Near-Real Time Monitoring of Season Start and Dryness with Scatterometer based Soil Moisture Data in Africa
Gangkofner, Ute; Haas, E. M.; Militzer, J.
Near-real time information about the start of the wet Season (SoS) and extraordinary dry periods are of importance especially in semi-arid and sub-humid regions, where vegetation growth is significantly driven by soil moisture availability. Decadal (10 day) Soil Water Index data (SWI) values show the course of soil moisture through the years, reflecting the onset, duration, strength and number of the rainy period(s). Agricultural early warning systems require that such information is available in near-real time. The development, production and dissemination of the SWI based indicators was designed accordingly and realised in the frame of the ESA Global Monitoring for Food Security (GMFS) project.
The SWI data used in the project are derived from surface soil moisture data, which are based on scatterometers on board of the ERS and MetOp satellites. They represent the soil moisture content in the first meter of the soil in relative units (0-100%) ranging between wilting level and field capacity. A soil moisture time series is available from 1992 to present, however data gaps do not permit to have a full time series. In the frame of the ESA Climate Change Initiative (CCI) the most complete and consistent global soil moisture data record based on active and passive microwave sensors will be created, which will boost also such application studies.
The primary advantages of scatterometer derived soil moisture data in the context of an agricultural early warning service are that the data is not corrupted by cloud cover and can therefore deliver indicators where or when NDVI based methods for monitoring dryness and agricultural productivity fail. The high temporal resolution of the scatterometer data and the direct relation of soil water to plant water availability are further aspects making soil moisture data a valuable information source in this context. A thorough validation based on reports, field data and remote sensing derived biophysical datasets on rainfall and vegetation condition showed that the derived SoS and dryness indicators are reliable and have the potential to be used in operational early warning systems.