A review and synthesis of habitat use by breeding birds in agricultural landscapes of Iowa.
More than 190 species of birds are known to breed in Iowa (Kent and Bendorf, 1991). The types and relative occurrences of the habitats available to these breeding birds have changed dramatically over the last several decades as a result of intensification and specialization of agriculture (Ferris et al., 1977). Cropland diversity has been reduced by shifts in crops from a mixture that included pasture, hay and small grains to row-crop monocultures of primarily corn and soybeans. Small farms have been consolidated into larger cropping units, with a concomitant decrease in noncrop habitats such as marshes, woodlands, fencerows and farmsteads. To maintain or enhance avian biodiversity in agricultural landscapes of Iowa and other parts of the Midwest, a better understanding is needed of how changes in the composition (and quality) of habitat types and their spatial arrangement affect bird species composition and abundance (Merriam, 1988; Fry, 1991). A focus on agricultural landscapes is warranted because in Iowa 93% of the land is in farms (U. S. Dep. Agric., 1994).
To date there have been numerous reports of bird use of the diverse habitats that compose the agricultural landscapes of Iowa (see Literature Cited). However, these studies have usually been conducted independently, using a variety of census methods, measures of abundance and study designs of differing levels of rigor (census count frequency, number of study sites, years studied, etc.). There is a need to standardize existing bird abundance information to improve understanding of habitat use by birds and to provide insight into the impacts of agriculture and other land-use practices on Iowa's avifauna. The objectives of our study, therefore, were to (1) review and summarize the existing data on bird abundances and nesting status in habitats typically found in the agricultural landscapes of Iowa and (2) to synthesize the bird habitat-use data into a standardized format and evaluate relative bird use of the various habitat types and landscape scenarios.
Existing information on the species composition and abundance of birds was compiled for habitats typical of the agricultural landscapes of Iowa. Data were derived from studies conducted within Iowa (32 references) and in portions of adjacent states with habitats similar to those in Iowa, specifically, northeastern Kansas (9), eastern Nebraska (3), southern Minnesota (2), southwestern Wisconsin (3), Illinois (5) and northern Missouri (2). The literature search included scientific periodicals, technical reports, theses and dissertations, state and federal agency publications and professional conference proceedings. Only bird abundance information obtained during the breeding season (May, June and July) was used to characterize habitat use. Data were compiled only for bird species that normally breed in Iowa; extremely rare breeders (Dinsmore et al., 1984) were excluded.
The bird abundance data obtained from the literature search were derived from studies that used a variety of census procedures and that reported abundance values in several different formats. To facilitate comparison among these data sets, the data were standardized by converting reported values to numbers of individual birds observed per census count per 100 ha. When values were originally reported as the number of males or territories per plot, the males were assumed to be monogamously paired and the reported values were doubled. Although such an assumption would not apply in all cases (e.g., polygynously mated males, unmated males), it was considered the best approximation in the absence of confirmatory data. In some instances, the data were inadequate to calculate bird abundances per unit area (e.g., frequency of occurrence, nest counts, total birds observed over an unspecified number of census counts or area) but did provide qualitative documentation of bird species presence in some habitats. Sometimes, reported nest densities were used, in combination with other abundance data, to set a lower bound on the bird abundance determinations.
References that provided data on bird habitat use also were reviewed for documentation of the bird species nesting in the various habitats. Additional data (Nolan, 1963; Gates and Gysel, 1978; Becker, 1981; Shalaway, 1985) on nesting habitats, particularly old fields and shrubland, that had not been reported for Iowa or its neighboring states were obtained from studies conducted elsewhere (Michigan and Indiana). Documented nesting in the various habitats also included personal observations by the authors.
Bird abundances and nesting data were compiled for 20 habitats typical of the agricultural landscapes of Iowa (Table 1). Bird use was characterized only for habitat types represented by data obtained from at least five and usually more than 10 sites, and in most instances, data for a given habitat type were available from more than one source. These data were subsequently consolidated into a single set of values for the species using each habitat type. To accomplish this, decision criteria were developed to determine the relative importance of the various sources of information in computing the standardized bird abundance values.
Table 1.--Description of major habitat categories Tilled row crops--Cornfields or soybean fields managed by wing various tillage practices Small grains Oat or wheat fields Wooded farmstead--Farm yards and buildings with scattered to moderate tree and shrub cover Farmstead shelterbelt--Parallel rows of trees and shrubs planted adjacent to farmsteads as windbreaks Herbaceous fencerow--Narrow fencelines between fields with herbaceous (primarily grass) but negligible woody vegetation Wooded fencerow--Fencelines between fields with scattered to continuous shrub and tree cover Grassed waterway-Shallow channels in row crop fields planted to grass to reduce erosion Railroad right-of-way--Active or abandoned rights-of-way with vegetation ranging from grass to scattered shrubs and trees Herbaceous roadsid--Road verges covered with herbaceous (primarily grass) but negligible woody vegetation and bounded by a fenceline and cropland CRP (Conservation Reserve Program)--Formerly cultivated land now seeded to perennial grasses (usually smooth brome) Alfalfa hayfield--Alfalfa fields mowed periodically for hay Grass hayfield--Fields with grass (cool or warm season) or grass/forb mixtures that are mowed periodically for hay Pasture--Fields with grass (cool or warm season) or grass/forb mixtures that are grazed by livestock Prairie--Tall-grass prairie consisting of native warm-season grasses and forbs Floodplain forest--Lowland deciduous forest adjacent to streams Upland forest--Deciduous forest in upland areas Shrubland--Scattered to moderately dense shrubs and small trees Old field--Early successional stages after land abandonment consisting of a mixture of fortes, grasses, and scattered shrubs and small saplings Natural marsh--mipermanent cattail marshes with about 50% open water Restored marsh--Marshes formerly drained and cultivated that have been allowed to refill and revegetate
The decision criteria were: (1) The relative rigor of studies was judged on the basis of the number of sites sampled, the area sampled and the number of census counts taken. (2) When data were available from multiple sources (multiple sites or years within the same study, or multiple studies) and the sources were of comparable rigor, the values were averaged to determine a single set of values. If the sources differed in rigor, then values from the most rigorous source (i.e., year or study) were used. (3) When bird abundance data for a particular habitat were available from both Iowa and an adjacent state, and the rigor of the data from Iowa was comparable to or better than that from the adjacent state, the values obtained from Iowa were used. (4) When a species was not recorded in a particular habitat type in any Iowa study (i.e., the habitat was censused) but was recorded in that habitat in studies in adjacent states, and the species is known to occur in Iowa during the breeding season (Dinsmore et al., 1984), the out-of-state abundance values were reduced by one half and used for Iowa. This adjustment was an attempt to approximate the probable abundance in Iowa, which is most likely less than that in adjacent states. (5) Sometimes, "pluses" or "minuses" were added to bird abundance values to indicate that the values probably were underestimates or overestimates, respectively. These were used when species abundance estimates for a given habitat were either low or high compared with those for other similar habitats or when values were derived from a source that reported abundance values either consistently high or low relative to those from other studies.
