Applications of l-band missions for environmental research

Resumen

Human activity leads to environmental changes which threaten the vegetation. Global warm ing trends, deforestation and land cover, or modified Earth's surface albedo, are sorne examples of these environmental changes. Additionally, natural hazards such as droughts and wildfires may increase their frequency and intensity in sorne regio ns of the globe due to climate shifts. In this context, remole sensing missions are needed to further our knowledge of the Earth's system, to monitor global ecosystems, and ultimately quantify changes in vegetation, carbon stocks and water resources. In particular, satellite pass ive microwave sensors provide a unique mea ns to regularly obtain global maps of the Earth's surface soil moisture (SM) and of the attenuation of the soil m icrowave em ission through the vegetation canopy (represented by the vegetation optical depth parameter; VOD). VOD is linked to the vegetation water content (WVC) and can also be related to the above-ground carbon density lmportantly, low microwave frequencies (L-band) are more sensitive to high densities of vegetation canopy and to soil moisture than higher frequencies. The first two space missions carrying an L-band radiometer on-board for measuring the Earth's SM have been launched in the last decade: the European Space Agency's (ESA) Soil Moisture and Ocean Salinity (SMOS, in 2009) and the National Aeronautics and Space Administration's (NASA) Soil Moisture Active Passive (SMAP, in 2015). The main goal of this Thesis is to assess the applicability of L-band satellite retrievals of SM and VOD on vegetation studies. The first part of this Thesis applies SM information -complementarily to other data sources- to assess and preven! drought­derived impacts on vegetation in the lberian Península: Soil moisture and land surface temperature (LST) data are applied to study the soil conditions before fire ignitions in the region. A tire risk index is developed. The forest decline in Catatonia after a severe drought in summer 2012 is modelled as a function of species, climate variables and SM. The second part of this Thesis presents the capacity to infer vegetation conditions and carbon density from VOD with two dedicated research studies: The sensitivity of L-band VOD to carbon stocks in tropical regions is compared with VOD at higher frequency bands and with optical-infrared vegetation indices. The L-band VOD shows enhanced sensitivity to carbon density in most regions. Seasonal metrics of VOD have been proposed for crop yield assessment. Results over the US Corn Belt show that VOD explains up to 78% of yield variance. The results presented in this Thesis contribute to improve present capabilities of predicting wildfires, forest decline and crop yield, as well as of monitoring forest carbon stocks, using new global information from L-band Earth observation sensors. SM and VOD are very valuable indicators to support food security and ecosystem preservation decision making, as well as to enhance present capabilities of mapping carbon stocks.