Home

secrdesignapp 1.0

This Shiny app runs in a web browser. It delivers much of the functionality of the R package secrdesign in a more accessible form.

The main purpose is to forecast the precision of estimates of animal population density. Precision is measured by the relative standard error RSE(D-hat) = SE(D-hat)/D-hat. RSE(D-hat) depends on the sample size (number of captures and recaptures) which in turn depends on the detector (trap) configuration, sampling duration, and underlying population parameters (density etc.).

We can mostly trust spatially explicit capture--recapture estimates to be unbiased or nearly so. Pathological (highly biased) results can be expected from some study designs, but these are usually designs that give very imprecise estimates and would be avoided for that reason alone.