southwell, P.H. and Rothwell, T.M. (1977) Report on Analysis of Output!Input Energy Ratios
of Food Production in Ontario. March 31, 1977. School of Engineering, University of Guelph,
Ontario, Canada. Contract serial number OSW76-00048.
Stehman, S.V. and Overton. W.S. (1987) Estimating the variance of the Horvitz-Thompson
estimator in variable probability, systematic samples. Proceedings of the Section on Survey
Research Methods of the American Statistical Association. American Statistical Assoc.,
Alexandria, VA.
Stevenson, F.J. and Elliott, E.T. (1985) Methodologies for assessing the quantity and quality
of soil organic matter. In: Dynamics of Soil Organic Matter in Tropical Ecosystems, ed. D.C.
Coleman, J.M. Oades and G. Uehara. University of Hawaii Press, Niftal Project.
Stewart, J.W.B. and Bettany, J.R. (1982) Mercury. In: Methods of Soil Analysis, Part 2.
Agronomy Monograph No. 9, Second Edition. ASA-SSSA, Madison, WI.
Stinner, B.R. and Crossley, D.A., Jr. (1982) Nematodes in no-tillage agroecosystems. In:
Nematodes in Soil Ecosystems, ed D.W. Freckman. University of Texas Press, Austin, pp. 14-28.
L-9
-------
Stone, J.R., GilUam, J.W., Cassel, D.K., Daniels, R.B., Nelson, L.A. and Kleiss, HJ. (1985)
Effect of erosion and landscape position on the productivity of Piedmont soils. Soil Sci. Soc. Am.
J. 49:987-991.
Stone, R. (1978) Storie Index Rating. Special Publications, Division of Agricultural Sciences,
University of California.
Stout, B.A. (1990) Handbook of energy for world agriculture, Elsevier Applied Science,
Elsevier Science Publishers Ltd., London and New York.
Suter, G.W.H. (1990) Endpoints for regional ecological risk assessments. Environ. Manage. 14:9-
23.
Temple, P.J. (1990) Growth form and yield responses of four cotton cultivars to ozone. Agron
J. 82:1045-1050.
Thomasson, AJ. (1978) Towards an objective classification of soil structure. /. Soil Science
29:38-46.
Tisdale, S.L. and Nelson, W.L. (1975) Soil Fertility and Fertilizers, 3rd Edition. MacMillian
Publishing House, New York.
Tivy, J. (1990) Agricultural Ecology. Longman Scientific & Technical; Harlow, Essex, England.
Turner, M.G. (1989) Landscape ecology: The effect of pattern on process. Annual Review of
Ecolog. System. 20:171-197.
Turner, M.G. (1987) Land use changes and net primary production in the Georgia, USA,
landscape: 1935-1982. Environ. Manage. 11:237-247.
Turner, R.S., Olson, RJ. and Brandt, C.C. (1986) Areas having soil characteristics that may
indicate sensitivity to acidic deposition under alternative forest damage hypotheses.
Environmental Sciences Division Report No. 2720, Oak Ridge National Laboratory, Oak Ridge,
Tennessee.
Turner, M.G. and Gardner, R.H. (1991) Quantitative Methods in Landscape Ecology. Spinger-
Verlag, Berlin.
Tyler, G., Balsberg Pahlsson, A.-M., Bengtsson, G., Baath, E. and Tranvik, L. (1989) Heavy
metal ecology of terrestrial plants, microorganisms and invertebrates. Water, Air Soil Pollut.
47:189-215.
L- 10
-------
United States Department of Agriculture. (1991) Agricultural Chemical Usage: 1990 Field
Crops Summary. USDA National Agricultural Statistics Service and USDA Economic Research
Service. U.S. Government Printing Office: 1991, 282-964/40207. Washington, DC.
United States Department of Agriculture. (1989) The Second RCA Appraisal. Soil, Water, and
Related Resources on Nonfederal Land in the United States. Analysis of Conditions and Trends.
USDA Misc. Publ. 1482. Washington, DC.
United States Department of Agriculture. (1981) Will There Be Enough Food?: The 1981
Yearbook of Agriculture. U.S. Government Printing Office: 1981 O-354-445.
United States Department of Agriculture. (1975) Soil taxonomy: A basic system of soil
classification for making and interpreting soil surveys. USDA Soil Conservation Service
Agricultural Handbook 436.
United States Department of Agriculture (1961) Land capability classification. USDA/SCS
Agricultural Handbook 210. Washington, DC.
United States Department of Agriculture, National Agricutural Statistics Service. (199la)
Agricultural Surveys: Interviewer's ManualJune 1991 - May 1992. March 1991. Agricultural
Statistics Board. Washington DC. 20250. (internal document)
United States Department of Agriculture, National Agricutural Statistics Service. (1991b) 199/
Soybean Objective Yield Survey: Interviewer's Manual. May 1991. Agricultural Statistics Boar a.
Washington DC. 20250. (internal document)
United States Department of Agriculture, National Agricultural Statistics Service. (199Ic)
Objective Yield S&E [Supervising and Editing] Manual. June 1991. (internal document)
United States Department of Agriculture, National Agricultural Statistics Service. (1991d)
Area Frame Design Information. Area Frame Section, Fairfax, VA. (internal document)
United States Department of Agriculture, National Agricutural Statistics Service. (1990) NASD A
Supervisory Enumerator Handbook, (internal document) '
United States Department of Agriculture, Soil Conservation Service. (1990) The Universal Soil
Loss Equation with Factor Values for North Carolina. Technical Guide Section I-C. Raleigh, NC.
United States Department of Agriculture, Soil Conservation Service. (1983) National Soils
Handbook. 430-VI-NSH. U.S. Government Printing Office, Washington, DC. July 1983.
United States Department of Agriculture, Soil Conservation Service. (1981) Land Resource
Regions and Major Land Resource Areas of the United States. USDA/SCS Agricultural
Handbook 296, Washingtion, DC.
L- 11
-------
United States Department of Agriculture, Soil Conservation Service. (1970) Soil Survey: Wake
County, North Carolina. November 1970. In cooperation with North Carolina Agricultural
Experiment Station. U.S. Government Printing Office, Washington, DC.
United States Environmental Protection Agency. (199la) Environmental Monitoring and
Assessment Program: Quality Assurance Program Plan. December 1991. Environmental
Monitoring Systems Laboratory. Cincinnati, Ohio.
United States Environmental Protection Agency, Region IV. (1991b) Environmental Compliance
Branch Standard Operating Procedures and Quality Assurance Manual. U.S. EPA, BSD, Athens
Environmental Research Lab, College Station Road, Athens, Georgia.
Veitch, P.P. (1902) The estimation of soil acidity and the lime requirements of soils. /. Amer.
Chem. Sac. 24:1120-1128.
Warren, H.V, Delavault, R.E. and Barasko, J. (1966) Some observations on the geochemistry
of mercury as applied to prospecting. Econ. Geol. 61:1010-1028.
Wasilewska, L. (1979) The structure and function of soil nematode communities in natural
ecosystems and agrocenoses. Polish Ecol. Studies 5:97-145.
Webster, R. and Oliver, M.A. (1990) Statistical methods in soil and land resource survey.
Oxford University Press, Oxford.
Wiegand, C.L., Richardson, A.J., Escobar, D.E. and Gerbermann, A.H. (1991) Vegetation
indices in crop assessments. Remote Sens. Environ. 35:105-119.
Williams, J.R., Jones C.A. and Dyke, P.T. (1984) A modeling approach to determining the
relationship between erosion and soil productivity. Trans. ASAE X: 129-144.
Wischmeier, W.H. and Smith, D.D. (1978) Predicting Rainfall Erosion Losses-A Guide to
Conservation Planning. Agriculture Handbook 537, U.S. Department of Agriculture, Washington,
DC.
World Resources Institute. (1991) Moving toward eco-development: generating evnironmental
information for decisionmakers. WRI Issues and Ideas. August 1991:1-8.
Yassoglou, NJ. (1987) The production potential of soils: Part II. Sensitivity of the soil
systems in Southern Europe to degrading influxes. In: Scientific Basis for Soil Protection in
the European Community, ed. H. Barth and P. L'Hermite. Elsevier Applied Science, New York.
pp. 87-122.
Yeates, G.W. (1971) Feeding types and feeding groups in plant and soil nematodes.
Pedobiologia 11:173-179.
L - 12
-------
Yeates, G.W. and Coleman, D.C. (1982) Nematodes in decomposition. In: Nematodes in Soil
Ecosystems, ed. D.W. Freckman. University of Texas Press, Austin, pp. 55-80.
L- 13
-------
-------
APPENDIX 1
AGROECOSYSTEM RESOURCE GROUP MEMBERS
Steering Committee
Walter W. Heck, Chairman
USDA/ARS
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail HECK.WALTER
Fax (919) 515-3593
Thomas J. Moser
MET!
200 Southwest 35th Street
Corvallis, OR 97333
(503) 754-4463
E-Mail
Fax (503) 754-4338
C. Lee Campbell, Assoc. Chairman
DepL of Plant Pathology
NC State University
Box 7616
Raleigh, NC 27695-7616
(919) 515-6816
E-Mail HECK.WALTER
Fax (919) 515-7716
John O. Rawlings
Department of Statistics
NC State University
Box 8203
Raleigh, NC 27695-8203
(919) 515-2535
E-Mail HECK.WALTER
Fax (919) 515-7591
Robert P. Breckenridge
INEL
P.O. Box 16258
Idaho Falls, ID 83415-2213
(208) 526-0757
E-Mail BRECKENRIDGE.R
Fax (208) 526-0603
Charles N. Smith
U.S. EPA-ERL
College Station Road
Athens, GA 30613
(404) 546-3175
E-Mail SMITH.C
Fax (404) 546-3340
Ray Halley
USDA-NASS
Room 4151 South Bldg.
14th Independence, SW
Washington, DC 20250-2000
(202) 720-2248
E-Mail
Fax (202) 720-0507
Al - 1
-------
Exofficio fTechnical/Administration)
Bruce Jones
U.S. EPA
EMSL-LV, BAD
P. O. Box 93478-3478
Las Vegas, NV 89193-3478
(702) 798-2671
E-Mail JONES.BRUCE
Fax (702) 798-2637,2638 or 2654
Ann M. Pitchford
U.S. EPA
EMSL-LV, BAD
P. O. Box 93478
Las Vegas, NV 89193-3478
(702)798-2366
E-Mail PITCHFORD.ANN
Fax (7.02) 798-2454 or 2221
Members (Current)
Gerald E. Byers
Lockheed Engineering
& Sciences Company
Environmental Monitoring Department
980 Kelly Johnson Drive
Las Vegas, NV 89119
(702) 897-3337
E-Mail
Fax (702) 897-6641
Allen S. Heagle
USDA/ARS
1509 Varsity Drive
Raleigh, NC 27606
(919)515-3311
E-Mail HECK.WALTER
Fax (919) 515-3593
Roy E. Cameron
Lockheed Engineering
& Sciences Company
Environmental Monitoring Department
980 Kelly Johnson Drive
Las Vegas, NV 89119
(702) 897-3318
E-Mail CAMERON.R
Fax (702) 897-6641
Craig M. Hayes
Dept of Agric. Statistics
NC Department of Agriculture
1 West Edenton Street
P. O. Box 27761
Raleigh, NC 27601
(919) 856-4394
Fax (919) 856-4139
George R. Hess
DepL of Plant Pathology
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919)515-3311
E-Mail EMSLRT::HESS
Fax (919) 515-3593
Virginia M. Lesser
DepL of Statistics
Oregon State University
(503) 737-3366 (office)
Al -2
-------
Members Continued
Steve Manheimer
USDA-NASS
Rm 4801, South Building
14th & Independence Ave, SW
Washington, DC 20250-2000
(202) 720-0684
E-Mail
FAX (202) 720-8738
Gail L. Olson
INEL
P.O. Box 16258
Idaho Falls, ID 83415-2213
(208) 526-1870
E-Mail BRECKENRIDGE.R
Fax (208) 526-0603
Julie R. Meyer
METI
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919)515-3311
E-Mail MEYERJULIE
Fax (919) 515-3593
Steven L. Peck
DepL of Statistics
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail HECK.WALTER
Fax (919) 515-3593
Michael J. Munster
Dept of Plant Pathology
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail MUNSTER.MIKE
Fax (919) 515-3593
Dorothy E. Sherrill
DepL of Plant Pathology
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail HECK.WALTER
Fax (919) 515-3593
Deborah Neher
DepL of Plant Pathology
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail NEHER.DEB
Fax (919) 515-3593
Mark Tooley
DepL of Plant Pathology
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail TOOLEY.MARK
Fax (919) 515-3593
Al -3
-------
Consultant
Alva L. Finkner
5425 Ironwood Lane
Raleigh, NC 27613
(919) 787-1483
Fax (919) 515-7591 (Rawlings)
Exoffico
John A. Dunning
USDA/ARS
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-2778
515-2779
Fax (919) 515-3593
Susan Spruill
Dept of Statistics
NC State University
1509 Varsity Drive
Raleigh, NC 27606
(919) 515-3311
E-Mail MEYERJULIE
Fax (919) 515-3593
Karl Hermann
METI
P. O. Box 12313
2 Triangle Drive
RTF, NC 27709
(919) 541-4205
E-Mail HERMANN.KARL
Fax 919-541-1486
Robert Smith, Jr.
USDA-SCS
P.O. Box 2890
South Agricultural Bldg.
Washington, DC 20013
(202) 720-4452
E-Mail
Fax (202) 690-2019
Douglas G. Lewis
DEHNR
Division of Planning and Assessment
P.O. Box 27687
Raleigh, NC 27611-7687
(919) 733-6376
E-Mail
Fax (919) 733-2622
Al -4
-------
APPENDIX 2
List of N.C. Counties Sampled in the 1992 Pilot Project
List of counties being sampled for the Agroecosystem 1992 Pilot. Designations following
county name indicate sample design which selected the county and the number of segments
selected in the county by each design. Hexagon segments chosen by EMAP hexagon 1991
sub-sample; NASS segments chosen using NASS rotational panel design, total segments chosen
in each county.
