Spatial Upscale and Temporal Downscale of GEDIs Leaf Area Index with Sentinel-1 and 2

Abstract:

The large-scale LiDAR products of the Global Ecosystem Dynamics Investigation (GEDI) with its high repetition rate offer promising new possibilities for a large number of ecological studies. Especially the provided leaf area index (LAI) can be an important base for a variety of ecological questions, e.g. in the context of phenology monitoring.

Due to the scattered footprints of GEDI products, no spatially continuous information can be provided by GEDI. For slowly changing areas, combining data of different overflights offers the possibility to improve coverage by time. However, studies for mid-latitude areas need to take pronounced seasonal changes into account and therefore can not assemble data over long periods of time. At the same time, spatio-temporally continuous products (e.g. LAI) would be desirable for those areas, since knowledge about seasonal changes are quite relevant for ecosystem monitoring. This study aims at integrating multisensor information to enhance the applicability of GEDI LAI data for ecological research questions.

We assume that GEDI LAI is correlated with optical and radar information that are provided by the Sentinel-1 and 2 satellites with a spatial resolution of up to 10m and a repetition about every 5 days. We use a heterogeneous region in Germany as a study area to test the possibility of using these data to develop a statistical model that is able to spatially predict the GEDI LAI information at the overflight of the Sentinel satellites. Hence, we are aiming at providing a spatio-temporally comprehensive GEDI LAI dataset with the spatial resolution and temporal repetition rate of Sentinel data, meaning a simultaneous spatial upscale and temporal downscale of the GEDI LAI, which is required for ecological monitoring.