Advice on using Côte d'Ivoire Cocoa maps #7
Description
Hello, and thank you very much for the work you have done on this!
I’m using the 2020 and 2023 Côte d'Ivoire cocoa probability maps from the 2024a model in Google Earth Engine as part of my work with Trase, a non-profit focused on mapping supply chains and deforestation. When applying a threshold of 0.5, I’m seeing substantial areas classified as cocoa in 2020 that are not classified as cocoa in 2023. I’m trying to understand whether this reflects actual land use change or uncertainty in classifications.
Could you offer any advice on:
- Whether you think it's wise to use these thresholded maps for annual cocoa deforestation estimates between 2019–2024?
- Whether you believe the maps for one year may be more accurate than another?
- If the 2025a model is likely to be more accurate than the 2024a model?
Thanks again.