COUNTY
ALAMANCE
ALEXANDER
ALLEGHANY
ANSON
ASHE
AVERY
BEAUFORT
BERTIE
BLADEN
BRUNSWICK
BUNCOMBE
BURKE
CABARRUS
CALDWELL
CAMDEN
CARTERET
CASWELL
CATAWBA
CHATHAM
CHEROKEE
CHOWAN
CLAY
CLEVELAND
COLUMBUS
CRAVEN
CUMBERLAND
CURRITUCK
DARE
DAVIDSON
DAVIE
DUPLIN
DURHAM
EDGECOMBE
FORSYTH
FRANKLIN
GASTON
GATES
GRAHAM
GRANVILLE
GREENE
GUILFORD
HALIFAX
HARNETT
HAYWOOD
HENDERSON
HERTFORD
HOKE
HYDE
HEXAGON
PLAN
2
I
I
ROTATIONAL
PANEL PLAN
2
1
TOTAL
3
1
1
1
2
1
1
1
2
1
1
1
2
1
1
1
1
1
1
1
1
2
1
1
2
1
1
2
1
1
1
1
3
1
2
1
1
1
1
A2 - 1
-------
IREDELL
JACKSON
JOHNSTON
JONES
LEE
LENOIR
LINCOLN
MCDOWELL
MACON
MADISON
MARTIN
MECKLENBURG
MITCHELL
MONTGOMERY
MOORE
NASH
NEW HANOVER
NORTHAMPTON
ONSLOW
ORANGE
PAMLICO
PASQUOTANK
PENDER
PERQUIMANS
PERSON
PITT
POLK
RANDOLPH
RICHMOND
ROBESON
ROCKINGHAM
ROWAN
RUTHERFORD
SAMPSON
SCOTLAND
STANLY
STOKES
SURRY
SWAIN
TRANSYLVANIA
TYRRELL
UNION
VANCE
WAKE
WARREN
WASHINGTON
WATAUGA
WAYNE
WILKES
WILSON
YADKIN
YANCEY
2
1
1
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
1
1
1
2
1
1
1
1
1
1
1
2
1
1
1
2
1
1
2
2
1
1
1
1
1
2
1
1
2
3
3
3
2
3
1
3
1
1
1
2
1
1
1
1
1
3
1
1
2
NUMBER OF SEGMENTS:
NUMBER OF COUNTIES SAMPLED:
51
49
65
55
116
82
A2 - 2
-------
APPENDIX 3
Expected Data Summaries from the Agroecosystem 1992 North Carolina Pilot
The information contained in this appendix shows how the ARG expects to summarize some of the
data obtained from measurements (indicators) obtained to quantify the assessment endpoints planned
for the 1992 Pilot Project. The indicator (measurement) data obtained, the planned summary statistic
and the type of summary expected are listed in this appendix for the five primary assessment
endpoints planned for the 1992 Pilot.
Indicator
Summary statistic (SI units)
Summary type
land area in a
given use class
hectares (ha) of each JES land use
category
ha of each JES crop type/ ha cropland
total ha cropland (all JES crops)
ha of each Fall survey land use category
for 1992
population estimate+std error
population estimate+std error
population estimate+std error
population estimate+std error
yield by crop
kg/hectare for each crop
CDF (ha crop x yield crop)
fertilizer use
fuel use
use of a given
pesticide class (e.g.
phenoxy
herbicides)
(classes to be
defined)
kg N applied
kg P applied
ha cropland treated with N
ha cropland treated with P
ha cropland treated with N or P / ha
cropland
hectares treated with municipal sludge
liter/hectare
ha cropland fuel rate used / ha cropland
kg active ingredient (or pesticide class)
applied
hectares cropland treated with active
ingredient (or class)
hectares cropland treated (with each
class) / hectares cropland
population estimate+std error
population estimate+std error
population estimate+std error
population estimate+std error
CDF (ha cropland x N/ ha)
CDF (ha cropland x P^a)
population estimate+std error
population estimate+std error
CDF (ha cropland x liter
fuel/ha)
population estimate+std error
population estimate+std error
population estimate+std error
A3 - 1
-------
land area managed
with soil
conservation
methods
(tillage and other
erosion control
methods- 7 total)
hectares managed with each specific
conservation method
hectares managed by each specific
conservation method / ha cropland
hectares managed by each specific
conservation method / ha cropland
managed by one or more conservation
methods
population estimate+std error
population estimate+std error
population estimate+std error
land area managed
with non chemical
pest controls (3
total)
land area managed
with pest control
advice (4 total)
hectares managed with each specific
nonchemical pest control method
hectares managed with each specific
nonchemical pest control/ ha cropland
hectares managed with each specific
nonchemical pest control/ ha cropland
managed with one or more nonchemical
pest control methods
population estimate+std error
population estiimate+std error
population estimate+std error
hectares managed with each specific pest population estimate+std error
control advice method
hectares managed with each specific pest
advice method/ ha cropland
hectares managed with each specific pest
advice method/ ha cropland managed
with one or more pest advice methods
population estimate+std error
population estimate+std error
land area irrigated
hectares cropland irrigated
hectares irrigated / ha cropland
population estimate+std error
population estimate+std error
amount of irrigation
water used
type of irrigation
system
volume water applied / hectare
volume water applied (in SI unit)
volume water applied by each specific
method / volume water applied
land area irrigated by each specific
method / hectares irrigated cropland
CDF (ha cropland x vol
water/ha)
population estimate+std error
population estimate+std error
population estimate+std error
A3-2
-------
source of irrigation
water
volume water obtained from each
specific irrigation water source / total
irrigation water applied
hectares cropland irrigated with water
from each specific source / ha irrigated
cropland
population estimate+std error
population estimate+std error
clay
organic carbon
available water
capacity
porosity
base saturation
exchangeable acidity
% by weight
% by weight
% by volume
% by volume
% by weight
cmol (+)/kg (centimoles positive
CDF (ha cropland x % clay)
CDF (ha cropland x % org C)
CDF (ha cropland x
% avail water cap)
CDF (ha cropland x
% porosity)
CDF (ha cropland x
% base satur'n)
CDF (ha cropland x acidity)
% exch. sodium
pH
electrical
conductivity
extractable
aluminum
cadmium
charge/kg)
% by weight
pH units
dS/m (decisiemens/meter)
cmol(+)/kg
mg/kg soil
CDF (ha cropland x ESP)
CDF (ha cropland x pH)
CDF (ha cropland x EC)
CDF (ha cropland x Al)
CDF (ha cropland x Cd)
A3-3
-------
-------
APPENDIX 4
METHODS - SOILS ANALYSES
1. SAMPLE PREPARATION AND STORAGE
Air-dry samples by spreading the soil out in aluminum pans. Once dry, grind the samples using
a hammer mill (2-4 minutes). Homogenize soil thoroughly by shaking sample in an inflated
plastic bag for 15-20 seconds. Store soil in excess of volumes necessary for analysis laboratories
at room temperature until all data is received and passes DQO standards (Tables 5.2-12 and 5 2-
14).
2. SOEL MOISTURE
Record empty weight of can with lid. Fill the can with 50 ml fresh soil and record the sample
code. Record weight of the can + lid + moist soil. With the lid propped underneath the can,
oven-dry the samples at 90 C for 48 hr or until constant dry weights are achieved. Place lid on
can immediately after removing the cans from the oven. Allow cans with soil to cool to
approximately room temperature before weighing. Record the oven-dry weight of the can + lid
+ dry soil. Record all weights to the nearest 0.1 g, rounding up if > 0.05 and down if < O.OSg.
Calculation:
WET=(AIR-DRY SOIL + CAN) - CAN
DRY=(OVEN-DRY SOIL + CAN) - CAN
% WATER CONTENT = WET - DRY * 100
DRY
Reference:
Hillel, D. 1982. Introduction to Soil Physics. Academic Press, Inc., New York. 364 pp.
3. SOIL TEXTURE ANALYSIS
Air-dry and grind samples using hammer-mill, then oven-dry soil at 90 C for 48 hr before
analysis.
1. Weigh 50 g (oven dry) of soil and place in a blender cup and add 100 ml calgon (or
equivalent) stock solution.
2. Blend mixture for 20 seconds.
3. Transfer the soil suspension to a sedimentation cylinder and make volume up to 1000
ml with distilled water.
A4- 1
-------
4. Shake or stir suspension vigorously. Place cylinder on table and record the time. At the
end of 20 seconds, carefully insert the hydrometer and read the hydrometer at the end
of 40 seconds from the time stirring ceased. Record the reading on the data sheet
5. Remove the hydrometer from the suspension. Record the temperature of the suspension
and the time at which the readings were taken.
6. Take a reading at the end of 6 hours. Insert hydrometer just before the 6-hr reading is
made. Also record temperature.
7. To make up a "control" add 100 ml of calgon (or equivalent) stock solution to a
sedimentation cylinder and make volume to 1000 ml by adding distilled water. Take
a hydrometer reading each time the 40-sec and 6-hr readings are taken. For each time,
subtract the hydrometer reading of the control from the hydrometer reading of the soil
suspension.
8. To further correct the hydrometer readings for temperature, for each degree above 68
F, add 0.2 to the reading to get the corrected hydrometer reading. For each degree less
than 68 F, subtract 0.2 from the reading.
9. Calculate the percent sand in the sample. The hydrometer is calibrated so that the
corrected reading gives the grams of soil material in suspension. The sand settles to the
bottom of the cylinder within 40 seconds, therefore, the 40-sec hydrometer reading
actually gives the amount of silt and clay in suspension. Tne weight of sand in the
sample is obtained by subtracting the corrected hydrometer reading from the total
weight of the sample. The percentage sand is calculated by dividing the weight of sand
by the weight of the sample and multiplying by 100.
10. Calculate the percent clay in the sample. At the end of 6 hr, the silt in addition to the
sand has settled out of suspension. The corrected hydrometer reading at the end of 6
hr represents the grams of clay in the sample.
11. Calculate the percent of silt in the sample. Find the percent silt by difference. Subtract
the sum of percentage of sand and clay from 100 to get the percent silt.
* For organic soils, if < 50 g, record sample weight processed, so a correction can be made in
calculation of proportions of soil that are sand, silt, and clay. Hydrometer readings are divided
by weight of the sample, as described in steps 9 and 10.
Calgon stock solution:
35.7 g hexametaphosphate (NaPO3)6
2.1 g NaCO3
1000 ml water
A4-2
-------
pH should be 8.3-8.5; adjust with Na2CO3
use within 2 weeks
4. ELECTRICAL CONDUCTIVITY (EC)
Procedure .
1. Measure 25 g into a 150 ml beaker, add 50.0 ml deionized water, stir thoroughly with
a glass stirring rod and allow suspension to settle for at least 30 minutes or long
enough for the solids to settle. For organic soils, use 5.00 g.
2. Pour supernatant into centrifuge tube and centrifuge at high rpm. Transfer the
supernatant using a pipet to a container to read EC.
3. Rinse cell with one or more portions of sample. Draw supernatant into the
conductivity pipette to slightly above the constricted part of pipette. Avoid drawing
liquid into rubber bulb. If this occurs, rinse bulb before continuing with the next
sample.
4. Adjust instrument to proper range and record the reading.
5. Rinse cell between samples with deionized water.
Calculation:
Electrical conductivity (EC) of the soil extract is calculated as follows:
EC in dS/m at 25 C = 1.4118 x R.,_.
^Standard
where the value of 1.4118 is the EC of the standard 0.01 M KC1 solution in dS/m at 25 C and
^standard and Rextnci refer to resistance in ohms of the standard (0.01 M KC1) solution and extract,
respectively. Report EC values in dS/m.
Alternate method of calculation: After the cell constant () has been determined, EC of the soil
extract can be obtained from the relationship,
EC, in dS/m at 25 C =J>
R
where is the determined cell constant and R is the resistance in ohms per cm of the soil extract.
5. pH
1. Measure 5 cm3 of soil into 1-oz. cups
2. Add 5 ml deionized water and let sit for 30 minutes
3. Standardize the pH meter:
a. uncover vent hole on electrode
A4-3
-------
b. immerse electrode in pH 7 buffer and set thumbwheel to 7.00
c. when button lights and remains on, press button and hold until meter reads 7.000
d. rinse electrode and immerse in pH 4.0 buffer and turn thumbwheel to 4.01—again
press the button when lighted and hold until 3 decimal places appear.
e. rinse and leave electrode in pH 7.0 buffer between trays of samples
4. Stir each sample using a glass rod before reading
5. Add 10 ml buffer (pH 7) to each sample
6. Cover each sample with plastic wrap and shake buffers for 10 minutes on slow speed.
7. Read buffered pH values using a pH meter
6. EXCHANGEABLE CATIONS IN MEHLICH HI EXTRACT
Exchangeable Ca, Mg, K, Na and Al extracted using Mehlich's double-acid method (Sabbe et
al. 1974, Tucker and Hight 1990) followed by direct current plasma.
1. 1 cm3 sample of soil is extracted with 10 ml of Mehlich HI extractajit (0.2 N acetic
acid, 0.015 N ammonium fluoride, 0.015 N nitric acid and 0.002 N EDTA) by shaking
for 5 minutes at high speed (280 exc/min) and filtering.
2. Dilute sample with LiCl buffer to obtain a final concentration of 3750 ppm Li.
3. Determine cation concentration using a dc plasma spectrophotometer.
Mehlich Extracting Solution
To make 20 liters:
1. 400.2 g ammonium nitrate (NH4NO3)
2. 80 ml stock solution (see below)
3. 228 ml acetic acid (CH3COOH)
4. 16.5 ml nitric acid (HN03)*
*Amount varies from one bottle of HNO3 to another, so concentration must be determined prior
to use and adjustments made to the protocol.
Adjust pH to 2.5 + 0.1. Use nitric acid to lower pH and ammonium hydroxide to raise pH.
Stock solution
To make 1 liter:
1. 138.0 g ammonium fluoride (NH4F)
2. 36.53 g EDTA
Dissolve the NH,,F in deionized water and pour into plastic volumetric flask, then add EDTA.
