Social Media Icons DEQ Facebook DEQ Twitter DEQ Instagram DEQ LinkedIn

Virginia CZM Program: 2011 Coastal Grant Project Description and Final Summary

Project Task:VA CZM logo



Old Dominion University Research Foundation 

Project Title:

Bird Density in Forested Migration Stopover Sites on the Lower Delmarva Peninsula and a Comparison with Radar-based Predictive Models 

Project Description as Proposed:

All three major bird conservation plans (Brown et al. 2001; Kushlan et al 2002; Rich et al. 2004) recognize the importance of stopover habitat, and acknowledge that in many cases habitat use during migration is poorly understood. For example, habitat use may be constrained by extrinsic factors (e.g., proximity to the coast) that limit access to habitats with high food availability (Buler and Moore 2011).

The Delmarva Peninsula is a particularly important area for migratory birds in North America (Watts and Mabey 1994, Buler and Dawson 2012). The lower Delmarva Peninsula includes 5 of the top 12 designated Important Bird Areas within VA (Weldon 2007). Identifying important stopover sites is a critical step in the development of a comprehensive regional conservation plan for migratory landbirds. Historically, migratory landbird use of the Lower Delmarva has been assessed using avian surveys. In 1991, a large multistate study was undertaken in order to assess neotropical migratory song bird use of the Cape May and Delmarva peninsulas (Mabey et al. 1993). This Virginia CZM-funded study relied upon hundreds of volunteers and biologists in each of the four states involved (NJ, MD, DE, and VA). Survey efforts of this magnitude and scope are very difficult to coordinate and fund.

More recently, biologists and land managers have shifted their attention to alternative methods of assessing migratory bird use of habitats over large geographical scales. One such method is to use existing NEXRAD weather radars to monitor and assess migratory stopover sites along the Atlantic flyway (Buler and Dawson 2012). By using these existing radars, biologists have been able to map high and low density sites along much of the Eastern Seaboard in order to build predictive models. These predictive models have been developed based on NEXRAD radar data in combination with landscape habitat composition, elevation, and geographic location. The advantage of these predictive models is that land managers may be able to predict stopover density and use of migrants in areas outside the coverage of NEXRAD radars. These predictive models, however, may not be able to distinguish among neotropical migratory songbirds (a conservation target) , temperate songbirds, and in some instances waterfowl or shorebirds.

The Lower Delmarva, an internationally significant neotropical migratory songbird stopover area, is ironically outside of the coverage of NEXRAD radars (Figure 1). A radar-based predictive model has, however, been recently developed by Buler and Dawson (2012) to make predictions about bird use of hardwood forests on the Cape May and Delmarva peninsulas. Because these predictive models have never been validated, this project sets out to ground validate the models through observation at sites on the Lower Delmarva during fall migration. Predicted patterns suggest some differences from historical patterns of stopover use, which include scrub-shrub habitat. Previous survey work has demonstrated that the southern tip of the Lower Delmarva is one of the most important neotropical migrant stopover sites. Radar-based predictive models that examine only hardwood forest types, lump neotropical and temperate migrants, and do not consider “funneling effects” of peninsulas suggest other high use sites might exist in Accomack County (Figure 1).

The objectives of this project are to: 1) Assess and validate existing NEXRAD-based predictive statistical models of important hardwood forest stopover sites for the Lower Delmarva using ground survey observations; 2) Assess habitat use of migrants in relation to food abundance and habitat and landscape features. 3) Compare, where possible, data collected as part of the 1991 Neotropical Migratory Songbird Coastal Corridor Study to look for long-term changes in the distribution and abundance of neotropical songbirds over the Lower Delmarva.

Field surveys of migrant land birds will be conducted on the Lower Delmarva in both Accomack and Northampton Counties. We will repeatedly sample 10 hardwood forest sites (>4 ha) throughout the fall migration. Sites will be stratified to incorporate low to high predicted radar reflectivity. Sampling at the sites will allow validation of predictive models. Because radar-based predictive models of migratory songbird stopover habitat are currently limited to hardwood forest sites, this study will not assess scrub-shrub habitat, another highly used neotropical migrant habitat type over the Lower Delmarva.