Because of substantial variation in the rigor and sampling intensity of the studies used as reference sources, the results were then simplified into abundance categories ranging from 0 through 5 (Table 2). The range of values included within each category was determined after a review of the distribution of abundances in the standardized data sets. Some species were shifted up or down one abundance category for certain habitats on the basis of the proximity of their abundance values (birds/census count/100 ha) to the class division lines and the species' pattern of abundance in other similar habitats. Frequency of occurrence data were sometimes considered when assigning species to abundance categories. When only such data were available for some species for a particular habitat, the species were assigned to category 1 (to at least indicate presence).
TABLE 2.--Categories used to index bird abundance Category Birds observed/census count/100 hectares Qualitative description 0 0 Absent 1 >0-5 Rare 2 >5-25 Occasional 3 >25-100 Common 4 >100-250 Abundant 5 >250 Very abundant
A final step in refining the bird habitat-use determinations consisted of critical reviews of the bird species habitat-use matrix by three of the authors who are professionals familiar with farmland bird communities and/or Iowa bird life. K E. Freemark and L. B. Best have studied farmland bird communities for several years in Canada and Iowa, respectively. J. J. Dinsmore is senior author of a book on birds of Iowa (Dinsmore et al., 1984). All three individuals are also familiar with bird census methods and their limitations. Each individual independently scrutinized the bird abundance values by species and by habitat; particular attention was paid to outlier values and to inconsistencies that did not seem to have a biological basis. Values were changed (13% of all values, see Appendix) only if there was consensus among the three reviewers. Furthermore, values were never changed by more than one category (e.g., changing "3" to "2" or "0" to "1").
In many instances species were added as users of various habitats where they had not previously been reported to occur. Several factors potentially could have contributed to a species being missed during census counts, and these factors were considered when adding species as users of a habitat. Additionally, some species had been documented to occur in particular habitats but their abundance values were adjusted up or down to more accurately reflect their actual abundance. The following criteria/factors were considered in making adjustments:
(1) The level of confidence in the sources of information from which abundance values were derived was taken into account in the adjustment process. Greater reliance was placed on expert opinion for those habitats or species where habitat use was less well-documented (i.e., less intensively/extensively studied).
(2) Some adjustments were made after comparing species' abundance patterns across various habitats. In some instances, species were judged to use a particular habitat even though not previously recorded there because they had been observed to use other similar habitats. For example, woodland bird species documented to use upland forest also likely use floodplain forest. Also, if a bird species had been recorded in forest and old-field habitats but not in shrubland, it was assumed to occur in shrubland as well because of the intermediate nature of this habitat. Furthermore, its abundance in shrubland was assumed to be intermediate between that reported for forests and old fields. A few bird species were judged not to use a particular habitat even though recorded there (i.e., their abundance values were changed from 1 to 0) if their occurrence in that habitat was thought to be extremely rare (i.e., anomalous). Additionally, abundance values were changed to zero in instances where the species' presence was likely the result of an "extraneous" feature not normally present in the habitat (e.g., woody vegetation in certain herbaceous strip cover habitats).
(3) Some bird species' abundances were known to differ between Iowa and adjacent states because of geographical range limits [e.g., Henslow's sparrow (scientific names of birds are given in the Appendix)]. When abundance data for such species were only available from the neighboring state(s), geographical range information was used to adjust values up or down to reflect likely abundances in Iowa more accurately. Also, when the abundance of a particular species was known to differ across Iowa because of geographical range (e.g., tufted titmouse) and the available abundance data were restricted to only a portion of the state, the values were adjusted to reflect the species' overall occurrence statewide. For some species with restricted geographical ranges. in Iowa, habitat use data were not available because census counts had not been reported for regions of the state where the species occur. Two examples are the western kingbird and blue grosbeak; both species are confined to counties along the western border of Iowa. Expert judgement based on personal observations was used to assign abundance values for these species.
(4) The occurrence of species in particular habitats often depends upon the landscape context of those habitats (e.g., Best et al., 1990). This information is frequently missing in census reports. For habitats that have not been censused extensively (e.g., small grains, wooded fencerows, railroad rights-of-way), the potential enhancement of habitat use resulting from proximity to different habitats would not have been documented completely. Consequently, certain bird species that could potentially use a particular habitat would not have been recorded there. As an example, mallard and blue-winged teal were not recorded in many of the grassland habitats where they potentially could occur, probably because marshland was not near the censused sites. In instances where a species had not been previously recorded in a given habitat, but it was judged that the species would likely use the habitat in certain landscape contexts, the species was considered a user (i.e., abundance category 1) of the habitat.
