United States National Health and Environmental EPA/600/R-96/061
Environmental Effects Research Laboratory Sept^ber 1996
Protection Agency Corvallis, OR 97333
Conceptual Framework for a
Synoptic Assessment of the
Prairie Pothole Region
-------
United States National Health and Environmental EPA/600/R-96/061
Environmental Effects Research Laboratory September 1996
Protection Agency Corvallis, OR 97333
%
Conceptual Framework for a
Synoptic Assessment of the
Prairie Pothole Region
-------
EPA/600/R-96/061
SerAmber 1996
JeoAmt
Conceptual Framework for a Synoptic *
Assessment of the
Prairie Pothole Region
by
Lynne S. McAllister, Barbara E. Peniston,
Jeffrey Hyman, and Brooke Abbruzzese
Dynamac Corporation
Corvallis, OR 97333
Contract 68-C6-0005
Project Officer
Scott G. Leibowitz
U.S. Environmental Protection Agency
National Health and Environmental Effects Research Laboratory
Western Ecology Division
Corvallis, OR 97333
U.S. Environmental Protection Agency
Office of Research and Development
National Health and Environmental Effects Research Laboratory
Western Ecology Division
Corvallis, OR 97333
-------
September 1996
NHEERL-WED, Corvallis, OR
Abstract
A conceptual framework is presented for conducting a synoptic assessment of the Prairie
Pothole Region (PPR), including portions of North and South Dakota, Minnesota, Iowa,
and Montana. A synoptic assessment is an inexpensive, rapid assessment technique that
provides a flexible, ecologically based framework for producing relative rankings of
environmental risks for landscape subunits for the purpose of region-wide prioritization of
restoration effort. It was developed specifically for cases in which time, resources, and
information are limited; therefore, it makes use of readily available data. It is intended as
a management decision tool to be used in conjunction with other approaches to most
effectively preserve valued wetland functions on a landscape scale. This conceptual
framework presents the ecological background and conceptual models that will serve as
the basis for conducting the Prairie Pothole assessment and outlines the processes
proposed for carrying out the assessment. Three wetland functions valued by humans
are selected for the Prairie Pothole assessment: wildlife habitat quality, water quality
improvement, and floodwater attenuation. Seven criteria are proposed for evaluating the
functions, and conceptual models are developed for each criterion. Models show the
factors that influence the criterion, ecological processes and human stressors that affect
those factors, linkages among model components, and the indicators and data sources
that will be used to evaluate or represent model components. The conceptual framework
proposes approaches for deriving indicator values, combining indicators into indices, and
producing ordered groupings for indices and criteria to represent ranks for prioritizing
restoration.
Key words:
prairie pothole, synoptic assessment, index, indicator, wetland function, wetland values,
restoration, prioritization, conceptual model, wildlife habitat, water quality, floodwater
attenuation, landscape
Preferred citation:
McAllister, L.S., B.E. Peniston, J. Hyman, and B. Abbruzzese. 1996. Conceptual
Framework for a Synoptic Assessment of the Prairie Pothole Region. EPA/600/R-96/061.
Corvallis, OR: U.S. Environmental Protection Agency, National Health and Environmental
Effects Research Laboratory, Western Ecology Division.
Notice:
The U.S. Environmental Protection Agency through its Office of Research and
Development funded and managed the researchdescribed here under Contract 68-C4-
0019 to ManTech Environmental ReseaghaS^ifes Corp. and Contract 68-C6-0005 to
Dynamac Corporation. It has been^m^f^lb tne Agency's peer and administrative
review and approved for publication*? an EPA document. Mention of trade names or
commercial products does not constitute endorsement for use.
ii
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Contents
Introduction 1
Part 1: The Synoptic Approach 4
Part 2: The Prairie Pothole Region Synoptic Assessment 6
2.1 Assessment Goals and Objective 6
2.2 Intended Use 7
2.3 Accuracy Needs 7
2.4 Assessment Constraints and Uncertainty 8
2.5 Assessment Area Description 9
2.5.1 Wetland Types 9
2.5.2 Natural Setting 11
2.6 Wetland Functions 13
2.6.1 Habitat Support for Bird Abundance and Diversity 13
2.6.2 Water Quality Improvement 14
2.6.3 Flood Attenuation 15
2.6.4 Future Loss of Wetland Function 15
2.7 Wetland Values to Humans 15
2.7.1 Habitat Support for Bird Diversity and Abundance 16
2.7.2 Water Quality Improvement 16
2.7.3 Flood Attenuation 16
2.8 Wetland Impacts 17
2.8.1 Artificial Drainage 17
2.8.2 Tilling and Planting ; 17
2.8.3 Sedimentation 17
2.8.4 Nutrient Inputs 18
2.9 Wetland Replacement Potential 18
2.10 Landscape Subunits 18
2.11 Assessment Criteria 19
Part 3: Conceptual Models and Selection of Indicators 20
3.1 Risk of Future Loss of Wetland Function 21
3.1.1 Risk from Sedimentation 27
3.1.2 Risk from Draining and Cropping Wetlands 27
3.1.2.1 Likelihood of Intensifying Cropping 28
3.1.2.2 Relative Ease of Draining or Cropping Wetlands 28
3.1.2.3 Opportunity Cost 28
3.2 Value to Humans of the Habitat Support Function for Bird Diversity and
Abundance 28
3.2.1 Total Bird Abundance and Diversity 31
3.2.1.1 Upland Habitat Quality 31
3.2.1.2 Wetland Habitat Quantity 31
3.2.1.3 Habitat Quality of the Landscape 32
3.2.1.4 Pothole Habitat Quality 32
3.2.2 Population of Users and Attractiveness of the Subunit for Use ... 33
3.3 Replacement Potential for the Habitat Support Function for Bird Abundance
iii
-------
September 1996
NHEERL-WED, Corvallis, OR
and Diversity 34
3.3.1 Wetland Quantity Available for Replacement 34
3.3.2 Likelihood of Full Functional Replacement 37
3.4 Value to Humans of the Water Quality Improvement Function 37
3.4.1 Attenuation of Potential Pollution to Streams 37
3.4.1.1 Potential Input of Pollution to Streams 40
3.4.1.1.1 Chemical Application 40
3.4.1.1.2 Chemical Application Transported 40
3.4.1.2 Wetland Capacity to Improve Water Quality 41
3.4.2 Population of Users and Cost of Substitutes 41
3.5 Replacement Potential for the Water Quality Improvement Function 41
3.5.1 Wetland Quantity Available for Replacement 41
3.5.2 Likelihood of Full Functional Replacement 41
3.6 Value to Humans of the Flood Attenuation Function 44
3.6.1 Attenuation by Wetlands of Potential Downstream Flooding 44
3.6.1.1 Potential Runoff Volume to Streams 47
3.6.1.2 Wetland Capacity to Attenuate Runoff 47
3.6.2 Population of Users and Cost of Substitutes 48
3.7 Replacement Potential for the Flood Attenuation Function 48
Part 4: Deriving Subunit Priority Rankings 51
4.1 Example 51
Part 5: Summary 65
Appendix A . 71
iv
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Tables
Table 1. Steps in conducting a synoptic assessment 3
Table 2. Example of a completed data table for the replacement potential for the
habitat support function 23
Table 3. Table format for entering indicator values for the risk of future loss of wetland
function 26
Table 4. Table format for entering indicator values for the value to humans of habitat
support 30
Table 5. Table format for entering indicator values for the replacement potential for
habitat support (bird abundance and diversity) 36
Table 6. Table format for entering indicator values for the value to humans of water
quality improvement 39
Table 7. Table format for entering indicator values for the replacement potential for
water quality improvement 43
Table 8. Table format for entering indicator values for the value to humans of flood
attenuation 46
Table 9. Table format for entering indicator values for the replacement potential for
flood attenuation 50
Table 10. Example of a completed table of index values for the replacement potential for
the habitat support function 55
Table 11. Example of a completed ranking table for the components of replacement
potential for the habitat support function 57
Table 12. Example summary table of ranks for the replacement potential for the habitat
support function, including the final rank for each of 25 HUCs in South
Dakota 60
Table A-1. Data sources and indicators used for criteria evaluation 72
v
-------
September 1996
NHEERL-WED, Corvallis, OR
Figures
Figure 1. Prairie Pothole Region of the United States 10
Figure 2. Method for producing conceptual models and data tables for synoptic
assessment 22
Figure 3. Conceptual model for the risk of future loss of wetland function due to crop
agriculture and urbanization 25
Figure 4. Conceptual model for the value to humans of habitat support function of
wetlands (bird abundance and diversity) 29
Figure 5. Conceptual model for replacement potential for habitat support function of
wetlands (bird abundance and diversity) 35
Figure 6. Conceptual model for the value to humans of the water quality improvement
function of wetlands 39
Figure 7. Conceptual model for replacement potential for water quality improvement
function of wetlands 42
Figure 8. Conceptual model for value to humans of flood attenuation function of
wetlands 45
Figure 9. Conceptual model for replacement potential for flood attenuation function of
wetlands 49
Figure 10. Method for producing final ranking tables for synoptic assessment 52
Figure 11. Mapped index ranks of physical availability, a component of replacement
potential for the habitat support function, in PPR subunits (HUCs) within South
Dakota 61
Figure 12. Mapped index ranks of socio-economic availability, a component of
replacement potential for the habitat support function, in PPR subunits (HUCs)
within South Dakota ' 62
Figure 13. Mapped ranks of the crop to upland area ratio, an indicator for the likelihood of
full functional replacement, a component of the replacement potential for the
habitat support function, in PPR subunits (HUCs) within South Dakota. ... 63
Figure 14. Combined ranks for replacement potential for the habitat support function in
PPR subunits (HUCs) within South Dakota 64
VI
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Introduction
The U.S. Environmental Protection Agency's (EPA) Wetlands Research Program
(WRP) is developing a comprehensive framework that will allow wetland restoration and
protection efforts to be focused on areas in which the greatest benefit can be achieved.
The WRP's Landscape Function Project takes a holistic view of wetland protection, by
focusing on regional-scale impacts and valued functions, and is developing approaches
for classifying wetland landscapes at the regional scale. These landscape classifications
are designed to facilitate prioritization of wetland protection and restoration efforts. Three
distinct approaches, representing a range of effort and accuracy, have been designed to
meet diverse management needs and resource constraints: 1) the formalized Best
Professional Judgement (BPJ) approach (Adamus 1992, Leibowitz et al. 1992b); 2) the
empirical approach (Abbruzzese et al. 1994, Leibowitz et al. 1992b); and 3) the synoptic
approach (Abbruzzese et al. 1994, Leibowitz et al. 1992a, 1992b).
The BPJ process was designed to assess wetland functions at the broadest spatial
scale and at the lowest level of effort. Using information collected from the literature and
regional experts, the formalized BPJ process makes a priori determinations of the relative
probability of loss of valued wetland functions (Adamus 1992). The empirical approach
represents the highest level of effort and accuracy. It produces a more rigorous,
quantitative assessment based upon established ecological relationships (Abbruzzese et
al. 1994). The synoptic approach (Leibowitz et al. 1992a) is conducted at a level of effort
intermediate between BPJ and the empirical approach. It is an inexpensive, rapid
assessment technique that provides a flexible, ecologically-based framework for
producing relative rankings of environmental risks between landscape subunits (e.g.,
counties, drainage basins) for the purpose of prioritizing region-wide wetland restoration
or protection measures (Leibowitz et al. 1992a). Conducting a synoptic assessment
involves following five steps (Table 1), which requires a team composed of managers,
resource specialists, and technical analysts to: define the assessment objectives, indices,
and indicators; evaluate and utilize available data; and update the assessment as better
data become available.
The remainder of this document presents a conceptual framework for carrying out a
synoptic assessment of the Prairie Pothole.Region (PPR). Part 1 presents background on
the synoptic approach. Part 2 introduces background information for designing a synoptic
assessment for the PPR. It includes descriptions of the assessment area, PPR wetland
functions and values, landscape stressors, the landscape subunits selected for inclusion
in the assessment, and a list of the functional criteria that will be evaluated in the
assessment. Part 3 summarizes the development of conceptual models, which provide
the scientific basis for combining data, and selection of indicators and data sources. Part
4 describes how indicator values will be assigned to subunits, ranked, and/or combined
drab
1
-------
September 1996
NHEERL-WED, Corvallis, OR
into indices (mathematical combinations of indicator values or ranks). An example of the
process shows how ranks for subunits can be mapped and interpreted to serve as a
management tool for prioritizing subunits of the PPR for wetland restoration. Part 5
summarizes our status and progress on the synoptic assessment.
This document is not the actual assessment; it presents our proposed conceptual
framework for the PPR synoptic assessment. The purpose of this assessment is to further
develop the synoptic methodology and provide potential users with a realistic case study.
The synoptic assessment will be conducted at the EPA National Health and
Environmental Effects Research Laboratory in Corvallis, Oregon. Results will be
tabulated and mapped and provided to potential users in the PPR region or in regulatory
branches of state or federal government. The proposed assessment is not targeted to any
specific user, although we are trying to make the assessment realistic and useful by
including PPR managers in the process of defining the assessment objectives, indices,
and indicators. This conceptual framework presents one example of how data might be
combined, displayed, and used to rank areas of the PPR for priority in wetland restoration.
2
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
Table 1. Steps in conducting a synoptic assessment.
September 1996
Steps
1. Define goals and criteria
Procedures
1.1 Define assessment objectives
1.2 Define intended use
1.3 Assess accuracy needs
1.4 Identify assessment constraints
2. Define synoptic indices
2.1 Identify wetland types
2.2 Describe natural setting
2.3 Define landscape boundary
2.4 Define wetland functions
2.5 Define wetland values
2.6 Identify significant impacts
2.7 Select landscape subunits
2.8 Define combination rules
3. Select landscape indicators
3.1 Survey data and existing methods
3.2 Assess data adequacy
3.3 Evaluate costs of better data
3.4 Compare and select indicators
3.5 Describe indicator assumptions
3.6 Finalize subunit selection
3.7 Conduct pre-analysis review
4. Conduct assessment
4.1 Plan quality assurance/quality control
4.2 Perform map measurements
4.3 Analyze data
4.4 Produce maps
4.5 Assess accuracy
4.6 Conduct post-analysis review
5. Prepare synoptic reports
5.1 Prepare user's guide
5.2 Prepare assessment documentation
3
-------
September 1996
NHEERL-WED, Corvallis, OR
Part 1: The Synoptic Approach
The synoptic approach provides a rapid means of obtaining a region-wide picture of
wetland functions and impacts. It was developed specifically for cases in which time,
resources, and information are limited; therefore, it makes use of readily available
information, including the BPJ of local experts. Since it is qualitatively based, it allows
both scientifically based and subjective factors (e.g., societal values) to be considered at
different scales. The method is intended neither to provide a precise, quantitative
assessment of environmental risks within an area nor to assess the environmental risks of
specific impacts. Rather, it allows the assessment of functional criteria in landscape
subunits relative to one another and can be used to rank the subunits for priority in
protection or restoration. The synoptic approach is flexible and offers multiple options for
processing and displaying information so that managers can meet specific needs. It is
intended to be applied easily so it can augment BPJ and other decision-making
processes.
Using the synoptic approach, wetland managers are able to produce regional or
statewide maps that rank subunits of the landscape on the basis of indicator values and/or
values derived from aggregation of synoptic indicators (i.e., indices). Comprehensive
examination of the distribution of ranks will facilitate decision-making and allow the setting
of priorities for protection or restoration efforts among subunits in a region, complementing
assessment on a permit-by-permit basis (Leibowitz et al. 1992a).
