Local temperature estimation under climate change

Abstract

A methodology to estimate the space-time distribution of daily mean temperature under climate change is developed and applied to a central Nebraska case study. The approach is based on the analysis of the Markov properties of atmospheric circulation pattern (CP) types, and a stochastic linkage between daily (here 500hPa) CP types and daily mean temperatures. Historical data and general circulation model (GCM) output of daily CP corresponding to 1 × CO2 and 2 × CO2 scenarios are considered. The relationship between spatially averaged geopotential height of the 500 hPa surface — within each CP type — and daily mean temperature is described by a nonparametric regression technique. Time series of daily mean temperatures corresponding to each of these cases are simulated and their statistical properties are compared. Under the climate of central Nebraska, the space-time response of daily mean temperature to global climate change is variable. In general, a warmer climate appears to cause about 5°C increase in the winter months, a smaller increase in other months with no change in July and August. The sensitivity of the results to the GCM utilized should be considered.