Modeling land surface fluxes using a sparse canopy model and radiometric surface temperature measurements

Land surfaces with fractional vegetation cover possess distinctive micrometeorological properties that complicate the use of energy balance models to simulate surface fluxes. The use of thermal infrared (TIR) measurements to drive surface energy balance models over such surfaces is further complicated because TIR sensors view both soil and canopy elements. A land surface energy balance model specifically designed for sparse vegetation canopies is presented that addresses both of these issues. The model consists of a one-dimensional, two-layer energy balance formulation based on a potential-resistance representation of the soil-canopy-atmosphere system. In this model, land surface TIR radiance is treated as being composed of soil and canopy components with different temperatures and emissivities. Model simulations demonstrated that modeled fluxes agreed well with observed fluxes and that deviations between observed versus modeled values were generally within the measurement error of the observed fluxes. Sensitivity analysis showed that model stability is sensitive to several input parameters, especially soil and canopy emissivity, and canopy stomatal resistance. Examination of modeled within-canopy temperature and moisture regimes demonstrated that variation in canopy leaf area index exerted strong controls on the sources and sinks of latent and sensible heat within the soil-canopy system. Finally, model simulations illustrated systematic differences between soil and vegetation temperatures that help to explain the complex relationship between surface radiometric and aerodynamic temperatures, and clearly demonstrated the need to distinguish between canopy and soil background temperatures in surface energy balance models that use TIR data to estimate land surface fluxes.