During research and development of the synoptic approach, assessments in different
regions of the United States have been produced to address a variety of hypothetical
objectives. For example, a synoptic assessment was conducted for the Pearl River Basin
in southern Mississippi and Louisiana to provide data for hypothetical evaluation of
cumulative impacts in the 404 permit review process. A Louisiana synoptic assessment
was targeted to the identification of environmental problems within subunits and the
identification of landscape functions that would benefit from restoration. In the state of
Washington, the stated goal was to provide information on future risk of valued habitat
loss-particularly with respect to rare, threatened, and endangered species-that could
contribute to state-wide planning for habitat protection. In Illinois, the goal was to identify
areas where riparian wetland restoration would provide the greatest benefit for reducing
the -impacts of nitrogen loading on human water supply and on non-degraded fish
communities (Leibowitz et al. 1992). These earlier pilot studies contributed to several
refinements of the approach but were not applied to actual resource management
problems.
In summary, a synoptic assessment should
produce regional maps of risk and priorities for protection and restoration;
be flexible so that diverse objectives can be met;
appeal intuitively to managers and scientists;
4
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
produce conceptual models, indicators, and indices that are theoretically justified,
ecologically appropriate, and meaningful to managers;
contain explicit documentation of assumptions; and
be developed so that assessment components are represented, whenever possible, by
one or more indicators which can be rapidly estimated using data that are
a) existing (i.e., the approach attempts to circumvent the effort and cost associated
with collection of additional data in the field);
b) readily accessible;
c) uniformly available;
d) collected at a landscape level; and
e) integrative.
The synoptic approach makes use of the best available data; however, assessments
are limited by data that meet the above criteria. Often, indicators must be selected as
surrogate measures of landscape concepts and functions, and assumptions about the
relevance of selected existing data to the concepts and functions are necessary. The
assessment is done at a very large scale; indicators selected are often very general and
produce rough estimates of functional criteria per subunit. If applied uniformly to subunits
throughout a region, however, comparisons of subunits relative to one another can be
made. The objective is to obtain indicators and indices that can be assessed relative to
each other in order to set priorities. The synoptic assessment is intended only as a
general framework or tool for making decisions involving complex, landscape-level
considerations.
5
-------
September 1996
NHEERL-WED. Corvallis, OR
Part 2: The Prairie Pothole Region Synoptic Assessment
The PPR was selected in WRP's Five-year Research Plan as the focus for a case
study to test and continue development of synoptic approaches and to rank areas for
priority in restoration (Leibowitz et al., 1992b). This selection was based on
the national significance of wetlands in the PPR,
the applicability of the results to other regions with similar wetland types (i.e.,
palustrine emergent),
the need for additional research to improve overall wetland protection, and
the possibility of collaborating with other federal agencies.
The proposed synoptic study in the PPR will produce an assessment of seven criteria
associated with the value to humans of prairie potholes, replacement potential, and the
risk of future loss of wetland function. For each subunit, indicators will be combined when
possible and placed into one of five ranking categories. Ranks will then be combined to
produce maps showing relative priority for wetland restoration in subunits of the region.
The assessment has been through several stages of development. The findings of an
extensive, formal BPJ assessment of the PPR (Adamus 1992) served as the basis for our
original development of synoptic indices. The BPJ study ranked the stressors which pose
the greatest threat to valued wetland functions. Artificial drainage, tillage removal of
vegetation near wetlands, and sedimentation and tillage within wetlands were considered
to be the most significant stressors. The BPJ study identified two wetland functions that
are considered most valuable by State and Federal resource managers: waterfowl
production capability and surface water quality improvement. Relevant information from
the BPJ assessment and other sources was used to develop indices and produce a draft
assessment strategy, which was peer reviewed in 1994. Concerns from the review
indicated a need for further input and consensus from regional managers. A workshop
was organized to obtain input regarding these concerns and to help us refine the indices.
The workshop was held in Bismarck, North Dakota, in May 1995 and attended by 12
regional scientists and resource managers. Input from the workshop and re-evaluation of
the approach by in-house staff resulted in a synoptic assessment strategy that has been
modified slightly from the original proposed methodology (Leibowitz et al. 1992a). This
document incorporates the modifications and represents the revised approach. Attendees
at the workshop also agreed that flood water attenuation is an important function of PPR
wetlands and worth evaluating. We therefore included it as a third wetland function in this
version of the assessment conceptual framework and are continuing to develop and refine
the conceptual models and indicators for evaluating it.
2.1 Assessment Goals and Objectives
The goal of the synoptic assessment is to produce ecologically based information in a
timely manner to be used for setting geographic priorities for the restoration of wetland
6
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
functions in the PPR. Results of the assessment should be useful to resource managers
and should assist them in selecting areas where restoration and protection is most likely to
benefit wetland landscape functions. The exercise will serve as a prototype that further
tests, develops, and refines methods for synoptic assessment being developed by the
EPA.
The objectives of the PPR synoptic assessment are
to produce relative rankings of landscape subunit functional criteria,
to produce maps showing a prioritization ranking of wetland restoration and protection
for subunits of the PPR, and
to develop scenarios for combining data and ranks and procedures for assigning final
subunit priorities for restoration.
2.2 Intended Use
State and Federal programs, e.g., the "Swamp Buster" provisions of the 1985 and
1990 Food Securities Act (the "Farm Bill") and the U.S. Department of Agriculture (USDA)
Wetland Reserve Program, were implemented to curtail the rapid loss of wetlands and to
find other ways to help reverse past wetland degradation. State and Federal programs
are restoring previously existing wetlands to replace lost wetland functions. Restoration
efforts that are currently being conducted by State and Federal agencies are often driven
by opportunity, e.g., when a private citizen decides to enroll in a wetlands subsidy
program. Unfortunately, this process does not provide priorities for restoring the
maximum benefit to either the wetland resource or the functions and values derived
throughout the landscape. This synoptic assessment should serve as an additional tool
that State and Federal wetland managers can employ to determine which subunits will
provide the most benefit per unit effort used to restore wetlands within them. Experience
and knowledge gained during the assessment process will be useful to EPA for designing
and developing synoptic assessments.
2.3 Accuracy Needs
The level of accuracy needed in a synoptic assessment depends on the intended use
of the data and results. Accuracy is influenced by the characteristics of the environmental
data used to derive the indicators (Leibowitz et al. 1992a). Because the synoptic
assessment is intended to be a broadly scaled assessment that provides a framework for
decision-making, the level of accuracy needed is much less than that for a more rigorous,
empirical analysis.
The PPR assessment is intended to augment procedures that are currently in place for
selecting wetlands for restoration. Priority rankings of subunits should not be used as the
sole decision criterion. They are intended to be used as a framework, along with BPJ and
other decision-making processes, to identify areas of the PPR where restoration should be
focused. Because the final product is a relative ranking of the subunits, the actual values
7
-------
September 1996
NHEERL-WED, Coivallis, OR
of the indices are less important than the values relative to each other. Using data that
are uniformly available for the entire PPR will help to assure precision in the ranks.
Subunit rankings may be aggregated (e.g., divided into quintiles) in the final analysis to
identify broader regions where restoration might be most beneficial for preserving wetland
functions. Managers can then use BPJ to select individual wetland sites for restoration
within selected subunits.
It is important, to the extent possible, to control factors influencing the accuracy of the
assessment, as this can be influenced by the uniformity of data received from the
reporting units (political or jurisdictional units, such as counties, from which data are
compiled and reported). If the selected subunit (e.g., watershed) is different from the
reporting unit, data from two or more reporting units comprising the subunit are prorated,
based on the proportion of the subunit occupied by each. The availability of data might
vary among reporting units, depending upon the resources available for collecting and
managing data. Missing or uneven data in one part of an assessment model might bias
results for some subunits. Data management procedures, collection techniques, and
laboratory techniques can also vary among personnel in different subunits. Selection of
index components and the ways in which the components are combined into indices can
also result in inaccuracies and magnification of uncertainty and error. To minimize these
potential effects, we have evaluated the available data and have selected those data sets
that are the most uniformly available and that provide the most accurate and reliable
estimates of the ecosystem components of interest to us.
Information on data collection, analysis, and management procedures will be obtained
with each database when available. This information should include the source of the
data, dates of collection, periodicity of collection, data collection techniques used, map or
air photo scales (if applicable), summaries of data quality or quality control procedures
used, description of data entry and verification procedures, and statements of
assumptions or limitations for data use. Missing data and outliers will be noted for each
reporting unit and considered when summarizing results from the synoptic assessment.
Because the synoptic assessment is not designed to be a highly rigorous modeling
procedure, the indices are intentionally simplified in order to reduce the magnification of
error when combining index components and to minimize uncertainty in the estimates.
2.4 Assessment Constraints and Uncertainty
Assessment constraints relate to data availability and uniformity among reporting units
(e.g., counties, states). The PPR synoptic assessment will not require a high level of
precision in data collection, rigorous data collection in the field, or data verification
because data required for calculating indices already exist and are easily acquired.
Quality assurance procedures have been performed on the data at their place of origin.
Completing the assessment, therefore, will require compiling existing data, confirming that
quality control procedures have been followed, and managing and manipulating
8
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
databases using statistical analyses and geographic information systems (GIS).
The analysis may be influenced by variation in the degree of data coverage among
reporting units, the accuracy with which data represent conditions throughout a subunit,
the comparability of methods used to collect data (if more than one), the dates of data
collection, and the degree to which data represent present conditions (see previous
section on Accuracy Needs).
The assessment may also be influenced by the design of the conceptual models and
by the uncertainty of our assumptions for the region as a whole. We are limited to using
indicators and developing indices that are representative of a large geographic area. This
type of analysis requires making broad assumptions and is therefore inherently subject to
uncertainty. An effective means of assuring certainty is to make reasonable assumptions
that are broadly applicable and documented in the literature. A user of a synoptic
assessment accepts uncertainty, but the tradeoff is that the assessment is simple, rapid,
and inexpensive. Many of the assumptions of linkages in the conceptual models we
present in this document are supported by previous research or BPJ. Empirical analyses
based on systematically collected field data or remotely sensed data can also be designed
to validate the assumptions related to linkages between concepts and indicators.
2.5 Assessment Area Description
The boundary for this synoptic assessment is the PPR (Figure 1) of the United States.
Ideally, the landscape boundary for a synoptic assessment should include the natural
setting and all factors which affect wetland function. For water quality function, the
boundary should contain at a minimum the entire drainage basin and include all related
processes, such as the transport of nutrients and sediments. For wetland habitat function,
the boundary ideally should be determined by the ranges of all wetland dependent species
that use PPR wetlands as habitat. Although birds are mobile and their ranges often
include areas outside the PPR, the assessment is limited to the PPR, and we cannot
assess the contribution to habitat from surrounding or distant areas. The assessment will
be limited, therefore, to only those portions of the breeding ranges of birds that lie within
the PPR. For the flood attenuation function, the boundary ideally should be the PPR as
well as any area downstream of the PPR where flooding occurs (e.g., the Mississippi
River Delta). This assessment, however, will be limited to downstream flooding within the
PPR and will assume that the cumulative effects of PPR wetlands on regions downstream
is constant.
2.5.1 Wetland Types
Wetlands in the PPR can be divided into four general types, according to the water
permanency classification system of Stewart and Kantrud (1971):
Permanent: The center of the wetland basin contains surface water during all years;
these may also be classified as lakes.
9
-------
Figure 1. Prarie Pothole Region (PPR) of the United States showing the
PPR boundary (Mann 1974), HUCs (Hydrologic Unit Code) (USGS
1976) clipped by the PPR boundary, and county boundaries.
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Semipermanent: The center of the wetland basin contains surface water during most
years. Over an idealized wet-dry cycle, occurring over several years, the vegetation in
semipermanent wetlands goes through four stages: dry marsh, regenerating marsh,
degenerating marsh, and lake marsh, with associated consequences for functions and
values.
Seasonal: The center of the wetland basin contains surface water through midsummer
during most years of normal precipitation.
Temporary: The center of the wetland basin contains surface water for less than two
weeks during most years of normal precipitation.
Over one third (36%) of the wetlands are 2.0 ha or less in size and nearly half (49%)
are 4.0 ha or smaller. Temporary and seasonal wetland basins are believed to be most
numerous, whereas seasonal and semipermanent wetland basins probably constitute the
largest acreage. Potholes within the PPR commonly occur at densities exceeding 7.7 per
square km and may exceed 38.6 per square km in some undrained areas (Adamus 1992).
2.5.2 Natural Setting
The PPR of North America covers over 777,000 square km, including parts of Canada,
and 274,540 square km in the States of North Dakota, South Dakota, Minnesota, Montana
and Iowa (Figure 1). Two major physical subdivisions occur in this area: 1) the North-
Central Lake-Swamp--Moraine Plains; and 2) the Dakota-Minnesota Drift and Lake-bed
Flats (USGS 1970). These areas are approximately coincident with the Great Plains and
Central Lowlands regions, respectively, described by geobotanists. The Missouri
Escarpment separates these two regions. The Missouri Coteau (52,000 km2) is the only
physiographic region of the PPR that is part of the Great Plains (Kantrud et al. 1989).
The PPR is an extensive Wisconsin-aged (9,000-14,000 years old) glacial terrain
(Richardson et al. 1994). Relatively fine-textured depositional glacial deposits of till form a
mantle over stratified sedimentary rocks of Mesozoic and Cenozoic age (Bluemle 1971,
Winter 1989). The bedrock is composed mostly of limestones, sandstones, siltstones, and
shales (Winter 1989). The region is bounded by the Canadian Shield to the northeast, the
Cordilleran to the west, and the limit of Wisconsin glacial advance to the south (Winter
1989).
The morphology of the glacial drift overlying the bedrock consists of end moraines,
stagnation moraines, ground moraines, outwash plains, and lake plains. Drift also fills an
extensive network of valleys. The end and stagnation moraines are the most striking
physiographic features of the glaciated prairie. These have steep slopes and can be 10 m
to over 100 m above the flatter plains. Most potholes occur within depressions in end
moraines and ground moraines, where the relief is not as dramatic (Winter 1989). The
resulting landscape is a mosaic of closed-system, kettle-shaped depressions that vary in
size, topographic position, and relationship to the groundwater (Richardson et al. 1994).
The glacial till contains dolomite and calcite which create slightly alkaline soils
11
-------
September 1996
NHEERL-WED, Corvallis, OR
(Richardson et al. 1994). Nearly all soils in the region are Mollisols (Soil Survey Staff
1975). Most soils are dense and are developed on silty, clayey till and on lake clays, and
thus do not permit rapid infiltration of water. Soil frost is deep (1-1.3 m), so spring
snowmelt and some spring rainfall do not readily infiltrate but run into depressions (Winter
1989).
The PPR has a cool, continental climate with cold winters, hot summers, and extremes
in temperature and precipitation. The precipitation regime varies from semiarid in the
west to humid in the south and east. Overall, the climate is dry, and droughts and pluvial
cycles are common. Most of the precipitation occurs in the spring and summer. The
interactions between precipitation, temperature, and evapotranspiration are important
factors in the water budget of wetlands and can influence wetland frequency on the
landscape (Richardson et al. 1994).
In morainal areas, a natural drainage network has not developed and the depressions
are not connected by an integrated drainage system (Winter 1989). PPR wetlands usually
form relatively large, complex wetland systems connected to each other by groundwater
flow (Richardson et al. 1994). Groundwater flow is therefore an important component of
the PPR wetland water balance (Winter 1988, 1992; Richardson et al. 1992).
