vvEPA
             United States
             Environmental Protection
             Agency
             Office of Research and
             Development
             Washington DC 20460
EPA/620/R-94/023
August 1994
Assessing the
Suitability of
Windbreaks as
Wildlife Habitat

1994 Pilot Plan
             Environmental Monitoring and
             Assessment Program

-------

-------
                                                       EPA/620/R-94/023
                                                       August 1994
  ENVIRONMENTAL MONITORING AND ASSESSMENT PROGRAM

                    Agroecosystem Resource Group



Assessing the Suitability of Windbreaks as Wildlife Habitat -  1994 Pilot Plan

                                by

                          George R. Hess
                            Jeff M. Bay
                    North Carolina State University
                    Raleigh, North Carolina  27606
                         Technical Director
                          C. Lee Campbell
                United States Department of Agriculture
                    Agricultural Research Service
                    Raleigh, North Carolina 27606
                          Project Officer
                         Susan E. Fransori
                   Monitoring Sciences Division
           Environmental Monitoring Systems Laboratory
                  Las Vegas, Nevada  89193-3478
         Environmental Monitoring and Systems Laboratory
                Office of Research and Development
               U.S. Environmental Protection Agency
                   Las Vegas, Nevada  89193-3478
                                                     Printed on Recycled Paper

-------
                                       Notice
      This research has been funded by the United States Environmental Protection Agency
(EPA) through its Office of Research and Development under Interagency Agreements
#DW12934170 with the U.S. Department of Agriculture (USDA), Agricultural Research
Service (ARS) and #DW 12934747 with the USDA National Agricultural Statistics Service
and by the USDA ARS.  It was conducted with our research partners under the management
of the Environmental Monitoring Systems Laboratory - Las Vegas in support of the
Environmental Monitoring and Assessment Program. Neither U.S. EPA nor USDA-ARS
endorses  or recommends any trade name or commercial product mentioned in this document
to the exclusion of others. They are mentioned solely for the purpose of description or
clarification.

Proper citation of this document will be:

Hess, G.R. and J.M. Bay. 1994. Environmental Monitoring and Assessment Program -
Agroecosystem Resource Group - Assessing the Suitability of Windbreaks as Wildlife
Habitat - 1994 Pilot Plan. EPA/xxx/R-yy/zzz.  U.S. Environmental Protection Agency,
Washington, D.C.
                                          11

-------
                                   Executive Summary
       The Environmental Monitoring and Assessment Program's (EMAP) Agroecosystems
Resource Group is developing a program to monitor and evaluate the ecological condition of
United States agricultural lands.  We are evaluating agricultural lands from several
perspectives: productivity; quality of air, water, and soil; and biodiversity. This pilot focuses
on the potential biodiversity value of windbreaks in the Great Plains, and is designed to test
the feasibility of applying the U.S. Fish and Wildlife's Bird Species Richness Index for
windbreaks on  a regional basis.
       Windbreaks are an important non-crop element in the Great Plains, an extensive
agricultural landscape. Although planted to protect fields, crops, livestock, and farmsteads
from the prevailing winds, windbreaks also provide some of the scarce wooded habitat for
birds and other wildlife.  For example, less than 2% of Nebraska is covered by trees; and
approximately 25% of that wooded cover is provided by windbreaks.  Windbreaks may also
have a negative impact on wildlife species  that require large, uninterrupted areas  of grassland
habitat; this issue is not addressed by this research.

       We are working in cooperation with the United. States Department of Agriculture,
National Agricultural Statistics Service (NASS).  During June,  1994, NASS enumerators will
collect information about the location of windbreaks in Nebraska.  This information will be
used to draw a probability sample of,windbreaks.  NASS enumerators will visit the sample
windbreaks later in the year and collect the data needed to calculate the Bird Species
Richness Index for windbreaks. The Index gives the number of different bird species
expected in a windbreak and is based on four habitat characteristics of the windbreak.  The
habitat characteristics needed to calculate the Index are windbreak area, average height of the
tallest row of trees in the windbreak, foliage height diversity, and snag density.  An
associated Wildlife Habitat Suitability Index can also be  calculated from these data; this  index
reflects a  windbreak's value as habitat for birds and  small mammals, but has been subjected
                                           iii

-------
to less extensive testing.

       We will produce a report documenting the results of this study, including:  (a)
estimated extent (number and area) of windbreaks in Nebraska, with 95% confidence; (b)
estimated cumulative distribution of the Bird Species Richness Index of windbreaks in
Nebraska, with 90% confidence intervals; and (c) estimated cumulative distribution of
windbreak area in Nebraska, with 90% confidence.