Field surveys will be conducted by 2 trained observers at each site. Bird use (migrant and resident) of each forested site will be quantified using five 100 m length randomly placed (without overlap) transects with distance sampling. Distance sampling involves walking a transect and estimating the distance of the observed bird from the transect. This accounts for differences in detection probability and allows density estimates to be made, something that cannot be done with point counts (Buckland et al. 1993). Observers will also document foraging strata (canopy, understory, or ground) for each bird observation. Each site will be sampled roughly every three days (sunrise + 4 hours) for 12 weeks (15 Aug-7 Nov). Fruit and insect availability along each transect will be quantified using observations and branch clippings from each of the three vertical strata with every visit. Vegetation composition and habitat structure will be sampled once per season.

Federal Funding:


Project Contact:

Eric Walters, 757.683.5461; 

Project Status:

4/15/2013 - 9/30/2014; Project Completed 

Final Product Received:

Bird Density in Forested Migration Stopover Sites on the Lower Delmarva Peninsula and a Comparison with Radar-based Predictive Models (PDF)

Project Summary Provided by Grantee:

Fall migration is a time when many migratory birds undergo one of the most physically demanding and most vulnerable periods of their annual cycle. Stopover habitats along their route provide areas for rest, refueling, and shelter from predators and inclement weather. It is thus critical that important stopover habitats be identified and protected to ensure the long-term survival of migratory bird species. Ironically, many stopover habitats have largely been ignored due to their ephemeral use. Recent advances in radar-based technology have allowed for continental scale tracking of migratory species. The ability to use radar technology to develop predictive models has the potential to revolutionize our understanding of bird movement patterns over large regions of the United States.

Virginia’s Eastern Shore is a critical stopover area for millions of migrants but very little is known about site selection and the relative quality of habitats available to migrants along the Delmarva peninsula. By being able to use predictive models in regions where radar coverage is lacking, particularly globally significant areas like the Delmarva peninsula, land managers would have a valuable tool to add to their decision making regarding habitat acquisition and protection. Studies such as the type outlined here allow for a rigorous assessment of how stopover habitats are being used by migrants and a determination of what factors might be important to avian life histories. By using advanced survey methods, a more accurate assessment of migrant use can be determined.

This study, conducted at 12 hardwood-dominated stopover sites in the fall of 2013 (15 Aug to 07 Nov), found that migrant use of stopover sites varies considerably over the course of the season. Neotropical migrants fluctuated from day to day but were evident over much of the first half of the migration season. Temperate migrants, on the other hand, exhibited large increases in numbers in late October but were present in small numbers throughout the fall season. Researchers attempted to evaluate the best predictors of migrant visitation to each stopover site.

Maintaining stopover habitats that support high fruit and insect abundance during migration is important, especially for migratory birds using an area for refueling purposes, and who must consume both fruits and invertebrates to adequately meet their dietary demands. Ripe fruit was abundant and available throughout the season. Fruit availability was predicted to decline over the season as more birds moved through the habitat but there was no evidence of a decline with season. Migrants were also predicted to feed on arthropods during stopovers. But, again, researchers did not find any evidence of a decline in arthropods over the season. There appeared to be just as many arthropods in August as in late October.

This study has demonstrated the utility of using radar-based predictive models to determine potentially important stopover sites, particularly in areas where current weather radar coverage is lacking. More detailed information needs to be obtained, both at a large and fine scale, to truly start to decipher the selection and use of forested habitats by migratory landbirds. Gaining a better understanding of the needs of migrant birds, particularly as it relates to key stopover areas during large-scale movements that often span several countries, should be a priority for land managers internationally.

Disclaimer: This project summary provides the federal dollars initially awarded to the grantee. Due to underexpenditure or reprogramming of grant funds, this figure may change. For more information on the allocation of coastal grant funds, please contact Laura McKay, Virginia Coastal Program Manager, at 804.698.4323 or email:

A more detailed Scope of Work for this project is available. Please direct your request for a copy to

footer divider
footer divider
footer divider
Virginia Department of
Environmental Quality
P.O. Box 1105
Richmond, VA 23218
(804) 698-4000

Some resources on this website require Adobe Reader and Flash Player, Microsoft Word, PowerPoint or Excel. If you wish to receive this content in an accessible format pursuant to Section 508 of the federal Rehabilitation Act of 1973, as amended (29 U.S.C. ยง 794 (d)), please call 800-592-5482. In addition, this website includes hyperlinks to websites neither controlled nor sponsored by DEQ or the Commonwealth of Virginia. Links may open in a new window. If you wish to receive content from a website which is neither controlled nor sponsored by DEQ or the Commonwealth, please contact the host of that website directly.

Privacy Statement | Terms Of Use | WAI Compliance | Contact Us