A4-4
-------
8700 ppm Li Solution
For 20 liters, use 1062.0 g of LiCl. When using a new lot number, new standards must be
diluted.
Mehlich HI high standard
To make 1 liter:
Zn 10 ppm
P 60 ppm
Mn 30 ppm
Fe 40 ppm
Cu 4.0 ppm
Mg 240 ppm
Ca 1000 ppm
K 60 ppm
Na 10 ppm
B 10 ppm
References:
Evans, C. E., and McGuire, J. A. 1990. Comparison of soil test extractants on Alabama soils.
Commun. in Soil Sci. Plant Anal. 21:1037-1050.
Mehlich, A. 1984. Mehlich-3 soil test extractant: A modification of Mehlich-2 extractant.
Commun. in Soil Sci. Plant Anal. 15:1409-1416. (original method).
Sabbe, W.E., W.L. Brekard, J.B, Jones, Jr., J.T. Cope, Jr., and J.D. Lancaste. 1974. Procedure
used by state soil testing laboratories in the southern region of the United States. Southern
Cooperative Series Bulletin 190, Alabama Agriculture Experiment Station, Auburn,
Alabama. 23 pp.
Tucker, M.R., and Hight, P.T. 1990. A comparison of the results from three soil testing
laboratories using the Mehlich-3 extractant on southeastern Coastal Plain soils. Commun.
in Soil Sci. Plant Anal. 21:2197-2208.
7. EXCHANGEABLE ACIDITY
Exchangeable acidity is a measure of the amount of exchangeable acidic cations on the soil
cation exchange complex.
Use BaCl2 extraction. The extracts are then titrated, and the results expressed as milliequivalents
exchangeable acidity per 100 g soil. The extraction and titration procedures are performed with
automated equipment using a mechanical extraction. This method is modified from Thomas
(1982) and USDA/SCS (1984).
BaCl2-TEA buffer solution for mineral soils:
Dissolve 61.07 g BaCl2»2 H2O and 14.92 g TEA in CO2-free, deionized water and dilute to 1.00
A4 - 5
-------
L. Adjust pH to 8.2 with 10% HC1. Protect solution from CO2 contamination by attaching a
drying tube containing ascarite to the air intake of the storage vessel.
BaClj-TEA buffer solution for organic soils:
Dissolve 61.07 g BaCl2«2 H2O and 29.8 g TEA in CO2-free, deionized water and dilute,to 1.00
L. Adjust pH to 8.2 with 10% HCL Protect solution from CO2 contamination by attaching a
drying tube containing ascarite to the air intake of the storage vessel.
Replacement solution (0.5 N with respect to BaClj):
Dissolve 61.07 g BaCl2»2 H2O with 5 ml of the appropriate BaCl2-TEA buffer solution and dilute
to 1.00 L with deionized water.
Acidity by BaCl2-TEA
Mineral soils:
1. Tightly compress a 1-g ball of filter pulp into the bottom of a syringe barrel with a
modified plunger. (To modify the plunger, remove the rubber portion and cut off the
plastic protrusion.) Tap the plunger and syringe assembly on a tabletop several times.
2. Weigh 2.00 g air-dry mineral sample into small glass tube and record exact weight
Place sample tube in upper disc of extractor and connect to inverted extraction syringe,
with the syringe plunger inserted in the slot of the stationary disc of the extractor.
Attach pinch clamp to delivery tube of syringe barrel. Add 10.00 ml BaClz-TEA buffer
solution for mineral soils to the sample. Stir the sample mixture with a glass stirring
rod for 10 seconds. Leave stirring rod in syringe. Allow sample to stand for 30
minutes.
3. Set extractor for a 30-minute rate and extract until 0.5 to 10.0 cm of solution remains
above each sample. If necessary, turn off extractor to prevent soil from becoming dry.
4. Add a second 10.00-ml aliquot of BaCl2-TEA buffer solution and continue extracting
until nearly all solution has been pulled through sample. Add replacement solution
from pipettor in two 20-ml aliquots, passing the first aliquot through the sample before
adding the next. Total time for replacement should be approximately 30 minutes.
Quantitatively transfer extract to an Erlenmeyer flask. Record the total volume of
buffer plus replacement solutions.
NOTE: Deionized water may be used at this point to aid in the quantitative transfer.
The final volume of deionized water should be 100 mi-see Step 5.
5. Titration--Add 100 ml deionized water to extract in Erlenmeyer flask. Use an
automatic titrator to titrate with 0.050 N HC1 to a 4.60 pH endpoint. Record volume
A4-6
-------
and normality of titrant. If the volume of titrant of any sample is less than 5% of that
measured for the blank, resolve the problem before further analysis.
Organic soils
Tightly compress a 1-g ball of filter pulp into the bottom of a syringe barrel with a
modified plunger. (To modify the plunger, remove the rubber portion and cut off the
plastic protrusion.) Tap the plunger and syringe assembly on a tabletop several times.
Weigh 2.00 g air-dry organic sample into small glass tube and record exact weight.
Add 5.0 ml BaQ2-TEA buffer solution for organic soils to the sample, cap, and shake
the tube and contents for 1 hour on a reciprocating shaker. Place sample tube in upper
disc of extractor and connect to inverted extraction syringe, with the syringe plunger
inserted in the slot of the stationary disc of the extractor. Attach pinch clamp to
delivery tube of syringe barrel. Quantitatively transfer contents of small glass tube to
sample tube with 5.00 ml buffer solution.
NOTE 1: Five to 10 ml of buffer solution may be used to transfer soil to syringe—see
Step 4,
NOTE 2: Some organic soils have very high acidity, which may require reducing the
amount of soil to 1.00 g to stay in the mid-range of the titration procedure.
Set extractor for a 30-minute rate and extract until 0.5 to 10.0 cm of solution remains
above each sample. If necessary, turn off extractor to prevent soil from becoming dry.
Add a second 10.00-ml aliquot of BaQ2-TEA buffer solution and continue extracting
until nearly all solution has been pulled through sample. Add replacement solution
from pipettor in two 20-ml aliquots, passing the first aliquot through the sample before
adding the next Total time for replacement should be approximately 30 minutes.
Quantitatively transfer extract to an Erlenmeyer flask. Record the total volume of
buffer plus replacement solutions.
NOTE 1: If 10-ml was used in Step 2, then 5 ml must be used here. Total buffer used
must equal 20.00 ml. A second extraction is essential.
NOTE 2: Deionized water may be used at this point to aid in the quantitative transfer.
The final volume of deionized water should be 100 mi-see Step 5.
Titration—Add 100 ml deionized water to extract in Erlenmeyer flask. Use an
automatic titrator to titrate with 0.100 N HC1 to a 4.60 pH endpoint. Record volume
and normality of titrant If the volume of titrant of any sample is less than 5 percent
of that measured for the blank, resolve the problem before further analysis.
A4-7
-------
Calculation:
Acidity in Bad,,
(meq/lOOg) ~
/"mean blank _ Titrant
I volume (ml) volume (ml
D^. Normality
ofHCl
* 100
MOIST = % water content of soil sample.
References:
Thomas, G.W. 1982. Exchangeable cations. Pages 159-165. In: Page, A.L., R.H. Miller, and
D.R. Keeney (eds.) Methods of soil analysis. American Society of Agronomy, Madison, WI.
U.S. Department of Agriculture/Soil Conservation Service. 1984. Soil Survey Laboratory
Methods and Procedures for Collecting Soil Samples. Soil Survey Investigations Report No.
1. U.S. Government Printing Office, Washington,. D.C.
8. CATION EXCHANGE CAPACITY
Calculated. The concentrations (meq/lOOg) .of the exchangeable cations (K, Ca, Mg, Na) plus
exchangeable acidity should approximate the cation exchange capacity (CEQ.
9. BASE SATURATION
Calculated. Base saturation is given as the total amount of exchangeable base cations (Ca2*, Mg2*,
K*. and Na+) divided by the CEC.
10. MINERALIZABLE NITROGEN
Procedure
1. Place 12.5 ± I ml of water in a 16 mm x 150 mm test tube, and add 5.00 g of air-
dried, sieved (< 2mm) soil. For organic soils use 1.25 g.
2. Stopper the tube, shake, and place it in a constant-temperature cabinet at 30 C for 2
weeks.
3. At the end of this period, shake the tube for about 15 sec and transfer the contents to
a 150-ml distillation flask designed for use with the steam distillation apparatus
described by Bremner (1965).
4. Complete the transfer by rinsing the test tube three times with 3-5 ml of 4 N KC1 using
a total of 12.5 + 1 ml of this reagent.
5. Add 0.25 + 0.05 g of heavy, carbonate-free MgO.
A4-8
-------
Analysis by distillation and titration (Bremner)
Determine the amount of ammonium-nitrogen in the incubated soil sample by collection and
titration of the ammonia-nitrogen liberated by steam distillation of the soil-potassium chloride
mixture for 4 min using the distillation apparatus and technique described by Bremner (1965).
Note: In this technique, the rate of distillation is approximately 7.5 ml per min, and the ammonia
liberated by distillation is collected in a 50-ml Erlenmeyer flask containing 5 ml of boric acid-
indicator solution and is determined by titration of the distillate with standardized 0.100 N HC1.
Analysis by automated distillation-titration
1. Remove sample tubes and quantitatively transfer each sample to a 250-ml digestion
tube. To remove the sample, blow the filter pulp and soil out of the syringe by using
a gently flow of compressed air. Wash with a minimum amount of deionized water.
Use a rubber policeman to complete the transfer.
2. Add 6-7 g NaCl to the digestion tube, spray silicone antifoam solution into the
digestion tube and connect it to the Kjeltec Auto 1030 or similar Analyzer.
3. Follow instructions in manual regarding safety and operation of the analyzer and titrate
to a pH 4.60 endpoint. .
4. Read ml titration and record with the normality of titrant: NH4OAc.
Calculation
Min. N (meq/lOOg) =
Titrant volume x normality of H2SO4
sample wt. x (1-(MOIST/(100+ MOIST)))
x 100
MOIST = % water content of soil sample
References
Waring, S. A. and J. M. Bremner. 1964. Ammonium production in soil under waterlogged
conditions as an index of nitrogen availability. Nature 201:951-952.
Bremner, J.M. 1965. Nitrogen availability indexes. In: C.A. Black et al. (ed). Methods of Soil
Analysis. Part 2. Agronomy 9:1324-1345. Am. Soc. of Agron., Madison, WI.
Keeney, D. R. 1982. Nitrogen-Available Indices. Pages 711-733 in: Methods of Soil Analysis.
Part 2. Chemical and Microbiological Properties. Agronomy Monograph No. 9 (2nd ed.).
ASA-SSSA, Madison, WI.
A4-9
-------
11. EXTRACTABLE PHOSPHOROUS USING BRAY II
1. 1 cm3 of soil is extracted with 10 ml of P2 extracting solution (0.03 N NH4F in 0.1 N
HC1) by shaking for five minutes at high speed (280 exc/min).
2. The extract is then filtered with #2 filter papers and phosphorus quantified using a
direct current plasma spectrophotometer.
Regression equation to compare Mehlich HI and Bray II:
Mehlich-m-P (ppm) = -13 + 0.79 Bray-H-P (ppm), r=0.95", n=59 (Tran et al. 1990)
**: P<0.01
Reference:
Tran, T. S., Giroux, M., Guilbeault, J. and Audesse, P. 1990. Evaluation of Mehlich-m extractant
to estimate the available P in Quebec soils. Commun. in Soil Sci. Plant Anal. 21:1-28.
12. ORGANIC CARBON
1. 1 cm3 of soil is used (assumed weight of 1.2 g of soil per 1 cm3).
2. Ash samples at 360 C for 2 hours.
3. Calculate the percent weight loss with ashing.
4. Percent weight loss by combustion can be transformed to the percent organic matter
determined by the Walkley-Black procedure a regression equation (Storer 1984, 1992).
% OM (Walkley-Black) = 68.4 (weight loss) - 0.5
r = 0.90
Reference: •
Storer, D. 1984. A simple high sample volume ashing procedure for determination of soil
organic matter. Commun. in Soil Sci. Plant Anal. 15:759-772.
Storer, D. 1992. An improved high sample volume ashing procedure for determination of soil
organic matter. Commun. in Soil Sci. Plant Anal, (in preparation)
13. MERCURY
The cold-vapor atomic absorption method, is based on the absorption of radiation at the 253.7-nm
wavelength by mercury vapor. The mercury is reduced to the elemental state and aerated from
solution in a closed system. The mercury vapor passes through a cell positioned in the light path
of an atomic absorption spectrophotometer. Absorbance (peak height) is measured as a function
of mercury concentration. The typical detection limit for this method is 0.0002 mg/L.
A4- 10
-------
Reagents
1- ASTM Type II water: Water should be monitored for impurities.
2. Aqua reeia: Prepare immediately before use by carefully adding three volume of
concentrated HC1 to one volume of concentrated HNO3.
3. Sulfuric acid. 0.5 N: Dilute 14.0 mL of concentrated sulfuric acid to 1 liter.
4. Stannous sulfate: ,Add 25 g stannous sulfate to 250 mL of 0.5 N sulfuric acid. This
mixture is a suspension and should be stirred continuously during use. A 10% solution
of stannous chloride can be substituted for stannous sulfate.
5.
6.
Sodium chloride-hvdroxylamine sulfate solution: Dissolve 12 g of sodium chloride and
12 g of hydroxylamine sulfate in Type II water and dilute to 100 mL. Hydroxylamine
hydrochloride may be used in place of hydroxylamine sulfate.
Potassium permanganate, mercury-free, 5% solution
permanganate in 100 mL of Type II water.
Dissolve 5 g of potassium
7.
8.
Mercury stock solution: Dissolve 0.1354 g of mercuric chloride in 75 mL of Type II
water. Add 10 mL of concentrated nitric acid and adjust the volume of 100.0 mL (1 0
mL = 1.0 mg Hg).
Mercury working standard: Make successive dilutions of the stock mercury solution
to obtain a working standard containing 0.1 pg/mL. This working standard and the
dilution of the stock mercury solutions should be prepared fresh daily. Acidity of the
working standard should be maintained at 0.15% nitric acid. This should be added to
the flask, as needed, before adding the aliquot.