       We are also planning a follow-up visit to the sample windbreaks during the Spring,
1995, to survey the birds present and test the Bird Species Richness Index.  As part of that
study, we will also  collect multi-scale remote sensing data for the study sites in an attempt to
determine the spatial scales  of habitat to which the bird community composition is most
closely correlated.  Details of these future efforts are not covered in this plan.
                                            IV

-------
                                    Table of Contents
 Notice   	               ji

 Executive Summary	                  ^

 List of Figures	      vii

 List of Tables  	      vii

 Acknowledgements  	

 1. Introduction: Why Windbreaks?	          i

 2. Approach and Objectives  	             3

 3. Design  	                           2

4. Logistics	                           g
      4.1. Equipment Purchases	       g
      4.2. Training	          6
      4.3. Field Data Collection	         6
      4.4. Data Collection Schedule	         7

5. Quality Assurance  	                              o
      5.1. Training	                                    g
      5.2. Audits   	                                                        o
                                  	   o
      5.3. Repeat Measures	                      g
      5.4. Data Consistency	„	„	                          9
                                           v

-------
       5.5. Measurement Variability  	    9
       5.6. Sensitivity Analysis	   10

6. Data Analysis and Reporting	   ! *
       6.1. For each windbreak	   11
       6.2. Regional estimates	   19
       6.3. Spatial Variability	   19
       6.4. National Resources Inventory Data  	   20
       6.5. Pilot Report	   20

7. Information Management  	   21

8. Future Efforts  	•	   23

9. Definitions	   24

 10. References	   29
                                            VI

-------
                                     List of Figures
Figure 1: Overview of calculations needed to derive Bird Species Richness Index from
       field data	   11
Figure 2: Calculating raw and scaled values for windbreak area	   12
Figure 3: Calculating tree heights from photographs.	   13
Figure 4: Calculating raw and scaled values for average height of tallest row of trees. ...   15
Figure 5: Calculating raw and scaled values of foliage height diversity.	   16
Figure 6: Calculating raw and scaled values of snag density	   17
Figure 7: Calculating Bird Species Richness Index from scaled values of windbreak
       attributes. .  .	   18
Figure 8: Information management for windbreak data	  21
Figure 9:  Examples of when and how to delineate  separate windbreaks	  26
Figure 10: Examples of when and how to delineate separate windbreaks	  27




                                     List of Tables

Table 1: Estimated forest and windbreak cover in Nebraska. .	   1
                                           VII

-------
                                  Acknowledgements
       Thanks to Len Stefanski (North Carolina State University) for statistical consultation
and Brian Schumacher (United States Environmental Protection Agency) for input on quality
assurance.  Bruce Wight (United States Department of Agricultural, Soil Conservation
Service) and Jim Brandle (University of Nebraska at Lincoln) helped us formulate a clear
definition of a windbreak; and Lynn Gentrup (United States Department of Agriculture,
National Agricultural Statistics Service) kept asking great questions that made us realize just
how difficult it is to define something that everyone intuitively recognizes. Phil Cook (Kansas
State University) and Rick Schroeder (U.S. Fish and Wildlife Service) patiently answered
questions about the best ways to collect data for the Bird Species Richness Index. Raymond
O'Connor (University of Maine at Orao) and Tom Martin (U.S. Fish and Wildlife Service)
helped provide the ornithological expertise and insight that we lacked.

       Thanks to Dan Fiscus (North Carolina State University) and Charles Harper (North
Carolina State University) for their help in early field-testing the data collection procedure.

       Jim Brandle, Lee Campbell (United States Department of Agriculture, Agricultural
Research Service), Bruce Jones (United States Environmental Protection Agency), Michele
Schoeneberger (United States Department of Agriculture Forest Service), Brian Schumacher
(United States Environmental Protection Agency),  and Llew Williams (United States
Environmental Protection  Agency) commented on  an earlier version of this plan.
                                           vm

-------
 1. Introduction: Why Windbreaks?
       The Environmental Monitoring and Assessment Program's (EMAP) Agroecosystems
Resource Group is developing a program to monitor and evaluate the ecological condition of
United States agricultural lands (Heck et al 1991, 1993; Campbell et al. 1994).  We are
evaluating agricultural lands from several perspectives: productivity; quality of air, water, and
soil; and biodiversity.  This pilot focuses on the potential biodiversity value of windbreaks  in
the Great Plains, specifically bird diversity.

       Agricultural lands - fields, pastures, and orchards - are managed to produce food and
fiber for people. When land is converted from other cover to agricultural use, the landscape
is changed. Changes in the amount and arrangement of different parts of the landscape -
forests, grassland, fields, woodlots,  windbreaks, and so forth - may affect populations of
plants and animals that live there (e.g. Forman and Godron 1986). Windbreaks are an
important non-crop element in the Great Plains, an extensive  agricultural landscape.  Although
planted to protect fields,  crops, livestock, and farmsteads from the prevailing winds,
windbreaks also provide  some
of the little wooded habitat      Table 1: Estimated forest and windbreak cover in Nebraska.
for birds and other wildlife.
For example, less than 2% of
Nebraska is covered by trees
and approximately 25% of
that wooded cover is provided
by windbreaks (Table 1).
Hunting, wildlife observation
and hiking are all enriched by
the diversity of life in
windbreaks (Johnson et al.
1991).
Area of...
Forest1
Windbreaks2
Nebraska1
1982
(sq-km)
2965.0
935.3
200504.6
1987 '
(sq-km)
2946.4
909.5
200504.6
1. USDA Soil Conservation Service National Resources
Inventory (USDA-SCS 1987, 1989).
2. Bruce  Wight, personal  communication,  based  on
National Resources  Inventory  data.   The  "windbreak"
and "forest" categories are mutually exclusive.