Landscape and climate influence interactions between groundwater and surface water
in the PPR. An equilibrium level of moisture develops to support hydrophytic vegetation,
where there is a balance between the input components of precipitation, overland flow,
and seepage inflow and the output components of evapotranspiration and seepage
outflow (Richardson et al. 1994). Because precipitation and temperature are so variable
in the PPR, the exact elevation where saturation persists long enough for hydrophytes to
grow is extremely variable and can change several feet in horizontal distance from year to
year (Richardson et al. 1994). Topographic features influence discharge and recharge
areas in the landscape, as well as flow-through systems (e.g., one part of a basin can be a
discharge, while a different portion can be a recharge area with the center of the basin
acting as a flow-through area). Flow-through systems are typical of semipermanent
wetlands in the PPR (Winter 1989). Temporary, seasonal, and semi-permanent wetlands
are often groundwater recharge wetlands, while semi-permanent and permanent wetlands
are typically discharge wetlands.
Because of the dry climate, wetlands in the northern prairie have a negative water
balance with respect to the atmosphere. Precipitation minus evaporation ranges from -10
cm in Iowa to -60 cm in southwestern Saskatchewan and eastern Montana (Winter 1989).
During a growing season, as much as 60% to 80% of the water loss from wetlands can be
attributed to transpiration by marginal vegetation and soil-water evaporation. The loss of
water is directly related to length of shoreline per unit area (Winter 1989).
The dominant native prairie vegetation has been grass for the last 6,000 years
(Richardson et al. 1994). The moisture retention capacity of soils allows a lush and
diverse prairie vegetation to develop (Winter 1989). Prairie grasses create dark-colored,
12
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
fertile soils that are high in organic matter (Richardson et al. 1994). More recently,
however, most of the native prairie has been converted to agricultural cropland.
2.6 Wetland Functions
Wetland functions are the physical, chemical, and biological processes and/or resulting
products that arise from the existence of a wetland. Three general categories of functions
include habitat, water quality, and hydrologic functions (Leibowitz et al. 1992a). Functions
of wetlands are present without consideration of the benefits they provide to society. As
described above, functions we originally selected for evaluation in the PPR, based on the
BPJ report (Adamus 1992), were habitat support for bird abundance and diversity and
water quality improvement. Since the BPJ report was completed, the Mississippi River
floods of 1993 signified an urgent need for addressing the role of wetlands in flood control.
Participants of the Bismarck workshop agreed that inclusion of the flood water attenuation
function would be valuable to the PPR synoptic assessment. We therefore added it as a
third function for synoptic assessment. In this section, we describe each function selected
for this assessment and how it is potentially influenced by natural and anthropogenic
factors. In addition, we will consider the risk of future loss of wetland function.
2.6.1 Habitat Support for Bird Abundance and Diversity
The degree of wetland habitat function in the PPR depends in part upon the spatial
extent and diversity of wetland types, which provide a multitude of niches that are needed
for plant and animal populations. Characteristic groups of birds are found in wetlands of
the PPR, and several bird species are locally, seasonally, or even exclusively dependent
on wetlands (Adamus 1992). A diversity of habitats helps maintain and enhance diversity
of associated bird species. Studies have shown that the number of PPR bird species
increases with wetland area (Brown and Dinsmore 1986, 1991; Hemesath and Dinsmore
1993). Most birds are not restricted to a single wetland basin for their habitat needs.
Complexes of wetlands sustain waterfowl populations over the course of a breeding
season (Krapu and Deubbert 1989) and have been shown to support more species and
greater abundances of birds than isolated wetlands (Brown and Dinsmore 1986). A
complex of wetlands may be able to provide a wider variety of habitat requirements and a
more constant food supply in a seasonally yarying environment (Brady and Pendleton
1983, Krapu and Deubbert 1989). Bird species richness in the PPR is greater in
landscapes with abundant wetlands than in landscapes with few wetlands (Kantrud 1981).
Diversity of plant species vegetation structure also contributes strongly to avian diversity
within wetlands (Adamus 1992).
Bird abundance is affected by landscape and wetland characteristics as well as land
use. Habitat heterogeneity at the landscape scale enhances bird abundance and diversity
(Adamus 1992). The primary breeding habitats of many species of waterfowl and non-
waterfowl birds are associated most often with seasonal and semipermanent wetlands
13
-------
September 1996
NHEERL-WED, Corvallis, OR
(Kantrud and Stewart 1984, Ruwaldt et al. 1979, Stewart and Kantrud 1973). Bird usage
of different types of wetlands varies by species over the year and is influenced by the
water regime (Krapu and Deubbert 1989). A significant component of the relationship with
season and water regime may be the availability and timing of emergence of invertebrates
(Brady and Pendleton 1983). Upland land use, which is typically some form of agriculture
in the PPR, can also have a profound effect on the biota and bird abundances in wetlands
(Kantrud and Stewart 1984, Stewart and Kantrud 1973, Krapu and Deubbert 1989,
Swanson et al. 1988). Uplands also provide critical nesting habitats for many species of
ducks; management practices and land cover can affect food availability and protection
needed by upland nesters.
Interest in waterfowl tends to dominate most land management activities in the PPR.
Waterfowl are extensively dependent on PPR wetlands for courtship, nesting, and rearing
of young. The PPR in North America and Canada produces 40-60% of the total
continental waterfowl (Batt et al. 1989); the U.S. portion of the PPR, which constitutes
36% of the total PPR, produces 25-30% of the continental waterfowl (Mineau 1987).
2.6.2 Water Quality Improvement
Water quality is improved by wetlands that trap overland flow laden with nutrients,
sediments, and pesticides, substances that would otherwise enter the region's streams
and rivers. Water quality is improved through the assimilation of nutrients by plants,
microbial conversions (e.g., denitrification) in the wetland, and adsorption of nutrients and
contaminants by sediments. Although the interception of sediments by wetlands is
recognized, it is not a sustainable function and eventually results in the filling of wetland
depressions. Only chemical application and transport will be targeted for index and
indicator development for this synoptic assessment. One process of chemical transport is
through adsorption onto sediments which can be transported in surface flows, so
sediments will be considered in this respect.
Phosphorus borne by overland runoff or incoming surface waters can be held for long
periods of time within the sediments of a wetland basin. While phosphorus is being
retained within a basin, it can be converted from one form to another, e.g., from organic to
inorganic form, or from oxidized to reduced form.
Nitrogen, transported via overland runoff, incoming surface waters, and groundwater,
can be processed by the wetland in a variety of ways. Inorganic nitrogen (nitrate, nitrite,
and ammonium) can be 1) removed from the immediate landscape as a result of
denitrification, which is the conversion of fixed inorganic nitrogen to gaseous forms, or 2)
retained for long periods of time within the sediment of a wetland basin. While nitrogen is
being retained within a basin, it can be converted from one form to another, e.g., from
organic (animal or plant proteins) to inorganic.
14
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
2.6.3 Flood Attenuation
The storage capacity of wetlands in a subunit helps attenuate the peak of the storm
hydrograph. Floods generally occur when runoff is not attenuated in the upland and
heavy pulses of water produce flow volumes that are greater than river capacity. In the
PPR, peak flow volumes are influenced by wetland storage capacity in the uplands,
infiltration, and runoff flow rate. All of these factors are affected by a number of natural
and anthropogenic stressors. The probability of flooding increases when wetlands are
filled or drained for agriculture, thereby reducing storage capacity. Tillage practices, land
cover type, the extent of wetland ditching and channelization, soil permeability, and slope
influence the amount and delivery time of runoff to streams from the upland (Scientific
Assessment and Strategy Team 1994). As a result, flooding can occur along streams or
in farm fields before the water reaches a stream.
2.6.4 Future Loss of Wetland Function
Functional loss represents the cumulative effects on a particular valued function that
have occurred within a subunit. It includes complete loss of function from conversion,
where the ecosystem is changed to a different ecosystem or land use (e.g., drainage),
and partial loss through degradation, where the impact does not change the ecosystem
type but alters function (e.g., tillage) (Leibowitz et al. 1992a). Historically, wetland loss in
the PPR has been due primarily to the intensification of agriculture (OTA 1984), and this
trend may continue at least in the near future. Agricultural practices can increase the
intensity of sedimentation, wetland drainage, and tillage within temporary and seasonal
wetlands. Loss can also result from urban growth. The amount of functional loss
associated with wetland area loss in a subunit depends on the characteristics of the
impact (its type, magnitude, timing, and duration), and ecosystem resistance, or the
relative sensitivity of the ecosystem to the impact (Leibowitz et al. 1992a).
2.7 Wetland Values to Humans
Wetland values are considered services realized by "users," human or non-human,
who benefit from wetland functions within a subunit. Benefits can be divided into use and
nonuse benefits. Use benefits can be tangible, such as clean water and habitat for
hunting, or intangible, such as aesthetics. Tangible benefits are generally associated with
consumptive activities, while intangible benefits are associated with nonconsumptive
activities. Nonuse benefits are those that are realized by nonhumans, and we consider
them therefore to be covered under the habitat support function. Values can be realized
within and outside of a subunit. Consequently, wetland value may need to be assessed at
large spatial scales. Because different users will value the same function differently,
determining wetland value includes subjective choices (Leibowitz et al. 1992a).
We consider two representations of value: 1) the total benefit associated with the
15
-------
September 1996
NHEERL-WED, Corvallis. OR
current level of wetland function in a subunft and 2) the marginal benefit associated with
the addition of a unit of wetland function onto the current level. Benefit is defined as a
positive change in welfare (Freeman 1979). An inverse relation between the marginal
benefit of (or marginal willingness to pay for) environmental improvement and the current
level of environmental quality is often assumed by economists because of the way that
humans tend to respond to scarcity. For example, adding one ha of wetland to 100 ha
usually has a higher benefit to humans than adding one ha of wetlands to 10,000 ha.
2.7.1 Habitat Support for Bird Diversity and Abundance
The value of wetland habitat support for bird diversity and abundance is found at both
the regional and local levels. At the local level, waterfowl and other wetland bird species
are valued by hunters and nonconsumptive users, such as bird-watchers. The North
American Waterfowl Management Plan (U.S. Fish and Wildlife Service and Canadian
Wildlife Service 1986) recognizes the PPR as the highest priority region for protection.
Migratory waterfowl produced by PPR wetlands support hunting in all states lying within
the Mississippi flyway. Waterfowl hunting and the economy it supports were valued at $21
million per year in North Dakota in 1986. North Dakota has the greatest number of
waterfowl hunters per capita of any state (Adamus 1992). At the regional level, bird
diversity is valued as an intrinsic indicator of ecosystem function and condition. Both
human and animal populations benefit from an ecosystem that is in good condition and
providing benefits.
2.7.2 Water Quality Improvement
Wetlands can improve water quality by intercepting materials from incoming waters
and 1) reducing downstream pollutant loads or 2) reducing pollutant loading to
groundwater in recharge systems. Wetland capability to improve water quality is valued
by society, which relies on surface and groundwater for household, commercial, and
agricultural uses.
2.7.3 Flood Attenuation
Cumulatively, wetlands in the upland can attenuate downstream flooding (Preston and
Bedford 1988). This function is valued by people living in the floodplain and others who
are financially or otherwise affected by the flood damage, which can potentially include all
taxpayers. Although precautions have been taken to reduce flooding (e.g., levee-building,
stream channelization), these measures often are ineffective against large floods, which
can cause millions of dollars in damage to large regions. Although the probability of a
large flood is low, the economic risks are very high. Consequently, those who are affected
by flooding value opportunities for flood abatement. The importance of wetlands in
attenuating floods was stressed after the Mississippi River floods of 1993, which caused
16
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
more than $15 billion in property damage (WWF 1994). Although these floods were of a
very large magnitude, they were not an unusual event (WWF 1994).
2.8 Wetland Impacts
Historically, and continuing to the present time, a number of stressors associated with
agriculture have impacted wetlands and wetland functions in the PPR (Adamus, 1992).
We describe four of these in the subsections below.
2.8.1 Artificial Drainage
Artificial drainage consists of ditches, drainage wells, or subsurface (tile) pipes placed
in wetlands or wet soils. Drainage networks are installed to lower seasonal water tables
for agriculture and to improve human access. Ditches are commonly placed within prairie
pothole wetlands, and several ditches may connect temporary and seasonal wetlands that
collectively drain into more semipermanent wetlands, lakes, or streams. Subsurface tile
drains also are used to lower water tables; they are prevalent in Iowa and southern
Minnesota. The installation of drainage ditches can completely convert a wetland to
upland or result only in partial conversion (Weller 1981, cited in Adamus 1992). Large
numbers and areas of wetlands in the PPR have been converted or partially drained for
agriculture or highway construction. The USFWS Regional Wetland Concept Plan
(USFWS 1990) describes artificial drainage as an important cause of wetland degradation
and loss in all PPR states.
2.8.2 Tilling and Planting
Temporary and seasonal wetlands in the PPR are most often the wetland types that
are tilled for planting, especially in flat landscapes or during drought. In addition, the
upland area surrounding a wetland is also tilled, leaving the extent of natural buffer zones
a matter of discretion by farmers. Many of these cultivated wetlands are not drained.
Tillage is more likely in areas with fertile soils; it is less common in areas owned or
managed by private conservation groups or government agencies, which usually prohibit
or restrict wetland tilling. Agricultural practices associated with preparing the land for
planting have been identified as significant causes of wetland degradation and loss.
2.8.3 Sedimentation
Sedimentation impacts all wetland functions through several different mechanisms,
e.g., impeding water circulation and light penetration, filling wetlands, and delivering
adsorbed contaminants to wetlands (Adamus 1992). The degree of functional impairment
depends upon several factors, including soil type and erodibility, wetland water regime,
landscape relief, and specific agricultural practices in the surrounding area.
Sedimentation problems in PPR wetlands are suspected to be increasing as conversions
17
-------
September 1996
NHEERL-WED. Corvallis, OR
from pasture to row crops occur (Kantrud et al. 1989).
2.8.4 Nutrient Inputs
Excessive nutrient inputs can substantially alter the functions of a wetland. Sources of
nutrients include fertilizers applied to crops, livestock waste, dry deposition, or
sewage/septic systems. "Excessive" inputs are considered to be those that wetlands are
incapable of assimilating without long-term detrimental impacts on wetland functions.
Crop fertilizers probably are the main source of excessive nutrients in runoff in the eastern
PPR, particularly when applied at greater than recommended rates. In northwestern parts
of the PPR, livestock waste probably contributes more to aquatic enrichment than does
fertilizer runoff from cropland.
2.9 Wetland Replacement Potential
The feasibility of restoring wetlands and wetland functions (e.g., supporting bird
species diversity and abundance) depends upon several factors (Adamus 1992, Leibowitz
et al. 1992a): 1) the availability of appropriate sites, 2) the specific types of sites available,
3) the degree of alteration of landscape processes (e.g., hydrology) necessary for
replacing the function, and 4) the extent of practical, physical constraints. Available sites
generally are those where wetlands previously existed or have been partially drained, the
locations and extent of which can be obtained from the Natural Resource Inventory (NRI)
database. Restoration probably can be accomplished most easily on wetjands
unsuccessfully drained or drained by ditches, where restoration is simply a matter of
stopping further drainage by "plugging" the ditch or channel.
2.10 Landscape Subunits
In a synoptic assessment, the landscape is divided into subunits, which are the areal
units for reporting results, making comparisons, and establishing ranks. Typically,
subunits follow natural ecological or political demarcations or a combination of the two.