(5) Some species are extremely wary or secretive (e.g., buteo hawks, wild turkey, waterfowl, rails) and not easily approached or detected by an observer and, consequently, are frequently overlooked. Such species often are not recorded within the boundaries of transects, circular plots, etc., during census counts. Even when recorded in census counts, the abundance values for these species tend to be underestimated. Thus species wariness was considered when making adjustments.
(6) Aerial foragers (e.g., swallows, hawks) may not have been included in some census counts because they were not observed on the ground or on vegetation, but rather flying overhead. We considered aerial foragers to be users of a particular habitat if they could be expected to search for prey while flying over that habitat.
(7) Some species (e.g., common nighthawk, whip-poor-will, owls) are active during times of day (dusk, night) when census counts are not normally conducted and thus would often not be recorded. Accordingly, low abundance values were sometimes adjusted up.
(8) Some species were added as present (i.e., abundance category 1) or nesting in some habitats where not previously recorded on the basis of personal observations by one or more of the authors.
The data on nesting habitats also were incomplete; thus expert judgment was used to supplement the available documentation in determining the habitats used by each species for nesting. Nesting (not previously recorded) was added to the classification for some species in certain habitats on the basis of three criteria: (1) Nests of some species are difficult to locate (e.g., ruby-throated hummingbird) and thus less likely to be recorded for habitats where they occur; (2) nesting was added for some species m certain habitats by the process of elimination: that is, if a breeding species only occurs in one habitat type (e.g., forest or marsh), it obviously has to nest there; (3) if the documented abundance for a species was high for a particular habitat and nest sites typical for the species occur in the habitat, the species was assumed to nest there even though nesting had not been recorded previously.
Bird abundance data were summarized into a table (Appendix) to facilitate evaluation of individual species' habitat-use patterns and to compare species richness (i.e., the number of species) among habitats. Principal components analysis (SAS Institute Inc., 1988) was used as a means to assess the relative similarities in use of the 20 habitats by the assemblage of breeding birds in Iowa. In principal components analysis a multivariate data set can be summarized into one or a few components (in our case, two). Abundance values of the 144 bird species for each habitat (Appendix) were used as the input data set in the analysis.
Lastly, the numbers of bird species likely to be present during the breeding season were predicted for four combinations of habitats (landscape scenarios). To be included in the count for each scenario, bird species had to nest in at least one of the habitats included in the scenario. Such a qualification is reasonable because the synthesis was restricted to breeding birds. Four different landscape scenarios were selected that range from a row-crop monoculture to a diverse mosaic of crop and noncrop habitats. All four landscape scenarios can be found in Iowa, and they represent a progression from conditions characteristic of intensively farmed areas to those associated with more sustainable and environmentally compatible farming (Altieri and Letourneau, 1982; Pimentel et al., 1992).
RESULTS AND DISCUSSION
The quantity and quality of census data available for the various habitat types and species differed considerably. Thus, some estimates of bird abundance and nesting status are more reliable than others. The most reliable census data came from studies that sampled particular habitats types over multiple sites and years and that repeated census counts within each season. Despite the limitations of the existing database, the habitat-use profiles presented in detail in the Appendix and summarized in Table 3 are the first attempt to standardize and synthesize these type of data for agricultural landscapes such as those in Iowa. Having gone through such an exercise, one gains a greater appreciation for well-designed field studies and for the value of standardized census procedures.
[TABULAR DATA 3 OMITTED]
Species richness.--As expected, bird use of the various habitats found in the agricultural landscapes of Iowa differs greatly. Total numbers of bird species are greatest in floodplain and upland forests, and lowest in herbaceous fencerows and small grains (Table 3). Habitats with woody cover are particularly important for nesting birds. Forests, railroad right-of-way, farmstead shelterbelts and wooded fencerows have the greatest numbers (>40) and proportions (>70% of the species using the habitat) of nesting species. The fewest bird species (<14) nest in tilled row crops, herbaceous fencerows and small grains. Only 16% of the species that use tilled row crops nest there; most species visit these fields primarily to feed.
Species abundance.--Bird species are more abundant in strip-cover habitats such as wooded fencerows, railroad rights-of-way, farmstead shelterbelts and grassed waterways than in other habitat types. At least 30% of the bird species that use each of these habitats are common to very abundant (abundance categories 3-5; Table 3). Each strip-cover habitat has at least two species categorized as very abundant. Such habitats provide nest sites, particularly shrubs and trees, that usually are unavailable in the surrounding cropland, and nest densities in strip-cover habitats typically are much greater than in the adjacent cropland (Wooley et al., 1985). Furthermore, the abundance and species diversity of arthropods are greater in uncropped, strip-cover habitats with perennial vegetation than in cultivated cropland (e.g., Thomas et al., 1991; Rodenhouse et al., 1992), and bird foraging tends to be concentrated in and near such areas, particularly when crop plants are absent or small (Rodenhouse and Best, 1994, and references therein). Arthropods are a primary food source for most farmland birds during the breeding season (Martin et al., 1951). Even when avian foods are available throughout crop fields, foraging is often concentrated near the field borders (especially early in the growing season) where better protective cover is nearby. This is particularly true for woodland species that venture out into cropland to feed. The greater avalability of nest sites, food and protective cover in strip-cover habitats all contribute to the high bird abundances and species richness found there.