Procedure
1- Sample preparation: Weigh triplicate 0.2-g portions of untreated sample and place in
the bottom of a BOD bottle. Add 5 mL of Type II water and 5 mL of aqua regia.
Heat 2 min in a water bath at 95°C. Cool; then add 50 mL Type n water and 15 mL
potassium permanganate solution to each sample bottle. Mix thoroughly and place in
the water bath for 30 min at 95°C. Cool and add 6 mL of sodium chloride-
hydroxylamine sulfate to reduce the excess permanganate.
CAUTION: Do this addition under a hood, as C12 could be evolved. Add 55 mL of
Type n water. Treating,each bottle individually, add 5 mL of stannous
sulfate and immediately attach the bottle to the aeration apparatus.
Continue as described under step 7.4.
2. An alternative digestion procedure employing an autoclave may also be used. In this
method, 5 mL of concentrated H2SO4 and 2 mL of concentrated HNO3 are added to the
0.2 g of sample. Add 5 mL of saturated KMnO4 solution and cover the bottle with a
A4- 11
-------
piece of aluminum foil. The samples are autoclaved at 121°C and 15 Ib for 15 min.
Cool, dilute to a volume of 100 mL with Type II water, and add 6 mL of sodium
chloride-hydroxylamine sulfate solution to reduce the excess permanganate. Purge the
dead air space and continue as described under step 7.4.
3. Standard preparation: Transfer 0.0-, 0.5-, 1.0-, 2.0-, 5.0-, and 10-mL aliquots of the
mercury working standard, containing 0-1.0 pg of mercury, to a series of 300-mL BOD
bottles. Add enough Type II water to each bottle to make a total volume of 10 mL.
Add 5 mL of aqua regia and heat 2 min in a water bath at 95°C. Allow the sample
to cool; add 50 mL Type n water and 1 mL of KMnO4 solution to each bottle and
return to the water bath for 30 min. Cool and add 6 mL of sodium chloride-
hydroxylamine sulfate solution to reduce the excess permanganate. Add 50 mL of
Type n water. Treating each bottle individually, add 5 mL of stannous sulfate
solution, immediately attach the bottle to the aeration apparatus, and continue as
described in step 7.4.
4. Analysis: At this point, the sample is allowed to stand quietly without manual
agitation. The circulating pump, which has previously been adjusted to a rate of 1
L/min, is allowed to run continuously. The absorbance, as exhibited either on the
spectrophotometer or the recorder, will increase and reach maximum within 30 sec.
as soon as the recorder pen levels off (approximately 1 min), open the bypass valve
and continue the aeration until the absorbance returns to its minimum value. Close the
bypass valve, remove the fritted tubing from the BOD bottle, and continue the aeration.
5. Construct a calibration curve by plotting the absorbances of standards versus
micrograms of mercury. Determine the peak height of the unknown from the chart and
read the mercury value from the standard curve.
6. Analyze all EP extracts, all samples analyzed as part of a delisting petition, and all
samples that suffer from matrix interferences by the method of standard additions (see
Method 7000, Section 8.7).
7. Duplicates, spiked samples, and check standards should be routinely analyzed.
8. Calculate metal concentrations: (1) by the method of standard additions, (2) from a
calibration curve, or (3) directly from the instrument's concentration read-out. All
dilution or concentration factors must be taken into account Concentrations reported
for multiphased or wet samples must be appropriately qualified (e.g., 5 pg/g dry
weight).
References:
Methods for Chemical Analysis of Water and Wastes, EPA-600/4-82-055, December 1982,
Method 245.5.
A4- 12
-------
Gaskill, A., Compilation and Evaluation of RCRA Method Performance Data, Work Assignment
No. 2, EPA Contract No. 68-01-7075, September 1986.
14. SOIL MOISTURE DETERMINATIONS WITH PRESSURE PLATE
Pack soil to a given bulk density in the rings used with the pressure plate apparatus.
Nitrogen in pressurized tanks are used to achieve pressure within the pressure plate apparatus.
Apply pressure (tension) to the pressure plate to achieve equilibrated matric potentials of -5, -10,
-33, and -1500 kPa (equal -0.050, -0.10, -0.3, and -15 bars, respectively).
Soil water content must be determined at each soil matric potential. Soil water content at soil
saturation must be known for proper calibration of water content of unsaturated soils.
Express the pore volume available at each matric potential as a function of soil water content (%
soil volume). A water release curve can be drawn by plotting the soil water content as the y-axis
and the soil matric potential as the x-axis.
NOTE: Soil moisture retention in a low-suction range (0-100 kPa) is strongly influenced by soil
structure and pore-size distribution. Hence, measurements made on disturbed samples cannot be
expected to represent field conditions (Hillel 1982).
Reference:
Hillel, D. 1982. Introduction to Soil Physics. Academic Press, Inc., New York. 364 pp.
A4- 13
-------
-------
APPENDIX 5
NASS SURVEY QUESTIONNAIRES
The first part of this Appendix contains the complete NASS questionnaire that will be
administered in November 1992 for the Agroecosystem component of EMAP. The survey
questionnaire is in draft form at this time and is not for distribution, as NASS had not yet given
approval for distribution.
The second part of this Appendix is the subset of the June Enumerative Survey which will
be used on the segments selected by the Hexagon Design. It contains the eight extra questions
which the ARQ, with the concurrence of NASS, has added to the regular JES specifically for the
Agroecosystem Program. Segments selected by the Rotational Panel Design will recieve the full
JES, including the eight exiia questions (7a, 10, 51; 51a, 51b, 52, 52a, 52b).
A5- 1
-------
NATIONAL
AGRICULTURAL
STATISTICS
SERVICE
US Department
of Agrkurtute
Washington, DC
202SO
1992
EMAP PILOT SURVEY
Form Approved
OMB Number 0535-Om
Expiration Date 09/30/93
PROJECT CODE »20
North Carolina
StMe
Stratum
Segment
00000
Tract i Subtr.
CONTACT RECORD
DATE
TIME
NOTES
3
9
COMPtl
- COMPLETED
-REFUSAL
- INACCESSABLE
TION CODE
001
; INTRODUCTION |
i[lntroduce yourself and ask for the operator. Rephrase in your own words.] ' \
i > i
| The National Agricultural Statistics Service in cooperation with North Carolina State ]
! University is conducting a survey of farm chemical use and cropping practices as they relate to !
'the environment. Information from this and other surveys will be used to monitor the1
J agricultural and environmental conditions within North Carolina. This information will be \
i used only for environmental analysis. Authority for collection of this data is Title 7, Section ,
' 2204 of the U.S. Code. Response to this survey is confidential and voluntary. •
i We encourage you to use your farm records during the interview. ,
BEGINNING TIME (MIUTARY\.
063
A5 - 2
-------
-2-
FiELD IDENTIFICATION
[Show aerial photograph to respondent and identify sample field.]
1. Did you make any of the day-to-day farming decisions for this field
in 1992?
YES - [Enter Code 1]
[If NO, conclude the interview, and ask for the respondent's
assistance in locating the correct operator.]
2. How many acres are in this field? (Include woods, waste, etc.),
3. How many acres in this field are considered cropland? .
coot
4. Do you {Does this operation) own this field or rent it?
[Enter code 1 for OWNED;
enter code 2 for RENTED, LEASED or USED KENT FREE.]
064
ACRES
069
ACHES
060
coot
073
CROP AND LAND USE CODES
ALFALFA. HAY
APPLES
ASPARAGUS
BARLEY
BEANS.
DRY
SNAP, ALL
GREEN LIMA
BEETS
BROCCOLI
BUCKWHEAT
CABBAGE, ALL
CANTALOUPS
CARROTS
CAULIFLOWER
CHERRIES, ALL
CHRISTMAS TREES
. CLOVER
CORN.
» FIELD
SILAGF
aWEET
CUCUMBER, ALL
c
E
c
c
c
c
c
c
c
112 EGGPLANT
10 FORAGE, ALL
63 GRAPES, ALL
311 GRASSES OTHER
THAN CLOVER
116 GREENS
, 11 HAY, ALL OTHER
117 LETTUCE. ALL
93 MUSHROOMS
95 NURSERY &
FLORAL CROPS
c
c
c
c
c
c
c
c
a
c
c
c
c
OTHER CROPS (Specify)
C
C
L.
c
r~
i_
c
c
15 OATS
118 OKRA
ONIONS.
120 DRY
119 GREEN
ORIENTAL VEG..
148 ALL
121 PARSLEY
68 PEACHES, ALL
16 PEANUTS
69 PEARS, ALL
PEAS.
122 GREEN
123 OTHER
PEPPERS.
126 BELL
127 ALL OTHER
71 PLUMS
19 POPCORN
20 POTATOES, IRISH
128 PUMPKINS
too CAmtucc
ity KAUIbnti
140 RASPBERRIES
22 RYE
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
f-
1—
c
c
c
c
c
r
c
SORGHUM.
25 GRAIN
24 SILAGE
26 SOYBEANS
132 SPINACH
133 SQUASH
74 STRAWBERRIES
30 SUNFLOWERS
31 SWEET POTATOES
32 TOBACCO
134 TOMATOES, ALL
145 TURNIPS
33 WATERMELONS
34 WHEAT, ALL
OTHER LAND USES
301 PASTURE
2/f) f Q D
3U/ w.ft.r.
303 SET ASIDE
304 IDLE CROPLAND
OA1 PAI 1 f\\Af
y\J I rMLLv/W
305 WOODLAND
313 WETLAND
314 RANGELAND
306 NON-AG
OTHER (Specify)
A5 - 3
-------
-3
LAND USE and TILLAGE HISTORY
B
1. Now I'd like to obtain the land use history for this field for the past three
years. Please report all crops grown, including cover crops. Let's start with
the 1992 crop year. What was the field used for in 1992?
[Use a separate line for each use of the field each year.]
10
When
was
harvest
completed?
How much
did the
[unit recorded
in column 7]
weigM
I rf reported iratci
pourttb enter I]
When was
this crop
planted?
[Record
reported
unit]
What
was the
average
yield
P«r
acre?
How many
acres were
planted?
\tfcolumn 2 utt a
not*crop, record
jcre* of report erf
bndust \
How many
acres were
harvested?
coot
a. [If soybeans were reported in 1992J
What variety of soybeans were grown? {Enter Code}..
241
A5 - 4
-------
-4-
B
land use and tillage history—continued
B
For the remainder of this interview we will be asking for information for only the
1992 crop year.
2. [Ask only if crops, idle cropland and/or government program land
was reported in item 1.]
Now I'd like to obtain the tillage history for this field for the 1992
crop year.
1
CROP
OR
LAND USE
[Write in]
2
CROP
CODE
245
246
247
3
What type of tillage was
used on this field in 1 992?
2 NO-TILL
3 RIDGE-TILL
4 MULCH-TILL (OR OTHER
CONSERVATION TILLAGE)
5 CONVENTIONAL
(MOLD8OARD PLOW)
6 OTHER CONVENTIONAL
251
252
253
4
What erosion control
methods were used on this
field?
1 NONE
2 TERRACING
3 CONTOUR CROPPING
ORPIOWING
4 STRIP CROPPING
5 GRASSED WATERWAYS
6 OTHER (SPECIFY)
1st
261
262
263
2nd
271
272
273
3rd
281
282
283
3. Has the Soil Conservation Service evaluated this field?
Q YES -[Enter Code land continue.]
D NO - [Skip to Section C, page S.]
a. Has the Soil Conservation Service classified this field as
"Highly Erodible"?
D YES - [Enter Code 1 and continue.]
D NO - [Skip to Section C, page S.]
coot
265
CODE
266
A5 - 5
-------
-5-
FERTILIZER USAGE HISTORY
SOIL TESTING
1. Were any soil tests made for this field:
a. in 1992? DvES-IfnferCodel]...
DNO
b. in 1991? D YES-[Enter Code 1]
DNO
1992
300
1991
301
SLUDGE USAGE
2. Has municipal sludge been applied to this field at any time during
the last five years?
D YES-[£nterCode1] :
D NO
COO€
309
MANURE USAGE
3. Was manure applied to this field at any time during the 1992
crop year? (Excludesludge.)
D YES -[Enter Code 1 and continue.]
D NO- [Skip to item 5.]
4. Now 1 need to get some specific information about the manure
applications for alt crops grown in this field this year.
COO€
310
1
CROP OR
LAND USE
{Write In]
2
CROP
CODE
391
392
393
3
What kind
of manure
was applied
during 1992
crop year?
\tHff*COOf\
311
312
313
MANURE TYPES
1 CATTLE
2 HOG
3 SHEEP
4 GOATS
5 CHICKENS
6 TURKEYS
7 HORSES
8 OTHER
{Specify)
4
How much
was applied
per acre?
321
322
323
5
UNIT
CODE
LB5 . 1
CWT » 2
TON •• 3
331
332
333
A5 - 6
-------
fertilizer usage history—continued
5. Was any Lime or Gypsum used on this field for any crop in 1992?
[] YES - (Enter Code 1 and Complete table.]
D NO - [Go to item 6 on Page 7.]
1200
1
CROP
OR
LAND USE
[Write In]
2
CROP
CODE
395
396
397
398
3
MATERIAL
UME
GYPSUM
UME
GYPSUM
4
How many tons
were
applied per
acre?
370
371
373
373
5
How many
total acres
were
treated?
ACHES
376
377
378
379
Notes and Calculations:
A5 - 7
-------
-7-
fertilizer usage history—continued
COMMERCIAL FERTILIZER USAGE
6, Were commercial fertilizers applied to this field at any
time during the 1992 crop year?
0 YES - (Enter Code 1 and continue.]
coot
320'
Q NO - [Skip to Section D, page 8.}
7. For each fertilizer applied to this field in the past year, I
need some information on the analysis applied and the
amount applied. What was the first fertilizer you applied ?
(Include sidedressing.) [Complete table.]