-------
       Much of the wooded habitat in the Great Plains is associated with riparian corridors.
Windbreaks may serve as stepping stones between riparian areas, as well as for birds that
migrate through the area each year. The EMAP Surface Waters Resource Group is
developing techniques to monitor habitat and birds along riparian corridors (Larsen and
Christie 1993).  The EMAP Forest Resource Group is monitoring birds in forest cover
(Martin et al.  1994).  How  these data will be integrated among resource groups is  still
unclear.  Although the USDA Soil Conservation Service evaluates the condition of
windbreaks with respect to  soil erosion (through the National Resources Inventory), there is
no current effort to monitor their condition for biological diversity. In this pilot, we will
evaluate the extent and condition of windbreaks in Nebraska as they relate to biodiversity,
particularly bird species diversity.  In a later effort, we will  survey the bird species present in
the windbreaks.

       Agriculture in  the Great Plains has eliminated much  of the grassland habitat that
existed before the area was settled by Europeans.  Some native grassland birds - such as
meadowlarks, dickcissels, and greater prairie chickens - need large, unbroken areas of grass.
Some of these birds have suffered declines, and windbreaks may provide habitat for other
birds and mammals that prey on grassland birds.  Thus, while windbreaks may help  some
wildlife species - those that depend on wooded habitat - they may be detrimental to others
(e.g.  Knopf 1992). This trade-off must be kept in mind when the overall condition and
sustainability of biodiversity in the Great Plains is considered; it is not considered in this
plan.

-------
2. Approach and Objectives

       We are working in cooperation with the United States Department of Agriculture,
National Agricultural Statistics Service (NASS) to evaluate the potential biodiversity value of
windbreaks. During the 1994 June Agricultural Area Survey, NASS enumerators will collect
information about the location of windbreaks in Nebraska. This information will be used to
draw a probability sample of windbreaks.

       In July, 1994, NASS enumerators will visit the sample windbreaks  and collect  the data
needed to calculate the U.S Fish and Wildlife's Bird Species Richness Index (BSRI) for
windbreaks (Schroeder et al.  1992). The Index is the number of different bird species
expected in a  windbreak,  based on correlations of the bird species present in windbreaks with
four habitat characteristics: windbreak area, average height of the tallest row of trees in the
windbreak, foliage height diversity, and snag density.  Schroeder et al.  (1992) developed the
Bird Species Richness Index during a three-year study of 34 windbreaks in Kansas. By using
the four windbreak characteristics they were able to explain a large portion of the variability
in the bird species present in the  windbreaks (R2 = 0.893). An associated Wildlife Habitat
Suitability Index can also be calculated from these data; this index reflects a windbreak's
potential value as habitat  for birds and small mammals, but has been subjected to less
extensive testing (Schroeder 1986).

       This pilot project is designed to test the feasibility of applying the U.S. Fish and
Wildlife's Bird Species Richness Index for windbreaks on a regional basis. Specifically, we
will:

       (1) Determine if the data needed to calculate the  BSRI can be collected successfully
       by NASS enumerators.
       (2) Evaluate several components of variability in the data used to calculate the index:
              (a) Among-team variability.  If two teams of NASS  enumerators measure the
              same windbreak, how variable are the results?  We recognize that this includes
                                            3

-------
             same-team variability but will not attempt to disentangle it from among-team
             variability. Because of the long term nature of EMAP, among-team variability
             is the more likely source of problems.
             (b) Measurement variability.  Is the number of sampling stations within
             windbreaks sufficient?  Can we use fewer without loss of information?
             (c) Spatial variability.  Are windbreaks in the same area more likely to be
             similar to one another than to windbreaks further away?
      (3) Develop data processing and information management techniques to use data
      collected at each windbreak to calculate a BSRI for the windbreak.
      (4) Develop statistical methods to use data collected at sample windbreaks to make
      regional estimates of the extent and BSRI of windbreaks.
      (5) Develop and evaluate logistics and quality assurance procedures.
      (6) Use this information to develop a sampling design suitable for monitoring
      windbreaks as wildlife habitat.
       This work focuses on the biodiversity value of individual windbreaks, specifically for
birds.  The BSRI provides information about the condition of windbreak habitat; it does not
provide information about the condition of bird populations and communities. Trends in the
BSRI over time will indicate whether the condition of available windbreak habitat is
improving, degrading, or remaining the same.  Other information - from the ongoing
Breeding Bird Survey (Robbins et al. 1986), future EMAP monitoring, or future efforts of the
National Biological Survey - will be needed to track the condition of bird populations and
communities.

-------
3. Design

       The EMAP Agroecosystems Resource Group is using the USDA National Agricultural
Statistics Service's (NASS) June Agricultural Area Survey as the basis for sampling (Cotter
and Nealon 1987; Campbell et al  1994).  In Nebraska, NASS defines seven strata based on
land use in the surrounding area.  The June Agricultural Area Survey is  a stratified random
sample of land areas, called segments, with sampling intensity based on  the amount of
agriculture in the surrounding area.

       The June Agricultural Area Survey consists of five non-overlapping replicate-years,
each covering the same geographic area (Nebraska, in this case). Each year  the oldest
replicate is removed from the Survey  and a new replicate is added to the Survey and remains
in the Survey for five years.  Within each replicate and stratum, segments are numbered in a
serpentine manner across the state. To sample windbreaks, NASS will draw a systematic
sample of 60 windbreaks from a list, ordered by segment number, of all  windbreaks identified
for the newest replicate-year.  Ordering by segment number within a single replicate-year
ensures, as much as possible, uniform geographic coverage.  If there are  fewer than 60 total
windbreaks in the newest replicate year, then the list will contain all windbreaks from the two
newest replicate-years. If necessary, replicates will continue to be added to the list until there
are at least 60 windbreaks in the list.  Based on NASS's past experience in Nebraska, it is
expected that our final sample will be about 20-30% percent smaller than planned due to
non-response.