Altogether, 17 species are either abundant or very abundant in one or more of the strip cover habitats (Appendix). The brown-headed cowbird and red-winged blackbird have the most consistently high abundances within the strip-cover habitats. Vesper sparrows also are either abundant or very abundant in strip-cover habitats in and adjacent to row-crop fields (fencerows, grassed waterways and roadsides). The house sparrow, mourning dove, American robin and song sparrow are either abundant or very abundant in wooded strip-cover habitats (farmstead shelterbelts, wooded fencerows and railroad rights-of-way). Other species attain their greatest abundances in herbaceous strip cover (e.g., dickcissel and meadowlarks in grassed waterways).
Because strip-cover habitats are all narrow, birds would be expected to be concentrated in the relatively small area available. Our bird abundance estimates in these habitats should be used with caution, however, as abundance values may be inflated to some degree by extrapolating values derived from small, linear study plots to expressions of birds per 100 ha. Although the abundance estimates for some species in strip-cover habitats may be overestimated, the tendency for birds to concentrate in such areas is a biological reality.
Although a relatively large number of bird species use agricultural habitats, most of these species occur in low abundances (i.e., classified as rare or occasional; Table 3). No species would be considered abundant or very abundant in pastures or tilled row crops, and species abundances also are low for small grains. Only one species (red-winged blackbird) is very abundant (category 5) in an agricultural habitat and only in grass hayfields (Appendix). Only two species are classified as abundant (category 4)--the bobolink in grass hayfields and the red-winged blackbird in alfalfa hayfields, small grains and CRP fields.
Several factors contribute to the comparatively low abundance of birds in agricultural habitats. (1) Most agricultural habitats are subjected to repeated anthropogenic disturbances which include tillage, planting, cultivation, pesticide and fertilizer application, and mowing. Grazing by domestic livestock is an additional disturbance in pastureland (Jensen et al., 1990). Such disturbances destroy nests and may adversely alter habitat (e.g., Rodenhouse and Best, 1983; Best, 1986; Frawley, 1989). Furthermore, the lack of permanence of many agricultural habitats (due to crop rotation, annual set-aside, etc.), coupled with poor nesting success, disrupt site fidelity behavior in birds (Best and Rodenhouse, 1984) and may result in populations not reaching abundance levels attainable in less disturbed habitats. (2) Because of land-use practices designed to maximize crop production, the plant communities of croplands characteristically have low species diversity and generally relatively simple structure. Both may contribute to reduced bird abundance (Rotenberry, 1985). (3) Use of pesticides (insecticides and herbicides) in agricultural habitats also can reduce bird numbers either directly through acute or chronic exposure to toxic compounds or indirectly through a reduction in food resources or by habitat alteration (Grue et al., 1983; Mineau, 1991; Freemark and Boutin, 1995).
Abundance patterns for bird species in natural habitats (forest, marsh and prairie) are intermediate between those in agricultural and strip-cover habitats (Table 3). Of the natural habitats, marshes have the most species classified as abundant or very abundant. Contributing to this is the fact that pothole marshes in Iowa are typically habitat islands surrounded by agricultural cropland that is either unsuitable or marginal for wetland birds. Food resources also are likely to be more abundant in marshes than in some cropland (Voigts, 1976; Mitsch and Gosselink, 1986). The very abundant bird species in marshes are the marsh wren, common yellowthroat, yellow-headed blackbird and red-winged blackbird (Appendix). Dickcissels are abundant in prairies, and house wrens are abundant in both upland and floodplain forests.
Species habitat-use patterns.--Habitat-use patterns differ considerably among species (Fig. 1, Appendix). Species such as the red-winged blackbird and brown-headed cowbird are essentially ubiquitous and occur in relatively high abundances in many habitats. Other species (mourning dove, American robin, common yellowthroat, song sparrow, dickcissel, meadowlarks, common grackle and American goldfinch) will use all 20 habitats, but their occurrences in some of those habitats are rare.
The horned lark and killdeer are examples of species that selectively use and nest in open, sparsely vegetated habitats such as tilled row crops and pasture (Fig. 1, Appendix). The highly disturbed landscapes created by intensive agriculture have expanded the availability of habitats attractive to these species. Grassland birds, such as the grasshopper sparrow and bobolink, avoid heavily wooded areas but will use and nest in habitats created by agricultural practices that simulate native prairie (e.g., hayfields, pasture, CRP and grassed waterways) (Fig. 1). In contrast, species such as the house wren, northern cardinal and indigo bunting prefer to use and nest in woody strip cover and forest habitats and tend to avoid the more open grassland and cropland habitats.
Some bird species, particularly those inhabiting forests and marshes, exhibit very restrictive habitat use. Fourteen species use only marsh habitats, and 24 species use forest habitats exclusively. If shrublands are included with forests, the list of bird species exclusive to these wooded habitats increases to 30. No other habitats are used exclusively. The restricted use of forest and marsh habitats is even more evident if the number of nesting species is considered. Forty bird species nest exclusively in forest habitats (43 if shrublands are included), and 17 species nest only in marshes.
Several factors probably contribute to the restricted habitat-use patterns of the bird species that use either marshes or forests exclusively. Two, in particular, come to mind. (1) Many of these species are known to be area-sensitive in their habitat requirements (see review by Freemark et al., 1995) and will not use habitat patches unless they exceed a certain minimum size. Such area requirements would explain the absence of many woodland species in narrow strip-cover habitats such as wooded fencerows and farmstead shelterbelts. (2) The relative uniqueness of a particular habitat (vegetation structure and composition, food resources, etc.) compared with other surrounding habitats also influences the degree of restriction in habitat use. This is particularly true for marsh habitats in the agricultural landscapes of Iowa.
