Influence of agricultural practices on micrometerological spatial variations at local and regional scales

Soil-vegetation-atmosphere transfers significantly influence interactions and feedbacks between vegetation and boundary layer, in relation with plant phenology and water status. The current study focused on linking micrometeorological conditions to cultural practices at the local and regional scales (lower than 100km2), over an agricultural region in South Western France. This was achieved considering observation and modelling tools designed for characterising spatial variabilities over land surfaces. These tools were the ASTER high spatial resolution optical remote sensing data, and the SEBAL spatialised surface energy balance model. Surface bidirectional reflectance and brightness temperature were first derived from ASTER data through solar and thermal atmospheric radiative transfer codes, and next used to infer surface radiative properties required for model simulations. Assessing model consistency in terms of air temperature simulations gave satisfactory results when intercomparing against weather station data, although basic model assumptions were not systematically verified in terms of spatial variability. Next, spatialised simulations of evapotranspiration and air temperature were analysed at the regional and local scales, in relation with pedology, land use, and cultural practices. It was shown model estimates were consistent with the considered crops and the related cultural practices. Irrigation appeared as the main factor amongst others (soil, landuse, sowing date) explaining the micrometeorological variability. Although interesting and promising in terms of linking micrometeorological conditions to cultural practices, the results reported here emphasised several difficulties, specially about capturing subfield scale variability and monitoring the considered processes at an appropriate temporal sampling.