Whooping Crane Migrational Habitat Use Draft (2)

Home Browse Search

Study area selection of characteristics measured on the ground Study area selection of land cover was modeled as a discrete choice in time (Manly et al. 2002). In this model, a used location was considered a choice, and each location was considered to have been chosen from a"choice set" consisting of decoy locations on conservation lands and surveyed during the same survey season (the available sample). This analysis only pertains to in- channel use locations because characteristics measured on the ground were only measured at in- channel locations. A discrete choice model will account for seasonal and temporal differences in habitat characteristics, while modeling of the underlying relationships between selection and on the ground predictor variables. The model for in-channel opportunistic locations had sample sizes of 131 used and between 2 and 14 available locations in each choice set. The model for in-channel systematic iocations had sample sizes of 37 used and between 2 and 14 available locations in each choice set. There were 8 variables considered as candidates for each of these models (Table 3), and forward model selection proceeded as described above. Study area selection of flow dependent characteristics Study area selection of flow dependent habitat characteristics was modeled as a discrete choice in time (Manly et al. 2002). In this model, a used location was considered a choice, and each location was considered to have been chosen from a"choice set" consisting of points in the available sample. A discrete choice model will account for temporal differences in flow characteristics around each the crane observation, while modeling the underlying relationships between selection and flow dependent predictor variables. The model for systematic locations had sample sizes of 55 used and 165 available transects in each choice set. The model for systematic and opportunistic locations combined had sample sizes of 155 used and 165 available locations in each choice set. There were 4 variables considered as candidates for each of these models (Table 4), and forward model selection proceeded as described above. Local area selection of flow dependent characteristics Local area selection of land cover was modeled as a discrete choice in time and space (Manly et al. 2002). In this model, a used location was considered a choice, and each location was considered to have been chosen from a"choice set" consisting of points in the available sample (the 165 base transects of the HECRAS model). A discrete choice model will account for spatial and temporal differences in habitat characteristics, while modeling of the underlying relationships between selection and flow dependent predictor variables. The model for systematic locations had sample sizes of 55 used and approximately 13 available tocations in each choice set. The model for systematic and opportunistic locations combined had sample sizes of 155 used and from 9 to 19 available locations in each choice set. There were 4 variables considered as candidates for each of these models (Table 4), and forward model selection proceeded as described above. 9