Technical advances for the incorporation of remote sensor images of Sentinel-2 to monitor ecosystems function in the National Parks Network of Spain.
Montserrat Escudero Clares
Nowadays, global environmental changes are a fact. Because of this, the management focused on strategies based on reduce and/or adapt to this changes is paramount. In order to achieve this goal a useful tool is monitoring the areas to be managed via sensors to focus on the strengths and weaknesses of the monitored area. This sensors pick up radiation emission data from the land surface related to the ecosystem functions. Based on this data, estimation of functional attributes which allow the characterization of features such as primary production, seasonality and vegetation phenology. This conceptual and technological framework allowed the design of a Spanish National Parks Network monitoring system. This system, called REMOTE was made in collaboration with “Organismo Autónomo de Parques Nacionales” (OAPN), the “Centro Andaluz para la Evaluación y Seguimiento del Cambio Global” (CAESCG), and TRAGSATEC. REMOTE is still improving, having emerged the need to incorporate new procedures to adequately inform about the changes in Nationals Parks and ecosystems of limited extension undergo and also inform about other aspects of ecosystemic functioning like the carbon cycle dynamic in the vegetation. Specifically, since REMOTE is based on the use of images of the MODIS sensor with a spatial resolution of 250 x 250 m and a temporal time of 16 days, it is considered convenient to incorporate images from the Sentinel-2 sensor with a higher spatial (10 x 10 m) and temporal resolution. (5 days). This work aims to evaluate the possibilities offered by these images for the REMOTE goals. For this purpose, Google Earth Engine (GEE), a new cloud computing tool that allows working with large volumes of information without downloading images, has been used. In addition, along with the Normalized Difference Vegetation Index (NDVI), which characterizes the dynamics of carbon gains, the processing of the land surface water index (LSWI) has been incorporated. This index reports on the water content in ecosystems. The processing work has been carried out in the Natural Space of Sierra Nevada, a park that we consider a model due to the diversity of ecosystems and environmental gradients it offers.
Keywords: Google Earth Engine, LSWI, NDVI, Spatial resolution, Teledetection.