The most appropriate subunit depends upon the application of the assessment results and
the regional ecology. Ecological boundaries (e.g., drainage basins) often enclose and
isolate related landscape processes and impacts within an area and may therefore be
optimal from an ecological perspective. Management decisions, however, usually are
made within political units, such as States or counties. Data often are available for
political units as well. In order to consider simultaneously the ecology, management
objectives, and data availability, we propose to use ecologically defined subunits to
conduct the PPR assessment and, if necessary, to weigh or prorate political unit data
based on the proportion of the ecological subunit covered by each political unit (Appendix
A).
The boundary for the PPR study area will be defined using the outer boundary of
Mann's ecoregions (Figure 1) (Mann 1974). The landscape subunits within this boundary
18
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
will be the hydrologic cataloging units (hereafter, hydrologic units) of the USGS
classification (Seaber et al. 1984). A hydrologic unit is a geographic area representing
part or all of a surface drainage basin, a combination of drainage basins, or a distinct
hydrologic feature (Seaber et al. 1984). According to this classification, each hydrologic
unit is given a numeric code (hydrologic unit code, or HUC) and is commonly referred to
as a HUC (Figure 1). When the data that will be used to derive the synoptic indicators are
available only at the county level, the proportion of each county within a HUC will be used
to weight the county data to derive a representative value for the HUC.
2.11 Assessment Criteria
To prioritize subunits for restoration, we will evaluate each function in two contexts: its
value to humans, and its replacement potential. The risk of functional loss will be
evaluated for all functions combined. The result will be seven general evaluations, which
we will refer to as assessment criteria. The following criteria will be evaluated for the PPR
synoptic assessment:
Risk of future loss of wetland function,
Value of bird abundance and diversity to humans,
Replacement potential for bird abundance and diversity,
Value of water quality improvement to humans,
Replacement potential for water quality improvement,
Value of flood attenuation to humans, and
Replacement potential for flood attenuation.
These criteria will be evaluated using indicators. Indicators, described in Part 3, are
metrics for which data exist; they allow the assignment of values to criteria for each
subunit. Subunits can then be ranked according to each criterion. Ranks can be
examined comparatively to help predict the areas where the most benefit to landscape
Wetland functions might be achieved through restoration. Ranks can thus be used to
prioritize subunits where restoration effort and funds might be best spent.
19
-------
September 1996
NHEERL-WED, Corvallis, OR
Part 3: Conceptual Models and Selection of Indicators
To conduct the PPR synoptic assessment, we need a way to place quantitative values
on each of the criteria listed in Section 2.11 so that they can be compared and ranked.
Evaluation will be accomplished through the use of indicators, which are measurable
surrogates for ecosystem processes and stressors. We will refer to ecosystem processes
and stressors that cannot be directly measured as conceptual quantities. Conceptual
models provide a framework for describing the patterns and linkages among conceptual
quantities and indicators that are important in evaluating assessment criteria.
Conceptual models consist of relatively few conceptual quantities considered
necessary for quantifying criteria on a very simple level. Simplicity is desirable because it
minimizes the multiplication of errors, or unexplained variation in the indicators, when
the index values are calculated;
minimizes the magnification of uncertainty in combining index components by
lessening overall the chance of making errors in the mathematical expression of their
relationships; and
produces expressions that are more intuitive and easier to understand, calculate, and
use for making general assessments.
If conceptual quantities are complex and cannot be represented by a single indicator,
indices can be constructed on the basis of linkages defined in conceptual models. Indices
are aggregations of data and can be represented by more than one indicator, other
indices, or a combination of both. Indices can be data-derived or ranks-derived. If they
are data-derived, the index is modeled according to an explicit, known functional
relationship among the index components. If they are ranks-derived, the explicit functional
relationship is not known, and the index is derived through some combination of ranked
indicator values. Index ranks can be used to address and prioritize management
concerns within subunits of a particular environmental setting. They allow the evaluation
of assessment criteria in a particular landscape subunit compared with other subunits.
For example, chemical transport to wetlands is a function of rainfall, sediment
transport, and upland assimilation, all of which can be quantified separately with indicators
(e.g., amount of precipitation, erodibility factors, slope, soil permeability, and land cover).
The direction of the relationships (e.g., inverse or direct) is known; however, the explicit
nature and form of the relationships is unknown. A ranks-derived index is then formulated
for chemical transport by mathematically combining the ranks of the indicators. If there
were a specific model describing the relationships between chemical transport and the
indicators, a data-derived index could be developed using the actual indicators.
Conversely, some conceptual quantities do not need to be estimated by combining data or
ranks. Chemical application in the upland is an example of a conceptual quantity that can
be estimated directly with one indicator; an index is not necessary.
20
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
For a synoptic assessment, indicators need to be measures for which data exist
uniformly across the entire region of interest. If no data exist, the ecological concept and
linkage is retained in the conceptual model, but no indicator is attached to it. Choosing
indicators often requires making simplifying assumptions because of limited information,
time, and money. The index and the indicator are defined separately so that it is clear that
the indicator is not necessarily the management concern but is only being used to
evaluate it. For example, agricultural nonpoint source nutrient loading is a management
concern, while agricultural area might be the landscape indicator selected to represent it.
The approach is not a fixed procedure that always uses the same data sources and
produces a standard set of end products. Unique indices and indicators are defined for
each synoptic assessment to make the assessment specific to the particular region of
interest and for particular uses. This approach to index development strives to minimize
index complexity (i.e., only one indicator is used to represent a process or relationship
when possible), avoid combination of terms into equations until necessary, and
ecologically justify the combination of terms. A flow diagram of the procedure is given in
Figure 2.
The final products of the PPR assessment will be ranking tables for each criterion.
The tables will contain the conceptual quantities of models that are meaningful to
managers and the final tier of indices and indicators. Indicator values will be entered into
the table for each subunit. To illustrate the procedure, Table 2 is a partially completed
table for the replacement potential for habitat support (i.e., only 14 PPR subunits are
included to demonstrate the procedure). Combining indicators into indices and deriving
the actual subunit ranks for each criterion is discussed in more detail in Section 4.
In this section we present the conceptual models and indicators for seven assessment
criteria, which will provide the basis for conducting the PPR synoptic assessment. We
also present the tables in which values will be entered as we conduct the assessment.
Conceptual quantities are the concepts or nonmeasurable model components necessary
for evaluating criteria. They can be quantified with indicators. In the models included in
this section, conceptual quantities are enclosed in ovals, while indicators and their data
sources are enclosed in rectangles. Arrows point in the direction of influence of one
concept on another; solid arrows denote positive influence, and dashed arrows denote
negative influence. Appendix A lists expanded database names, the indicators from each
database that are used in conceptual models, and the indicators' units.
3.1 Risk of Future Loss of Wetland Function
The conceptual model for this criterion is presented in Figure 3; indicator values will be
tabulated in Table 3. In developing this model, we assumed that the risk of future loss of
wetland function is due primarily to human activities that result in the draining or filling of
wetlands.
Two human activities were considered in the model: crop agriculture and urban growth.
21
-------
Figure 2. Method for producing conceptual models and data tables for synoptic assessment.
22
-------
Table 2. Example of a partially completed data table for the replacement potential for the habitat support function. Indicator values are shown for the
25 HUCS in South Dakota.
ro
CaJ
Criterion:
Replacement Potential for Habitat Support: Bird Abundance and Diversity
Conceptual
Wetland Quantity Available for Replacement
Likelihood of Full Functional Replacement
Quantities
Physical Availability
Socio-economic Availability
Degradation of Upland
Habitat Quality
Degradation of Wetland
Habitat Quality:
Pothole Component
Current Wetland
Availability
Historic
Wetland
Availability
Indicators
Area of Wetland
(square
kilometers)
Area of
Hydric Soil
(square
kilometers)
Extent of
Private Land
Ownership
(percent)
Property Value
Per Square
Kilometer
(dollars)
Cropland:Upland Ratio
(ratio)
Data
Source
NRI
NRI
NRI
Census of
Agriculture
NRI
Subunit:
- .Ic: Ķ"
Ķ "
K
. : -s: ,f:, - -Vris-f
7020001
6.43
9.90
94.1
74200
.62
7020003
9.03
0.86
96.9
33000
.69
9020101
3.03
4.82
96.3
35100
.89
9020105
4.27
5.49
96.4
10600
.77
10130102
2.38
1.57
97.7
35900
.49
10130105
5.03
1.78
96.3
76700
.65
10130106
4.56
3.87
94.8
48700
.45
10140101
7.98
4.49
93.4
71000
.54
10140103
1.90
1.75
92.6
64800
.55
10140105
1.80
1.73
97.5
59200
.36
10160003
9.20
12.82
97.1
32800
.69
10160004
2.90
4.25
93.8
28100
.55
10160005
0.43
1.18
99.8
88100
.84
10160006
8.16
9.38
98.4
84000
.57
10160007
3.72
3.68
96.2
63200
.39
-------
Table 2, cont. Example of a partially completed data table for the replacement potential for the habitat support function. Indicator values are shown for
the 25 HUCS in South Dakota.
Indicators
Area of Wetland
Area of
Extent of
Property Value
Cropland:Upland Ratio
Hydric Soil
Private Land
Per Square
(square
(square
Ownership
Kilometer
kilometers)
kilometers)
(percent)
(dollars)
(ratio)
Data
NRI
NRI
NRI
Census of
NRI
Source
Agriculture
Subunit:
. i ' V!/ ij r'-y /'V,:
10160008
4.36
4.97
97.1
69300
.48
10160009
3.39
3.45
98.2
61500
.49
10160010
7.50
5.22
94.1
82500
.51
10160011
9.67
13.73
98.8
107500
.61
10170101
2.55
2.27
97.9
99200
.67
10170102
9.12
12.34
98.2
159300
.77
10170103
1.96
1.37
97.5
99100
.78
10170201
6.23
6.40
94.0
103000
.73
10170202
4.50
6.15
96.1
126100
.65
10170203
4.95
8.59
95.8
162100
.76
-------
KEY
M Conceptual Quantiy
Indicator (Measured Quantiy)
Wet
Risk of Wetland Function Loss
Duo to Sedimentation
7"
Agricultural
Intensity
Tillage
Practices
Risk of Future Loss of Wetland Function
1
Risk of Future Loss of Wetland Function
Due to Crop Agriculture
Risk of Future Loss of Wetland Function
Due to Urban Growth
1
Risk of Draining/Cropping Wetlands
on Cropped Land
I
Population
Growth Rate
U.S. Census
Likelihood of Intensifying Cropping
or Converting Land to Cropping
1
Relative Ease of j
Draining/Cropping WetlandsJ
Soil/Land
Productivity
I
Land Capability
Class
NRI
X
Opportunity Cost
of Accepting Subsidy
to Not Drain/Crop Wetlands
Size and Class
of Wetlands
Ratio of Corporate
to Family Farms
Agricultural
Growth Rate
U.S. Census
of Ag.
Percent
Wetland Area
Seasonal and
Temporary
NRI
Average Size
of Farmsteads
U.S. Census of Ag.
Figure 3. Conceptual model for risk of future loss of wetland function due to crop agriculture and urbanization.
-------
Table 3. Table format for entering Indicator values for the risk of future loss of wetland function.
Criterion:
Risk of Future Loss of Wetland Function
Conceptual
Quantities
Risk of Future Loss of Wetland Function Due to Crop Agriculture
Risk of Future
Loss of
Wetland
Function
Due to Urban
Risk of Wetland Function Loss Due
to Sedimentation
Risk of Draining/Cropping wetlands on Cropped Land
Agricultural Intensity
Likelihood of Intensifying
Cropping or Converting
Land to Cropping
Relative Ease of
Draining/Cropping
Wetlands
Opportunity
Cost of
Accepting
Subsidy to Not
Growth
Tillage
Practices
Soli/Land
Productivity
Size and
Class of
Wetlands
Ratio of
Corporate to
Family
Farms
Drain/Crop
Wetlands
Indicators
Crop:
Upland
Ratio
(no
Indicator
available)
Erodlblllty
Land
Capability
Class
. Agricultural
Growth Rate
(no Indicator
available)
Average
Size of
Farmsteads
(no Indicator
available)
Population
Growth Rate
Data Source
NRI
NRI
NRI
US Census
of Agriculture
US Census
of
Agriculture
US Census
Subunit:
ĶĶĶĶmi
1
2
3
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
The risk due to urban growth will be quantified with a single indicator-population growth
rate. Data will be obtained from the U.S. Bureau of the Census (U.S. Bureau of the
Census 1982, 1992). Evaluating the risk due to crop agriculture is more complex and can
be further subdivided into: 1) the risk due to sedimentation and 2) the risk due to draining
and cropping wetlands.
3.1.1 Risk from Sedimentation
The risk from sedimentation is the risk of a wetland filling with sediments to the extent
that it no longer holds sufficient water to function hydrologically. Wetland depressions
have been known to fill with sediments during a single storm event (L. Cowardin, personal
communication, Northern Prairie Science Center, Jamestown, ND, 1993), particularly
when soil is exposed or if sediment is highly erodible. Sediments can be dislodged and
transported by wind and/or runoff water to wetland basins. We will use the soil erodibility
variable from the 1992 Natural Resources Inventory (NRI) (U.S. Department of Agriculture
1992) database as one of the indicators of risk from sedimentation. We assumed that soil
exposure is primarily a factor of agricultural intensity, which is influenced by tillage
practices (e.g., no-till, conservation tillage, timing of tillage, the amount of time fields are
left fallow) and the proportion of land that is used for growing crops. Tillage practices are
difficult to quantify, and we did not identify a quantifiable indicator for that particular
conceptual quantity. Consequently, agricultural intensity will be quantified by the ratio of
cropland to upland, which will be obtained from the NRI database. This is a very
simplified model because it does not consider differences in land management practices
(e.g., maintaining a cover crop to prevent soil erosion; practicing conservation tillage) and
the differential effects of those practices on sedimentation. However, we assumed that
the mix of management practices are relatively uniform over the landscape. The
relationship of subunit values relative to each other thus will be preserved so that subunits
can be ranked.
3.1.2 Risk from Draining and Cropping Wetlands
We considered the risk from draining and cropping wetlands to be a- function of three
factors: 1) the likelihood of intensifying cropping or converting land to cropping; 2) the
relative ease of draining or cropping wetlands; and 3) the opportunity cost of accepting a
government subsidy for not draining or cropping wetlands. Draining wetlands to plant
crops or planting crops directly in wetlands causes hydrologic and habitat alterations that
compromise natural wetland functions. Cultivation modifies infiltration and drainage
patterns, which alters the hydrologic functioning of the wetland and can lead to increased
runoff in downslope areas. Replacement of wetland plant species with agricultural crops
alters the habitat support function by reducing the species and structural diversity of the
wetland and eliminating food sources important to wildlife.
27
-------
September 1996
NHEERL-WED. Corvallis, OR
3.1.2.1 Likelihood of Intensifying Cropping
We presume that the likelihood of intensifying cropping depends on the agricultural
growth rate and on the productivity of the land, i.e., fanners are more likely to convert land
that is known to be productive and will yield a profitable crop. Data for the agricultural
growth rate will be obtained from the 1992 Census of Agriculture (U.S. Department of
Commerce 1992). Land productivity will be quantified with the land capability class.
Indicator data for land capability will be obtained from the 1992 NRI database.
