Northern Prairie Wildlife Research Center
This report is a comprehensive analysis of existing observational data (1943-99) and site evaluation data (1977-99) for locations used by whooping cranes (Grus americana) during migration through the United States portion of the Wood Buffalo-Aransas flyway. The apparent migration path, as outlined by the distribution of whooping crane observations, is very similar to that delineated in earlier reports, following a relatively straight line north-northwest from Aransas National Wildlife Refuge (NWR) to central North Dakota then curving northwest along the Missouri Coteau to the North Dakota-Saskatchewan border. The distribution of spring and fall observations were relatively similar, except for the higher frequency of fall observations on Quivira NWR and Cheyenne Bottoms State Wildlife Area in Kansas, Salt Plains NWR in Oklahoma, and in Texas. Timing of spring and fall migrations also appears similar to that described earlier and shows no changes over the 57-year period of data collection. Regardless of season, most sightings included 1-3 whooping cranes, but groups with as many as 14 and 19 cranes have been sighted in spring and fall, respectively.
The complete site evaluation database contained 1060 observations. We examined characteristics of 3 types of stopover habitats: 1) roost sites, 2) feeding sites, and 3) dual-use sites (i.e., where cranes were observed using a site for both roosting and feeding). Characteristics of sites examined included: wetland type or class (system, class, and regime), wetland size, river width, water depth, water quality, wetland substrate, wetland shoreline slope, dominant emergent vegetation, distribution of emergent vegetation, primary adjacent habitat, similar habitat within 16 km (10 mi), site descriptions, distance to feeding sites, primary potential food sources, observed foods consumed, distance to human development or to utility lines, visibility, other bird species present, site ownership, and site security.
Results revealed some new insight into whooping crane habitat use. Palustrine wetlands accounted for >75% of records in all states except Nebraska; in that state, the proportions of observations occurring on palustrine and riverine systems were both high (56.0 and 39.6% of state records, respectively). Roost sites were most common on riverine systems only in Nebraska, primarily the Platte, Niobrara, and North and Middle Loup rivers. Most of the whooping cranes found on riverine roosts were single cranes or nonfamily groups, particularly on the Platte, but we found no strong pattern in social groups on riverine roost sites or on feeding and dual-use sites. Whooping cranes were most commonly observed on wetlands having seasonal and semipermanent water regimes. Cranes were observed on a wide range of wetland sizes in both spring and fall, with no apparent pattern relative to social groups. Cranes used portions of rivers that ranged in width from 27 to 457 m and averaged 267 ± 87 (SD) m.
Many of the results in this study concur with earlier reports. Maximum depths of wetlands on which cranes were observed averaged 50.8 ± 41.4 cm (20.0 ± 16.3 inches), while specific sites within wetlands where cranes were observed feeding or roosting averaged 18.0 ± 10.7 cm (7.1 ± 4.2 inches). Most wetland shorelines were classified as having a slight slope (1 to <5% slope). In riverine systems, roosting cranes were more often observed on unvegetated sites than on vegetated sites, but palustrine roost sites had a broad range of emergent vegetation types. Most feeding sites were described as upland crops, whereas dual-use sites were more often wetlands. On upland crop sites, 83% of grain stubble was wheat stubble, 75% of row-crop stubble was corn, and 80% of green crops was winter wheat. Habitats adjacent (≤1.6 km [1 mi]) to roost sites were most frequently described as cropland (73.8%) and upland perennial cover (69.5%). Woodland habitat occurred adjacent to >70% of riverine roost sites but <8% of palustrine roost sites. There was little difference in the frequency distribution of social groups among permanent wetlands, cropland, and upland cover. We detected no patterns in distance between roost and the closest feeding sites. More than two-thirds of sites where cranes were observed were <0.8 km (0.5 mi) of human developments. Nearly half of the roost sites and two-thirds of feeding sites had unobstructed visibility of <0.40 km (0.25 mi). Private ownership accounted for >60% of all sites used by whooping cranes and >80% of feeding sites, which reflected the high use of crop fields.
The 57 years of data in the whooping crane sightings database provide a descriptive summary of whooping crane migration over a large geographic region. As an observational database, however, it is subject to a number of limitations, including 1) seasonal and regional differences in the distribution, density, and activity of individuals reporting crane sightings, 2) area or regional differences in the distribution and strength of interest of biologists to confirm observations and collect additional habitat data, and 3) varying landscape patterns that may hinder observation of cranes in some habitats. Such spatial and temporal differences affect the chance of detecting whooping cranes and therefore can bias the data set so that particular regions and habitat types may be over- or under-represented relative to actual use by migrants. Therefore, results presented here should be interpreted and used with caution. We provide recommendations on data limitations, needs for further information, and survey protocol for future monitoring of whooping cranes during migration.