USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Conservation Reserve Program:
Benefit for Grassland Birds in the Northern Plains

Study Area and Methods


For duck nest success, our study area included that portion of North and South Dakota east or north of the Missouri River (Figure 1). This area corresponds roughly to the Prairie Pothole Region of North and South Dakota and also was studied by Klett et al. (1988). For nonwaterfowl bird population trends, our study area included all of North Dakota.

Figure 1: Map of North Dakota and eastern South Dakota showing locations of study sites.
Figure 1.  Locations of study sites (dots) in eastern South Dakota and North Dakota used to estimate duck nest success. Study area used for nonwaterfowl bird population trends analyses is indicated by hatching.

Duck Nest Success

In the spring/summer 1992 and 1993, we located duck nests (scrape or bowl containing ≥1 egg) in CRP cover and planted cover on WPAs using methods described by Klett et al. (1986). For our study area, we obtained a sample of 2-mile by 2-mile (3.2 km by 3.2 km) blocks from another study (see Cowardin et al. 1988). From this sample, we selected blocks that met two criteria: (1) a minimum of 40 acres (16.2 ha) of CRP cover; and (2) sufficient ponds to attract ≥20 breeding pairs of mallards estimated from a pair-pond regression model similar to that described by Cowardin et al. (1988).

For each block, we selected the nearest WPA (farthest WPA was 9 miles [14.5 km]) that had ≥40 acres (16.2 ha) of planted cover. Planted cover on WPAs consisted of a vegetation mix similar to CRP cover. Each block and associated WPA constituted a study site (Figure 1). For each study site, fields to be searched by treatment (CRP or planted cover) were selected randomly from all fields available until the last field selected resulted in ≥200 acres (80.9 ha) of that treatment for the study site. For study sites with ≤200 acres of a treatment, all fields of that treatment were searched. Each field was searched three times.

We calculated daily nest survival rates (DSR) for each treatment by study site the modified Mayfield method of Johnson (1979). We modeled DSR as a function of treatment, year and location of each study site. For spatial effects, we considered linear and quadratic terms involving universal transverse mercator (UTM) coordinates of Easting and Northing values and their cross products. Four regression models, corresponding to the four combinations of treatment and year, were developed and statistically compared for each species. The method of weighted least squares, with weights equal to the number of exposure days for each DSR estimate, was used to fit the model.

To compare our results with those of Klett et al. (1988), we required overall nest success estimates for the North Dakota portion of our study area (Klett et al. 1988 did not report estimates for the remainder of our study area). We used our model of DSR to estimate the average DSR for each of the nine Wetland Management Districts (Districts) in North Dakota, using the centroid UTM coordinates for each District as an explanatory variable. Some of the nine Districts (four in 1992 and two in 1993) had UTM coordinates that were farther west than our westernmost study sites (Easting = 482296 and 369094, respectively). To avoid extrapolating beyond the geographic range of our data, we truncated our models at the westernmost study site and used the estimated DSR for that study site as a constant for Districts farther west. Nest success by District was estimated by raising the District DSR to the power equal to the mean laying plus incubation period for successful clutches (Klett et al. 1986). Nest success by year for North Dakota then was estimated by weighting each District's nest success by the proportion of breeding pairs that occurred in that District, as estimated from surveys conducted by the FWS, and averaging. Overall nest success was estimated by averaging the individual year values weighted by population estimates in each year.

Nonwaterfowl Bird Populations

We estimated population trends from the BBS (Peterjohn and Sauer 1993) during the pre- (1966-86) and post-CRP (1987-92) periods for grassland-nesting nonwaterfowl birds regularly observed in CRP fields in North Dakota. We used all species reported in Table 2 of Johnson and Schwartz (1993b). We estimated trends (a percentage change per year estimated as a weighted average of slopes of linear regression on each route [Geissler and Sauer 1990]) for each species in each period.

To evaluate the effects of CRP on these species, we classified them into two groups: species primarily nesting within grassland habitats whose trends may be associated directly with the availability of CRP grasslands (CRP-influenced species), and species nesting in various habitats in addition to grasslands whose trends may not be associated necessarily with the availability of CRP grasslands (CRP-neutral species, Table 1). We then used t-tests to determine whether the difference in mean trends (trendPost-CRP - trendPre-CRP) for CRP-influenced species was similar to those of CRP-neutral species. Rejection of this null hypothesis in favor of a one-sided alternative hypothesis would indicate that CRP-influenced species were more likely to have positive population changes than CRP-neutral species between the pre- and post-CRP periods.

Table 1.  Trends in abundance of grasslands-nesting nonwaterfowl birds for the periods 1966-86 and 1987-92 in North Dakota with associated significance levels and sample size (n of routes)
Species 1966-86 1987-92
Trenda n Trenda n
CRP-influenced species
    Bobolink (Dolichonyx oryziverus) –0.15 37   –1.46 42
    Western meadowlark (Sturnella neglecta) –0.94 37     1.98 43
    Chestnut-collared longspur (Calcarius ornatus) –0.86 29     8.26 32
    Savannah sparrow (Passerculus sandwichensis) –3.14 35   –6.03 41
    Baird's sparrow (Ammodramus bairdii) –1.19 25 –15.29 24
    Grasshopper sparrow (Ammodramus savannarum) –7.86 37   10.06 42
    Dickcissel (Spiza americana) –6.51 25   –9.86 24
    Lark bunting (Callamospiza melanocorys) –7.70 29   21.73 35
CRP-Neutral species
    Killdeer (Charadrius vociferus)   1.92 37 –13.02 43
    Mourning dove (Zenaida macroura)   4.17 37     6.51 43
    Eastern kingbird (Tyrannus tyrannus)   5.13 37     7.79 43
    Western kingbird (Tyrannus verticalis)   5.85 36   12.61 43
    Horned lark (Eremophila alpestris)   1.09 37     0.74 43
    Brown-headed cowbird (Molothrus ater)   4.15 37   –2.98 43
    Red-winged blackbird (Agelaius phoeniceus) –1.38 37 –10.89 43
    Vesper sparrow (Pooecetes gramineus)   2.59 37     3.22 42
    Clay-colored sparrow (Spizella pallida) –3.88 36   –1.46 42
    Barn swallow (Hirundo rustica)   5.09 37   –4.68 43
    Common yellowthroat (Geothlypis trichas) –0.36 37   –2.86 43
    Sedge wren (Cistothorus platensis) –1.15 22 –10.07 23
a Percentage change per year, () P<0.10; () P<0.05; () P<0.01.

Previous Section -- Introduction
Return to Contents
Next Section -- Results

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo USA.gov logo U.S. Department of the Interior | U.S. Geological Survey
URL: http://www.npwrc.usgs.gov/resource/birds/benefit/study.htm
Page Contact Information: Webmaster
Page Last Modified: Friday, 01-Feb-2013 18:12:22 EST
Sioux Falls, SD [sdww54]