Laura Beri, Master's Candidate, Deparment of Statistics, Colorado State University

Monday, April 7, 2008

11:30 a.m.; 223 Weber


Since 1979, the U.S. Bureau of Land Management (BLM) and the U.S. Minerals Management Service (MMS) have conducted aerial line transect surveys in the Beaufort and Chukchi Seas in order to study the endangered bowhead whale population. The main reason for this survey has been to determine the effect of Beaufort Sea oil and gas lease sales on this whale population. To investigate this question using the line transect data, one must first estimate the detection function, which estimates the sightability of whales at distances out from the trackline. Distance sampling techniques (Buckland et al., 2001) were used to estimate the detection function, g(y). This estimates the probability of detecting an animal at distance y from the trackline, given that it is present. Three covariates were considered during the model selection phase: Beaufort sea state, visibility and sky condition. These covariates were binned into different groups and detection functions were fit for various combinations of these covariates. Model selection methods were then used to choose the best model. The results show that visibility was the only significant covariate for estimation of sightability using only the on-transect data. For on-search data (sightings made during deviations from the trackline and flight time beyond the scheduled survey region), the results show that both Beaufort sea state and visibility were significant for estimation of sightability. Potential future uses of these results include spatial-temporal relative abundance estimation.




  home for studentspeoplecoursesdistance degreeresearchconsulting seminars resources
PRIMESSTARMAPemail us equal opportunitydisclaimerweb master • last modified: August 11 2003