3.1.2.2 Relative Ease of Draining or Cropping Wetlands
The relative ease of draining or cropping wetlands is considered a function of the size
and class of wetlands and the ratio of corporate to family farms. The percent of wetland
area comprised of seasonal or temporary wetlands will be used as an indicator for the
former. It is often easy for farmers to till through seasonal and temporary wetlands and
plant crops because draining or other hydrologic modifications are not necessary. The
percent of wetlands in a subunit that are seasonal or temporary will be estimated from the
NRI database. We assumed that the ratio of corporate to family farms is an indication of
profit and income and that corporate landowners are more likely to be able to fund
drainage operations, including those involving larger wetlands. We chose the average
size of farmsteads as the indicator for this conceptual quantity. Data will be obtained from
the U.S. Census of Agriculture.
3.1.2.3 Opportunity Cost
The opportunity cost of accepting a subsidy for not draining or cropping wetlands is the
income that a landowner foregoes by not farming. We assumed that, if the opportunity
cost is high, it is more likely that the landowner will reject the subsidy and farm the land,
which increases the risk of draining or cropping wetlands. Because the value of a
particular crop, and therefore the opportunity cost, is more unpredictable from year to
year than is a subsidy, we simplified the model by assuming that the future risk is
primarily a function of the change in the subsidy. Because no data source for agricultural
subsidies exists, and we do not know what the future subsidy would be, we will not
attempt to evaluate this conceptual quantity.
3.2 Value to Humans of the Habitat Support Function for Bird Diversity and
Abundance
The conceptual model of the value to humans of the habitat support function is shown
in Figure 4; indicator values will be tabulated in Table 4. We considered value to be a
function of current value and potential marginal benefit with restoration. Each of these
components can be represented by current total bird abundance and diversity (i.e., current
function), the population of users, and the attractiveness of the subunit for use by humans.
Their values, however, differ because current bird abundance and diversity are positive
28
-------
Figure 4. Conceptual model for value to humans of habitat support function of wetlands (bird abundance and diversity).
-------
Table 4. Table format for entering indicator values for the value of habitat support to humans.
CO
O
Criterion: Value of Habitat Support to Humans: Bird Abundance and Diversity
Conceptual
Quantities
Population
of Users
Attractiveness
of Subunit for
Use
Total Bird Abundance and Diversity
Access
to
Users
Aesthellc
Quality
Quantity
of
Wetland
Quality of Wetland Habitat: Pothole
Quality of Wetland
Habitat:
Landscape
Quality of Upland
Habitat
Input of Chemicals and Sediment
Soil/Land
Productivity
Diversity
of
Wetland
Types
Intensity
of
Draining
Intensity
of
Cropping
SolM_and
Productivity
Chemical
Application
Delivery
Intensity of
Cropping
Intensity
of
Draining
Pre-
clp-
Ita-
tlon
Erodl-
blllty
Slope
Soil/Land
Produc-
tivity
Crop:
Upland
Ratio
Ditch-
ing
Indicators
Population In
Focal and
Surrounding
Subunlta
(no
Indica-
tor
avail-
able)
(no
Indica-
tor
avail-
able)
Wetland
Area
Land
Capability
Class
Soil
Capability
Class
(no
Indicator
available)
Ditching
Crop:
Upland
Ratio
Soil
Capability
Class
Data
Source
US Census
USGS
NRI
NRI
NRI
NRI
PRI
SM
NRI
NRI
NRI
NRI
NRI
Subunit:
1
2
3
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
influences on current value and negative influences on potential increase in value. The
negative association results because, as noted above, humans tend to place a higher
value on the addition of one unit of a commodity when that commodity is scarce. Thus, if
current bird abundance and diversity are low, the potential marginal benefit with
restoration will be high; if bird abundance and diversity are currently high, an urgent need
for increasing it is not perceived, and the increase in value with restoration is low.
3.2.1 Total Bird Abundance and Diversity
We considered the total current bird abundance and diversity to be a function of upland
habitat quality, wetland habitat quantity, the quality of wetland habitat over the landscape,
and pothole habitat quality. Each of these is discussed below.
3.2.1.1 Upland Habitat Quality
We defined the quality of upland habitat as a function of the intensity of cropping
and soil productivity. Soil productivity is determined by the cover type, the food resource,
and the spatial arrangement of habitat patches. Cover provides protection from predation
for nesting, feeding, and brood-rearing activities and is most effective when its structural
diversity is greatest. Cover can also serve as food (e.g., seeds) and/or as substrate for
prey such as aquatic invertebrates. The diversity and structural forms of vegetation and
invertebrate food items present depend to a large degree on soil productivity. Birds often
select areas where productive soils result in increased plant and invertebrate abundance
in wetlands, even if the uplands are planted in row crops. Higher soil productivity will also
result in denser and more robust crop growth, which provides better cover than the same
crop on unproductive soils. The indicator we will use to quantify soil productivity is land
capability class, which is part of the NRI database.
Intensive agriculture can result in wetland drainage and often is associated with row
cropping, which typically provides less adequate cover for bird nesting than a dense cover
of prairie grasses or grain crops. The higher the cropping intensity, the lower the quality
of upland habitat. We will use the ratio of cropland to upland (NRI data) as an indicator of
the intensity of cropping. Although some crops are more labor intensive than others (e.g.,
require more fertilizers or pesticides, necessitate more tillage or bare soil), for simplicity
we assumed that the mix of crop types is a minor influence compared to the presence or
absence of cropping activities.
3.2.1.2 Wetland Habitat Quantity
The quantity of wetland habitat can be represented directly by wetland area from the
NRI database.
31
-------
September 1996
NHEERL-WED, Corvallis, OR
3.2.1.3 Habitat Quality of the Landscape
The third factor that influences current breeding bird abundance and diversity is the
quality of wetland habitat over the landscape, which refers to the heterogeneity and spatial
pattern of wetlands. Most wetland birds are not associated with isolated or single
wetlands; instead they require a diversity of habitats and wetland types for support across
their life history (Krapu and Deubbert 1989, Ruwaldt et al. 1979). The proximity of
wetlands to each other within a complex, the connectivity of complexes over the
landscape, and the diversity of wetland types (e.g., permanent, seasonal) within
complexes are also important considerations for breeding and migrating birds. Numerous
metrics have been used to describe spatial patterns such as proximity and connectivity
(e.g., contagion, shape index, fractal dimension, patch area, patch perimeter, nearest
neighbor distance), but their predictive capabilities have not been consistently shown. We
could not find data for the region on wetland type. We therefore show the diversity of
wetland types in the conceptual model as a conceptual quantity and will not attempt to
quantify it. In Figure 4 we show that the intensity of draining also influences the habitat
quality of the landscape. Drainage diminishes wetland habitat in a piecemeal fashion, and
ditches can fragment a landscape. We will use the extent of ditching as an indicator of the
intensity of draining. Data will be obtained from the NRI.
3.2.1.4 Pothole Habitat Quality
The quality of pothole wetland habitat, the fourth factor that influences current breeding
bird abundance and diversity, is considered a function of soil productivity (as described
above) and the input of chemicals and sediment. Chemicals and sediment transported in
runoff to wetlands can introduce numerous stresses and imbalances to pothole functioning
including the contamination of biota, reduction in biodiversity, eutrophication, choking by
excessive plant growth, changes in invertebrate composition and abundance, reduction in
wetland area, and burial of plants. All of these stresses can lower the quality of pothole
wetland habitat and, therefore, have an overall detrimental effect on current bird
abundance and diversity. The amount of input depends on the amount of chemicals
applied in the upland and the amount delivered to wetlands.
Delivery comprises erodibility, precipitation, and the intensity of cropping and draining.
We will use PRISM (Daly et al. 1994) precipitation data and the soil erodibility factor from
the NRI database, described in Section 3.1, as indicators for delivery. The erodibility
factor represents the likelihood that sediments are dislodged during storm events. Its
value incorporates the potential for sediments and sediment-laden chemicals to be moved
down slope. The amount and intensity of precipitation are used as indicators of the
likelihood of transport of chemicals and sediments in overland flow. We will use the extent
of ditching and the ratio of cropland to upland area, both obtained from the NRI data base,
as indicators of the intensity of cropping and draining. Ditching will be used as an
indicator of the amount of wetland drainage and hydrologic modification that has occurred.
32
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Wetland drainage can alter drainage patterns, leading to the accumulation of runoff and
the concentration of chemicals in fewer wetland basins. Drainage also increases
connectivity among basins, resulting in a "flushing effect," whereby chemicals and
sediments are transported in rapidly moving surface water rather than being absorbed in
the upland. The intensity of cropping can also influence delivery of chemicals and
sediments to potholes. We assumed that a greater proportion of cropland would mean
that soil is exposed to a greater extent and at a higher frequency than on uncropped land,
and it may be compacted by tillage, which leads to a greater runoff volume and greater
potential for delivery of chemicals and sediments to wetlands. The indicator for cropping
intensity will be the ratio of cropland to upland in a subunit, which will be obtained from the
NRI database.
Chemical application and delivery work simultaneously to influence the input of
chemicals to wetlands. For example, a subunit may be characterized by a high chemical
application rate, but the effect on wetland basins could be small if the delivery of those
chemicals to wetlands is low. We could find no reliable, consistently available data for
estimating chemical application over the course of a year. We assumed, however, that
chemical application is associated with the intensity of agriculture. The cropland to upland
ratio, obtained from the NRI database, will be used as an indicator of agricultural intensity.
Hydrologic modification, the third component of the habitat quality of the wetland basin
habitat, is also affected by the intensity of cropping and draining. Cropping can alter
runoff patterns and the amount of surface flow. Draining eliminates the hydrology
necessary for maintaining wetland functions. It also changes the hydrologic regime in
down-slope areas by connecting wetlands and creating surface water flow channels,
resulting in increased water delivery to wetlands at the ends of channels. This hydrologic
modification can negatively impact plant communities, aquatic organisms, and wildlife use
of wetlands. The extent of ditching, from the NRI database, will be used as a measure of
the intensity of draining (hydrologic modification).
3.2.2 Population of Users and Attractiveness of the Subunit for Use
On the right side of the conceptual model (Figure 4), the population of users and the
attractiveness of the subunit for use are shown to positively influence both the current
value and the potential marginal benefit with restoration. Value is directly proportional to
the.number of people who receive the benefits (i.e., users). We assumed that the majority
of use in a given subunit is by people who live in that subunit and in adjacent subunits.
We will use the total population in focal and surrounding subunits as an indicator of
potential users (consumptive and non-consumptive) in the focal subunit, assuming that the
number of users is proportional to the total population. The population will be estimated
from U.S. Bureau of the Census data.
Attractiveness of the subunit for use is influenced by aesthetic qualities and
accessibility to users, both of which are depicted as conceptual quantities with no
33
-------
September 1996
NHEERL-WED, Corvallis, OR
associated indicators. The value to users is recreational, rather than intrinsically
ecological; therefore, it is directly proportional to the number of people (users) who
benefit.
3.3 Replacement Potential for the Habitat Support Function for Bird
Abundance and Diversity
The conceptual model for this criterion is presented in Figure 5; indicator values will be
tabulated in Table 5. We considered replacement potential for habitat support to be a
function of wetland quantity available for replacement and the likelihood of full functional
replacement.
3.3.1 Wetland Quantity Available for Replacement
Wetland quantity available is a function of the physical availability and the socio-
economic availability. Physical availability refers to the area that was once wetland and
could be restored by replacing the original hydrology. It is essentially the difference
between historic wetland area and current wetland area. Current wetland area will be
estimated from the NRI data base. We assumed that, historically, wetlands occurred
where hydric soils occur. Although many wetlands have been drained, the location and
extent of hydric soils have not changed and can serve as the historic record. Data will be
obtained from the NRI data base.
This approach assumes that hydric soils can be used to estimate historical wetland
extent and that NRI data represent current wetland area adequately. However, hydric soil
can be underestimated, and wetland extent can be inaccurate because of the sample-
based estimation technique of NRI. The NRCS soil surveys, from which NRI data are
produced, depict predominant soil types, but soil surveys often do not include small
patches of hydric soils that lie within a larger matrix of non-hydric soils. These smaller
patches are often referred to as hydric inclusions. Regional rankings may be heavily
influenced if hydric inclusions are not considered, but given the constraints of data
availability, we could not find an efficient method for estimating the area of inclusions. In
addition, NRI estimates might include wetlands that are actually partially drained, which
could result in an underestimation of the wetland area available for restoration. We
acknowledge the above biases as limitations and possible sources of error in the
assessment. To evaluate the potential effect of these errors on the assessment results,
we plan to analyze alternative sources of data and/or conduct field reconnaissance
studies to estimate the variability in area of small wetlands and hydric inclusions for a
subset of subunits. If the variability is large, it might result in inaccuracies in the rankings.
If it is small, we can assume a uniform influence over the landscape, which should not
greatly affect the relative rankings.
Socio-economic availability will be quantified by two indicators: the extent of private
land ownership and the per acre property value. We presume that wetland restoration will
34
-------
Figure 5. Conceptual model for replacement potential for habitat support function of wetlands (bird abundance and diversity).
-------
Table 5. Table format for entering Indicator values for the replacement potential for habitat support.
Criterion:
Replacement Potential for Habitat Support: Bird Abundance and Diversity
Conceptual
Wetland Quantity Available for Replacement
Likelihood of Full Functional Replacement
Quantities
Physical Availability
Socio-economic Availability
Degradation of Upland
Habitat Quality
Degradation of Wetland
Habitat Quality:
Pothole Component
Current
Wetland
Availability
Historic
wetland
Availability
Indicators
Area of
Wetland
Area of
Hydric Soil
Extent of
Private
Land
Ownership
Property
Value
Crop: Upland Ratio
Data Source
NRI
NRI
NRI
US Census of
Agriculture
NRI
Subunit:
' * *
1
2
3
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
be less likely on private land than on public land because of the need for landowner
approval. Restoration will also be less likely where property values are high because of
the expense of buying the land beforehand and because high property values are
associated with high land productivity. We assume that it would be more difficult to
persuade a landowner to sell or ease land for restoration and forego the profits from crop
production if property values are high.
3.3.2 Likelihood of Full Functional Replacement
The goal of restoration is not only to replace the water but to ensure, to the greatest
extent possible, subsequent development of the function (wetland habitat support) present
historically. Replacing quality basin habitat requires the development of wetland
vegetation, colonization by decomposers and other biota, and water fluctuation necessary
to maintain biological processes on a long term basis. The extent to which these
processes have been hindered by human modifications to the landscape, and the extent
to which this degradation persists after restoration, affects replacement potential. We
assumed that the greater the degradation of the landscape, the lower the likelihood of fully
replacing the lost function. We also assumed that natural ecological factors that affect
function (such as soil type/productivity, rainfall, erodibility) have remained relatively
constant and need not be components in this part of the conceptual model because they
are not components of degradation. Landscape alteration can be the result of
degradation of upland or wetland basin habitat, including the complete elimination of
wetlands. In the PPR, both are largely the result of agricultural activity. We will use the
ratio of cropland to upland area as the indicator for both types of degradation. Data will be
obtained from the NRI database.
3.4 Value to Humans of the Water Quality Improvement Function
The conceptual model for this criterion is presented in Figure 6; indicator values will be
tabulated in Table 6. The structure of this conceptual model is similar to that of the model
for value to humans of habitat support. Value is considered to be a function of current
value and the potential marginal benefit with restoration. Both of these are functions of
the attenuation of potential pollution to streams, the population of users, and the cost of
substitutes such as chemical treatment of water.
3.4.1 Attenuation of Potential Pollution to Streams
Lessening pollution to streams affects the current value in a positive direction and the
potential marginal benefit with restoration in a negative direction. We presume that
pollution attenuation is influenced mainly by the potential input of pollution to streams per
unit time and wetland capacity to improve water quality per unit time.
37
-------
GO
00
Figure 6. Conceptual model for value to humans of water quality improvement function of wetlands.
