GeoRice has been selected by ESA in 2015 as a Data User Element (DUE) Innovators III project, for a two-year duration from March 2015 to February 2017 (contract n° 4000113388/15/I-NB).
The project is motivated by the need to monitor rice production in near real-time, in a context of human and climate-induced production variability, threatening food security. The project is designed to provide valuable inputs to country stakeholders and international players, including Asia-RiCE, a component of the Group on Earth Observations Global Agricultural Monitoring (GEOGLAM) initiative.
In the first phase of GeoRice (2015-2017), the project was conducted in the Mekong Delta, one of the major rice production region in the world. The consortium was led by CESBIO for the algorithm development, with the participation of Capgemini for the implementation to a computation and visualisation platform. The following products were delivered, based on Sentinel-1 time-series imagery: seasonal rice map, cropping intensity map, and phenological stages map.
In 2018, ESA awarded a 1.5 year extension of the project (October 2018-April 2020) to CESBIO (PI) and GlobEO, to extend the work to 5 full countries in South-East Asia (Vietnam, Laos, Cambodia, Thailand, and Myanmar).
The GeoRice-extension project aims at the up-scaling of the successful proof-of-concept for rice monitoring in the Mekong delta to a regional scale demonstration.
GeoRice-extension main objectives are:
- Demonstrate large scale rice monitoring in five countries in South-East Asia (Vietnam, Lao, Cambodia, Thailand, Myanmar)
- Update and maintain the GeoRice processing system in a cloud computing environment for large scale processing
- Make the GeoRice processing system available as open source and facilitate the user up-take in the community
The project is being conducted in 4 tasks:
- Regional demonstration: the regional demonstration shall cover 5 countries in South-East Asia: Vietnam, Laos, Cambodia, Thailand and Myanmar. The processing system will be installed on an appropriate cloud-computing platform made accessible to the involved user entities. The demonstration will deliver the following EO products:
- Dynamic rice crop maps and annual anomalies
- Dynamic maps of phenological stages
- GeoRice system update : the processing system will be optimised and updated in a cloud environment
- User uptake : make available the GeoRice processing system to the user community as installation kit as well as open source code over a GeoRice website and Github. Organise one user workshop to engage the wider user community.
- Data cube support: assess the benefits to implement GeoRice algorithms in a data cube concept and identify appropriate algorithms for data cube implementation. Support the CEOS data cube in Vietnam
Why using satellite radar imagery?
One reason is the fact that in monsoon Asia, clouds, rain and haze limit the use of optical sensors, especially for fine resolution and near-real time monitoring. On the contrary, cloud cover is invisible to C-band SAR sensors such as Sentinel-1.
Another reason is the specific response of the radar backscattering of inundated vegetation, leading to specific signatures of rice crops in C-band SAR images compared to other vegetation types, as was shown in numerous publications.