-------
4. Logistics

4.1. Equipment Purchases

       All necessary equipment will be assembled at the EMAP-Agroecosystems headquarters
in Raleigh, NC. NASS has indicated that they will use 10 teams of enumerators to perform
the windbreak sampling. We will purchase equipment to make up 11 kits - one for each team
and a spare for the NASS Nebraska office.  Equipment assembly will begin in late March,
and all equipment will be at the NASS Nebraska office by 1 July 1994. A list of equipment
is included in the Windbreak Data Collection Manual (USDA-NASS 1994).

4.2. Training

       A training session will be hosted by NASS in Grand Island, Nebraska on 13  July
 1994.  George Hess (indicator lead) will lead the session; Brian Schumacher (Quality
Assurance Officer) will also attend. Training will consist of a two-hour classroom session that
will provide relevant background information and a detailed overview of the procedure.  The
classroom session will be followed by a 3-4 hour field session, during which techniques will
be demonstrated at an actual windbreak.

4.3. Field Data Collection

       Teams of two NASS enumerators will visit each of the  windbreaks.  At each
 windbreak, data will be collected to calculate four windbreak attributes.  These four values
 are used to calculate the Bird Species Richness Index for the windbreak.  Detailed data
 collection procedures are described in the Windbreak Data Collection Manual (USDA-NASS
 1994).
        A brief description of the data items that will be collected follows.

-------
1) Windbreak area. The area covered by a windbreak is an important indicator of its ability to
support biological diversity.  Larger windbreaks provide more habitat.  The enumerator will
measure the length and width of the windbreak by calibrated pacing.

2) Average height of tallest row of trees in windbreak. Height is the third dimension of a
windbreak — taller windbreaks provide more habitat.  One enumerator will take several full-
height photographs of the windbreak while another holds a standard of known height at the
edge of the windbreak.  These photographs will be used to determine the  average height of
the tallest row of trees in  the windbreak.

3) Foliage height diversity.  Different birds prefer different parts of the windbreak for nesting
and foraging.  Some like to nest on the ground, others in small trees and bushes,  and others
in the canopy.  A windbreak with many different kinds of habitat can support a more diverse
group of animals than one with only a few kinds of habitat.  Foliage height diversity is a
measure of the different kinds of habitat available in the vertical direction. At each sample
point, the enumerator will use a marked 6-meter standard to determine if there is  any
vegetation at each of three levels:  less than 1 meter (the ground vegetation), between  1 and 6
meters (mid-story), and above 6 meters (canopy).              '

4) Snag density.  Snags are dead but standing trees and branches.  They provide yet another
important type  of habitat - homes  for birds that build their nests by hollowing out dead trees.
Enumerators  will count all snags in the windbreak so that a number of snags per unit area can
be calculated.

4.4. Data  Collection  Schedule

       NASS enumerators will carry out all field work for the 60 windbreaks in the' sample.
Field work will be performed on  16, 18, 19, 20 July and 1-5 August.  Sampling activities will
be coordinated by the NASS state office.  NASS enumerators are responsible for
transportation to and from field sites,  locating sample windbreaks, completing all
                                           7

-------
measurements, mailing completed data sheets to the NASS state office, and shipping exposed
film to the EMAP-Agroecosystems headquarters via Federal Express.

       Any field work not completed by 5 August will be completed in the autumn of 1994.
The only problem we anticipate is failure to obtain the photographs needed to measure the
height of the windbreaks.  The sampling period is fairly late in the growing season, and some
crops (e.g., corn) may already be taller than the enumerators, making it impossible to obtain
photographs. In this case, the enumerator will make a note and photographs will be obtained
in a re-visit  after harvest.
5. Quality Assurance

5.1. Training

       Ten teams of NASS enumerators will be trained by George Hess (indicator lead)
immediately before the field measurement period (see Logistics). Brian Schumacher (Quality
Assurance Officer)  will attend the training session.

5.2. Audits

       George Hess and Brian Schumacher will perform field audits on approximately 5% of
the sampled windbreaks.  They will confirm that data collection is being completed in
accordance with the instructions in the "Windbreak Data Collection Manual" (USDA-NASS
 1994).

5.3. Repeat Measures

       During the enumerator training session multiple teams of enumerators will measure the
 same windbreak. Field training will take place in an area in which five windbreaks are in
                                            8

-------
 close proximity. Two teams will measure each windbreak simultaneously in a round-robin
 fashion.  Then teams will be shifted around and the process of two teams measuring each
 windbreak simultaneously will be repeated. This will be done three times, so that each of the
 five windbreaks will be'measured by six teams; this provides 25 degrees of freedom for
 estimating among team variability.

 5.4. Data Consistency

       Data consistency checks have been built into the data collection procedure (USDA-
 NASS 1994).

 5.5. Measurement Variability

       The  variability of width, height, and foliage height diversity within windbreaks will be
 measured during data analysis.

 Width.  Windbreak width is measured at multiple points in the windbreak (USDA-NASS
 1994).  Variability will be determined from these measures.

Height. Windbreak height is measured from photographs. One-fifth of the photographs will
be re-measured by the same observer so that we can calculate same-observer variability. If
more than one person is used for these analyses, we will also measure among-observer
variability.
Foliage height diversity.  Foliage height diversity is measured at multiple points in the
windbreak (USDA-NASS  1994). Variability will be determined from these measures.

