Remote Sensing for Agricultural and Environment Outcomes

Resources

Slides (PDF)

Lesson 3 Slides: Geospatial Impact Evaluations and Remote Sensing of Agriculture and Environment

Code (GitHub)

https://github.com/aiddata/gie-toolkit

This GitHub repo contains code demonstrations for the entire course. This code includes examples of analysis in Stata as well as utilizing remote sensing and earth observation data through Google Earth Engine.

Related Reading

Burke, Marshall, et al. "Using satellite imagery to understand and promote sustainable development." Science 371.6535 (2021): eabe8628. Note: the article synthesizes the growing literature that uses satellite imagery to understand development outcomes, with a focus on approaches that combine imagery with machine learning. This both offers a good survey of recent satellite-based data sources, as well as of machine learning approaches to deal with the paucity of ground-based data from which to train these algorithms. The paper does not focus on impact evaluation approaches, and includes substantial detail on specific approaches, so please feel free to read only the first page and/or skim the remaining sections.

Gómez, Cristina, Joanne C. White, and Michael A. Wulder. "Optical remotely sensed time series data for land cover classification: A review." ISPRS Journal of Photogrammetry and Remote Sensing 116 (2016): 55-72. Note: One of the major types of outcome data used in GIEs is land cover classification (i.e., categorizing raster cells or land areas based on the type of land use and land cover observed). This paper offers a review of classification approaches, with an emphasis on time series options for such classification. Again, this paper includes substantial detail, so please feel free to read only the first page and/or skim the remaining sections.

Li, Shuang, et al. "High-quality vegetation index product generation: A review of NDVI time series reconstruction techniques." International Journal of Applied Earth Observation and Geoinformation 105 (2021): 102640. Note: One of the other frequently used types of outcome data in GIEs is satellite-derived vegetation indices, including the normalized difference vegetation index (NDVI). This paper offers a review of how such NDVI time series measures are typically constructed, and a good overview of the potential satellite options for NDVI measurement. Again, feel free to read only the first page and/or skim the remaining sections.