Data Fusion in a Data-Rich Era

The application of precision agriculture requires the estimation of space-time variability at a very fine scale. A very wide variety of both remote and proximal sensors is used to supplement the sparse information from sampling. However, this poses problems for the joint analysis of an often huge amount of data with different characteristics. Geostatistical principles to data fusion of heterogeneous spatial data are illustrated. A detailed geostatistical approach to data fusion is also described followed by application to a real case. Finally, in remote sensing, the SAR/optical data fusion techniques will be shown highlighting their uniqueness. In the last decade, SAR/optical data fusion has gained a new impetus, mainly due to two important developments: firstly, the increasing availability of imagery with high spatial and spectral resolution and the global mapping of the Earth's surface; secondly, the implementation of new international space missions with the launch of SAR and optical sensors such as ESA's Copernicus program.