Data quality in thermal summation development models for forensically important blowflies.

To highlight some issues regarding data quality that are significant in estimating post-mortem intervals (PMI) from maggots, the developmental constants of thermal summation models for development of Chrysomya megacephala Fabricius (Diptera: Calliphoridae) were calculated from incidental data gathered from 12 published studies, and from data generated specifically for the purpose in a single experiment. The focused experiment involved measuring the timing of five developmental landmarks at nine constant temperatures with a sampling resolution of 6-12 h, which is characteristic of other published studies. Combining data from different studies produced inconsistent results because of statistical noise introduced by (at least) disparities in temporal precision, descriptive statistics, geographical location and rearing diets. A robust experimental design to estimate a developmental model should involve at least six constant temperatures, starting at about 7 degrees C above the relevant developmental zero (D0) and going almost to the upper critical temperature, and a temporal sampling interval with a relative precision of about 10%, which requires sampling about every 2 h until hatching, about every 3 h until first ecdysis and about every 6 h until second ecdysis.