Extracting operative temperatures from temperatures of physical models with thermal inertia

Abstract

Temperatures of operative temperature models, particularly those of thick-walled models of larger ectotherms, lag behind and are more restricted in range than the operative temperatures they estimate. Algorithms are provided to extract estimates of instantaneous operative temperatures from model temperatures. A simple deconvolution method can be used when wind speeds are constant. An iterative estimation method must be used when wind speed varies during the monitoring period. The iterative method is sensitive to measurement error, and so uses a smoothing filter to limit instabilities. The smoothing also limits the short-term fluctuations in the estimated operative temperature. Iterative estimates of operative temperature suggested time lags of up to 90 min between predicted operative temperatures and model temperatures for desert tortoises (mass=3 kg). Differences this large could affect estimates of time available for foraging.