FORECASTING MINIMUM TEMPERATURES ON CLEAR WINTER NIGHTS IN AN ARID REGION

Abstract

Minimum temperature formulas for clear nights in December, January, and February are developed by the Young method and tested on both original and test data. The results of these tests lead to a discussion of some weaknesses of the ?method of arbitrary corrections.? Jacobs' graphical adaptation of Brunt's equation is tested (a) using experimentally determined values for the local soil constants to compute the ?effective? soil factor, and (b) using a soil factor determined empirically for local meteorological data. With Brunt's equation as a model, the physical justification for Young's method is discussed, and a more direct approach suggested using the evening dry bulb and wet bulb temperatures and a modern method of data analysis. There is further discussion of some implicit assumptions in the Young method of analysis and an attempt is made to see if these assumptions are satisfied by the analysis performed in combining the evening dry bulb and wet bulb temperatures and the expected morning minimum using modern methods. In the Appendix, the application of the latter method is extended to cloudy nights and performance comparisons with official forecasts are made and discussed.