A simple method for estimating daily and monthly mean temperatures from daily minima and maxima

Historical near-surface air temperature data from many locations are available as time series of daily minimum and maximum temperatures. However, estimation of the daily and monthly mean temperatures as the average of the minimum and the maximum is prone to large errors. Therefore, a simple method for estimating the mean daily temperature on the basis of minima and maxima is presented. The proposed method accounts for the temperature trend by including the minimum temperature on the following day in addition to the extremes on the day in question. This and a few conventional methods are tested on the temperature time series from a European flat terrain site and from an Alpine valley site. The proposed algorithm approximates the value of the actual mean daily and monthly temperatures much better than the average of extreme temperatures. Furthermore, its bias and root mean squared error on daily and monthly timescales approach those of the algorithms that require several temperature records per day. Copyright (C) 2006 Royal Meteorological Society.