Estimation of Vegetation Functioning in a Drought Episode from Optical and Thermal Remote Sensing

Accurate estimation of vegetation functioning, i.e. canopy photosynthesis [gross primary production (GPP)] and evapotranspiration (ET) in a drought episode is of great importance. We investigated the capability of Landsat (TM5 and ETM7) optical/thermal observations for estimating daily GPP and ET of annual C3 grasses at a Fluxnet site (US-Var) during the 2004 prolonged drought episode in California. Vegetation properties information [i.e., Leaf Area Index (LAI), leaf chlorophyll content $(C_{\mathrm{ab}})$ , leaf water content $(C_{\mathrm{w}})$ , leaf dry matter content $(C_{\mathrm{dm}})$ , the leaf inclination distribution function (LIDF) and the senescent material content $(C_{\mathrm{s}})]$ were obtained by inversion of the optical radiative transfer routines (RTMo) of the Soil-Canopy-Observation of Photosynthesis and Energy fluxes (SCOPE) model, against the optical bands of Landsat. The values of the maximum carboxylation capacity $(V_{\mathrm{cmax}})$ , the Ball-Berry stomatal conductance parameter $(m)$ and soil resistances ( $r_{\mathrm{ss}}$ and $r_{\mathrm{bs}}$ ) were obtained by inversion of the energy balance and thermal radiative transfer routines (RTMt) of SCOPE, against the thermal band of Landsat. Finally, the retrieved properties were linearly interpolated over time and used together with half-hourly meteorological variables in the SCOPE model to estimate time series of daily GPP and ET. The results demonstrate that estimates of daily GPP and ET during a drought episode can be improved considerably by exploiting the information contained in both the optical and thermal domain together.