A novel method is proposed for estimating ground heat flux, sensible heat flux and evaporation at the surface from time series of soil variables and atmospheric variables. The method eliminates the need of estimating gradients required by most existing algorithms to obtain surface fluxes. It achieves this objective by exploiting diffusion analogies to the vertical movement of heat and moisture in the soil and atmosphere. In such cases the vertical gradient of a scalar variable can be expressed as the half order time derivative of the variable, which in turn is a weighted integral of the time series of the variable in question at the interface between the soil and the atmosphere. Proof-of-concept has been achieved but there are many outstanding issues and need for extensive verification with field data. Some of the questions yet to be addressed include: 1. properly representing parameters that may be soil, vegetation or state dependent; 2. expanding the concept to estimating fluxes over regions, and properly handling the heterogeneity of properties in that case; 3. explore the robustness of the assumptions of the method. T'his work promises to: 1. enhance our ability to utilize remotely sensed observations to quantify surface hydrometeorological fluxes; 2. improve ability to quantify regional energy balances; 3. suggest better parameterizations of global atmospheric and land models.