T-TYPE
2
TABLE
001
MATERIAL UNIT CODES
1 Pounds of materials
12 Gallons of materials
15 Ounces
19 Actual nutrients (pounds)
L
N
E
01
02
03
04
05
06
07
oa
09
10
11
12
13
14
15
1
CROP
OR
LAND USE
[Write In]
2
CROP
CODE
MO
080
080
080
oeo
080
080
OftO
080
080
oeo
080
080
080
080
3
MATERIAL USED
[Enter percent inttyfii or
taial pounds of pltnt nutrients
ipplitdptr ten.]
N
082
082
082
082
082
082
082
082
082
082
082
082
082
082
082
P
083
083
083
083
083
083
083
083
083
083
083
083
083
083
083
K
084
084
084
084
084
084
084
084
084
084
084
084
084
084
084
4
How much
was applied
per acre per
application?
(Leave ttm
coluTTTn blank if
actual nutrients
were reported
085
085
085
085
085
085
085
085
085
085
085
085
085
085
085
5
[Enter
Unit
Code]
066
086
066
086
0(6
OK
0*6
046
086
086
086
086
086
086
086
fi
HOW
many
acres
were
treated?
M7 ._
"7 •__
W7 ._
M7 ._
M7
0»7 ._
087 .,_
087
0«7
087 ._
087 ._
087
087
087 . _
087
T-TYPE
0
TABLE
000
LINE
00
OFFICE USI
007
A5 - 8
-------
-8-
PEST MANAGEMENT
D
1. Were any pesticides (such as herbicides, insecticides, fungicides, nematkides,
defoliants or growth regulators) applied to this field in 1992?
CODE
D YES - [E nter Code 1 and complete table.]
n NO- [Go to item 2, page 9.]
APPLICATION METHODS
1 Broadcast (Ground) S Band In/Over Row
2 Broadcast (Air) 6 Directed Spray
3 In Furrow 7 Chiseled/Knifed-in
4 Irrigation Water 8 Foliar Application
9 Spot Treatment
T-TYPE
3
TABLE
002
L
1
N
E
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
1
Crop
or
Land Use
[Write In]
2
Crop
Code
•90
no
•90
no
no
no
no
no
no
no
no
no
no
no
no
3
What
pesticides
were
applied?
IfnrerCocfel
092
092
092
092
092
092
092
092
092
092
092
092
092
092
092
4
How
many
acres
were
treated?
ACRES
093
093
093
093
093 .
093
093
093
093
093 .
093
093
093
093
093
5
How much
was appliec
per acre
(per
application}
RATE
094
094
094
094
094
094
094
094
094
094
094
094
094
094
094
6
\CnterUnit
Code]
1 Pound
12 Gallon
13 Quart
14 Pint
15 Ounce
09S
095
OSS
09S
09S
09S
095
095
095
095
095
095
095
095
095
7
How was
it
applied?
IfnferCodtl
096
096
096
0%
09*
09t
096
09C
096
096
096
096
0%
096
096
8
Number
of times
applied?
099
099
099
099
099
099
099
099
099
099
099
099
099
099
099
[ENUMERATOR NOTE: If any chemical is reported for which no code is on the listing sheet,
complete the appropriate line in the table above (leaving out the unknown product code), and
record the name and a description of the chemical below.*
LINE NUMBER
CHEMICAL NAME & FORMULATION
LIQUID OR DRY PRODUCT EPA NUMBER
A5 - 9
-------
-9-
D
pest management-continued
Now I'll be asking about pest management and services for crops grown in
this field. Consider the management and services you used for insect
management, weed control, etc.
Considering the crops grown in this field,
have you consulted with any of the
following for pest management in 1992--
a. Hi red Staff? (Indude only those trained in pest
management, entomology, etc)
DNO
b. Local extension service/university/state/federal?
NO
c. Chemical dealer, supplier or store?
Quo
d. Professional scouts?
(Exclude scooting provided by a chemical supplier.)
D YES* [Enter Cod* 1] .....
DNO
T-TYPE TABLE LINE
0 000 00
CROP OR LAND USE
(Enter Code]
574
575
576
577
578
540
541
542
543
S**
545
546
547
548
549
3. Now I need to ask you about some specific pest management practices
you may have used for the crops harvested from this field this year.
a. Was the specific variety of the crop(s) you planted this year chosen for
pest or disease resistance?
L!YES-l£nterOod«1]
DNO
587
b. Did you use pheromones or insect traps for monitoring and/or controlling pests?
D YES-[EnterCode11
DNO
589
c. How about crop rotations?
D YES- [fnterCodel] ..,
DNO
593
A5 - 10
OFFICE U${
009
-------
- 10-
FIELD OPERATIONS
Now I'd like to find out how many gallons of fuel were used in this field for the 1992 crops.
To do this we'll collect information about each piece of equipment and machinery used on
the field and the amount of fuel it used. Let's begin with the first operation performed after
the 1991 crop harvest.
I
1
N
E
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
1
What crop
was this
for?
2
Crop
Code
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
3
What type of
operation
was done?
[Write In)
4
Machine
Code
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
5
What
was the
PTO
horsepower
of the
tractor
used?
[Code]
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
6
What type
of fuel
did this
tractor use?
1 DIESEL
2 GASOLINE
3 IP GAS
4 OTHER
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
005
7
How many
total gals.
of fuel or
How many
gals, of fuel/
acre were used?
Total Q Gals/
Gals Acres
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
035
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
IENUMERA TOR NOTE: If an operation is reported for which no code is on the listing sheet,
complete the appropriate line in the table above (leaving out the unknown machine code).
Record the line number, the name and a description of the machine below. J
LINE NUMBER MACHINE NAME & DESCRIPTION
Off ICE USE
010
A5 - 11
-------
-11-
IRRIGATION and DRAINAGE
1. Was this field irrigated for any crop harvested in 1992?
D YES- [EnterCode 1, and complete table.].
Q NO- [Go to Item2.]
CODE
ess
IRRIGATION SYSTEM CODES
SPRINKLER
CENTER PIVOT
LATERAL MOVE
HAND-MOVE
END-TOW
WHEEL MOVE
6 SOLID-SET OR PERMANENT
7 REEL TYPE OR TRAVELING GUN
8 OTHER SPRINKLER (Of SCR/Bf)
GRAVITY
9 OPEN DITCH WITH CUT OUT
10 OPEN DITCH WITH SIPHON TUBES
11 GATED PIPE
12 GATED PIPE WITHSURGE CONTROL
13 CABLEGATION
14 OTHER GRAVITY (Of SC/f/Sf)
DRIP OR TRICKLE
15 DRIP WITH BUBBLERS
16 DRIP WITHOUT BUBBLERS
17 SUBIRRIGATION
It OTHER (D£5G?/B£)
WATER SOURCE COOES
1 PURCHASEDWATER
2 WELLS
3 PONDS
4 LAKES. RIVERS. CANALS
5 RETURN. WASTE WATER
AND OTHER WATER
1
CROP OR
LAND USE
[Write In]
2
CRO?
CODE
£97
698
£99
3
How many acts
were irrigated in
1992?
ACHES
661
•
662
•
663
•
4
What type of
irrigation system
was used?
IfWTfRCOOf)
664
66S
666
5
What was the
main source of
irrigation wat(sr7
N
IFWTfRCODfl
667
668
669
6
What was the
average number
of inches of
water applied
per acre in 1992?
INCHES
670
671
672
2. How many acres in this field are drained by subsurface (tile) drains?.
ACRES
696
OFFICt USt
Oil
A5 - 12
-------
-12-
IRRIGATION
1. Was this field irrigated for any crop harvested in 1992?
D YES •[Enter Code 1]
D NO - [Enter Code 2 and go to
Section G. Conclusion.l
2. How many acres of each crop were irrigated in this field? &*p.LH
Crop 2
Oop 3
3. What type of irrigation system was used? [Enter code for each crop.} OOP v
Crop]
OopJ
iODt
4. What was the source of irrigation water? [Enter code for each crop.]
Ml5^K«(J*o t&llttr Gtr« by ic^cz-
Oopi
Oop 2
Oop]
a
n
n
5. How many acre-inches of water were applied to each crop? <>?j?.1..
OopJ
OopJ
*Ott-
-------
-12-
CONCLUSION
This concludes our interview. Thank you for your cooperation.
[Review this questionnaire.]
RESPONDENT
OPERATOR/MANAGER «»
SPOUSE =2
OTHER *3
RESPONDENT'S NAME.
PHONE
coot
105
[Did respondent use farmlranch records fo report the majority of this data ?
ENDING TIME {MILITARY].
ENUMERATOR
Fimuzt*
PESTODEA
CHEMICAL
AFFIXATIONS
•06
SUPflCMtNTS USED
•07
•09
DATE
MMDDVY
{VALUATION
EltUMERATOKO
110
•11
•12
[ENUMERATOR NOTE:
If other people (custom applicators, contractors, etc.) were
contacted for assistance in completing this questionnaire,
please record their names and phone numbers below.
Also, use this space for any additional notes or comments.]
A5 - 14
-------
NATIONAL
AGRICULTURAL
STATISTICS
SiRvtci
US Department
of Agrkutture
WiiMngton. D.C.
202SO
1992
JUNE EMAP SURVEY
Authority lor collection of information on the June (MAP Survey
it Title 7. Section 22M of the U S Code The information will be uitd
to prepare agricultural estimates Individual leporu tit confidential
Response if voluntary
form Approved
O.M.I. Number OS1VMX1
Approval tipjfes S/Z1/93
Are* Version
NORTH CAROLINA
Project Cod* »20
Segment Number:_
Tract Letter:
.County:.
State
Stratum
Segment
00000 .__
Tract No
. 00
OfFICt USE - OMKJNAl
407
4W
1. I need to make sure we have your (the operator's) correct name and address.
Name of Farm.
Ranch, or Operation:.
Name of Operator:
Ifvstl
Address:.
Iffot/tecrStrverl
ICryl
Telephone: ( )
ISutel
\Zift Codel
|A/«» CooW IMumtefi
2. On June 1. were the day-to-day decisions for this tract of land made by
an individual operator, by partners, or by a hired manager?
Q [Individual - enter 1J
Q JPartnen - enter number of partners, including operator]
Q [Hired manager • enter8]
A5 - 15
-------
r
PAGE 2 CROPS AND LAND USES
How many acres ar« inside this blue tract boundary drawn on the photo (map)? .
Now I would like to at* about each field mvo> this blue tract boundary and m use during
1992.
FIELD NUMBER
1 Total acre* m field
3 Ciop o» land ujw lipeciryi
3. Occupied farmjtead o> dwelling
4 Wood*, roadi. ditches, vacant farmstead, etc
Permanent-net me/op rotation
5. Pasture
Cropland-used on»» »o» pastu«»
7. Idle cropUnd- idle alt dining 1992
7*. idle cropland in government p/ograrm
8. Two crop* planted m Urn field to harvest
thrt year ot two urn of the lame crop
\lpetitylKond crop or use!
/Urn
t. Acres left to be planted
11 AotfSiirtgated and to b« irrigated
IS. Planted
Winter Wheat
U. Fotgram
17. Planted andto be planted
Ky.
It. Fotgraui
lj_ PUnted and to be planud
Oao
j» Fo* gr *n
2i Planted and to be planted
Kilty
ZJ. for orain
3j Planted and to be planted
Corn
J4. . Fo» 9«*n
2(, Planted and 10 be pUnted
Scwghumlfjic/ud*c/ou«»ifhJ«l*n| - -•"
J7 fa cj/wn
21. Other ut« o» grami pJanted U»e
{atxindoned.iilage.tu )
Acm
2) Alfalfa and aMallamixtixn
30. Hay Gram
(cut and to be tui) '
jj. Other hay
33 Planted and to b« planted
loybeam
jt FoBowing another crop
3Sc, iutley Attes
Tobacco
3i± Flue-ctned *«e«
3t Peanutl Planted and to b< planted
}(. upUndConon Planted and to txpUnltd
(Net jc/f ! if slip roivedl
4t. (ruh PoutOrt Planted and »b«pi*nt«J
47. Sweetpotato« Planted and to be pUnt*d
41. Other cropi Aciei planted ex m
-------
CROPS AND LAND USES
{Enter totil trtct acres)
FIELD NUMBER | 06
1. Total acres in held
2. Crop 01 land M» (specify)
4. Woods, roads, ditches. vacant farmstead, etc
Permanent-not in crop rotation
Cropland-used only for pasture
7. ldl« cropland • idle M during 1 992
7*. Idle cropland in government programs
1. Two cops planted in this held for harvest
this year or two us«s of the same crop
Itpec/rV lecono'crop or tat\
Acres
». Acres left to be planted
10. Acres irrigated and to b« irrigated
lincSude ooubfe crop acrei)
IS. Planted
Winter Wheat
14. For grain
1 7. Planted and to be planted
Hye
It. For gram
U. Planted and to b« planud
20. For grain
21. Planted and to be planted
22. For grawi
23. Planted and to be planud
Corn
24. For grain
26 Planted and to be planted
27. For grain
21. Other uses of grains planted Use
(aoanaonea. mage, etc.)
Acre*
29. Alfalfa and alfalfa mixtures
30. Hay Cram
32. Other hay
33. Planted and to be planted
34 Following another croc
3Sc- Burley Acres
tobacco
3Sd. Flue-cured Aon
34 Peanuts Planted tnd to be planted
38. Upland Cotton Planted and to be planted
INet ies I] No
M4
610
•
620
*
540
»
541
•
547
•
548
•
533
$34
535
536.
S30
531
570
571
.
653
656
6S4
600
602
732
315
690
524
8M
*
558
•
.
09
828
M1
•
M2
856
857
MS
1 !Y*s 1 I NO
M4
610
620
•
540
541
547
•
543
•
533
534
535
536
530
531
570
571
,
653
656
654
600
602
*
732
315
690
524
8M
558
PAGE 3
Office Use
Total Acres
00
840
•
A5 - 17
-------
PAGE 4
CROPS AND LAND USE (Continued)
[Refer to photo and point out blue tract boundaries]
51. Inside these blue lines, is there a pond, either constructed or naturally formed.
used to provide water for livestock, fish and wildlife, irrigation, or other related uses?
DYES D DON'T KNOW •
I DNO • 3)
[inter code then go to Item 52}
51a. How many?
51 b. At any time during 1992. will any of them be used for irrigation water?
