A method for estimating regional surface sensible heating using shelter-level air temperature and upper-air data

A diagnostic technique is developed to evaluate the 12-h time-integrated sensible heat flux at the land surface. The scheme relies on the sensitivity of the planetary boundary layer (PBL) height to the time-integral of surface sensible heating. In our prior work, it has only been possible to evaluate the PBL height and surface sensible heating at radiosonde locations. In this study, we use shelter-level (2 m) measurements of surface air temperature, available at relatively high horizontal resolution, combined with analyses of radiosonde upper-air data, in a technique for estimating both the height of the PBL and surface sensible heating at a sub-synoptic scale. The method is relatively simple and can be applied over large geographical areas. A daytime maximum of shelter-level potential temperature (theta(max)) from surface measurements, in combination with a 1200 UTC upper-air analysis constructed from synoptic upper-air reports, is used to diagnose a maximum daytime height of the planetary boundary layer. Using this height and a simple model describing the dependence of the time-change of the PBL height on the time-integral of surface sensible heating, we are able to make an evaluation of the 12 h sensible heating total at the land surface. The technique is tested in stages, first using data from the First ISLSCP Field Experiment (FIFE) and lastly, in an application to a region covering the Midwest and Central Great Plains of the United States. Surface sensible heating totals and PBL heights estimated in this manner compare well with observations made at the FIFE site for observation days in the Summer of 1987 and 1989 and the method was able to capture a 'dry down' period (an increased amount of sensible versus latent heating) which occurred at the FIFE site in August of 1989. The regional application of the method is able to depict the general east-west gradient of increasing sensible heat flux across the Midwest and Great Plains, as well as the spatial variability in both sensible heating and PBL heights caused by known variations in surface moisture. (C) 1997 Elsevier Science B.V.