Improvements in Estimation of Malaria Transmission Potential Modeling with Anthropogenic Climate Warming

This paper addresses improvements that can be made to malaria transmission potential modeling using precise environmental inputs. While many malaria models employ measures of monthly or daily temperature to drive transmission processes, use of diurnal temperature variation can more fully replicate the local ambient environment that drives organism development and behavior. Similarly, in climate change impact assessment, use of course-scale general circulation model (GCM) output lacks the spatial and temporal resolution necessary to derive robust biophysical impact projections. In this study, empirical downscaling (Hewitson and Crane 2006) was applied to a series of GCMs to predict location-specific malaria transmission potential for four Kenyan sites representing disparate climate conditions. Hourly temperatures used in driving the transmission model were derived from Parton and Logan's (1981) procedure based on daily maximum and minimum temperature. Results show raw GCM data underestimates the effect of climate change at hot and cold extremes, while overestimating it under intermediary temperatures. Use of mean monthly temperatures alone underestimates the rate of malaria development under cool conditions, but overestimates it for warmer conditions. Thus, techniques to derive high spatio-temporal resolution of modeling inputs have great potential for improving accuracy of malaria projections and likely other biophysical phenomena impacted by climate change.