Q YES • 1 n DON'T KNOW .2 QNO « 3 |£ntercode|
803
805
806
52. Are there any water wells, drilled or dug for any purpose, inside the blue lines?
DYES DOONTKNOW - 2 \_{lntercodelnen go to Total Acres Operated]
I UNO • 3/
52a. How many?.. —
52b. At any lime during 1992, will any of them be used for irrigation water?
Q YES - 1 n DON'T KNOW -2 QNO « 3 Jfntercode]
TOTAL ACRES OPERATED
[IF HIRED MANAGER CHECKED ON FACE PAGE (921 * I). CO TO ITEM 2[
1. Now I would like to ask about the total acres operated under this land arrangement.
Include farmstead, all cropland, woodland, pastureland, wasteland, and
government program land.
807
809
810
901
1a. On June 1. how many acres did this operation own? *. *—
902
1b. Rent from others? [Exclude /and used on an anima/ unit month (AUM) bnn\ 1- *—
905
1d. Rent to others? —- -" * -
900
1u TK»n ih» total jfrgs operated under this arrangement was ITEM It + Ib-ld : - - * 1 , ,
[GO TO ITEM 3]
2. Now I would like to ask about the total acres operated as a hired manager.
On June 1, how many acres were operated for others as a hired manager
under this land arrangement? .
3. Does this include the farmstead, all cropland, woodland, pastureland.
wasteland, and government program land? [If not. make corrections)
CONCLUSION
(Chec* type or" respondent and enter code]
D Operator /Manager » 1
DSpouse . • 2
D Other Jfnter name be/ow] « 3 >
QObsR « 4
DObsNR • 5
QPartlnt • <
[Record name of respondent if not the operator or spouse) enumerate*»
Enumerator: . 098
Date:
|A/ote$ about respondent's answers or other da la collection problems]
A5 - 18
May
26 149
29- ISO
30-ISI
31-152
June
01-153
02 IS4
Junt
Oi 155
04-1Sf
05 157
06-15*
07 159
M 160
09 161
1H62
11-163
U-164
13-165
14-166
15-167
Julian Dili
987
OHU* UM
Quality Ratin|
100
-------
APPENDIX 6
ENUMERATOR MANUAL FOR SAMPLING SOIL
I.
LOCATING THE 5-ACRE SAMPLING AREA IN THE FIELD
Vt
• '»
IM
«M
*t
M
% % %
fWFi
^ Lv I * I>L^fe
VMM
» ^'i' P IH H
^tSlM
UMH
H x L M
HMM
nnnnuH
HM
M
vfa>t>fc
MM
MM
M
M
M
M'k'te
V^lM
M^M
M
M
M
MM
M
M M M
MMM
M
M
M
M
M
M
M
M
M
Figure 1. Starting point in odd-shaped fields.
The point of entry into the field will be the first comer
of the field which is reached when approaching the
field. If the field has NO definite corners, enter the
field from the point which is most accessible by car.
Remember that the point or corner selected for entry
into the field MUST allow an opportunity for the units
to fall ANYWHERE within the sample field
boundaries (excluding Form A deductions) following
the procedures outlined in the manual. If the field has
been selected for more than one sample, the second
closest corner to the starting corner will be used as the
starting corner for the second sample number. Every
sixth field will have two samples and the second
sample will be double-sampled.
The following steps outline procedures to follow when locating and laying out sample units.
STEP 1 Determine the starting corner. This will be the first corner of the field
which is reached when approaching the field.
STEP 2
STEP 3
IMPORTANT*
*EXCEPTIONS*
Walk along the end of the field the required number of paces (steps).
This will be your entry point into the field.
Then walk the specified number of paces (steps) into the field. Start
your first pace about one and one-half feet outside the plowed edge of
the field. .
If you cross any of the acres deducted as "Other Uses" on the Form A
while you are counting paces, stop counting at the start of each such
area and resume counting at the other side. However, any blank or
unplanted areas in the field that were not deducted should be included
in the row and pace count.
1. "Bounce back". When pacing along the edge of the field, or pacing
into the field, you reach the opposite end or side of the field and still
have not taken the required number of paces, turn around and walk back
in the direction from which you came until the required number of
paces has been stepped off.
A6- 1
-------
r
STEP 4
STEPS
STEP 6
"IMPORTANT*
STEP?
STEP 8
*IMPORTANT*
2. Odd-Shaped fields. The bounce-back rule applies. However, as in
Figure 1, you should count paces only while walking in the initial
direction. If the field border takes an abrupt turn, follow the border, but
do not count paces that are not in the initial direction of walking. Then
count paces into the field in the usual way.
3. Edges. If the random starting point lands at a corner or an edge, turn
at 90-degree increments to your right until the sampling transect fits
within the field.
After you have taken the last of the required paces, place a yellow stake
at the toe of your shoe. Lay the right-angle on the ground with the red
point touching the stake. Place a second yellow stake at the right
corner of the right-angle.
Flip the right-angle 180° and place a third
the right-angle to form a straight
line with all three stakes (Figure
2).
Beginning at the center stake,
take two and one-half paces,
staying in a straight line with the
three yellow stakes. Place a red
stake at the toe of your shoe.
Consider this stake 1.
Bellow stake in the corner of
to* of'iho*
Carry the right-angle and 10 red
stakes with you will pacing off
transect for sampling soil for
later use in sampling soil.
Walk five paces from the stake
and place a second red stake at
the toe of your shoe.
Figure 2. Placement of yellow reference
stakes at center of transect
Repeat step 7 until 10 red stakes have been inserted into the soil since
the original center stake. The transect should be diagonal across rows
(Figure 3).
If you reach a border of the field while walking along the diagonal
transect, turn a 90-degree angle and proceed with your paces and
inserting stakes. Repeat the 90-degree rule for each border encountered
(Figure 4).
A6-2
-------
Figure 3. View of entire diagonal transect across the sampling area.
STEP 9
STEP 10
STEP 11
STEP 12
"IMPORTANT*
From the last stake (number 10), take 1 soil core 3' (marked red on the
right-angle) from the red stake measuring away from the yellow
reference stakes. Take 2 cores at each stake if the field is classified as
double-sampled. Pull the stake after taking the soil core.
Repeat step 9 for the remaining stakes walking toward the original
yellow reference stakes for all even-numbered stakes; otherwise take the
soil sample 1.5' from the red stake.
Stakes
1,3,5,7,9
2, 4, 6, 8, 10
Distance Soil Sample
Taken From Stake
1.5'
3'
Beginning at the center stake reverse your direction and repeat steps 6
through 10.
Remove all stakes and exit field.
Make sure all stakes, soil probe, and bucket are free of soil before
leaving the field area. Rinse all equipment thoroughly with water.
A6-3
-------
Bounce rules
Samp I Ing start t ng
point
pec«s
Figure 4. Bounce-rule of 90-degrees for each border encountered.
H. TAKING THE SOIL SAMPLE
For each core, push the soil probe straight down into the soil, without twisting, to the depth
that fills the entire length of the tube (8"). Pull up the tube and push it down onto the bolt
(in wooden block) to empty the core into the bucket; for some soils, a large screwdriver can
be used to scrape the core out of the tube. If the core is less than 8" in depth, take another
core within 6" of the same location. If it is impossible to reach 8" depth with the tube,
collect at least 4" deep cores and an additional core to result in the same volume that 20-8"
deep cores would provide; record the problem on the survey form. Do not accept cores less
than 4" in length. Combine all cores sampled per transect in the bucket (20 cores for regular
samples, 40-cores for double-samples).
NOTES:
In the probe set, three tips will be available for the core tube for sampling soil under a
range of conditions. The regular tip (2 notches), mini tip (1 notch) and super duty tip
(3 notches) are for sampling moist, dry, and stony soils, respectively. A "wrench" for
changing tips is included in each probe set
Discard any rocks larger than 1" diameter. Do not remove plant or other organic
debris from the soil surface, but keep as part of the sample.
A6-4
-------
ffl. LABELLING AND TRANSPORTING THE SAMPLE
When all cores have been deposited into the bucket for 1 composite sample, break up the
clumps gently (excessive pressure or mechanical abrasion may kill nematodes). Mix the soil
thoroughly. Fill the plastic beaker to the surface with soil and pour into a plastic bag marked
"A" and close the bag with a wire tag with the appropriate sample number.
Transfer the rest of the soil (large sample) into a plastic bag marked "B" and close with the
wire tag with the appropriate sample number. Note that double-samples will have two wire
tags.
"IMPORTANT* Record the date collected, date mailed, and enumerator code on the bag
labels and associated postcard for the field.
Store all samples in the cooler (in the shade!) at all times to avoid
temperatures lethal to nematodes!
Mail the beaker (small) sample in the small mailing envelope marked with an "A" and the
large sample in the large mailing envelope marked with a "B" using Federal Express
overnight-delivery. Mail one sample per container using the pre-addressed, postage-paid
envelopes provided. Use strapping tape to close the mailing envelopes. Samples can be
dropped-off at the county Soil Conservation Service office. You may also arrange to have
samples picked up at a residence or office address. All samples should be mailed on the
same day of sampling or first thing the following day. For pickup, call Federal Express at 1-
800-238-5355. The time of the latest pick-up time of a day is available from Federal Express
on a 24-hour a day basis by calling the same 1-800 number and providing the zipcode for the
pickup address.
For example, some pick-up deadlines are:
Raleigh M-F 6 pm, SAT 4 pm
Gates M-F 12 noon, no SAT pickup
Danbury M-F 12 noon, SAT 12 noon
Brevard M-F 4 pm, no SAT pickup
New Hope M-F 2:30 pm, no SAT pickup
It is important to keep the samples in a cooler until they are picked up. If you must store the
samples over a weekend, keep them indoors at room temperature or in a cooler.
Mail completed postcards at a nearby post office.
A6-5
-------
-------
APPENDIX 7
Sample Identification for QA/QC Procedures
The following table illustrates the information that will
accompany coordinates of each segment sampled in the Pilot. Each
sample sent to an analysis laboratory will be assigned an arbitrary
number between 1-447. This number will not reveal anything'about
the location where the sample was collected or whether the sample
is a duplicate or known blank for quality assurance determination.
This procedure is necessary to acquire unbiased results from
analysis. A database will include at least the parameters listed
in the table plus actual sampling coordinates to permit proper
identification of samples after analyses are completed.
Design"
-------
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
12
12
12
12
12
13
13
13
14
14
14
14
15
15
15
16
16
16
16
16
17
17
17
18
18
18
18
19
19
19
20
20
20
20
20
21
21
21
9999
22
22
22
22
23
23
23
24
24
24
24
24
25
25
25
26
26
26
26
27
27
27
28
34
35
36
36
36
37
38
39
40
41
42
42
43
44
45
46
47
48
48
48
49
50
51
52
53
54
54
55
56
57
58
59
60
60
60
61
62
63
9999
64
65
66
66
67
68
69
70
71
72
72
72
73
74
75
76
77
78
78
79
80
81
82
42
43
44
45
46
47 •
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
9999
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
9999
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
A7 - 2
-------
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
28
28
28
28
29
29
29
30
30
30
30
31
31
31
32
32
9999
32
32
32
33
33
33
34
34
34
34
35
35
35
36
36
36
36
36
37
37
37
38
38
38
38
39
39
39
40
40
40
40
40
41
41
41
42
42
42
9999
42
43
43
43
44
83
84
84
84
85
86
87
88
89
90
90
91
92
93
94
95
9999
96
96
96
97
98
99
100
101
102
102
103
104
105
106
107
108
108
108
109
110
111
112
113
114
114
115
116
117
118
119
120
120
120
121
122
123
124
125
126
9999
126
127
128
129
130
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
9999
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
9999
2
1
1
1
1
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
9999
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
9999
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
A7 - 3
-------
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
44
44
44
44
45
45
45
46
46
46
46
47
47
47
48
48
48
48
48
49
49
49
50
50
50
50
51
51
51
52
52
52
52
9999
52
53
53
53
54
54
54
54
55
55
55
56
56
56
56
56
57
57
57
58
58
58
58
59
59
59
60
60
131
132
132
132
133
134
135
136
137
138
138
139
140
141
142
143
144
144
144
145
146
147
148
149
150
150
151
152
153
154
155
156
156
9999
156
157
158
159
160
161
162
162
163
164
165
166
167
168
168
168
169
170
171
172
173
174
174
175
176
177
178
179
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1 ,
1
1
1
2
2
1
1
1
1
1
2
2
9999
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
9999
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
A7 - 4
-------
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
RP
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
60
60
60
61
61
61
62
62
62
62
63
9999
63
63
64
64
64
64
64
65
65
65
1
1
1
1
2
2
2
3
3
3
3
3
4
4
4
5
5
5
5
6
6
6
7
7
7
7
7
8
8
9999
8
9
9
9
9
10
10
10
11
180
180
180
181
182
183
184
185
186
186
187
9999
188
189
190
191
192
192
192
193
194
195
1
2
3
3
4
5
6
7
8
9
9
9
10
11
12
13
14
15
15
16
17
18
19
20
21
21
21
22
23
9999
24
25
26
27
27
28
29
30
31
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
2
2
2
1
1
1
1
1
2
2
1
9999
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
9999
1
1
1
2
2
1
1
1
1
0
1
2
0
0
0
0
0
0
0
0
9999
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
9999
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
A7 - 5
-------
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
11
11
11
11
12
12
12
13
13
13
13
14
14
14
15
15
15
15
15
16
16
16
17
17
17
17
18
18
18
19
9999
19
19
19
19
20
20
20
21
21
21
21
22
22
22
23
23
23
23
23
24
24
24
25
25 .