-------
Table 6. Table format for entering indicator values for the value to humans of water quality Improvement.
Criterion: Value to Humans of Water Quality Improvement
Conceptual
Quantities
Population of
Users
Cost of
Substitutes
(eg-.
Ireatment)
Attenuation of Potential Pollution to Streams
Potential Input of Pollutants to Streams
Wetland Capacity to
Improve Water Qualitv
Chemical Application Transported
Chemical
Application
Assimilation Capacity
of Wetlands
| Sediment Transport
Upland
Assimilation
Water Input
Indicators
Population In
Focal and
Downstream
Subunlts
(no Indicator
available)
Precipitation
Erodlblllty
Slope
Crop:
Upland
Ratio
Soil
Permea-
bility
Crop:
Upland
Ratio
Wetland Area
Data
Source
US Census
USGS
PRISM
NRI
NRI
NRI
NRI
NRI
Subunit:
" i-..' <' - jĢ:*
V'"
1
2
3
-------
September 1996
NHEERL-WED. Corvallis, OR
3.4.1.1 Potential Input of Pollution to Streams
Potential input can be described by the amount of chemicals applied and the
proportion of those chemicals that are transported to wetlands. The attenuation of
potential pollution conceptual quantity assumes that input exceeds storage capacity and
the attenuation should be input minus capacity. However, if storage capacity exceeds
input, then the percent attenuation will always be 100%.
3.4.1.1.1 Chemical Application
Chemical application varies depending on the crop, the rotation of crops within a year
and from year to year, soil type and nutrient content, climate, and other factors.
Calculating chemical application is very complex, and the level of accuracy implied by the
final result might be misleading if we chose a simple algorithm. For simplicity we assumed
that, on the average, chemical application is proportional to agricultural intensity, which we
will represent, as previously, as the ratio of cropland to upland area (data from NRI).
3.4.1.1.2 Chemical Application Transported
The amount of chemicals transported to wetlands is determined by water input,
sediment transport, and the amount of assimilation that occurs in the upland. We
assumed that most of the chemical delivery to wetlands occurs primarily through overland
water transport. Chemicals can be dissolved in water or adsorbed to sediments. Water
input incorporates rainfall amount and intensity per storm event and frequency of storm
events. Intense rainfall events are more likely to produce more surface runoff. Surface
runoff is also intensified with events of longer duration and greater frequency. We
acknowledge that all of these elements are important in influencing surface water runoff,
but development of a model to estimate surface runoff was beyond the scope of the
assessment. To simplify, we will use the average annual precipitation as an indicator to
quantify water input. Data will be obtained from the PRISM precipitation model.
Sediment transport is influenced in part by water input. The volume and intensity of
rainfall and the energy of overland flow influences the extent to which soil particles are
dislodged and transported. Sediment transport also incorporates soil erodibility, slope,
and land use. Transport is more likely to occur when conditions in the upland favor
overland flow over water retention. These conditions include high erodibility, steeper
slopes, and less ground cover. Cover with a low stem density is less able to attenuate the
water flow rate, resulting in less time for water infiltration in the upland. We assumed that,
on the average, uncropped land is likely to be pasture or preserves characterized by
denser cover such as grasses; cropped land typically consists of a larger percentage of
row crops, which are characterized by lower stem density. Indicators we will use for
sediment transport are erodibility (from the NRI data base), which incorporates soil type
and slope, and the ratio of cropland to upland area (NRI data).
Upland assimilation of chemicals is enhanced by dense cover (as described in the
40
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
previous paragraph) and high soil permeability. We will incorporate these factors using
NRI data.
3.4.1.2 Wetland Capacity to Improve Water Quality
Wetland capacity to improve water quality per unit time, the second component of the
attenuation of potential pollution to streams, is represented conceptually in Table 6 as the
assimilation capacity of wetlands. Assimilation by wetlands further reduces the
downstream transport of chemicals and helps ensure cleaner drinking water at the point
where it is extracted from a river for human use. Soil microbial decomposers assimilate,
break down, or transform agricultural chemicals. Chemicals that are adsorbed onto
sediments are removed from runoff when sediments are deposited in wetlands. Wetlands
therefore often act as chemical sinks. Because most chemical processes occur at the
sediment-water interface, a greater substrate area results in more processing. Similarly, a
greater catchment area in a subunit increases the chances of trapping and deposition of
chemical-laden sediments. Thus, the indicator we selected to represent assimilation
capacity is wetland area, which will be obtained from the NRI data base.
3.4.2 Population of Users and Cost of Substitutes
On the right side of the value model, the population of users will be estimated by the
same indicator as for the habitat value model (Section 3.2), the population in focal and
downstream subunits. Data will be obtained from the U.S. Census. The cost of
substitutes, such as intensifying artificial water treatment, also influences current and
potential increase in value, but it remains a conceptual quantity in this model because we
know of no uniformly available existing data that could be used to estimate it. It will thus
have no influence in the assessment.
3.5 Replacement Potential for the Water Quality Improvement Function
The conceptual model for this criterion is presented in Figure 7; indicator values will be
tabulated in Table 7. As with the conceptual model for replacement potential for habitat
support (Section 3.3), this model is a function of the wetland quantity available for
replacement and the likelihood of full functional replacement.
3.5.1 Wetland Quantity Available for Replacement
The left side of the model, which shows the process for quantifying the wetland
quantity available for replacement, is identical to the model for replacement potential for
habitat support (Section 3.3).
3.5.2 Likelihood of Full Functional Replacement
On the right side of the model, the likelihood of full functional replacement is a negative
41
-------
KEY
( J Conceptual Quantly
Indicator (Measured Quanlly)
Wmt
_ NegatNre Association
Posit Kre Association
Replacement Potential for
Water Quality Improvement
Wetland Quantity
Available for Replacement
/
Physical
Availability
Current
Wetland
Availability
T
Area of
Wetland
NRI
i
Historic
Wetland
Availability
I
Area of Hydrlc
Soil
NRI
\
Socio-economic
Availability
Extent of
Property
Private
Value
Land
Ownership
U.S. Census
NRI
ofAg.
Likelihood of Full Functional Replacement
c
Increased Input of Pollution to Streams
t
\ Percent of Chemical Application]
Transported
A
*
Sediment j
^ Upland ^
Transport J
Assimilation
jonj j Chemical ]
I Application!
Crop: Upland
Ratio NRI
Agricultural lntensity_J
Figure 7. Conceptual model for replacement potential for water quality improvement function of wetlands.
-------
Table 7. Table format for entering indicator values for the replacement potential for water quality improvement.
Criterion:
Replacement Potential for Water Quality Improvement
Conceptual
Wetland Quantity Avai
able for Replacement
Likelihood of Full Functional Replacement
Quantities
Physical Availability
Socio-economic Availability
Increased Input of Pollution to Streams
Current
Wetland
Historic
Wetland
Percent of Chemical
Application Transported
Chemical
Application
Availability
Availability
Sediment
Transport
Upland
Assimilation
Agricultural Intensity
Indicators
Area of
Wetland
Area of Hydric
Soil
Extent of
Private Land
Ownership
Property Value
Crop:Upland Ratio
Data Source
NRI
NRI
NRI
US Census of
Agriculture
NRI
Subunil:
I
II
888
Sgjg
{
sM
hmmm
BiiliiMM
1
2
3
-------
September 1996
NHEERL-WED, Corvallis, OR
function of the increased input of pollution to streams per unit time. The input of pollution
to streams is an indirect measure of the chemicals applied that are not assimilated in the
uplands or in wetlands. We assumed that a greater input indicates a more altered system,
and that the likelihood of fully replacing the lost function would be lower. We represented
the increased input of pollutants to streams as a function of the amount of chemicals
applied and the amount of those chemicals that are transported to streams. The amount
of chemicals applied that are transported is described by sediment transport and upland
assimilation. Application, assimilation and transport are influenced by agricultural
intensity, as described previously, and we will use the ratio of cropland to upland as the
indicator. The influence of agricultural intensity on sediment transport and chemical
application is positive, while the influence on upland assimilation is negative.
3.6 Value to Humans of the Flood Attenuation Function
The conceptual model for this criterion is shown in Figure 8; indicator values will be
tabulated in Table 8. The flood attenuation function presented some conceptual
difficulties not inherent in the functions already described; in particular, the difficulty of
assigning ranks when the subunits that contribute most to flooding are not necessarily the
subunits that receive the most flood damage. We thus present our current conceptual
models for the flood attenuation function, but we are continuing to develop and refine the
models as we acquire new information for hydrologically linking the subunits receiving the
greatest flood impact with the subunits that contribute most to those impacts.
The basic structure of this value model is similar to the other value models described
above. We separated value into current value and potential marginal benefit with
restoration. Each of these is described by the attenuation of potential downstream
flooding (the actual function), population of users, and cost of substitutes (e.g., dams,
levees). All concepts have positive relations with value except that the attenuation of
potential downstream flooding has a negative relation to potential marginal benefit with
restoration. We presumed that the latter is true because, as explained above, people
value commodities more when they are scarce, and marginal value increases with
scarcity. Thus, people in subunits with a higher percentage runoff attenuation are likely to
value an increase in the function less than would people in subunits with a low percentage
attenuation.
3.6.1 Attenuation by Wetlands of Potential Downstream Flooding
In the model, we describe attenuation of potential downstream flooding as a function of
the potential runoff volume to streams per unit time and wetland capacity to attenuate
runoff per unit time.
44
-------
Figure 8. Conceptual model for value to humans of flood attenuation function of wetlands.
-------
Table 8. Table format for entering Indicator values for the value to humans of flood attentuation.
-P*
CT>
Criterion: Value to Humans of Flood Attenuation
Conceptual
Quantities
Cost of
Substitutes
Population of
Attenuation by Wetlands of Potential Downstream Flooding
Users
Potential Runoff Volume to Streams
Wetland Capacity to Attentuate Runoff
Drainage
Area
Water Input
to
Landscape
Channelized
Runoff
Delivery
Upland
Permeability
Ability to
Desynchronize
Flow Volume
Ability to Reduce Flow
Volume
Storage
Capacity of
Wetlands
Infiltration;
Groundwater
Recharge
Evapo-
transpir-
ation
Rate
Indicators
(no
Indicator
available)
Population
In Focal and
Downstream
Subunlts
Area of
Upstream
Drainage
Precipitation
Relative
Extent of
Drainage
Ditching
Crop:Upland
Ratio
Wetland Area
Slope
Soil
Permeability
Grazing Intensity
Data Source
USGS
US Census
USGS
PRISM
NRI
NRI
US Census of
Aqri culture
NRI
Subunit:
'U \ )kt > jŦ- v.-;.;'''' _
iij'ljI.Y:::
Ay..y.'.v.'ss.y*. ,s.-.
1
2
? -
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
3.6.1.1 Potential Runoff Volume to Streams
An estimate of the potential runoff volume to streams can be obtained by considering
the components of the hydrologic cycle. Estimating these components is very complex
and usually requires assumptions about the watershed and considerations of the use of
the information. The components include precipitation, evaporation, evapotranspiration,
infiltration, overland flow, streamflow, and groundwater flow (Bedient and Huber 1988).
To simplify, we assumed that evaporation and evapotranspiration for the period of input
are uniform over the landscape and did not include them in the model. Lateral
groundwater flow is relatively slow and uniform overtime and thus has a smaller impact
on the timing and magnitude of peak stream flows than does overland flow and we thus
omitted it from consideration in the model. We were more concerned with the amount of
runoff delivered to streams and less concerned with actual streamflow, although we did
consider the flow within artificial drainage ditches which often connect drained
depressions. In addition to the basic hydrologic cycle components, soil moisture affects
the amount of precipitation that is able to infiltrate. Because soil moisture is variable and
unpredictable, we will not attempt to estimate it with an indicator, although it might be an
important determinant of runoff volume.
We thus will estimate potential runoff volume as a function of drainage area, water
input, channelized runoff delivery, and upland permeability. The drainage area in
upstream subunits will serve as an indicator of the potential runoff relative to the subunit's
position in the drainage system. Water input to the landscape will be estimated using the
PRISM precipitation model. The extent of drainage ditching (NRI data) will serve as an
indicator of channelized runoff delivery. Upland permeability is influenced by the soil type,
slope, cover, and livestock compaction. It will be derived by combining four indicators:
ratio of cropland to upland area (from the NRI data base), average slope (NRI data), soil
permeability (NRI data), and grazing intensity (from the Census of Agriculture).
3.6.1.2 Wetland Capacity to Attenuate Runoff
We describe wetland capacity to attenuate runoff per unit time as a function of the
ability to desynchronize flow volume and the ability to reduce flow volume.
Desynchronization is achieved through storage and gradual release of precipitation or
snow melt to streams and has the effect of widening and flattening the storm hydrograph
(Thomas and Benson 1970). The storage volume of wetlands in a subunit would be the
ideal measurement for storage capacity, but volume changes greatly from year to year
depending on climatic conditions, and detailed data are unavailable. We therefore
selected wetland area as the indicator of the storage capacity of wetlands with the
assumption that volume is generally proportional to area.
The ability of wetlands to reduce flow volume during the period of storage is influenced
by the evapotranspiration rate and the rate of infiltration and groundwater recharge.
Methods of measuring evapotranspiration are weak and subject to strong seasonal and
47
-------
September 1996
NHEERL-WED. Corvallis, OR
annual variation because of the unpredictability and variability of climatic factors and plant
populations in wetlands. The infiltration and evaporation rates, however, are related to the
area of wetland over which these processes can occur (i.e., the area of water in contact
with the substrate and with the air). We therefore link the wetland area indicator with
these conceptual quantities as well.
3.6.2 Population of Users and Cost of Substitutes
The right hand side of the conceptual model, comprising population of users and cost
of substitutes, is the same as for the model of the value to humans of water quality
improvement function (see Section 3.4.2).
3.7 Replacement Potential for the Flood Attenuation Function
The conceptual model for this criterion is shown in Figure 9; indicator values will be
tabulated in Table 9. As with the model of the value to humans of the flood attenuation
function (Section 3.6), we are still acquiring information for developing and refining the
attributes and indicators for this model; therefore, the model presented in this section
represents our preliminary ideas about flood control in the PPR.
The model is structured similarly to the other replacement potential models. The two
major contributors to replacement potential that we consider are wetland quantity available
for replacement, on the left side of Figure 9, and the likelihood of full functional
replacement, on the right.
The wetland quantity available for replacement, on the left side of the model, is
identical to the replacement potential models presented in Sections 3.3 and 3.5. We will
estimate the likelihood of full functional replacement (on the right side) by assuming that
physical and climatic factors have remained constant through time and that the extent of
human stresses that impact wetland function will have the greatest influence on whether
the function can be replaced to its historic levels. In other words, despite the replacement
of wetland acreage, the upland landscape might have been altered to the extent that it
cannot attenuate runoff sufficiently, resulting in overload to wetlands and a lower overall
flood attenuation capacity for the subunit. A decreased upland infiltration rate or
assimilation capacity would presumably result in a lower likelihood of full functional
replacement. In the model, we describe upland assimilation and infiltration as the extent
to which precipitation infiltrates or runoff is slowed (or desynchronized) before reaching
wetlands. Human impacts that influence these processes are primarily the result of
agricultural activities, namely cropping and grazing. Flooding-related impacts due to
cropping, as described previously, include reduced infiltration and increased overland flow
rate. Infiltration is also affected by soil compaction caused by grazing. Our selected
indicators for these processes are thus the ratio of cropped upland to non-cropped upland
(obtained from the NRI data base) and grazing intensity, which we will obtain from the
U.S. Census of Agriculture.
