How can We Predict the Crop Yield in a Certain Area? The Answer Lies in the Google Earth Engine
Published:
After learning about Google Earth Engine (GEE) from CASA0025: Building Spatial Applications with Big Data in 2024 from CASA, I was impressed by the vast potential of this tool. Not only does it provide access to extensive planetary data, but it also includes essential tools for in-depth analysis and prediction, which can be dynamically visualized and presented through its spatial application. It’s a truly full-stack process!
One of my products, created in collaboration with five other peers using GEE, is called cropinvest. This spatial application can predict crop yield for a specific crop in a designated field or area. When the application is opened, users can draw an area of interest (AOI) or select a specific area for a certain crop, and the machine will automatically calculate the yield the area has produced or will produce over a certain period. It’s simple yet powerful!
A full explanation of cropinvest can be accessed here
Creating cropinvest only required partitions of datasets from the GEE catalogue, which contains dozens of datasets that can be utilized. For this reason, it would be exciting to unlock the greater potential of GEE for another case using different datasets, such as those related to health, income inequality, climate change, or urban studies. This plan is intended for a future project.
Credit and copyright for cropinvest belongs to:
- Fitrah Ramadhan
- Zimeng Song
- Burhan Ahmad Wani
- Jiang Han
- Weixian Liang
- Fanyi Li
Housing has been an undeniable primary aspect of human life. Without a proper shelter, a person cannot fulfill his/her basic need in life. This need has given more pressure when we know that there is only a limited amount of land that can be utilized for it. On other side, the competition to obtain a proper housing is getting more intense overtime due to urbanization (more people are coming to urban area to settle in). That is the exact reason why the price for this resource is skyrocketed to the point of no return.
I was wondering if I could add one more variable to my previous bivariate map, and here they are: triangulating 1) Population Density (represented by point size); 2) Kilometers traveled per year by people who travel to work by Car or Van (gold color); and 3) Total carbon footprint per person in kgCO2e (blue color) into a trivariate map!