-------
5.6. Sensitivity Analysis

       Using the data collected during the survey we will perform a sensitivity analysis to
determine the effect error in each of the Index components has on the'BSRI values.
                                            10

-------
 6. Data Analysis and Reporting

        The EMAP Agroecosystems Resource Group is responsible for all data analysis and
 reporting.

 6.1. For each windbreak
        Various characteristics of each
 windbreak are combined into a single value
 to calculate Bird Species Richness Index
 (Figure 1) (Schroeder et al.  1992). Many
 individual data items will be combined to
 arrive at a raw value for each of four
 windbreak characteristics:

       (1) Area
       (2) Average height of tallest row of
       trees
       (3) Foliage  height diversity
       (4) Snag density
                  Bind  Species  Richness
Scaled value:     Area    Height   heS6   Sna9
                               diversity   density
                 III!
  Raw value:     Area    Height
                                        Snag.
                                        dens"y
     Hold data:
                          Data sheets,
                          photographs
Figure 1: Overview of calculations  needed to
derive Bird Species Richness Index from field
data.
The raw value for each of these characteristics is then scaled from 0-1, where 0 is the poorest
condition and 1  is the best condition, (Schroeder et al.  1992), and the 4 scaled values
combined to yield the Bird Species Richness Index.  The calculations are described below.
                                            11

-------
6.1.1. Area



Input Data.   Length of windbreak (paces); width of windbreak (paces) at 4 or 9 sample

stations along the windbreak.

Calculations. The calculations are detailed in Figure 2.
                                  _Q
                                  "D
                                            Width 1
                                         -^ Width 9
                Length (paces)
              x meters/pace

                Length (meters)
Sum  79  = Average
              Width
              (pace)

      x meters/pace
                                           Average Width  (meters)
            Raw value:
                   Length (m) x  Average  Width (m)  = Area  (ha)
                                10,000 m2/ha
            Scaled  value:

                    Minimum of (  1  , 17.4  x Area?359
            	42
           Figure 2: Calculating raw and scaled values for windbreak area.
                                       12

-------
 6.1.2. Average height of tallest row of trees

 Input data.    Photographs of the windbreak at 4 or 9 sample stations along the windbreak.
 Measurement of the height of the range pole (or a section of it) in the photograph and the
 height of the tree in the photograph.
                                               HT
                                                      Range
                                                       Pole
                                                                                  Photo
                                                                                  Point
                                               Tree
Obtaining height measures. Each photograph
will contain a calibrated range pole .  The
height of the tallest tree directly behind the
range pole can be calculated as follows
(Figure 3).

       The height of the calibrated range pole	n	
(HP) in the photograph is related to the actual   Figure 3:   Calculating   tree   heights  from
                                              photographs.
distance between the photo point and the
range pole (D0\  This relationship can be determined by photographing the range pole at
various distances, measuring the range pole on the photographs, and performing a regression
with a quadratic model  (this is a wide angle lens, so the relation is not linear). The
relationship is
                                  D0 = aHp2
                                                bHp + c
where a, b, and c are constants determined by the regression.  Since film processing
procedures are standardized, the relationship between the actual size of an object at a given
distance from the camera and its size in the photograph is constant.  This relationship (i.e.,
values for a,  b, and c) will be determined in advance at EMAP-Agroecosystem headquarters.

      The distance between the range pole and the base of the tree (DT~) is  assumed to be
half the width of the windbreak, which is known from field measures.  The  distance between
                                           13

-------
the photo point and the base of the tree, DB, is D0 + DT.  Given this distance, the scale height



of the range pole at distance DB can be calculated by solving
                                    £>„ = ah2  + bh + c
                                      D
for h.  The height of the tree can be measured in the photograph and the actual height of the



tree can be calculated as:
                     Actual Height of Tree _   Actual Height of Pole
                     Photo Height of Tree    Scale Height of Pole (h)
                                            14

-------
Calculation of average height.  The calculations are detailed in Figure 4 (based on Schroeder

et al. 1992).
                  ->  Photo 1
                  -* Photo 9
->  Height 1
                                  w

                                  &
                                  OS
                                  o
                                  •s
       +

    Height 9
                                             Sum 79  = Average
                                                          Height
                                                         (meters)
           Raw  value:   Average Height (meters)
           Scaled value:
                 Minimum of (  1  , 0.2 +     x Average Height  )
         Figure 4: Calculating raw and scaled values for average height of
         tallest row of trees.
                                      15

-------
6.1.3. Foliage height diversity



Input data.   Presence or absence of vegetation in each of 3 vertical layers at each of 4

sample points at each of 4 or 9 sampling stations.

Calculations. Calculations are detailed in Figure 5 (Phil Cook, personal communication;

Schroeder et al. 1992).
            .a
            -a
                             Ground
                             Cover
           Midstory   Canopy
                    Station 1
Measurements are taken  at
4 sample  points  at each  of
   the sampling stations.


       1  = Present
       0  = Absent
                  -*• Station 9
                              Cover
          Raw value:  FHD =
                                         Sum   +Sum
                                         Midstory  Canopy
                                        1.0
                                 Total
                                 Present
                 / Sum Ground i2 .  i Sum  Midstorv\2 + / Sum Canopy i
                 V    Total    i    \    Total    '    l     Total    I


           Scaled value:

                 Minimum of ( 1  , 0.2 + 0.26666 x  FHD  )	
         Figure 5:  Calculating  raw  and scaled values of foliage  height
         diversity.
                                          16

-------
6.1.4.  Snag density



Input data.    A count of all snags in the windbreak and windbreak area.