25
25
26
26
26
27
27
32
33
33
33
34
35
36
37
38
39
39
40
41
42
43
44
45
45
45
46
47
48
49
50
51
51
52
53
54
55
9999
56
57
57
57
58
59
60
61
62
63
63
64
65
66
67
68
69
69
69
70
71
72
73
74
75
75
76
77
78
79
80
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
9999
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2 ,
2
1
1
1
1
1
2
2
1
1
1
1
1
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
9999
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
A7 - 6
-------
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX,
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
:HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
27
27
27
28
28
28
29
29
9999
29
29
30
30
30
31
31
31
31
31
32
32
32
33
33
33
33
34
34
34
35
35
35
35
35
36
36
36
37
37
37
37
38
38
38
39
39
39
39
9999
39
40
40
40
41
41
41
41
42
42
42
43
43
81
81
81
82
83
84
85
86
9999
87
87
88
89
90
91
92
93
93
93
94
95
96
97
98
99
99
100
101
102
103
104
105
105
105
106
107
108
109
110
111
111
112
113
114
115
116
117
117
9999
117
118
119
120
121
122
123
123
124
125
126
127
128
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
2
2
2
1
1
1
1
1
9999
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
9999
2
1
1
1
1
1
2
2
1
1
1
1
1
0
1
2
0
0
0
0
0
9999
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
9999
2
0
0
0
0
0
0
0
0
0
0
0
. 0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
A7 - 7
-------
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
HEX
GRAND TOTAL
43
43
43
44
44
44
45
45
45
45
46
46
46
47
47
47
47
47
48
48
48
49
49
49
49
50
9999
50
50
51
51
51
51
51
129
129
129
130
131
132
133
134
135
135
136
137
138
139
140
141
141
141
142
143
144
145
146
147
147
148
9999
149
150
151
152
153
153
153
348
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
447
2
2
2
1
1
1
1
1
2
2
1
1
1
1
1
2
2
2
1
1
1
1
1
2
2
1
9999
1
1
1
1
2
2
2
59
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
9999
0
0
0
0
0
1
2
29
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0 .
:0
0
0
0
0
0
0
0
0
0
. 1
0
0
0
0
0
0
0
11
• Rotation Panel Design (RPJ or centroids of EMAP hexagons (HEX)
b Segment identification code
c Field number within a segment
d Sample code number for analysis laboratories
* Second composite sample collected within a field
-------
APPENDIX 8
Ecology and Epidemiology
Response of Two White Clover Clones to Peanut Stunt Vims and Ozone
Allen S. Heagle, Michael R. McLaughlin, Joseph E. Miller, and Ronald L. Joyner
Plant pathologist, USDA-ARS, Department of Plant Pathology, North Carolina State University, RaJeigh 27695; plant pathologist,
USDA-ARS, Crop Science Research Laboratory, Mississippi State, MS 39762; plant physiologist, USDA-ARS, Department of
Crop Science; and research assistant. Department of Plant Pathology, North Carolina State University, Raleigh 27695.
Cooperative investigations of the USDA-ARS and the North Carolina State University. Funded in part by the North Carolina
Agricultural Research Service.
The use of trade names in this publication does not imply endorsement of the products named, nor criticism of similar ones not
mentioned, by the North Carolina Agricultural Research Service or the USDA.
We thank Robert Philbeck for construction and maintenance of dispensing and monitoring systems; Fred Mowry for data acquisition
software; John Rawlings, Susan Spruill, and Thangam Arumugham for statistical advice and analyses; James Moyer and Lee
Campbell for technical advice; and Walter Pursley, Sam Iscnhower, Cathryn Grace, and Scott Karriker for technical assistance.
Accepted for publication 10 September 1991 (submitted for electronic processing).
ABSTRACT
Heagle, A. S., McLaughlin, M. R.. Miller, J. E.. and Joyner, R. L. 1992. Response of two white clover clones to peanut stoM virus and ozone
Phytopathology 82:254-258.
Effects of ozone (Oj) and peanut stunt virus (PSV) on two clones
of white clover (Trifolium reptru) were measured in open-top field
chambers. An O3-resistanl clone (NC-R) and an Oj-iensitive clone (NC-
S), with and without PSV infection, were exposed to O3 for 12-h day"1
for 111 days. The exposures were proportional to ambient Oj and resulted
in 12-h day"' mean concentrations of 26, 45, 64, and 76 nL L"' for
the 111 days of exposure. Plant shoots were harvested five times to measure
effects of O3 and PSV on foliar injury, foliar chlorophyll, and shoot
dry weight. Infection by PSV caused foliar chlorosis, which tended to
be more severe on NC-S than on NC-R. PSV infection suppressed shoot
dry weight accumulation of NC-R by 23% and of NC-S by 18%. Oj
also caused foliar chlorosis and suppressed shoot dry weight accumulation,
and jhe severity of the effects increased with increased O3 dose. Seasonal
jhoot weight of NC-S plants exposed in nonfillend air chambers to
ambient concentratioas of O, (45 nL L"') was 20% less than for NC-
S plants in charcoal-filtered air chambers (26 nL L"'). Shoot weight of
NC-R was not significantly affected by any of the Oj treatments. The
clone X Oj interaction was significant for all measwes for each harvest
except for the first harvest. Although the Oj coactntrations remained
relatively constant, toe differences between NC-S aad NC-R shoot weight
became greater as UK season progressed. There were no significant
interactions between O, and PSV for any of the response measures.
White clover (Trifolium repens L.) and tall fescue (Festuca
arundinacea Schreb.) are commonly grown together in the south-
eastern United States to provide high quality forage for livestock.
However, the clover usually persists for only a few years. Micro-
organisms, insects, poor management practices, plant competi-
tion, poor drought tolerance, and tropospheric ozone (O3) have
been suggested as causes for white clover decline (3,5).
Tropospheric O3 causes foliar injury and suppresses yield of
many crops (11,16), and white clover is among the most sensitive
(2,3,15,21). In a 2-yr field study with 'Tillman' white clover grown
with fescue at Raleigh, NC, ambient 03 suppressed shoot weight
production of the clover by 22% each year (3). In a subsequent
field study with 'Regal' white clover grown with fescue at Raleigh,
ambient O3 suppressed shoot weight production of the clover
by 8% the first year and by 44% in the second (15,24). Clover
plants that survived that study were propagated clonally to
determine whether selection for resistance to Oj had occurred.
More individuals from the population of plants that survived
exposure to high Oj levels were resistant to foliar injury induced
by short-term O3 exposure than were individuals from the popula-
tion that survived exposure to low Oj levels (13). Whether the
observed selection for resistance of foliage to injury from.short-
tcrm (acute) Oj exposures is related to resistance to growth effects
caused by long-term (chronic) O3 exposure is not known.
Peanut stunt virus (PSV) is one of the most prevalent viruses
of white clover in the southeastern United States (1,20). One field
study with two clover clones, propagated vegetatively from PSV-
infected plants, showed that PSV caused from 49 to 91% loss
in shoot weight accumulation, depending on the year and clone
(23). In another field study, in which seedlings were inoculated
This article is in the public domain and not copyricf table. It may be freely
reprinted with customary crediting of the source The American Phyto-
pathological Society. 1992.
with PSV before placement in the field, PSV caused a 28% suppres-
sion in shoot weight production, and the levd of reduction was
greater with increased duration of infection (8). Results from
growth chamber studies with PSV (10) were similar to those
reported for seedlings (8).
Virus infection often causes some protection from O3 injury,
and the type and degree of protection depends on the specific
host and vinis (4,6,7,22,26,27). However, there are exceptions
to this generality. O3 caused more injury OB tobacco infected
with tobacco streak virus than it did on uninfected tobacco (25).
Three hurley tobacco cuhivars infected with tobacco etch virus
. tended to show less O3-induced growth suppression than unin-
fected plants, but tobacco vein mottling vires tended to cause
the opposite effect (26). For both viruses, the response to O3
was dependent on the cubivar (26). The mechanisms for virus-
induced changes in plant response to pollutants are unknown,
although virus liter, plant age, and season of the year are important
factors. There, have been no studies to determine whether clover
viruses affect clover response to O3 or vice versa.
This study was done to determine: the differences in growth
response to long-term O3 exposure for two clover clones known
to be sensitive or resistant to injury from short-term O3 exposure;
the relative importance of tropospheric O3 and PSV in causing
growth decline of white clover, and whether PSV infection affects
clover response to 03 or vice versa.
MATERIALS AND METHODS
White clover plants that survived a 2-yr field study to determine
effects of chronic O3 exposure (15,24) were propagated and
screened for relative sensitivity to O3. One done survived 2 yr
of exposure to high O3 levels and subsequently was shown to
be resistant to O3, whereas the other had been exposed to low
levels of O3 and was very sensitive (13). The resistant clone (NC-
254 PHYTOPATHOLOGY
AS - 1
-------
R) »nd the sensitive clone (NC-S) were subsequently freed of
viruses by ihooHip meristem culture (13).
On 6 February 1989, cuttings of each clone were placed in
pots containing 0.22 L of 12:1:1 mixture of sandy loam topsoil,'
sandf'Mctro Mix 220 (W. R. Grace Co. Cambridge, MA) in a
greenhouse. Half of the plants (84 of each clone) were mechanically
inoculated with PSV on each of 3 days (13, 14, and 15 March
1989). On each date, upper leaf surfaces were rubbed with
expressed jap from PSV-infected white clover leaves in 0.03 M
sodium phosphate buffer at pH 7.4, containing 0.02 M 2-
mercaptoelhanol and 600 mesh Carborundum. On 23 March,
all plants were individually transplanted to pots containing 14
L of the 2:1:1 medium and were next inoculated with Rhiiobium.
Plants were cut to a height of 5 cm on 24 April.
The experimental design was a randomized complete block with
four blocks of four Oj treatments in open-top chambers (12) with
subplots of two clones (NC-S and NC-R) and two virus treatments
(plants with and without PSV inoculation). Each of the 16 cham-
bers contained 16 pots (four pots each of the NC-S and NC-
R clones, with and without inoculation with PSV). Plants were
transferred to open-top field chambers on 3 May and were watered
as needed to prevent moisture stress throughout the season. To
decrease the chances of spreading PSV to noninoculated plants,
all inoculated plants were randomly assigned to one side of each
chamber (east or west). This arrangement allowed a minimum
of 30 cm between inoculated and nomnoculated plants. The two
clones were arranged in two randomized 2X2 latin squares on
each side of the chamber.
An enzyme-linked immunosorbent assay (EL1SA) (19) (tone
on 22 May for aB PSV-inoculated plant* showed that 60 of the
64 NC-S plants were infected, but that only 33 of the 64 NC-
10 30 40 » 10 7» 10 10 ioo no
D*r>
Fij. 1. Daily 12-h per day (0*00-2000 h EST) ozone (Oj) concenirationj
in ambient air S km »outh of Rikifh, NC, durinj studiei to twarare
effect! of Oj and peanut stuat virui on two Udino clover clones. The
fijurt ihowj ambient O, concentrations for tie 111 days from 4 May
to 23 Aupat 19S9.
R plants were infected. However, there were • least two NC-
R plants that tested positive for PSV in all but three chambers
and four NC-S plants with PSV in all but four chamber!. Inocu-
lated plants that tested negative for PSV were not included in
any data analyses or interpretation of results bwt were retained
to maintain plot uniformity.
Oj dispensing and monitoring techniques have been described
previously (14). The O3 treatments, which began on 4 May, were
charcoal-filtered air, noniiltered air, and two nonfiltered air treat-
ments to which O3 was added for 12-h day"1 (0800-2000 h EST)
in amounts proportional to ambient Oj concentrations. The
seasonal (4 May to 23 August) 12-h day'1 meaa O3 concentra-
tions in ambient air and in the charcoal-filtered air, nonfiltered
air, and two Oradded treatments were 51, 26, 45, 64, and 76
nL L"1, respectively. The chamber fans were turned off from
2100 to 0500 h EST daily.
Plants were cut to a height of approximately 7 cm above the
soil level on five dates during the experiment: 11 May; 5-6 June;
28-29 June; 25-26 July, and 23-24 August. Stolons growing
outside of the perimeter of each pot were also cut. At each harvest,
the shoots (leaves, petioles and / or stolons and flowers) were placed
in paper bags, dried for 2 days at 55 C, and wcigted.
Estimates of foliar injury and foliar chlorophyll analyses were
performed one day before the second, fourth, aid fifth harvests
using five adjacent leaves on one stolon, starting with the youngest
fully expanded leaf. Visible foliar injury was estimated for each
leaf as the percentage of chlorosis and necrosis i» 5% increments
(0-100%). The same leaves were used for chlorophyll analyses
as described by Knudsen et al (18). Leaves were placed in approxi-
mately 70 ml of ethyl alcohol (one brown glaa container per
five-leaf sample) and placed in the dark. After 3 days, the volume
of alcohol for each container was increased to 100 ml, and the
amounts of chlorophyll a and b were measured sixxlrophotometricalry.
Starting on 19 May, all plants were sprayed at 2- to 3-wk
intervals with Capture (bifenthrin), 3.2 EC, 3.1 mL L"1 to prevent
infestation of aphids and decrease the potential for spread of
viruses. ELISA te$u were done on 7 August to determine whether
plants not inoculated in'March had become infected with PSV,
alfalfa mosaic virus, clover yellow vein virus, red clover vein
mosaic virus, or white clover mosaic vinii. Tfce results were
negative for all but five plants: two NC-R and two NC-S plants
(three separate plots) were positive for PSV, and one NC-R plant
was positive for clover yellow vein virus. Therefore, data from
these plants were discarded.
Data from inoculated plants that tested negative for PSV on
22 May were not used in statistical analyses, so the latin square
design was incomplete. Therefore, the design w«s reduced to a
split-split plot with unequal samples in the subplots. Analyses
of variance were done for shoot weight, chlorophjfl content, and
foliar injury for each harvest separately, and for total seasonal
shoot weight using S AS software (SAS Institute, Gary, NQ. Mild
heterogeneity of variance was found for the last two harvests,
but data transformations were not considered to be advantageous.
TABLE I, Mean squares from analyses of variance for effecu of ozone on shoot (leaves, petioles, and/or stolons
per pot), total chlorophyll (pj/ml), and foliu injury (mem penaiMft per k«f) for two while clover clones, with
slum virui*
Source
Block (B)
Ozone (0)
Virui (V)
VXO
Clone (Q
CXO
cxv
cxoxv
<1f
3
1
1
3
1
3
I
3
Harvest!