One way of assessing the relative uniqueness of various habitats from the birds' perspective is to perform a principal components analysis using the abundance values of the breeding bird species (Appendix) as the input data set. Such an approach evaluates the way in which the bird community as a whole responds to different habitats. The greater the similarity in the bird species' abundances between (among) habitats, the more "tightly" the habitats cluster in principal components space. The first two principal components plotted in Figure 2 accounted for 33 and 22% of the variance, respectively, in bird species composition and abundances among the 20 habitats. The first principal component separated the habitats primarily on the basis of the birds' affinities for woody vegetation; the second component expressed, in part, affinities for wetland/mesic environments.
Not unexpectedly, the bird communities in the two marsh habitats and in the two forest habitats were the most distinct from all other habitat types (Fig. 2). On the basis of bird abundances, the remaining wooded habitats (wooded fencerow, farmstead shelterbelt, railroad right-of-way, wooded farmstead, shrubland and old field) segregated from all other nonwooded habitats. The bird-use patterns of the herbaceous strip-cover habitats (roadsides, waterways and fencerows) and the other cropland (tilled row crops, small grains, and alfalfa and grass hayfields) and grassland (prairie, pasture and CRP) habitats were quite similar. This was particularly true for the two types of hayfields, pastures and CRP fields. Although bird species abundances are very similar among some habitats, the number of species that nest in each habitat differs (Table 3, Appendix). For example, the bird communities frequenting tilled row-crop fields and prairies are quite similar (Fig. 2), but the number of nesting species is over three times greater in prairies. Furthermore, the availability of suitable nest sites and nesting success may vary greatly among habitats because of differences in anthropogenic disturbances (i.e., mowing and grazing).
There are nine bird species listed as endangered in the state of Iowa (State of Iowa, 1994); three of these (bald eagle, red-shouldered hawk and king rail) are among the species that we evaluated. An inspection of the Appendix reveals why these species are having difficulties. As is typical of threatened and endangered birds, all three species occur in very low abundance (only abundance category 1) and have very restricted habitat-use patterns (either floodplain forest or natural marsh). The species abundance patterns for the 20 habitats provided in the Appendix also can be used to assess which species are most susceptible to habitat loss or to changes in land-use practices (i.e., used to generate tolerance indices to habitat disturbance, a la Stauffer and Best, 1980).
Landscape scenarios.--Based on field surveys of British farmland, Arnold (1983) found that the number of bird species increased from six in cropland to 19 in cropland interspersed with ditches, short and tall hedges, and woodland. We used a similar approach (see Methods) to assess effects of habitat simplification on avian biodiversity in agricultural landscapes such as those in Iowa by synthesizing the potential composition of nesting species across various combinations of habitats. Particularly in N-central Iowa, agricultural landscapes are commonly dominated by extensive fields of tilled row crops of corn and soybeans (Fig. 3, Scenario 1). Noncrop habitat consists primarily of narrow herbaceous fencerows and roadsides. Under this landscape scenario, a maximum of 18 bird species could be present and nesting. If crop diversity is expanded to include pasture and alfalfa, and a grassed waterway is included in the tilled row-crop field for soil erosion control, the maximum number of nesting species increases more than 40% to 26 species (Fig. 3, Scenario 2). With the addition of a natural (pothole) marsh, the potential number of nesting species, which now includes wetland birds, doubles to 52 species (Scenario 3). In the most diverse scenario (Scenario 4), the landscape mosaic includes two habitats with woody vegetation- woody fencerows and a farmstead windbreak. With the addition of habitats for woodland birds, the potential number of nesting species increases to 93, a fivefold increase over the first landscape scenario dominated by tilled row crops.
This analysis is a simplistic but useful approach for preliminary evaluations of landscape-level effects of habitat patterns on farmland birds. It provides "upper limits" to the number of bird species likely to be nesting in a given agricultural landscape because the input data were compiled over multiple sample sites and multiple years for most habitats. Fewer species (particularly rare species) are likely to be nesting in a given habitat type at a single site in a single breeding season.
More importantly, however, these landscape scenarios do not take into account aspects of landscape structure, other than habitat composition, that are known to influence the dynamics of bird habitat use (see reviews by Freemark, 1995; Freemark et al., 1995). Species richness, composition and abundance of birds within a given habitat are affected by patch size, the type and amount of edge, the quantity and quality of resources available, and how individuals and their resources are affected by natural disturbances and farming practices. The spatial arrangement or geometry of habitat elements in the landscape also significantly affects bird species distributions. For birds using natural habitats, less isolated patches support more bird species than more isolated patches. The nature of boundaries created by the juxtapostion of different habitats, and the presence of corridors that facilitate the movement of individuals among patches can also significantly affect bird species richness, composition and abundance. Comprehensive field studies are needed to examine interrelationships of landscape composition, landscape geometry, and farming practices and their effects on the species composition, abundance and productivity of birds (and other wildlife) in agricultural landscapes.
Concluding comments.--The habitat-use profiles presented in this paper represent the best information available given the current state of knowledge. Undoubtedly, some of our abundance estimates and nesting habitat determinations will need to be refined in the future as additional habitat-use data become available. Furthermore, the habitat types for which bird habitat-use data were summarized did not include all habitats available to birds in the agricultural landscapes of Iowa. The data presented herein were restricted to those habitats for which the best documentation of bird use exists (i.e., multiple sample sites). Several bird habitat-use characterizations are still needed, including those for small streams and ditches, terraces, and row crops under different tillage practices.
Our study has focused on an evaluation of bird habitat-use patterns in the agricultural landscapes of Iowa, but such an approach could also be applied to other states or regions. There is a need to synthesize the existing information documented in independent studies into standardized databases that would permit and improve comparative analyses of bird habitat use. Efforts such as ours facilitate evaluations of the effects of various land-use practices on the avifauna (i.e., ecological risk assessment). In addition, they provide valuable baseline data for landscape-level research and modeling that is of increasing interest to conservation and management agencies (Fahrig and Freemark, 1995; Freemark, 1995).
