The course aims at introducing remote sensing techniques for the retrieval of geophysical parameters related to the hydrological cycle, such as soil moisture, vegetation biomass and snow parameters, by using satellites SAR and optical sensors. A particular focus will be given to Artificial Neural Networks.
Course Prerequisites
Basic knowledge about physics (electromagnetic waves), remote sensing sensors (passive and active sensors) and data (digital numbers, processing levels).
Teaching Methods
Lectures and interactive discussion and experiments using open access and commercial software.
Assessment Methods
The students will have to pass two tests: one about processing of SAR data for using SNAP and ENVI/SARscape; the other one about the implementation of an Artificial Neural Networks algorithm.
Texts
The slides used by the teacher will be provided, together with a list of relevant books and literature papers.
Contents
1. Description of the available most common satellite optical sensors: operational characteristics and working modes 2. Processing of optical satellite data a. Pre-processing of optical data b. Generation of satellite images stack for data analysis 3. Sensitivity analysis to the ground parameters a. Comparison of acquired ground data with satellite images b. Methods followed for the sensitivity analysis of satellite measurements in comparison with ground-truth data c. Integrated use of SAR and optical images for classification of natural surfaces 4. Retrieval of soil, snow and vegetation parameters a. The direct problem: forward electromagnetic models b. The inverse problem: retrieval algorithms c. Combining optical and microwave data in Artificial Neural Networks algorithms 5. Validation of retrievals and generation of thematic maps a. Validation of results using ground-truth measurements, model simulations, and ancillary data b. Generation of classification and thematic maps of soil moisture, vegetation biomass, snow parameters.