Shoot
dry wt
47'
g
206"
3
I84~
4
16"
2
Shoot
dry wt
787"
599"
2,598"
40
4,081"
1,058"
388"
32
Harvest 2
Total
chlorophyll
215
188
197-
61
486"
378"
75"
10
Folimr
injury'
7
nr
59"
1
194"
84"
0
2
Hanoi 3
Shoot
drywt
290*
2475"
4.»»~
74
6,»17~
3j
-------
Regression analyses were done using SAS or Cricket Software
(Cricket Software, Malvern, PA) with shoot weight and chloro-
phyll content at the dependent variables and mean 12-h day'1
O3 concentrations (for individual growth periods and for the total
season) as the independent variable.
RESULTS
The daily 03 fluctuations (Fig. 1) and seasonal mean O3 concen-
trations in ambient air during this experiment were similar to
previous seasons at the site (IS). The weather during the study
was somewhat cooler and wetter than normal with daily mean
maximum temperatures of 20, 28, 30, 31, and 30 C and rainfall
of 6, 4, 17, 12, and 19 cm for growth periods 1-5, respectively.
Vina effects. Infection by PSV caused approximately 5-10%
foliar injury (chlorosis) on both the O3-senshive clone (NC-S)
and O3-resistam clone (NC-R) at each harvest (Tables 1,2). The
response of chlorophylls a and b to PSV and the O3 treatments
were similar, so chlorophyll responses will be presented in terms
of total chlorophyll. Total chlorophyll content of PSV-infected
NC-S plants (mean across all treatments) was 21 and 23% less
than for uninfected NC-S plants for harvests 2 and 4, respectively
(Table 2). The comparable numbers for NC-R were 5 and 8%,
respectively. The clone X PSV interaction for chlorophyll was
significant at harvest 2 (Table 1), but the PSV effect was similar
for both clones at harvest 5.
Fewer NC-R than NC-S plants became infected by PSV from
mechanical inoculation, and PSV generally caused smaller
decreases of NC-R chlorophyll than of NC-S chlorophyll. The
clone X PSV interaction was significant at harvest 2 (Table 1).
However, NC-R was more sensitive to growth effects of PSV
than was NC-S. For all chamber treatments combined, infection
by PSV suppressed seasonal ihoot weight production of NC-
R by 23% and of NC-S by 18% (Table 3; Fig. 2), and the clone
X PSV interaction for shoot weight was significant at each harvest
(Table 1). There were no significant PSV X Ojor three-way inter-
actions for shoot weight. The differences in shoot weight response
of the two clones to PSV at the different 03 levels at the individual
harvests (Table 3) were similar to those shown for seasonal shoot
weight production (Table 3; Fig. 2).
O3 effects. O) exposure caused foliar injury (chlorosis and
necrosis) and decreased foliar chlorophyll content (Tables 1,2).
The effects were significant except for chlorophyll at harvest 2.
The effects of O3 were much greater on the O3-sensitive NC-
S than on the O3-resistant NC-R (Table 2) and caused the
significant clone X O3 interaction for all harvests for both measures
(Table 1).
Except for harvest I (after 7 days of 03 treatment), the effect
of O3 on shoot weight was significant at all harvests (Table 1).
O} suppressed seasonal shoot production of NC-S more 'than
that of NC-R (Table 3; Fig. 2), and tie O3 X clone interaction
was significant at all harvests. The difference between iboot weight
of plants grown in charcoal-filtered air and shoot weight of plants
grown at higher O, concentrations increased as the season pro-
gressed. For example, shoot weight of NC-S in the nonfiltered
air treatment (45 nL L"1 of O3) was St, 88, 79, and 56% of that
in the charcoal-filtered air treatment (26 at L"' of Oj) for harvests
2, 3, 4, and 5, respectively. Likewise, shoot weight of NC-R in
the highest O) treatment (76 nL L"1) was 104, 102, 95, and 77%
of that at 26 nL L"' for harvests 2, 3, 4, and 5, respectively.
The standardized slopes (standardized to a maximum of I by
dividing the slope of each regression node! by its intercept) of
the shoot weight response increased for NC-S across all harvests
and increased for NC-R between harvests 4 and 5 (Table 3). The
same trends for increased response with successive exposure
occurred for PSV-infected plants of both clones, and there were
ao PSV X O3 interactions.
DISCUSSION
The present study showed that NC-S, which was more sensitive
to foliar injury from acute O3 exposan than NC-R, was also
more sensitive than NC-R to growth effects caused by chronic
Oj exposure. These results agree with previous studies showing
the relationship between Oj doses and forage production of white
dover (3,15,21). The present study also showed that the amount
of yield loss caused by PSV and ambient O3 was similar for
NC-S and corroborated a previous report of differences in sensi-
tivity to PSV between clones of white dowr (23). Because cultivars
of white clover are extremely heterozygous, cultivars probably
contain genotypes with a wide range in sensitivity to both stresses,
so the relative importance of 63 and PSV for a given cultivar
will presumably depend on the degree of sensitivity to both stresses
among the genotypes.
The results suggest that chronic exposure to O3 caused plants
to become more sensitive to effects of subsequent exposure. How-
ever, the differences in response could kave been caused by other
factors, including the influence of weather patterns or -physio-
logical effects related to onset of flowering. We can only surmise
as to the relative importance of these factors, became none was
specifically studied. There were no obrioos relationships between
temperature or rainfall and the change a response to Oj; tempera-
tures were relatively uniform, and plants were irrigated to prevent
moisture stress. Flowering of the two dones began at different
times (near the beginning of growth period 1 for NC-R and near
TABLE 2. Effects of chronic exposure (o different levels of ozone on foliar injury and chlorophyll content of u ozotc-resisum (NC-R) and an
ozone sensitive (NC-S) white clover clone with and without infection by peanut stunt virus (V)
Number of
Growth exposure
period' days
2 15
4 27
5 29
12-h mean
Ozone ozone conctnlratio
tnumeuf (nL L"1)
CF
NF
NF-1
NF-2
CF
NF
NF-1
NF-2
CF
NF
NF-1
NF-2
32
56
7»
97
25
43
63
72
23
43
63
73
Percentage of foliar injury"
NC-R
1
2
1
5
0
2
24
24
2
7
29
30
NC-S
4
18
32
41
1
31
50
65
5
41
56
66
NC-RV
10
14
16
14
17
10
33
33
19
27
35
46
NC-SV
12
24
44
60
5
38
64
74
22
48
65
64
Toul chlorophyll Gig/ ml)*
NC-R
29.1
29.1
32.7
33.0
19.5
20.8
18.0
18.0
20.6
18.5
13.8
16.2
ac-s
».7
31.7
24.1
D.I
22.8
K.7
11.9
7.9
23.3
15.0
S.1
4.9
NC-RV
30.4
29.6
29.1
28.4
18.4
22.7
16.2
13.0
14.8
13.9
13.5
10.2
NC-SV
25.9
26.9
20.9
14.2
17.1
15.4
7.6
5.5
17.9
12.1
6.7
5.9
'Each value is the mean injury per leaf or mean chlorophyll per five leaves for 20 leaves (five leaves on one plant in four blacks).
'Growth period 2 = from 11 May to 4 June; growth period 4 = from 28 June to 24 July; growth period 5 = from 25 July to 22 August.
' Plants were exposed for 12 h per day in open-top Held chambers to charcoal-filtered air (CF), nonfiliered air (NF). or to NF with different proportions
of ambient ozone added.
256 PHYTOPATHOLOGY
AS - 3
-------
TABLE 3 Dry weight of shoots (leaves, petioles, and or iiolora and flowers) of *D ozone-resistant (NC-R: i»d »n ozone-iensitive (NC-S) clone
of white clover with and without infection by peanut stuat virus (V) after exposure 10 differtnt level, of ozone*
Growth
penod
.
2
4
5
1-5
Days of Ozone
growth treatment*
7 CF
NF
NF-I
NF-2
25 CF
NF
NF-1
NF-2
23 CF
NF .
NF-I
NF-2
27 CF
NF
NF-1
1
NF-2
29 CF
NF
NF-I
NF-2
111 CF
NF
NF-1
NF-2
12-h mean
ozone concentration
(nL L",)
31
57
65
32
56
71
97
22
3S
55
65
25
43
63
72
23
43
63
73
26
45
64
76
Ozone dose
(nL L" h 100)
26
40
48
55
96
168
234
291
Standardized dope*
61
105
152
179
Standardized dope
81
139
204
233
Standardized dope
80
150
219
254
Standardized dope
344
602
857
1.012
Standardized dope
Shoot dry weight per plant (g)'
NC-R
10.0
10.1
10.5
10.1
354
368
37.1
36.9
+0.65
59.9
59.1
57.7
61.2
+0.15
50.4
50.5
48.7
48.1
-1.02
38.7
38.3
34.8
29.9
-3.78
194
195
189
186
-0.86
« the mean of 16 planu (four pots, four blocks [chambers!) except for two NC-S and three NC-R planu that
were discarded because of van. infotion; each value for NC-RV is the mean of six to nine planu (one to tircr pott, four block.); each value
for NC-SV is the mean of 13-16 planu (two to four pou, four block.).
'Denned u the slope of the linear response model adjusted by dividing the dope by the intercept. Models we
with the independent variable u Oj m n\. L"'.
using Cricket Software
NC-R • 100 - 8.173*
JJ.7) (00f»)
NC-RV . HI- 8.117X
(11.1) (COM)
NC-S . 200 • 1.72SX
0.3) (0.051)
MC-SV . in - i.4i«x
(44)
tO SO 40 M «0 70 (0 tO 100 lit
Oior* conctnlnfcn (II kour/day mtan - nUt)
Fit- *• Eftecu of chronic exposure to ozone (O3) on seasonal shoot dry
ueijhl production by two clover clones (NC-R and NC-S) wilt and
without infection by the peanit stunt vires (V). The regression models
show the relationships between seasonal production of shoot dry weight
per p!>n( (grams) and seasonal 12-h per day O} concentration (X) in
nL L"' Standard errors are sfcown in parenthesis. The models for NC-
S and NC-SV were statistically significant (slope different from 0 as shown
by an f lc.1), but the model, for NC-R and NC-RV were not significant.
the end of growth period 3 fot NC-S). Both clones produced
flowers until the end of the experiment. The greatest changes
in response to Oj occurred betwtxa growth periods 3 and 4 for
both clones. Thus, the onset of Hovering occurred near the time
of a Urge change in Oj response for NC-S but not for NC-R.
The most plausible explanation for the change in response to
a given O, dose is that the observed decrease in foliar chlorophyll
concentration (Table 2) was accompanied by a decrease in photo-
synthesis and, therefore, decreased energy reserves in stolons and
roots. O3 has been shown to decrease white clover root-shoot
ratios (20) and to decrease leveb rf starches in white clover roots
(23). A gradual decrease in energy reserves prtbably would be
accompanied by decreased capacity for detoxification or repair.
Because the effects of a grven Oj dose increased with successive
growth periods, a cumulative Oj dose metric would probably
be more appropriate as the independent variable in regression
analyses than a growth period mean. A cumulative Oj dose,
differentially weighted for successive growth periods, might be
suitable. Further research b required to clarify the role of weather
conditions and the level of cumulative effects for each clone.
The effect of PSV on shoot growth was variable over the season.
For NC-R in charcoal-filtered air, PSV decreased shoot weight
by 27, 26, 20, 12, and 38<5c- respecu->ely, for the five consecutive
harvests. The comparable values for NC-S were 19, 12, 18, 17,
and 32%. No gradual trend for increased effects of PSV occurred
with increased duration of infection. The large increase in PSV-
A8 - 4
Vol. 82. No. 3. 1992 257
-------
induced loss for both dones at harvest 5 may have been due
to increased duration of infection combined wki our practice
of harvesting stolons that grew outside of the pots. Infection with
PSV is known to reduce the root system in white clover (10),
and harvesting stolons that grew outside of the pots decreased
the establishment of secondary root systems.
The response of NC-S and NC-R, expressed as a ratio, could
be a useful indicator of ambient O3 levels and of the O3 effects
on other crop species. The usefulness for indicating ambient Oj
will depend on how much the two clones vary in response to
other factors that affect growth. Measurements of shoot weight
production indicated little or no effect of differences in weather
conditions on the relative growth of NC-S and NC-R in charcoal-
filtered air; the percentage of the total seasonal shoot weight
produced during each growth period was almost identical for
both clones. Further development of these clones as an O3 indi-
cator will require more data on their relative response to variation
in weather conditions, cdaphic factors, biotic diseases, as well
as other atmospheric factors such as carbon dioxide and sulfur
dioxide. Using the clover system to indicate, the effects of O3
on other crops will require knowledge of how given factors affect
clover response to O3 relative to how the same factors affect
response of other crops to O3. In other words, this will require •
information on how the clover clones and other crops respond
to Oj over a wide range of conditions.
Although PSV-resistaat germ plasms that are adapted to the
southeastern United States (9,17) are available, toeir reaction to
O3 is not known. Thus, characterization of the overall level of
resistance to O3 and PSV in white clover strains might be worth-
while as pan of the development of strains with improved
persistence in the Southeast.
LITERATURE CITED
1. Baracll, O. W., and Gteon, P. B. 197S. Identification and prevalence
of white clover viruja «nd the resistance of TrtfoSum ipecies to
these viruses. Crop Set. 15:32-37.
2. Becker. 1C.. Saiuer, M., Eater, A., and Fuhrtr, J. 1989. Seniitivity
of white clover to ambieat ozone in Switzerland. New Fhytol. 112:235-
243.
3. Blum, U.. Heagle, A. S., Burni, J. C., and Umbra, R. A., 1983.
The elTecii of Oj on fesoe-dover forage regrowth, yield, and quality.
Environ. Exp. Bot. 23:121-13.
4, Bisseiar, S., and Temple. P. V. 1977. Reduced ozoae injury on vinu
infected tobacco io tae field. Plant Dis. Rep. 61:31-34.
5. Blake, C. T., Cbambfcr, D. S.. and Woodhome, W. W., Jr. 1966.
Influence of some environmental and managemeal factors on the
persistence of Udioo clover in association with orchardgrass. Agron.
J. 58.487-489.
6. Breanan, E., and Leoar. I. A. 1969. Suppression
-------
------- |