EFFECTS OF REGIONAL TEMPERATURE, WIND-SPEED AND SOIL WETNESS ON SPATIAL STRUCTURE OF SURFACE AIR-TEMPERATURE

Climate parameters are usually collected on some grid or pattern that is supposed to represent the unobserved neighborhood. Spatial dependence is a measure of the extent to which observed data represent the unobserved neighborhood. Geostatistical analyses provide procedures for measuring spatial dependence. In this study, semivariograms were estimated from hourly observations of screen-height air temperature obtained from a dense meteorological observation network. The range and spatial component normalized by the sill were estimated from the semivariogram in order to obtain information on the spatial structure of the air temperature. Zones of spatial correlation were delineated, using the range of the semivariogram. Scales of spatial correlation depended on the regional mean air temperature and regional wetness at the ground. The range or spatial scale for data collected in winter was larger than those in summer. The range under wet conditions was larger than under dry conditions. Effects of regional wind speed on range were different, depending on the regional mean air temperature. The normalized spatial component was used as an index for measuring continuities on the interpolation surface. The normalized spatial component generally increased as the range increased.