Validation of the Texas Mesonet from the aspect of the site spacing density

Abstract

The nondeterministic character of the atmosphere is the main problem with observing and forecasting atmospheric phenomena accurately. Because of this problem, a large number of observation sites with different types of sensors are used for observing the atmosphere. However, observation sites are irregularly spaced and mostly incapable of producing the spatial and temporal resolution desired for nowcasting and short-range forecasting. For this reason, various site selection techniques such as the statistical approach, the entropy approach, and the dynamical approach have been used for filling the gaps in the observation systems. In this study, a site spacing procedure is discussed for the Texas mesoscale monitoring network (Texas Mesonet). The spatial correlation functions for pressure, temperature, humidity and wind are determined to understand the characteristics of observations and atmospheric variability over the domain. The spatial correlations are examined by spectral analysis in order to define the scales that govern the parameter variations. Power spectra of the parameters are used to determine the error in estimating Fourier coefficients for varying wave numbers. Finally, the error variance in estimating the true field is estimated for each parameter as a function of site spacing. Significant decrements in error variance percentage were observed between the current site spacing (200 km) and the proposed site spacing (50 km) for Texas Mesonet. The decrements obtained were 58.6 % for pressure, 50.2% for temperature, 71.4% for humidity, 45.4% for u component of the wind and 57.7% for v component of the wind.