jmigueldelgado / buhayra

Real-time water body extent in drylands from Sentinel-1

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Buhayra

Buhayra (from al-buhayra) is a prototype application aiming at obtaining water extent of small reservoirs in semi-arid regions from satellite data in real-time. It collects, filters and processes weekly reservoir extents from Sentinel-1 for northeast Brazil and stores this geo-referenced information in a structured data model. This work has been funded by the German Research Foundation DFG under project number 266418622 and runs on the compute server of the Institute of Environmental Sciences and Geography of the Unierstity of Potsdam.

Preliminary results can be found on this buhayra-app. Click on the lakes to obtain plots and current state.

Before you start...

Read about configurations and setup on the wiki and create and configure your location file accordingly (in buhayra/parameters/location.yml).

The scripts are suited to work on a PBS cluster or at least a dedicated machine with large RAM. There is a crontab that schedules the jobs to run once a week or more often. Although there are conda environment files to go with this repo, some libraries are quite machine specific and the currently used environments evolve a lot due to the experimental nature of this repo. Please contact me in case youi want to use any of this.

What it does

In short, the following steps are done sequentially:

Visualization is being provided by a demo dashboard under development.

example output

An evaluation of the results is given by valbuhayra

In progress:

  • combine the water extent collection with bathymetric survey from TanDEM-X

Some of our references

We were at ESAs mapping water bodies from space 2nd conference in Frascati (Rome), and at the World Water Forum in Brasília 2018.

Shuping's talk and Martin's poster in Frascati. My talk in Brasília.

About

Real-time water body extent in drylands from Sentinel-1

License:MIT License


Languages

Language:Python 95.8%Language:Shell 4.2%