Calculations.  Calculations are detailed in Figure 6 (based on Schroeder et al.  1992).
                   Number of
                    Snags  in
                   Windbreak
             Windbreak Area (ha)
                                    = Snag Density  (snags/ha)
       Raw value:      Snag Density



       Scaled  value:

             Minimum  of ( 1  ,  0.5 + 0.02273 x Snag  Density)
       Figure 6: Calculating raw and scaled values of snag density.
                                     17

-------
6.1.5. Calculating the Bird Species Richness Index
     The scaled values are combined to calculate the Bird Species Richness Index for the

windbreak (Figure 7) (Schroeder et al. 1992).
  Habitat  Suitability Index  (HSI) =
      Scaled
      Area
Scaled     Scaled
FHD     + Height
Scaled
Snag Density
   Bird Species  Richness  Index  = 5.34  +  41.1  x  HSI
  Figure 7: Calculating Bird Species  Richness Index from scaled  values of windbreak
  attributes.
                                  18

-------
 6.2. Regional estimates

       Once a Bird Species Richness Index is available for each of the sampled windbreaks,
 these values are used to generate regional estimates of windbreak condition.  Cumulative
 distribution functions (CDFs) will be estimated for the Bird Species Richness Index (BSRI)
 and for each of the four raw values which go into the BSRI.  An estimated CDF for these
 measures will give the estimated proportion of Nebraska windbreaks that have values of the
 measure less than or equal to any specified value (e.g., the proportion of windbreaks with a
 BSRI of 14 or less).  A CDF will also be estimated for the number of windbreaks per unit
 area, where unit area is defined as the average size of NASS segments.  This CDF will make
 use of windbreak counts from the complete June Agricultural Area Survey. Approximate
 90% pointwise confidence bands will be computed for each CDF to indicate  the precision of
 the estimated CDF.

       In addition to  the CDFs, extent estimates will be made for the number of windbreaks
 in Nebraska using the complete June Agricultural Area Survey, and for the total area of
 windbreaks in Nebraska using the subsample from the June Agricultural Area Survey.
 Approximate 95% confidence intervals will be included with the  estimates  of extent.

       NASS samples different strata with different intensities and, therefore, windbreaks
 from different strata have  different inclusion probabilities (Cotter and Nealon 1987).
 Estimates of regional  CDFs will be constructed using procedures  described in Lesser and
 Overton (1994), which allow for varying inclusion probabilities and are unbiased.

 6.3. Spatial Variability

       We will perform analyses to determine if the condition of windbreaks  is spatially
 autocorrelated to answer the question "Are windbreaks in the same  area more similar to one
another than they are  to distant windbreaks?". We will divide Nebraska into  several
geographic regions and determine the spatial variability of windbreaks within  and among
                                          19

-------
regions.
6.4. National Resources Inventory Data

       The USDA Soil Conservation Service has data on the length and area of thousands of
windbreaks collected through their National Resources Inventory (USDA-SCS 1987, 1989).
We would like to use these data to get a better measure of spatial variability and to improve
our regional estimates  of windbreak condition.  Our regional estimate can be improved using
these data if we find that the other components  of the Bird Species Richness Index are closely
correlated with windbreak area, and if we assume that this correlation holds for all
windbreaks  in Nebraska. Under these assumptions, we could use windbreak area to estimate
the Bird Species Richness Index of all windbreaks in the National Resources Inventory.  We
have been in contact with the Soil Conservation Service and should be  able to obtain these
data for analysis (Bruce Wight, personal communication).

6.5. Pilot Report

        We will produce a report documenting the results of this study,  including:
        (a) Estimated extent (number and area) of windbreaks in Nebraska, with 95%
        confidence.
        (b) Estimated cumulative distribution of the Bird Species Richness Index of
        windbreaks in  Nebraska (e.g., percent of windbreaks expected to contain S or fewer
        species of bird), with 90% confidence intervals.
        (c) Estimated cumulative distribution of the four raw values that comprise the BSRI,
        with 90%  confidence.
        (d) Estimated cumulative distribution of the number of windbreaks per unit area, with
        90% confidence.
        (e) Results of  variability studies.
                                            20

-------
7. Information Management
      Information management for this pilot is summarized in Figure 8.
                 Dafa/sheets'fJ';
                         NASS
                         data
                         entry
       EMAP-Agroecosystems
       photo
     ^analysis
        NASS responsbile

        EMAP-Agroecosystems
        responsible
 erge  by
Sample ID
                          K5%X-A»iH;i^:'rj, •&£*.*•*?££•' ^.Vi-
                                      Analysis:  for each
                                     f windbreak

                                       Analysis: regional
                                       estimates
                                 tf^M^M^iity-jS
   Figure 8: Information management for windbreak data.
                                      21

-------
       NASS will be responsible for tracking information in the field and collecting all
completed data sheets at their state office in Lincoln, Nebraska. NASS will provide a SAS
dataset containing the values for all measurements made on each windbreak, with the
exception of average height. Each windbreak will have a unique sample identification
number.