48
-------
KEY
CJ Conceptual Quantly
Wtt Indbator (Measured Quanlly)
f. Negative Association
f Positive Association
Replacement Potential for
Flood Attenuation
Wetland Quantity
Available for Replacement
L
X
Physical
Availability
Socio-economic
Availability
Current
Wetland
Availability
i
Historic
Wetland
Availability J
Area of
Wetland
NRI
Extent of
Private
Land
Ownership
NRI
Property
Value
U.S. Census
of Ag.
Likelihood of Full Functional Replacement
Decreased
Upland Infiltration
Rate;
Increased Runoffy*
Son ^^JSoil Compaction^
1
Grazing
Intensity
U.S. Census of Ag.
Figure 9. Conceptual model for replacement potential for flood attenuation function of wetlands.
-------
Table 9. Table format for entering indicator values for the replacement potential for flood attentuation.
Criterion: Replacement Potential for Flood Attenuation
Conceptual
Quantities
Wetland Quantity Available for Replacement
Likelihood of Full Functional Replacement
Physical Availability
Socio-economic Availability
Decreased Upland Infiltration Rate; Increased Runoff
Current
Wetland
Availability
Historic
Wetland
Availability
Soil Compaction
Crop:Upland Ratio
Indicators
Area of
Wetland
Area of
Hydric Soil
Extent of
Private
Land
Ownership
Property Value
Grazing Intensity
Data
Source
NRI
NRI
NRI
U S Census of
Agriculture
US Census of
Aqriculture
NRI
Subunit:
1
2
3
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Part 4: Deriving Subunit Priority Rankings
After indicator values are entered into tables, several methods may be used to
combine values into indices and to assign ranks to the indicators or indices. Indices may
or may not be combined, depending on the predilections of the user. As a general rule,
we will assign ranks by standard quintiles unless an alternative five group ordering is
suggested by other information about the actual indicator values or their frequency
distributions. We chose to rank into five groups because it is easier to read and interpret
maps with a limited number of mapping groups. In addition, the uncertainty of the
accuracy of many synoptic data does not give us the confidence to distinguish between
subunits divided into a larger number of ranking groups. After ranks are assigned, a
variety of weighting schemes may be applied, based on beliefs about the relative
importance of indices or indicators. Maps of the five ordered groupings or standard
quintiles of risk, value, and replacement potential or their component parts can be derived
from the ranking tables.
After indicator values are tabulated in the tables in Section 3, we will develop and
experiment with different methods of combining data. Ranking will require exploratory
work, requiring manipulations with the actual indicator data. We therefore will not propose
a rigid data analysis, but instead will show how a user can combine indicator data while
also preserving the original values so that the analysis is flexible and can be tailored to
address specific management goals. Figure 10 shows a simple flow diagram of one
possible method for combining data. The flow diagram consists of a series of questions
and recommended actions. In this section, we use the flow diagram in Figure 10 as a
guide to present an example for combining indicator values, using the South Dakota data
presented in Table 2 for the replacement potential for habitat support. This example
represents only one approach for assessing restoration priority in a region. Others will be
developed as the synoptic assessment proceeds.
4.1 Example
In step 1 of Figure 10, we asked the following:
Can or should the indicator values in Table 2 be re-expressed (e.g., standardized by
the maximum value across all subunits or the area of the subunit)?
Should the indicators be combined into indices of conceptual quantities? In particular:
a) Can we represent a functional relationship between two or more indicators
with an index?
b) Is the re-expression or combination of indicators legitimate mathematically?
c) Would the new index or indicator be useful to managers?
51
-------
START
Đ
Figure 10. Method for producing final ranking tables for synoptic assessment.
52
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Table 2 contains data from PPR subunits (HUCS) that lie completely or partially within
South Dakota. It consists of five indicators that we have proposed for evaluating
replacement potential for habitat support of wetlands. One of this function's conceptual
quantities of interest to managers is the quantity of wetland available for replacement. We
subdivided this into physical and socio-economic availability. Physical availability was
assumed to be the difference between historic wetland availability and current wetland
availability. Using NRI data, we estimated these quantities with the indicators hydric soil
area and current wetland area, respectively. Although the NRI variables are derived from
sample data, we felt that we could get a reliable estimate of the subunits relative to each
other by producing an index equal to the difference between the two indicators.
Physical Availability = S°" ' Area <* WMm<*
Subunit Size
Similarly, we assumed that socio-economic availability can be measured using the
percent of the subunit that is privately owned and the property value per unit area of
farms. For example, in Table 2, the private land ownership indicator value was derived
from NRI sample data. The property value indicator was taken from the Census of
Agriculture. We assumed that land that is more expensive is more difficult to acquire, and
restoration on private land requires the approval of the. landowner. Both indicators are
therefore negatively related to socio-availability. We could accommodate'this by
producing an index and ranking it in reverse order (highest ranks given to smallest
values). In this case, however, we made a simple mathematical correction by forming an
index for socio-economic availability. First, we standardized property value by dividing by
the maximum value across all subunits; then we subtracted from 100% the product of
percent private land ownership and standardized property value.
Socio-economic Availability = 100 -
Property Value
Property Value
x Percent Private Property
At this point, we considered combining physical availability and socio-economic availability
into a single availability index. However, there may be a significant difference in the
accuracy of the individual indices, based on the quality of the indicator data. Area of
hydric soil is the best indicator that we have for historic area of wetlands, but we have no
measure of its reliability. In fact, in some subunits, because of the discrepancy between
data sets used, the area of wetlands is greater than the area of hydric soil. Therefore,
physical availability is probably a much less accurate index than socio-economic
53
-------
September 1996
NHEERL-WED, Corvallis, OR
availability. Although mathematically legitimate, the combining of the two indices might
result in the loss of information which could be useful to managers.
Because the answers to the Step 1 questions were "yes," we proceeded to Step 2. In
Step 2, we used the above procedure to calculate index values for physical and socio-
economic availability (Table 10). We then proceeded to Step 3.
In Step 3 (Figure 10), we asked the following:
Is any threshold information about the indicator or index available?
Are the actual numbers qualitatively significant (e.g., good vs. bad or high vs. low with
respect to an ecological function) or can they be compared to specific standards?
We were unaware of any threshold information (e.g., reported minimum or maximum
values) ranges of values, or standard values for the ratio of cropland to upland area or
physical availability that are known to be associated with successful wetland replacement.
The answers to the Step 3 questions were therefore "no," so we proceeded to Step 5 (see
Figure 10).
In Step 5, we asked the following:
After examining frequency or cumulative frequency distributions of the actual indicator
data, are there are any "breaks" or clumping of observations that might reflect
important differences between groups of subunits and suggest rank assignments?
For any reason, does the manager desire to assign specific numbers of subunits in the
region to each ranking category?
We examined frequency distributions of the values for the physical availability and
socio-economic availability indices, and for the cropland to upland ratio indicator. For the
indices, the distribution of values appears to be approximately normally distributed, so we
decided to rank both by standard quintiles. The highest ranks were assigned to the
subunits with the highest indicator values, i.e., to the subunits with the highest levels of
physical and socio-economic availability. We assumed that the crop to upland ratio
indicator is negatively related to the likelihood of full functional replacement. We thus
assigned these ranks by quintiles, giving the highest ranks to the lowest indicator values,
i.e., to those subunits with the least degraded upland and wetland habitat.
The answer to the Step 5 question was 'no', so we proceeded to Step 7 (see Figure
10). In Step 7, we assigned standard quintiles to the subunit ranks (Table 11). We then
proceeded to Step 8 (see Figure 10).
In Step 8, we asked: Can ranks be meaningfully and usefully combined to help
managers prioritize subunits? Although in some circumstances a manager might find it
useful to consider the individual ranks without combining them, we assumed that in most
cases combined ranks and overlay maps would be meaningful and useful to resource
managers for decision-making. Because the answer to the Step 8 question was 'yes', we
proceeded to Step 9 (see Figure 10).
In Step 9, we combined ranks and produced tables and overlay maps. We then
assigned the combined ranks new rankings in groups of five (quintiles). For illustration we
54
-------
Table 10. Example of a completed table of index values for the replacement potential for the habitat support function. Values for physical and socio-
economic availability were derived from indicator data presented in Table 2 for the 25 HUCS in South Dakota.
in
CJ1
Criterion: Replacement Potential for Habitat Support: Bird Abundance and Diversity
Conceptual
Quantities
Wetland Quantity Available for Replacement
Likelihood of Full Functional Replacement
Degradation of Upland
Habitat Quality
Degradation of Wetland
Habitat Quality:
Pothole Component
Indicators
and
Indices
Physical Availability Index
(square kilometers)
Socio-economic Availability Index
(percent)
Cropland:Upland Ratio
(ratio)
Subunit:
7020001
3.5
57
.62
7020003
1.8
80
.69
9020101
1.8
79
.89
9020105
1.2
94
.77
10130102
-0.8
78
.49
10130105
-3.2
54
.65
10130106
-0.7
71
.45
10140101
-3.5
59
.54
10140103
-0.1
63
.55
10140105
-0.1
64
.36
10160003
3.6
80
.69
10160004
1.3
84
.55
10160005
0.7
46
.84
10160006
1.2
49
.57
10160007
0.0
63
.39
-------
Table 10, cont. Example of a completed table of index values for the replacement potential for the habitat support function. Values for physical and
socio-economic availability were derived from indicator data presented in Table 2 for the 25 HUCS in South Dakota.
Indicators
and
Indices
Physical Availability Index
(square kilometers)
Socio-economic Availability Index
(percent)
Cropland:Upland Ratio
(ratio)
Subunit:
Ķ .< i#:
- '
. VĶ --.ft ./"J- .jf-
10160008
0.6
59
.48
10160009
0.1
62
.49
10160010
-2.3
52
.51
10160011
4.1
34
.61
10170101
-0.3
40
.67
10170102
3.2
3
.77
10170103
-0.6
40
.78
10170201
0.2
40
.73
10170202
1.7
25
.65
10170203
3.6
4
.76
-------
Table 11. Example of a completed ranking table for the components of replacement potential for the habitat support function. Ranks were derived from
values in Table 10 for the 25 HUCS in South Dakota.
Criterion:
Replacement Potential for Habitat Support: Bird Abundance and Diversity
Conceptual
Wetland Quantity Available for Replacement
Likelihood of Full Functional Replacement
Quantities
Degradation of Upland
Habitat Quality
Degradation of Wetland
Habitat Quality:
Pothole Component
Indicators
and
Indices
Physical Availability
(rank)
Socio-economic Availability
^rank)
Cropland:Upland Ratio
(rank)
Subunit:
7020001
1
3
3
7020003
2
1
4
9020101
2
1
5
9020105
3
1
5
10130102
5
2
1
10130105
5
3
3
10130106
5
2
1
10140101
5
3
2
10140103
4
2
2
10140105
4
2
1
10160003
1
1
4
10160004
2
1
3
10160005
3
4
5
10160006
3
4
3
10160007
4
3
1
-------
Table 11, cont. Example of a completed ranking table for the components of replacement potential for the habitat support function. Ranks were derived
from values in Table 10 for the 25 HUCS in South Dakota.
Indicators
Physical Availability
Socio-economic Availability
Cropland:Upland Ratio
and
Indices
(rank)
(rank)
(rank)
Subunit:
M. " : ..*Ķ :"
10160008
3
3
1
10160009
3
2
2
10160010
5
4
2
10160011
1
5
3
10170101
4
5
4
10170102
1
5
5
10170103
4
4
5
10170201
3
4
4
10170202
2
5
3
10170203
1
5
4
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
simply assigned equal weight to physical availability, socio-economic availability, and
likelihood of full functional replacement. A manager more familiar with a specific district or
region might be better able to assign appropriate weightings and combine ranks to suit his
or her management objectives. We produced the final new ranks by simply summing the
individual ranks for each subunit. We then re-ranked the sums by assigning standard
quintile ranks. Table 12 contains the ranks of the indices, the sum of their ranks, and the
final new ranks. Figures 11 through 14 are maps of the quintile ranks for physical
availability, socio-economic availability, the likelihood of full functional replacement (as
evaluated with the cropland to upland ratio), and the final new ranks for replacement
potential, respectively. Maps are produced as a visual aid for looking at spatial patterns in
the ranks and to assist in selecting priority subunits.
The final new ranks in Table 12 are priority rankings for restoring the habitat support
function in South Dakota. A manager could use these ranks to help decide where
restoration funds and efforts should be directed. According to the results in this example,
seven subunits should receive the highest priority for wetland-habitat restoration projects,
while three subunits should receive the lowest priority. The remaining subunits fall in
between high and low priority and could be considered in decision-making, depending
upon other management constraints or objectives.
Alternatively, a manager could use the first three columns of Table 12 (the individual
indicator ranks) concurrently with or in place of the final ranks, depending upon specific
conditions in his/her jurisdiction. For example, a manager might surmise that socio-
economic availability and cropland to upland ratio do not have an impact on replacement
potential in his/her jurisdictional area and therefore might prioritize restoration efforts using
only the ranks for physical availability. Alternatively, a wildlife manager might see no
reason to allocate money for restoring habitat unless the wetland would be in a subunit
with a high percentage of upland to provide protective cover for nesting birds. He/she
might therefore wish to emphasize the cropland to upland ratio ranks to prioritize subunits
for restoration.
59
-------
September 1996
NHEERL-WED, Corvallis, OR
Table 12. Example summary table of ranks for the replacement potential for the habitat
support function, including the final rank for each of 25 HUCs in South Dakota,
Final ranks are 1-5, with 1 being high restoration priority for replacing habitat
function and 5 being low priority.
Cropland:
Subunit
Physical
Socio-Economic
Upland
Sum of Ranks
Final New Ranks
(HUC)
Availability:
Availability:
Ratio:
Rank
Rank
Rank
7020001
1
3
3
7
1
7020003
2
1
4
7
1
9020101
2
1
5
8
2
9020105
3
1
5
9
3
10130102
5
2
1
8
2
10130105
5
3
3
11
4
10130106
5
2
1
8
2
10140101
5
3
2
10
4
10140103
4
2
2
8
2
10140105
4
2
1
7
1
10160003
1
1
4
6
1
10160004
2
1
2
5
1
10160005
3
4
5
12
5
10160006
3
4
3
10
4
10160007
4
3
1
8
2
10160008
3
3
1
7
1
10160009
3
2
2
7
1
10160010
5
4
3
11
4
10160011
1
5
3
9
3
10170101
4
5
4
13
5
10170102
1
5
5
11
4
10170103
4
4
5
13
5
10170201
3
4
4
11
4
10170202
2
5
3
10
4
10170203
1
5
4
10
4
60
-------
MT
WY
ND
SOUTH
DAKOTA
i
Physical Availability (Rank)
Hi Medium-High
EES Medium
YZZA Med i u m -Lo w
I IL o w
MN
IA
Fiqure 11. Mapped index ranks of physical availability, a component °f r1ŪP,ice!nie.n.t
potential for the habitat support function in PPR subunits (HUCs) within
South Dakota. Physical availability is high in dark-shaded subunits and
low in light-shaded subunits.
-------
cr>
ro
ND
MT
WY
SOUTH
DAKOTA
1 -
Socio-economic Availability (Rank)
igh
SiSMedium-High
EEB3 Medium
EZE2 M ed i u m -Low
~ L o w
MN
IA
Figure 12. Mapped index ranks of socio-economic availability, a component of replacement
potential for the habitat support function in PPR subunits (HUCs) within South
Dakota. Socio-economic availability is high in dark-shaded subunits and low in
light-shaded subunits.
