Spatial extrapolation of agrometeorological variables

Agrometeorological variables are especially subject to variations in space- a fact that makes calculations of areal mean values for e.g. evaporation ex tremely elusive. Wind, temperature, and humidity cannot be measured continu ously in space. In this paper we present simple methods to determine values of those variables for non-ideal positions from measurements at a single i deal point and test these spatial functions by measurements at three non-id eal points in a catchment. For wind, we were able to obtain a relatively go od fit by taking the sheltering effects of the catchment rims into account. Temperature was not regionalizable with simple approaches, but needs to be regionalized with spatially and temporally differentiated models. Specific humidity can be assumed to be homogeneously distributed even though there were sinks and sources for water vapor in the study area. (C) 1999 Elsevier Science B.V. All rights reserved.