       Windbreak height measures will be obtained from photographs taken at the site.  A
label containing the sample identification number will be affixed to each camera.  The sample
identification number will also be photographed as the first and last frame on each roll of
film.  Recyclable cameras will be returned directly to EMAP-Agroecosystems' headquarters in
Raleigh, NC for processing and  analysis.  After heights are obtained by analyzing the
photographs, height measures will be merged with the other data by sample identification
number.
                                            22

-------
8. Future Efforts

       Schroeder et al. (1992) developed the BSRI on 34 windbreaks in Kansas, and
speculated that it will provide reliable estimates over most of the Great Plains region.  In the
spring of 1995, trained ornithologists will revisit the windbreaks sampled during July,  1994
and determine the species of birds present.  This will provide a test of the BSRI for our study
region.

       This Pilot focuses on the potential biodiversity value of individual windbreaks,
specifically for birds.  The Bird Species Richness Index does not account for surrounding
landscape features. Further efforts at multiple scales are needed to understand how
windbreaks and other types of habitat influence regional biodiversity (e.g., the effect of
wooded habitat on grassland birds and other grassland wildlife; the effect of nearby wooded
habitat on the biodiversity value of an individual windbreak).  Planning is underway to
expand our 1995 bird survey to include a larger area around the windbreak, and to correlate
these data with remotely sensed land cover data at multiple scales and  with U.S. Fish &
Wildlife  Service Breeding Bird Survey data (Robbins et al. 1986). This work will help us
identify appropriate monitoring scales for birds and their habitat, and ways in which to
integrate our results with those of other programs.
                                           23

-------
9. Definitions

Windbreak:  One or more rows of trees and shrubs planted either along the edges of fields
       and pastures, within fields and pastures, or near farmstead buildings and feedlots,
       designed to protect the field, pasture, building, or feedlot from the wind.

       Minimum height: None. Since windbreaks may be planted as young trees and grow
       over time, there is no minimum height (Jim Brandle, Bruce Wight, personal
       communication).

       Minimum length for windbreaks protecting fields or pastures: 300 feet.
       Field windbreaks  should be at least 10 times their mature height of 30 to 50 feet (Jim
       Brandle, personal communication).

       Minimum length for windbreaks protecting buildings or feedlots: None.

       Gaps may occur in a windbreak.  Although poor windbreaks do a poor job of
       providing wind protection, they are still windbreaks and we want  to include them in
       our sample .  Excellent windbreaks consist of a stand of trees and shrubs that are
       continuous from end to end and which present an almost unbroken barrier to wind.
       Gaps in a windbreak reduce their effectiveness. Poor windbreaks  lack continuity and
       consist of sparse, open stands of trees and shrubs in scattered clumps or  as individuals
       (Read 1958).  Small gaps of  15-50 feet are caused by failure of three to  eight
       consecutive trees; large gaps  may extend for up to several hundred feet.

       Separating windbreaks. Windbreaks are designed to provide protection from the wind.
       The decision to mark off a separate windbreak is  related to windbreak function rather
       than to  the occurrence of gaps in the windbreak.  However, changes in direction and
       some gaps may indicate that  an otherwise continuous row of trees should be defined
                                           24

-------
as two windbreaks (see Figures 9 and 10 and further definitions below).

Count each segment of an otherwise continuous row of trees and shrubs that is broken
by a road — paved or otherwise — that creates a 60 foot or larger gap as a separate
windbreak (e.g., Figure 9, Windbreaks 3 & 5).  Section roads - which run through
approximately 60 foot right-of-ways — are considered to separate an otherwise
continuous row of trees into two windbreaks (Jim Brandle, Bruce Wight, personal
communication).  We extend this concept to all roads. A  60 foot or larger gap  without
a road does not separate a windbreak into two.

Count each segment of a field or pasture windbreak that  changes direction as a
separate windbreak (e.g., Figure 9 Windbreaks 2 & 3, 4 &  5). The Soil Conservation
Service  considers a change in direction as a new windbreak for field windbreaks
(USDA-SCS 1991).  For example, a windbreak on the western edge of a field blocks
winds blowing from  the west; a windbreak on the northern edge of a field blocks
winds blowing from  the north.  Since these are two different functions, the windbreaks
are considered to be  separate.

Count all windbreaks around farmstead buildings or feedlots as a single windbreak,
regardless of changes in direction (e.g., Figure 9 Windbreak 6). This  is consistent
with the Soil Conservation Service definition (USDA-SCS 1991). The function of the
windbreak is to protect the farmstead buildings or feedlot.
                                    25

-------


1 L >IB9JQDUIAAI

Section Road






jwinaoreak d~1
Windbreak 2

Gap larger than 60 feet with
/ 	
Dirt Road_ . . .

in
j*:
CO
&
c
fe
Windbreak 4 1
Windbre
\
NNo road through gap,
or road and gap less than


a
1
road


• 	 1
ijO;.-?!^™.-^ _-„
l^gl— iO rr
n~i
60 feet
Section Road
yVindbreak 7 /'
p


1S1



ead


Figure 9:  Examples of when and how to delineate separate windbreaks.
                                     26

-------
                     Section Road
                       Windbffs
                       Section  Road
                               
-------
Dead tree: A tree with less than half the branches living (bearing foliage).

Live tree:  A tree with more than half the branches living (bearing  foliage).

