CGIAR Crop Yield Prediction Challenge
$3 000 USD
Can you predict maize yields on East African farms using satellite data?
646 data scientists enrolled, 195 on the leaderboard
AgriculturePredictionComputer VisionUnstructuredImageSatelliteSDG2
Kenya
21 October 2020—7 February 2021
110 days
With field locations could be possible to use data from Google Earth Engine and Nasa Power.
published 3 Dec 2020, 19:25

If we have the field locations, could be possible to use data from the open platforms Google Earth Engine (https://earthengine.google.com/) and Nasa Power (https://power.larc.nasa.gov/). These platforms are free, and everyone can access their data.

We'd love to have this open, but we didn't share locations on this for a couple of reasons:

- Not everyone has the skills or access to those platforms, so keeping the data consistent helps keep the playing field level

- We're launching another competition based on some of this data early next year, which would be more difficult if the raw locations had already been shared here

Plus, if we're honest, keeping to a fixed set of allowed data makes code review and validation of the winning solutions much easier. External data adds enough complexity that our poor code review team end up spending ages trying to understand and run everything while making sure nothing cheaty is happening :)

Do you have specific datasets on Earth Engine that you think would be useful for this task? We've just shared some better climate timeseries (TERRACLIM) and soil information (ISRIC SoilGrids) which combined with the image timeseries should be plenty of info to get started with. (see fields_w_additional_data.csv now in Data section)

Am a begginer. I am getting confused with this new csv file Can you please explain how fields_w_additional_data.csv will be important or can i still succed if i use the other data provided before