Challenges in the generation of LAI time series for West Africa: difficulties and potential improvements

Kurzfassung

The Leaf Area Index (LAI) is a key biophysical variable and is often used as a descriptive measure of plant canopies or vegetation in general. It is defined as the one-sided area of green leaves per unit ground area. Among various other applications, LAI can serve as a basis for the modeling and monitoring of vegetation phenology and productivity. In land surface and climate models, it is fundamentally important as an input parameter and representative of terrestrial vegetation. Processes, such as infiltration and interception of rainfall, absorption of radiation and photosynthesis are influenced by vegetation and, in particular, by the amount of foliage in plant canopies. Thus, LAI controls important exchange processes between biosphere and atmosphere. For studies with regional to global coverage, consistent LAI time series are needed which can only be derived from remote sensing. The proposed poster gives an overview over currently available LAI products and highlights strengths and weaknesses in their application in sub-humid to arid regions like West Africa. Furthermore, a compilation of the different aspects and potential improvements will be presented. At the moment, several remotely sensed LAI time series are available, for example the MOD15A2 product based on MODIS (Moderate Resolution Imaging Spectroradiometer), the LSA-SAF product based on SEVIRI (Spinning Enhanced Visible and Infrared Imager), the CYCLOPES product as well as the GEOV1 product, both based on SPOT-VGT (Satellite Pour l’Observation de la Terre-VEGETATION) and the GLOBCARBON product based on SPOT-VGT and ATSR/AATSR (Advanced Along Track Scanning Radiometer) data. For an application in sub-humid to arid regions like West Africa, all of these products have their advantages and disadvantages related to either the sensor characteristics or the underlying LAI algorithm. In areas with persistent cloud coverage like the coastal regions of West Africa, time series of sensors with sun-synchronous orbits are highly affected by data gaps. Since these sensors have a repeat cycle of 1 to 35 days, the probability of a cloud-free scene or composite is rather low in these regions. Due to the geostationary position over Africa and consequently very high temporal resolution of SEVIRI, the LSA-SAF product shows only a low fraction of data gaps. However, drawbacks are the short temporal coverage since 2006 and the rather coarse resolution of approximately 3 km. In addition, some products underestimate the LAI of densely vegetated areas such as evergreen broadleaved forests which could be attributed to a saturation effect of the applied algorithms. The complementary strengths of available sensors and LAI algorithms could result in an optimized LAI time series that is suitable for regional application in sub-humid to arid regions like West Africa. Different aspects that need to be considered for an optimized product will be presented and proposed improvements will be given. One could be the exploration of suitable data fusion approaches in order to combine advantages of different sensors. Another could be a regionally enhanced land cover base map instead of the commonly used global datasets. Furthermore, an adjusted smoothing algorithm might be needed for minimizing noise in the time series. Other aspects are the incorporation of a BRDF model or an improved cloud detection which could further enhance the quality of time series.