Snag:  A snag is a dead tree > 8" in diameter at breast height, or a live tree with one or more
       dead branches > 8"  diameter.
                                             28

-------
 10. References

 Campbell, C.LV J. Bay, C.D. Franks, A.S. Hellkamp, N.P. Helzer, G.R. Hess, M.J.
       Munster, D. Neher, G.L. Olson, S.L. Peck, J.O. Rawlings, B. Schumacher, and
       M.B. Tooley. 1994.  Environmental Monitoring and Assessment Program -
       Agroecosystem Pilot Field Program Plan -1993. EPA/620/R-93/014. U.S.
       Environmental Protection Agency, Washington, DC.

 Cotter, J. and J. Nealon. 1987. Area frame design for agricultural surveys. U.S.
       Department of Agriculture, National Agricultural Statistics Service, Research
       and Applications Division, Area Frame Section. Washington, DC.

 Forman, R.T.T. and M. Godron. 1986. Landscape Ecology. John Wiley and Sons, N.Y.

 Heck, W.W., C.L. Campbell, A.L. Finkner, C.M. Hayes, G.R. Hess, J.R. Meyer, M.J.
       Munster, D. Neher, S.L. Peck, J.O. Rawlings, C.N. Smith, and M.B. Tooley. 1993.
       Environmental Monitoring and Assessment Program Agroecosystem 1992 Pilot Plan.
       EPA/620/R-93/010. U.S. Environmental Protection Agency, Washington, D.C.

 Heck, W.W., C.L. Campbell, R.P. Breckenridge, G.E. Byers, C.M. Hayes, G.R.  Hess, V.M.
       Lesser, J.R. Meyer, T.J. Moser, D.A.  Neher, G.L. Olson, S.L. Peck, and  J.O. Rawlings.
       1991. Environmental Monitoring and Assessment Program Agroecosystem Monitoring
       and Research Strategy. EPA/600/4-91/013. U.S. Environmental Protection Agency,
       Washington, D.C.

Johnson, R.J., M.M. Beck, and J.R. Brandle. 1991. Windbreaks and wildlife. University of
       Nebraska Extension EC 91-1771-B, University of Nebraska,  Lincoln, NE.

Knopf, F.L. 1992. Faunal mixing, faunal integrity, and the biopolitical template for diversity
       conservation. Transactions of the North American Wildlife and Natural Resources
                                        29

-------
       Conference 57: 330-342.

Larsen, D.P. and S.J.  Christie (editors). 1993. EMAP-Surface Waters 1991 Pilot Report.
       EPA/620/R-93/003. U.S. Environmental Protection Agency, Washington, D.C.

Lesser, V.M. and W.S. Overton. 1994. EMAP Status Estimation: Statistical Procedures and
       Algorithms. EPA/620/R-94/008. U.S. Environmental Protection Agency, Corvallis, OR.

Martin, T.E., G.E. Canterbury, D.R. Petit, and L.J. Petit.  1994. Wildlife habitat. Pages 12.1-
       12.24 In T.E. Lewis and B.L. Conkling (editors), Forest Health Monitoring Southeast
       Loblolly/Shortleaf Pine Demonstration Interim Report. EPA/620/R-94/006. U.S.
       Environmental Protection Agency, Washington, D.C..

Read,  R.A. 1958. The Great Plains Shelterbelt in 1954. Great Plains Agricultural Council
       Publication  16, Station Bulletin 441, The Experiment Station of the University of
       Nebraska, Linclon, Nebraska.

Robbins, C.S., D. Bystrak, and P.H. Geissler. 1986. The breeding bird survey: its first fifteen
       years,  1965-1979. U.S. Department of Interior, Fish and Wildlife Service Resource
       Publication  157, Washington, D.C.

Schroeder, R.L. 1986. Habitat suitability index models: wildlife species richness in
       shelterbelts.  U.S. Fish and Wildlife Service Biological Report 82(10.128),
       Washington, DC.

Schroeder, R.L., Cable, T.T., and  Haire, S.L.  1992. Wildlife species richness in shelterbelts:
       test of a habitat model. Wildlife Society Bulletin  20: 264-273.

USDA-National Agricultural Statistics Service. 1994. EMAP Windbreak Sampling
       Interviewer's  Manual — July 1994. USDA National Agricultural Statistics Service,
                                            30

-------
       Washington, D.C.

USDA-Soil Conservation Service.  1987. Basic Statistics:'1932 National Resources Inventory.
       Statistical Bulletin No. 756. USDA-Soil Conservation Service, Washington, D.C.

USDA-Soil Conservation Service.  1989. Summary Report: 1987 National Resources
       Inventory. Statistical Bulletin No. 790. USDA-Soil Conservation Service, Washington,
       D.C.

USDA-Soil Conservation Service.  1991. Instructions for Collecting 1992 National Resources
       Inventory Sample Data. USDA-Soil Conservation Service, Washington, D.C.
                        *D.S. GOVERNMENT PRINTING OFFICE: 1994-650-006/00215 Region 5.
                                          31

-------

-------

-------
m

i
ro
o
33

I
ro
CO
o> 3
— CD
o c
^ w
-05'
Z3. 0)

S ^

?


cn
o>
o  o m  c
5'  CD  3  =J
S.  S ^.  s-'
3  CD  3  S.
                    ro

                    §
    S  5
    3  O
    3  c?
    CD  O

    S 5-
    92.3

    33 >
    CD  (Q
    CO  CD
    CD  3
    0)  O
              I

              I
              o
              3
                  3
                  1
                  O

-------