Acknowledgments.--The research described in this article was funded by the U.S. Environmental Protection Agency Laboratory in Corvallis, Oregon, through interagency agreements with the U.S. Fish and Wildlife Service (DW14935634-01-1) and Environment Canada (DWCN935524-01-0) and a cooperative agreement with Environment Canada (CR821795-01-0). It has been subjected to the Agency's peer and administrative review, and it has been approved for publication. C. McFarlane, C. Ribic and anonymous reviewers provided helpful comments on earlier drafts of the manuscript. This is Journal Paper J-15573 of the Iowa Agriculture and Home Economics Experiment Station, Ames, Project 2168.
[Figures 1 to 3 ILLUSTRATION OMITTED]
Altieri, M. A. and D. K. Letourneau. 1982. Vegetation management and biological control in agroecosystems. Crop Protect., 1:405-430.
Arnold, G. W. 1983. The influence of ditch and hedgerow structure, length of hedgerows, and area of woodland and garden on bird numbers on farmland. J. Appl. Ecol., 20:731-750.
Basore, N. L., L. B. Best and J. B. Wooley, Jr.. 1986. Bird nesting in Iowa no-tillage and tilled cropland. J. Wildl. Manage., 50:19-28.
Becker, B. W. 1981. Site selection and nesting success of oldfield songbirds. Ph.D. Dissertation, Michigan State Univ., East Lansing. 67 p.
Best, L. B. and B. J. Hill. 1983. Fencerows are for the birds. Iowa Bird Life, 53:16-21.
--. 1986. Conservation tillage: ecological traps for nesting birds? WildL Soc. Bull, 14:308-317.
-- and N. L. Rodenhouse. 1984. Territory preference of vesper sparrows in cropland. Wilson Bull., 96:72-82.
-- R. C. Whitmore and G. M. Booth. 1990. Use of cornfields by birds during the breeding season: the importance of edge habitat. Am. MidL Nat., 123:84-99.
-- and K. E. Freemark 1993. Landscape structure to enhance terrestrial wildlife in lowa ecosystems. Task 1: habitat characterization. Interim rep. to Environ. Protect. Agency, Environ. Res. Lab., Corvallis, Oregon. Interagency agreement identification no. DW14935634-01-1.
Bishop, R A., R. D. Andrews and R. J. Bridges. 1979. Marsh management and its relationship to vegetation, waterfowl, and muskrats. Proc. Iowa Acad. Sci., 86:50-56.
Boyd, R. L., C. L. Cink, D. Bryan and M. Joyce. 1982. Breeding populations of selected oak-hickory forests in northeastern Kansas. Am. Birds, 36:85-86.
-- and W. J. Stark 1983. Forty-sixth breeding bird census: oak-hickory forest VII. Am. Birds. 37:87.
Braband, L. A. 1979. Railroad right-of-way as wildlife habitat in Story Country, Iowa. M.S. Thesis, Iowa State Univ., Ames. 115 p.
Bryan, G. G. and L. B. Best. 1991. Bird abundance and species richness in grassed waterways in lowa row crop fields. Am. Midl. Nat., 126:90-102.
-- and -- . 1994. Avian nest density and success in grassed waterways in Iowa rowcrop fields. Wildl. Soc. Bull., 22:583-592.
Burgess, H. H., H. H. Prince and D. L. Trauger. 1965. Blue-winged teal nesting success as related to land use. J. WildL Manage., 29:89-95.
Burns, T. L. and R. B. Dahlgren. 1983. Breeding bird use of flooded dead trees in Rathbun Reservoir, Iowa, p. 99-101. In: J. W. Davis, G. A. Goodwin and R A. Oakenfels (eds.). Snag habitat management: proc. symposium. U.S. For. Serv., Gen. Tech. Rep. RM-99.
Camp, M. and L. B. Best. 1993. Bird abundance and species richness in roadsides adjacent to Iowa row crop fields. WildL Soc. BulL, 21:315-325.
-- and --. 1994. Nest density and nesting success of birds in roadsides adjacent to rowcrop fields. Am. MidL Nat., 131:347-358.
Cink, C. L. 1974. Thirty-eighth breeding bird census: floodplain tall grass prairie. Am. Birds, 28:1031-1032.
--. 1975. thirty-ninth breeding bird census: oak-hickory forest. Am. Birds, 29:1098.
-- and R. L. Boyd. 1979. Breeding bird populations of selected oak-hickory forests in northeastern Kansas. Am. Birds, 33:66-67.
-- and --. 1980. Breeding bird populations of selected oak-hickory forests in northeastern Kansas. Am. Birds, 34:104-105.
-- and -- . 1981. Breeding bird populations of selected oak-hickory forests in northeastern Kansas. Am. Birds, 35:61-62.
--, -- and W. J. Stark, Jr. 1983. Breeding bird populations of selected oak-hickory forests I in northeastern Kansas. Am. Birds, 37:86-87.
De Geus, D. W. and J. B. Bowles. 1990. Nest-box and habitat use by eastern screech owls in south- central Iowa. Iowa Bird Life, 60:57-59.
Delphey, P. J. 1991. A comparison of the bird and aquatic macroinvertebrate communities between restored and "natural" Iowa prairie wetlands. M.S. Thesis, Iowa State Univ., Ames. 85 p.
Dinsmore, J. J., T. H. Kent, D. Koenig, P. C. Petersen and D. M. Roosa. 1984. Iowa birds. Iowa State Univ. Press, Ames. 356 p.
