Mapping Frost-Sensitive Areas with a Three-Dimensional Local-Scale Numerical Model. Part I. Physical and Numerical Aspects

Abstract

Radiative frost is one of the most severe weather conditions that affects agricultural activities in many parts of the world. Since various protective methods to reduce frost impact are available, refinements of frost forecasting methodologies should provide economical benefits. In the present study, a three-dimensional numerical local-scale model for the simulation of the microclimate near the ground surface of nonhomogeneous regions during radiative frost events was developed. The model is based on the equations of motion, heat, humidity and continuity in the atmosphere and the equations of heat and moisture diffusion in the soil. Emphasis was given in establishing a refined formulation of energy budget equations for soil surface and plant canopy Additionally, an improved finite difference scheme procedure for approximating horizontal derivatives in a terrain-following coordinate system was introduced. The sensitivity of the model to various parameters that way affect the nocturnal minimum temperature near ground surface during radiative frost events was tested by using one- and two-dimensional versions of the model. This temperature was found to be sensitive to topography, plant cover, soil moisture content, air specific humidity and wind velocity.