Talk Title: NASA Harvest Africa Program: Advancing the Use of Earth Observations and Machine Learning for Agriculture Monitoring for Food Security in Africa
Abstract: Global food security is predicated on identifying sustainable production systems that can adapt to and mitigate climate change. Regions like Sub-Saharan Africa, where more than 60% of the population are smallholder farmers, are particularly vulnerable given challenges related to infrastructure, access to farming inputs and sources of information, and a rapidly changing climate. Assessing (e.g., mapping) and monitoring smallholder agriculture is critical to building sustainable farms that are as productive or even more productive than large-scale farms. Recent advances in Earth observation (EO) and Artificial Intelligence/Machine Learning (AI/ML) offer scalable, affordable methods to quantify, monitor, and analyze risks to the agricultural sector at unprecedented scales. This information can help understand, plan and manage climate risk and offer insights and a quantitative basis for policies toward climate-resilient food systems. In this talk, I will describe open problems for research in AI/ML methods and applications utilizing EO for agriculture. I will provide detailed examples of the utility of methods and products, discuss some pathways from deployment to impact, and include resources to datasets, repositories, and other information to get started. I will also discuss the challenges and gaps in supporting agricultural monitoring in smallholder, low-resource regions.