Estimation of latent heat fluxes with a modified Jackson model applied to sparse vegetation : application to the hapex Sahel experiment

The Jackson Model allows to estimate daily actual evaporation from the integrated daily net radiation and from the difference between surface temperature and air near-midday. Accepting a few assumptions and after mathematical manipulations of the initial model, the daily evapotranspiration can be obtained if the near-midday instantaneous sensible heat fluxes and the daily and near midday net radiation are known. This paper focus on the estimation of the instantaneous sensible heat fluxes. Two approaches are used to estimate them : the first uses a classical approach which represents the surface with a homogeneous single layer. The second model approximates the surface with a double-layer. In the latter model, one layer represents the soil and herbaceous vegetation while the second layer is assumed to represent the foliage of trees or shrubs (high vegetation) depending on the site. Finally, this double-layer model is simplified to allow its use by remote sensors observations. These three models are compared with field measurements during the Hapex-Sahel 92 experiment in Niger. The results show that the classical one-layer model is unable to express the latent heat fluxes of this sparse vegetation while the two double-layer models appear promising and need further developments.