Ducey, J. and L. Miller. 1980. Birds of an agricultural community. Neb. Bird Rev., 48:58-66.
Easterla, D. A. 1962. Avifauna of Tucker Prairie. M.S. Thesis, Univ. Missouri, Columbia. 144 p.
Fahrig, L. and K. E. Freemark. 1994. Landscape-scale effects of toxic events for ecological risk assessment, p. 193-208. In: J. Cairns, Jr. and B. Niederlehner (eds.). Ecological toxicity testing:, scale, complexity and relevance. Lewis Publishers, Boca Raton, Florida.
Farris, A. L., E. D. Klonglan and R C. Nomsen. 1977. The ring-necked pheasant in Iowa. Iowa Conserv. Comm., Des Moines. 147 p.
Frawley, B. J. 1989. The dynamics of nongame bird breeding ecology in Iowa alfalfa fields. M.S. Thesis, ; Iowa State Univ., Ames. 94 p.
-- and L. B. Best. 1991. Effects of mowing on breeding bird abundance and species composition in alfalfa fields. WildL Soc. Bull., 19:135-142.
Freemark, K E. and C. Boutin. 1995. Impacts of agricultural herbicide use on terrestrial wildlife in temperate landscapes: a review with special reference to North America. Agric., Ecosystems Environ., 52:67-91.
--, J. B. Dunning, S. J. Hejl and J. R. Probst. 1995. A landscape ecology perspective for research, conservation and management. In: T. Martin and D. Finsh (eds.). Ecology and management of neotropical migratory birds: a synthesis and review of the critical issues. Oxford Univ. Press., in press.
--. 1995. Assessing effects of agriculture on wildlife: developing a hierarchical approach for the U.S. EPA. Landscape Urban Planning, 31:99-115.
Fry, G. L. A. 1991. Conservation in agricultural ecosystems, p. 415-443. In: I. E Spellerberg, F. B. Goldsmith and M. G. Morris (eds.). The scientific management of temperate communities for conservation. Blackwell Scientific Publications, Oxford, United Kingdom.
Gates, J. E. and L. W. Gysel. 1978. Avian nest dispersion and fledging success in field-forest ecotones. Ecology, 59:871-883.
Graber, R. R. and J. W. Graber. 1963. A comparative study of bird populations in Illinois. 1900 - 1909 and 1956-1958. m. Nat. Hist. Surv. Bull, 28:383-528.
Gremaud, G. K 1983. Factors influencing nongame bird use of row crop fields. M.S. Thesis, lowa State Univ., Ames. 56 p.
Grue, C. E., W. J. Fleming, D. G. Busby and E. F. Hill. 1983. Assessing hazards of organophosphate pesticides to wildlife. Trans. N. Am. Wildl. Nat. Resour. Conf., 48:200-220.
Hayden, T. J., J. Faarborg and R. L. Clawson. 1985. Estimates of minimum area requirements for Missouri forest birds. Trans. Mo. Acad. Sci., 19:11-22.
Hemeseth, L. M. 1991. Species richness and nest productivity of marsh birds on restored prairie potholes in northern lowa. M.S. Thesis, lowa State Univ., Ames. 87 p.
Hoffman, R M. and D. Sample. 1988. Birds of wet-mesic and wet prairies in Wisconsin. Passenger Pigeon, 50:143-152.
Huser, B. and J. Probst. 1985. Breeding birds of the Sioux City Prairie. Iowa Bird Life, 55:42-45.
Iowa, State of. 1994. Iowa administrative code. Section 571, chapter 77.2, p. 1-4.
Jensen, H. P., D. Rollins and R. L. Gillen. 1990. Effect of cattle stocking density on trampling loss of simulated ground nests. Wildl. Soc. Bull, 18:71-74.
Karr, J. R. 1968. Habitat and avian diversity on strip-mined land in east-central Illinois. Condor, 70: 348-357.
Kendeigh, S. C. 1982. Bird populations in east-central Illinois: fluctuations, variations and development over a half-century. m. Biol. Monogr., 52:1-136.
Kent, T. H. and C. J. Bendorf. 1991. Official checklist of lowa birds 1991 edition. Iowa Bird Life, 61: 101-109.
Knutson, M. G. 1993. Avian diversity and demography in forested wetlands of the Upper Mississippi River, p. 20-22. In: lowa Cooperative Fish and Wildlife Research Unit Annual Report, Vol. 58. Iowa State Univ., Ames.
Lingle, G. R and W. Whitney. 1983. Forty-sixth breeding bird census: wetland sedge meadow I, II. Am. Birds, 37:101.
-- and -- . 1991. Wetland sedge meadow I, II. J. Field Ornithol., 62 (Suppl.):77-78.
Martin, A. C., H. S. Zim and A. L. Nelson. 1951. American wildlife and plants. McGraw-Hill Book Co., Inc., New York. 500 p.
Merriam, G. 1988. Landscape dynamics in farmland. Trends Ecol. Evol, 3:16-20.
Mineau, P. (ED.) 1991. Cholinesterase-inhibiting insecticides: their impact on wildlife and the environment. Elsevier, New York. 348 p.
Mitsch, W. J. and J. G. Gosslink. 1986. Wetlands. Van Nostrand Reinhold Co., New York. 539 p.
Mossman, M. J. 1988. Birds of southern Wisconsin floodplain forest. Passenger Pigeon, 50:321-337.
Nolan, V., Jr. 1963. Reproductive success of birds in a deciduous scrub habitat. Ecology, 44:305-313.
Patterson, M. P. 1994. Bird species abundance, composition and vegetation characteristics, and bird productivity in Conservation Reserve Program land in central lowa. M.S. Thesis, lowa State Univ., Ames. 54 p.
