Estimating monthly mean air temperature using a radiative cooling scale

A new method was developed to estimate monthly mean air temperature using two different data sets: data measured over a short period at estimation sites, and data measured by the permanent meteorological observatory adjacent to estimation sites. The method developed involves estimating air temperature from the air temperature difference at an estimation site and at a permanent observation site. The air temperature is converted to potential temperature for compiling more accurate meteorological grid data. Difference of the potential temperature (Δθ) is partitioned into two contributing factors: an impact degree influenced by radiative cooling intensity at the estimation site (TESC), and an impact degree influenced by radiative cooling intensity at the permanent observation site (TSSC). A new meteorological scale “Radiative Cooling Scale” (RCS) proposed in the present study represents strength of the radiative cooling intensity determined from difference in potential temperature between an upper air pressure surface and the ground surface, since values of TESC and TSSC are related to the strength of radiative cooling intensity. The model used to estimate TESC is a linear expression that uses RCS as a variable. TSSC is also correlated with RCS. However, the model used to estimate TSSC is a multiple regression equation that uses similar explanatory variates as those used in the RCS model for more accuracy, since meteorological data is measured directly. Using the new method, it was estimated the root mean square error (RMSE) of monthly differences of the potential temperature in hilly areas to be 0.21 K.