-------
cr>
OJ
MT
WY
ND
SOUTH
DAKOTA
Crop : Upland Area Ratio (Rank)
igh
i urn-High
EfflMedium
[z23Med i um-Low
~~Low
MN
IA
Figure 13. Mapped ranks for the crop to upland area ratio, an indicator for the likelihood
of full functional replacement, a component of the replacement potential for
the habitat support function, in PPR subunits (HUCs) within South Dakota. The
likelihood of full functional replacement is high in dark-shaded subunits and
low in light-shaded subunits.
-------
CT>
MT
WY
SOUTH
DAKOTA
i
i .
Replacement Potential for
Habitat Support (Rank)
Medium-High
S M e d i u m
Ģ22 Me d i u m Lo w
mn L o w
MN
IA
Figure 14.
Combined ranks for replacement potential for the habitat support function
PPR subunits (HUCs) within South Dakota. Replacement potential is high in
dark-shaded subunits and low in light-shaded subunits.
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Part 5: Summary
This document presents conceptual and ecological background and a procedure for
deriving ranks, data tables, and maps for prioritizing subunits of the PPR for wetland
restoration. Data tables, ranks, and maps will be derived for all assessment criteria as
was done in the example in Part 4. The procedure provides the information necessary for
decision-making about prioritizing restoration. The synoptic assessment is designed to
preserve information so that it can be used flexibly to address both regional and local
restoration issues. We anticipate that managers will use specific data in the tables that
are relevant to their particular management objectives. Interpretation and use of the
results by managers will influence the ways in which subunits are prioritized for
restoration; manager choices and decisions, therefore, play a role in the final stages of the
synoptic assessment.
As described in the introduction, the primary objective of the PPR synoptic assessment
is to develop and improve the synoptic methodology to produce a more effective product.
The resulting use by managers is a secondary objective at this time, although we
anticipate that continued development and improvement of the product will make it a more
useful and practical tool for managers. We are concurrently developing new approaches
and techniques for improving the accuracy and usefulness of the synoptic assessment.
For example, we are investigating methods for evaluating assessment quality and looking
at alternative ways of combining data.
We are developing a rigorous framework and set of methods to appraise the quality of
ecological assessments where indicators are only qualitatively linked to an unmeasured
endpoint. The framework is based on Structural Equation Modeling (SEM) (Bollen 1989,
Johnson et al. 1991), a graphical and statistical methodology that combines path analysis
with measurement models. Potential indicators are placed in the context of a graphical
model, and several measures of their relative quality are calculated based on the structure
of the graphical model and BPJ estimates of SEM parameters. The framework allows a
range of information, both qualitative and quantitative, to be used. It can be entirely BPJ-
based when little information is available, but also provides a statistical approach when
validation data are available. With these methods we can compare the quality of
alternative assessments, evaluate which data and how much effort are needed to produce
an assessment of given quality, and ultimately select and refine a cost-effective qualitative
assessment.
We are also exploring methods of testing assumptions made in the conceptual
modeling stage by employing SEM, by looking for relationships in empirical data collected
in the field, or by examining variability of results of more intensive analyses conducted
with existing data from a random subset of subunits.
We anticipate that these refinements to the synoptic approach, as well as the
conceptual modeling procedure presented in this document, will render a more useful and
65
-------
September 1996
NHEERL-WED, Corvailis, OR
flexible tool for resource managers and other individuals or agencies involved in the
preservation and restoration of wetland functions and for prioritizing wetland restorations.
Data analysis is currently being performed by researchers at NHEERL-WED. Preliminary
results, including data tables, ranks, maps, and documentation for all assessment criteria
are expected to be completed by December 1996. A final report with technical
documentation is expected by August 1997.
66
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Literature Cited
Abbruzzese, B., B. Peniston, arid S. Leibowitz. 1994. Evaluation of the landscape
function of wetlands in Illinois, with recommendations. EPA/600/R-94/200. U.S.
Environmental Protection Agency, Environmental Research Laboratory, Corvallis, OR.
Adamus, P.R. 1992. A process for regional assessment of wetland risk. EPA/600/R-
92/249. U.S. EPA-Environmental Research Laboratory, Corvallis, OR.
Batt, B.D. J., M.G. Anderson, C. D. Anderson, and F.D. Caswell. 1989. The use of prairie
potholes by North American ducks. Pages 204-227 IN A.G. van der Valk (Ed.), Northern
Prairie Wetlands. Iowa State University Press, Ames, IA.
Bedient, P.B. and W.C. Huber. 1988. Hydrology and Floodplain Analysis. Addison-
Wesley Publishing Company, Menlo Park, CA.
Bluemle, J.P. 1971. Depth to bedrock in North Dakota. N. Dakota Geologic Survey Misc.
Map 13.
Bollen, K.A. 1989. Structural Equations with Latent Variables. John Wiley and Sons, New
York, NY.
Brady, E.N. and B.A. Giron Pendleton. 1983. Aquatic bird use of wetlands in Brookings
County, South Dakota. Proceedings of South Dakota Academy of Science 62:148-153.
Brown, M., and J.J. Dinsmore. 1986. Implications of marsh size and isolation for marsh
bird management. Journal of Wildlife management 50(3):392-97.
Brown, M. and J.J. Dinsmore. 1991. Area-dependent changes in bird densities in Iowa
marshes. Journal of the Iowa Academy of Science 98(3):124-126.
Daly, C., R. P. Neilson, and D. L. Phillips. 1994. A statistical-topographic model for
mapping climatological precipitation over mountainous terrain. Journal of Applied
Meteorology 33:140-158.
Freeman, A.M. 1979. The Benefits of Environmental Improvement. Resources for the
Future, John Hopkins University Press, Baltimore, MD.
Hemesath, L.M. and J.J. Dinsmore. 1993. Factors affecting bird colonization of restored
wetlands. The Prairie Naturalist 25(1):1-11.
67
-------
September 1996
NHEERL-WED, Corvallis, OR
Johnson, M.L., D.G. Huggins, and F. DeNoyelles, Jr. 1991. Ecosystem modeling with
LISREL: a new approach for measuring direct and indirect effects. Ecological Applications
1:383-398.
Kantrud, H.A. 1981. Grazing intensity effects on the breeding avifauna of North Dakota
native grasslands. Canadian Field Naturalist 95:404-417.
Kantrud, H.A., G. L. Krapu, and G.A. Swanson. 1989. Prairie Basin Wetlands of the
Dakotas: A Community Profile. Biol. Rep. 85(7.28). U.S. Fish & Wildl. Serv., Washington,
D.C. 111 pp.
Kantrud, H.A. and R.E. Stewart. 1984. Ecological distribution and crude density of
breeding birds on prairie wetlands. J. Wildl. Manage. 48:426-437.
Krapu, G.L. and H.F. Duebbert. 1989. Prairie wetlands: characteristics, importance to
waterfowl, and status. In: R.R. Sharitz and J.W. Gibbons (eds.). Freshwater Wetlands and
Wildlife. CONF-8603101, Symposium Series No. 61. U.S. Dept. Energy, Oak Ridge, TN.
Leibowitz, S.G., B. Abbruzzese, P.R. Adamus, L.E. Hughes, and J.T. Irish. 1992a. A
synoptic approach to cumulative impact assessment: A proposed methodology.
EPA/600/R-92/167, U.S. Environmental Protection Agency, Environmental Research
Laboratory, Corvallis, OR. 129 pp.
Leibowitz, S.G., E.M. Preston, L.Y. Amaut, N.E. Detenback, C.A. Hagley, M.E.
Kentula, R.K. Olson, W.D. Sanville, and R.R. Sumner. 1992b. Wetland Research Plan
FY92-96: An Integrated Risk-Based Approach. EPA/600/R-92/060, U. S. Environmental
Protection Agency, Environmental Research Laboratory, Corvallis, OR.
Mann, G.E. 1974. The Prairie Pothole Region: a zone of environmental opportunity.
Naturalist 25(4):2-7.
Mineau, P. 1987. Pesticides and waterfowl on the Canadian prairies: A pressing need for
research and monitoring. Pages 133-147 IN A.W.D. Diamond and F. Filion (Eds.), The
Value of Birds. ICBP Technical Publication No. 6, International Council for Bird
Preservation.
Office of Technology Assessment (OTA). 1984. Wetlands: Their Use and Regulation.
OTA-0-206. U.S. Congress, OTA, Washington, D.C.
Preston, E.M., and B.L. Bedford. 1988. Evaluating cumulative effects on wetland
68
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
functions: A conceptual overview and generic framework. Environmental Management
12(5): 565-583.
Richardson, J.L., LP. Wilding, and R.B. Daniels. 1992. Recharge and discharge of
groundwater in aquatic conditions illustrated with flownet analysis. Geoderma 53: 65-78.
Richardson, J.L., J.L. Arndt, and J. Freeland. 1994. Wetland soils of the Prairie Potholes.
Advances in Agronomy 52:121-171.
Ruwaldt, J.J., L.D. Flake, and J.M. Gates. 1979. Waterfowl pair use of natural and man-
made wetlands in South Dakota. Journal of Wildlife Management 43(2):375-83.
Scientific Assessment and Strategy Team. 1994. Science for floodplain management into
the 21st century: A blueprint for change part V. Interagency Floodplain management
Review Committee, Washington D.C.
Seaber, P.R., F.P. Kapinos, and G.L. Knapp. 1984. State Hydrologic Unit Maps. Open-
file Report 84-708. U.S. Geological Survey, Reston, VA.
Soil Survey Staff. 1975. Soil Taxonomy. U.S. Department of Agriculture, Agricultural
Handbook 436.
Stewart, R.E. and H.A. Kantrud. 1971. Classification of Natural Ponds and Lakes in the
Glaciated Prairie Region. Publ. 92. U.S. Fish and Wildlife Serv. Washington, D.C.
Stewart, R.E. and H.A. Kantrud. 1973. Ecological distribution of breeding waterfowl
populations in North Dakota. J. Wildl. Manage. 37:39-50.
Swanson, G.A., T.C. Winter, V.A. Adomaitis, and J.W. LaBaugh. 1988. Chemical
Characteristics of Prairie Lakes in South-central North Dakota: Their Potential for
Influencing Use by Fish and Wildlife. Fish and Wildl. Tech. Rep. 18. U.S. Fish & Wildl.
Serv., Washington, D.C.
Thomas, D.M., and M.A. Benson. 1970. Generalization of streamflow characteristics
from drainage basin characteristics. Water-Supply Paper 1975. U.S. Geological Survey,
Reston, VA.
U.S. Bureau of the Census. 1982. Census of population and housing: 1980. U.S.
Government Printing Office, Washington, D.C.
69
-------
September 1996
NHEERL-WED, Corvallis, OR
U.S. Bureau of the Census. 1992. Census of population and housing: 1990. U.S.
Government Printing Office, Washington, D.C.
U.S. Department of Agriculture (USDA). 1992. Natural Resources Inventory. National
Cartography and GIS Center, USDA Soil Conservation Service, Fort Worth, TX.
U.S. Department of Commerce. 1992. 1992 Census of Agriculture. U.S. Department of
Commerce, Economics and Statistics Administration, Bureau of the Census, Washington,
D.C.
U.S. Fish and Wildlife Service (USFWS). 1990. Regional Wetland Concept Plan,
Emergency Wetlands Resources Act: Mountain-Prairie Region. USFWS, Lakewood, CO.
86 pp.
U.S. Fish and Wildlife Service (USFWS) and Canadian Wildlife Service. 1986. North
American Waterfowl Management Plan. U.S. Fish & Wildlife Service, Washington, D.C.
U.S. Geological Survey (USGS). 1970. The National Atlas of the United States of America.
Washington, D.C. 417 p.
U.S. Geological Survey (USGS). 1976. Hydrologic Unit Maps-States of Minnesota, Iowa,
North Dakota, South Dakota, Montana. U.S. Department of Interior, Geological Survey.
Reston, VA.
Weller, M.W. 1981. Freshwater Marshes: Ecology and Wildlife Management. University
of Minnesota Press, Minneapolis, MN.
Winter, T.C. 1988. A conceptual framework for assessing cumulative impacts on the
hydrology of nontidal wetlands. Environmental Management 12:605-620.
Winter, T.C. 1989. Hydrologic studies of wetlands in the northern prairie. Pages 16-54 in
Van der Valk (Ed.), Northern Prairie Wetlands. Iowa State University Press, Ames, IA.
Winter, T.C. 1992. A physiographic and climatic framework for hydrologic studies of
wetlands. Pages 127-148 in R.D. Robarts and M.L. Bothwell (Eds.), Aquatic ecosystems
in semi-arid regions. National Hydrol. Res. Symp. Ser. 7, Environment Canada, Ottawa,
Ontario.
World Wildlife Fund (WWF). 1994. The 1993 Mississippi River Floods: Implications for
Wildlife, Wetlands, and People. World Wildlife Fund, Washington, D.C.
70
-------
Conceptual Framework for a Synoptic Assessment of the Prairie Pothole Region
September 1996
Appendix A
To conduct the synoptic assessment, we will use data obtained from several nationally
maintained databases. Table A-1 contains a listing of these data sources, associated
indicators and their units, and the reporting units of the databases. A significant amount of
data will be derived from the National Resources Inventory (NRI). The NRI database
contains location variables identifying sample sites by various reporting units, including
major land research area (MLRA), county, and USGS hydrologic cataloging unit. We
selected the USGS hydrologic cataloging unit as our PPR synoptic assessment subunit.
Because the reporting unit is also an assessment subunit, it will be not be necessary to
make any calculations to adjust the areal representation of the NRI-based indicators.
Other synoptic indicators will be derived from databases that have a different reporting
unit, e.g., U.S. Census of Agriculture and USGS precipitation records. These data may be
one of two types: aggregate or intensity (Leibowitz et al. 1992a). Aggregate data (e.g.,
county-based U.S. Census of Agriculture) are based on the total number of objects found
in the unit, whereas intensity data (e.g., USGS precipitation records) represent an average
intensity or process rate measured across the entire unit. Both types of data will be
prorated to the hydrologic cataloging units, if necessary, using the following equation:
VALU ESynoptic subunit ^ VALUEreportjng unit
x (AREAjoint /AREA
reporting unit) >
where AREAjoint is the area of overlap between the subunit and the reporting unit.
71
-------
Table A-1. Data sources and indicators used for criteria evaluation.
DATA
SOURCE DESCRIPTION INDICATOR UNITS
NRI
1992 Natural Resources Inventory,
Cropland:upland ratio
(ratio)
Natural Resources Conservation Service
Percent wetland area
(percent)
(U.S. Department of Agriculture 1992)
Seasonal and temporary wetland area
acres
Wetland Area in surrounding subunits
acres
Ditching (relative extent of)
(percent)
Wetland diversity
Extent of private land ownership
acres
Soil binding capacity
Erodibility
Land capability class
(class)
Hydric area
acres
Slope
percent
Soil Impermeability
Census of
1982-1992 Census of Agriculture,
Agricultural growth rate
(percent)
Agriculture
U.S. Department of Commerce
Average size of farmsteads
acres
(U.S. Department of Commerce 1992)
Per acre property values (farms)
dollars
Grazing intensity
head/acre
U.S. Census
1980, 1990 U.S. Census
Population growth rate
humans
(U.S. Bureau of the Census 1982, 1992)
Population in focal and surrounding
subunits
USGS
U.S. Geological Survey,
Area of upstream drainage
acres
Hydrologic Unit Maps
(Daly et al. 1994)
PRISM
U.S. Precipitation Maps
Precipitation
inches
(Daly et al. 1994)
------- |