Pimentel, D., U. Stachow, D. A. Takacs, H. W. Brubaker, A. R. Dumas, J. J. Meaney, J. A. S. O'Neil, D. E. Onis and D. B. Corzillius. 1992. Conserving biological diversity in agricultural/forestry systems. BioScience, 42:354-362.
Rodenhouse, N. L. and L. B. Best. 1983. Breeding ecology of vesper sparrows in corn and soybean fields. Am. MidL Nat., 110:265-275.
-- and --. 1994. Foraging patterns of vesper sparrows (Pooecetes gramineus) breeding in cropland. Am. Midl. Nat., 131:196-206.
--, G. W. Barrett, D. M. Zimmerman and J. C. Kemp. 1992. Effects of uncultivated corridors on arthropod abundances and crop yields in soybean agroecosystems. Agric Ecosystems Environ., 38:179-191.
-- Rotenberry, J. T. 1985. The role of habitat in avian community composition: physiognomy or floristics? Oecologia, 67:213-217.
Sample, D. W. 1989. Grassland birds in southern Wisconsin: habitat preference, population trends, and response to land use changes. M.S. Thesis, Univ. Wisconsin, Madison. 588 p.
SAS Institute Inc. 1989. SAS/STAT users' guide: volume 1, version 6, 4th ed. Cary, NC. 890 p.
Shalaway, S. D. 1985. Fencerow management for nesting birds in Michigan. Wildl. Soc. Bull., 13:302-306.
Stauffer, D. F. and L. B. Best. 1980. Habitat selection by birds of riparian communities: evaluating effects of habitat alterations. J. Wildl. Manage., 44:1-15.
Strohmeyer, D. L. and L. H. Fredrickson. 1967. An evaluation of dynamited potholes in northwest Iowa. J. Wildl. Manage., 31:525-532.
Thomas, M. B., S. D. Wratten and N. W. Sotherton. 1991. Creation of `island habitats in farmland to manipulate populations of beneficial arthropods: predator densities and emigration. J. Appl. Ecol., 28:906-917.
United States Department of Agriculture. 1994. Farm numbers and land in farms. U.S. Dep. Agric., Natl. Agric. Stat. Serv. 15 p.
Voigts, D. K 1976. Aquatic invertebrate abundance in relation to changing marsh vegetation. Am. Midl. Nat., 95:313-322.
Weller, M. W. and L. H. Fredrickson. 1974. Avian ecology of a managed glacial marsh. Living Bird, 12:269-291.
-- and C. S. Spatcher. 1965. Role of habitat in the distribution and abundance of marsh birds. Iowa Agric. Home Econ. Exp. Sin. Spec. Rep. 43. 31 p.
Westemeier, R. L. and J. E Buhnerkempe. 1983. Responses of nesting wildlife to prairie grass management on prairie chicken sanctuaries in Illinois, p. 39-46: In: R Brewer (ed.). Proc. 8th North Am. Prairie Conf., W. Mich. Univ., Kalamazoo.
Wilson, M. F. 1974. Avian community organization and habitat structure. Ecology, 55:1017-1029.
Wilson, B. L. 1983a. Forty-sixth breeding bird census: farmstead. Am. Birds, 37:104.
--. 1983b. Forty-sixth breeding bird census: floodplain tall grass prairie. Am. Birds, 37:80.
--. 1983c. Forty-sixth breeding bird census: smooth brome pasture. Am. Birds, 37:104.
--. 1984a. Forty-seventh breeding bird census: farmstead. Am. Birds, 38:127-128.
--. 1984b. Forty-seventh breeding bird census: floodplain tall grass prairie. Am. Birds, 38:128.
Wooley, J., R. George, B. Ohde and W. Rybarczyk 1982. Nesting evaluations of native grass pastures and narrow-row soybeans, p. 5-6. In: R. B. Dahlgren (ed.). Proceedings of Midwest agriculture interfaces with fish and wildlife resources workshop. Iowa State Univ., Ames.
Wooley, J. B., Jr., L. B. Best and W. R. Clark. 1985. Impacts of no-till row cropping on upland wildlife. Trans. N. Am. Wildl. Nat. Resour. Conf:, 50:157-168.
Yahner, R. H. 1982. Avian nest densities and nest-site selection in shelterbelts. Wilson Bull., 94:156175.
--. 1983. Seasonal dynamics, habitat relationships, and management of avifauna in farmstead shelterbelts. J. Wild l. Manage., 47:85-104.
Zenner, G., T. LaGrange and S. White. 1990. Preliminary results-Iowa Prairie Pothole Joint Venture evaluation, 1989 and 1990. Breeding pair estimates and nest success. Unpubl. Res. Rep. Iowa Dep. Nat. Resour., Clear Lake. 7 p.
Zimmerman, J. L. and J. L. Tatschl. 1975. Floodplain birds of Weston Bend, Missouri River. Wilson Bull, 87:196-206.
LOUIS B. BEST, JAMES J. DINSMORE, and MARTHA CAMP, Department of Animal Ecology, Iowa State University, Ames 50011 KATHRYN E. FREEMARK,Canadian Wildlife Service, Environment Canada, Ottawa % U.S. Environmental Protection Agency, 200 SW 35th St., Corvallis, Oregon 97333
|Printer friendly Cite/link Email Feedback|
|Author:||Best, Louis B.; Freemark, Kathryn E.; Dinsmore, James J.; Camp, Martha|
|Publication:||The American Midland Naturalist|
|Date:||Jul 1, 1995|
|Previous Article:||Response of masked shrews (Sorex cinereus) to precipitation in irrigated and nonirrigated forests.|
|Next Article:||Abundance of gulls and other birds at landfills in northern Ohio.|