Premium
Model‐based Prediction In Ecological Surveys Including Those with Incomplete Detection
Author(s) -
Melville Gavin J.,
Welsh Alan H.
Publication year - 2014
Publication title -
australian and new zealand journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.434
H-Index - 41
eISSN - 1467-842X
pISSN - 1369-1473
DOI - 10.1111/anzs.12084
Subject(s) - estimator , wildlife , aerial survey , transect , sampling (signal processing) , generalized linear model , statistics , mathematics , distance sampling , geography , likelihood function , function (biology) , maximum likelihood , ecology , computer science , econometrics , cartography , biology , filter (signal processing) , evolutionary biology , computer vision
Summary This paper explores and develops model‐based predictors for surveys of plants and wildlife including those with incomplete detection. The methodology allows for estimating a detection function to account for objects which were not detected at the time of the survey. The model‐based theory utilises generalized linear models (GLMs) and is either new or adapted from other areas of sampling. A simulation study is used to validate the estimators and comparisons are made with an integrated likelihood approach. An aerial survey of kangaroos in western New South Wales is used to illustrate the theory. The area within 50m of the aircraft is treated as a strip transect and mark‐recapture methods are used to estimate the detection function.