Whooping Crane Migrational Habitat Use Draft (2)

Home Browse Search

Land cover parameters were obtained from the Bureau of Reclamation (BOR) land use/cover GIS based on 1998 photography (USBOR 2000). Amalgamation of the vegetation classes (VEG DESC) resulted in nine land cover categories. Agriculture included agriculture alfalfa, agriculture bare ground, agriculture corn, agriculture mown field, agriculture other crops, agriculture soy bean, and agriculture other crop. Development included development commercial, development residential, and development single dwelling. Grass included lowland grasses, mown lowland grasses, and upland grasses. Open water included open water pit, pond or lake, open water, open water slough, and open water canal. Shrub/forest included shrubs inside floodplain, shrubs outside floodplain, wooded river within floodplain, woody outside floodplain. Transportation included bridge, other road, railroad, gravel road, interstate, and paved road. Wetted channel included wetted channel and barren beach/bar. Wet vegetation included herbaceous riparian, emergents, and wet meadow mosaic. Wet grass included all land cover classes in the grass and wet vegetation categories. We extracted the distance from each point to the nearest location of each landscape category (meters) and the proportion of each of the landscape category within a 0.25 mile radius buffer. The distance variables were squared to allow for possible non-linear effects. For the models relating landscape level characteristics to crane group selection at a local scale, we defined the available set of points as the subset of the study area available sample of points that fell within 3.2 miles of the used location. We defined the available sample of points for the wetted channel models at a local scale as the subset of points that fell within the 'wetted channel' designation of the 1998 land cover/use GIS layer and 3.2 miles of the used location. The distance for the local area definition was obtained through the movement pattern summary of all the data (see above). For the models relating characteristics measured on the ground to crane group selection in the wetted channel, we defined the available set of points as the set of random locations where decoys were placed during the same survey season. On the ground measurements were obtained by survey crews at the time the profiles were measured. The percentage of fine sand (less than lmm), coarse sand (1-4.9mm), small gravel (5-14.9mm), and large gravel (greater than 15mm) were occularly estimated as the channel was waded. The distance to visual obstructions (objects (e.g., vegetation, bank, etc.) >1.5m above water line) was estimated as the average of the shortest distance to visual obstructions in four quadrants centered on the crane group location. We also estimated the area around the crane use location which was clear of visual obstructions by the product of the squared average distance and pi. Unobstructed width was taken as the distance between obstructions >1.5m in height through which an observer could not see (e.g., dense vegetation) along a line perpendicular to the channel and passing through the crane observation. This distance variable was squared to allow for possible non-linear effects. For the models relating flow dependent characteristics of the wetted channel to crane group selection, we employed the HECRAS 1D hydraulic model constructed by Mohammed Samad for the Platte River Unsteady Flow and Bank Storage Model (Randle and Samad 2007) to estimate the water surface differential for the purpose of adjusting flow dependent parameters. The water surface differential corrects for the change in water surface elevation between the time the river profiles were measured and the time of crane group use. Due to the endangered species status of the whooping crane, we were unable to take measurements of the channel while the birds were in