Synthetic Aperture Radar (SAR) imaging from space is in its golden age. With an increasing number of space-based SAR sensors in operation, and an increasingly growing temporal and spatial resolution, combined with the characteristics of SAR: independent from daylight, cloud coverage and weather conditions, we are seeing a fast growth of applications. One of the most important (and lucrative) application areas is agriculture.
The most important technology for using SAR data in agriculture is called ‘polarimetry’. This is a technique widely used to collect physical information of land surfaces.
The basic concept of SAR polarimetry is given by the 2×2 complex scattering matrix that describes the transformation of the two-dimensional transmitted polarization. The information obtained from this process includes classification and segmentation of vegetation types (texture based), soil moisture, bare surface roughness, crop types, crop phenology, and forest/non-forest classification.
Based on this technology we can generate agricultural maps that help farmers make decisions on activities. The maps help famers do seasonal monitoring, crop type mapping, resource utilisation, and yield estimates in different areas and different growing seasons. This information is of course critically important to improve yield while minimising the use of scarce resources like water and pesticides. This in turn helps improve the world’s food security, especially in vulnerable areas.
Figure 1 above shows us a polarimetric SAR image and its respective identification of crops. The crop map and the classification of crops are products of heavy data processing, giving us the status and the types of crops of an area.
SAR Indicators for Agriculture
In an earlier article we listed 20 Earth Observation-derived indices that help farmers, based on optical images. With the growing importance of SAR remote sensing in agriculture, we will learn about similar indices based on SAR polarimetry. In the new ESA SNAP 9.0 software, there is a section for SAR applications in agriculture, that you can find under the Radar Vegetation Index. Here we can find the following indexes:
- The Dual Pol Radar Vegetation Index (DPRVI) is an indicator of crop growth. It can be obtained from Sentinel-1 SAR data and its workflow can be found in Github (here)
- The Compact Pol Radar Vegetation Index (CPRVI) is another indicator of crop growth.
- The Generalized Pol Radar Vegetation Index (GPRVI) is another SAR index that gives information about crop growth, volume and height.
All these indexes were generated for the Microwave Remote Sensing Lab (MRSLab) Indian Institute of Technology Mumbai.
Also read: 20 space acronyms that help farmers
The following figures show us an example of an application of a SAR polarimetric images composite (left) and its derived crop type map (right). You can clearly see the many crop variations this analysis can provide. And it provides this data day and night, regardless of cloud cover. In many temperate and tropical areas with lots of clouds this means that SAR provides a much better temporal resolution than its optical counterpart. In agriculture, a day can make a big difference!
There are many indexes for optical and SAR images, supporting decision making processes in agriculture. Farmers can use this information to make decisions on when to plant what, where and when to irrigate, monitor crop growth and interfere where and when needed, predict harvests, monitor seasonal changes and generally improve sustainable development.
Satellite remote sensing will also help farmers manage natural hazards like droughts and flooding, or decline of land productivity due to soil degradation caused by excessive monoculture cultivation and improper irrigation.
This article is part of two series on Groundstation.Space: