Insect phenology models are used in the practices of pest and vectored-disease management and forensic entomology. The precision of phenology model output is limited by the resolution of data used in driving models. Besides enhancing phenology model output, increasing precision of input data to the scale relevant to insects enables the incorporation of behavioral mechanisms into the models. The development of an insect can be modeled as a function of the insect's body temperature through time, when body temperature is modulated by the insect according to behavioral rules. We introduce a phenology modeling approach that predicts development as a function of insect body temperature, but does not assume that body temperatures represent random samples from the relevant local environment. Instead, insect body temperatures are predicted as a function thermal landscapes, wherein the insects make behavioral decisions about sampling from those environments. Where environmental heterogeneity occurs and is exploited by insects, this approach results in improved phenological predictions, with downstream benefits to post-mortem interval estimation, to prediction of vector transmission dynamics in time, and to pest management that depends on action at phenologically optimal times.
