The Center on Food Security and the Environment is a joint effort of the Freeman Spogli Institute for International Studies and the Stanford Woods Institute for the Environment.
616 Jane Stanford Way, #E401, Encina Hall, Staford, CA- 94305
Sonal has been working at FSE since 2018 and prior to that she was working for non-profit organizations for under privileged kids. She has Masters in Economics and Finance from Eastern Michigan University.
High resolution satellite imagery and modern machine learning methods hold the potential to fill existing data gaps in where crops are grown around the world at a sub-field level. However, high resolution crop type maps have remained challenging to create in developing regions due to a lack of ground truth labels for model development. In this work, we explore the use of crowdsourced data, Sentinel-2 and DigitalGlobe imagery, and convolutional neural networks (CNNs) for crop type mapping in India. Plantix, a free app that uses image recognition to help farmers diagnose crop diseases, logged 9 million geolocated photos from 2017–2019 in India, 2 million of which are in the states of Andhra Pradesh and Telangana in India. Crop type labels based on farmer-submitted images were added by domain experts and deep CNNs. The resulting dataset of crop type at coordinates is high in volume, but also high in noise due to location inaccuracies, submissions from out-of-field, and labeling errors. We employed a number of steps to clean the dataset, which included training a CNN on very high resolution DigitalGlobe imagery to filter for points that are within a crop field. With this cleaned dataset, we extracted Sentinel time series at each point and trained another CNN to predict the crop type at each pixel. When evaluated on the highest quality subset of crowdsourced data, the CNN distinguishes rice, cotton, and “other” crops with 74% accuracy in a 3-way classification and outperforms a random forest trained on harmonic regression features. Furthermore, model performance remains stable when low quality points are introduced into the training set. Our results illustrate the potential of non-traditional, high-volume/high-noise datasets for crop type mapping, some improvements that neural networks can achieve over random forests, and the robustness of such methods against moderate levels of training set noise. Lastly, we caution that obstacles like the lack of good Sentinel-2 cloud mask, imperfect mobile device location accuracy, and preservation of privacy while improving data access will need to be addressed before crowdsourcing can widely and reliably be used to map crops in smallholder systems.
The ocean could produce up to 75 percent more seafood than it does today and drive sustainable economic growth, holding a key role in solving global hunger. Center on Ocean Solutions Co-Director Jim Leape joined Stanford Earth Professor Roz Naylor for a conversation about food security, the current COVID-19 pandemic and how global food policies can better integrate “blue foods” from marine and freshwater systems.
"COVID-19 is disrupting processed and widely traded seafood products, such as salmon, shrimp and tuna," states Naylor. "However, locally produced and consumed food systems are actually faring much better. This is especially true for some small-scale fisheries, where local fishing groups have taken the initiative to sell seafood locally and new markets are emerging during the COVID-19 period. Production and consumption have become more tightly connected as a result."
Both Leape and Naylor are part of the global Blue Food Assessment, the first comprehensive review of aquatic foods and their roles in the global food system. Naylor will discuss the assessment with collaborators during the Virtual Ocean Dialogues on June 3rd.
The pair also highlighted promising innovations for sustainable future food systems. "Illegal fishing defeats efforts to manage the resource sustainability and cheats the fishers who are playing by the rules," Leape explains. "And we can end it. Emerging technologies are bringing much greater transparency into the fishing industry."
"If we want healthy oceans in the future, we have to be thinking about a wide range of innovations, and the institutions, financial incentives, and public trust needed to turn these innovations into real market solutions," says Naylor.
Read the full Stanford ReportQ&A >
Researchers including David Lobell analyze how human-caused climate change has impacted a water deficit in Southern Africa and might contribute to a rising food security crisis in the region.
Join us for a talk with agricultural and development economist Christopher B. Barrett, this quarter’s visiting scholar with the Center on Food Security and the Environment. Barrett is the Stephen B. and Janice G. Ashley Professor of Applied Economics and Management and an International Professor of Agriculture with Cornell’s Dyson School of Applied Economics and Management.
Professor Barrett will discuss food systems advances over the past 50 years that have promoted unprecedented reduction globally in poverty and hunger, averted considerable deforestation, and broadly improved lives, livelihoods and environments in much of the world. He’ll share perspectives on the reasons why, despite those advances, those systems increasingly fail large communities in environmental, health, and increasingly in economic terms and appear ill-suited to cope with inevitable further changes in climate, incomes, and population over the coming 50 years. Barrett will explore the new generation of innovations underway that must overcome a host of scientific and socioeconomic obstacles.
Also a Professor of Economics in the Department of Economics, Barrett is co-editor in chief of the journal Food Policy, is a faculty fellow with David R. Atkinson Center for a Sustainable Future and serves as the director of the Stimulating Agriculture and Rural Transformation (StART) Initiative housed at the Cornell International Institute for Food, Agriculture and Development.
The advent of multiple satellite systems capable of resolving smallholder agricultural plots raises possibilities for significant advances in measuring and understanding agricultural productivity in smallholder systems. However, since only imperfect yield data are typically available for model training and validation, assessing the accuracy of satellite-based estimates remains a central challenge. Leveraging a survey experiment in Mali, this study uses plot-level sorghum yield estimates, based on farmer reporting and crop cutting, to construct and evaluate estimates from three satellite-based sensors. Consistent with prior work, the analysis indicates low correlation between the ground-based yield measures (r = 0.33). Satellite greenness, as measured by the growing season peak value of the green chlorophyll vegetation index from Sentinel-2, correlates much more strongly with crop cut (r = 0.48) than with self-reported (r = 0.22) yields. Given the inevitable limitations of ground-based measures, the paper reports the results from the regressions of self-reported, crop cut, and (crop cut-calibrated) satellite sorghum yields. The regression covariates explain more than twice as much variation in calibrated satellite yields (R2 = 0.25) compared to self-reported or crop cut yields, suggesting that a satellite-based approach anchored in crop cuts can be used to track sorghum yields as well or perhaps better than traditional measures. Finally, the paper gauges the sensitivity of yield predictions to the use of Sentinel-2 versus higher-resolution imagery from Planetscope and DigitalGlobe. All three sensors exhibit similar performance, suggesting little gains from finer resolutions in this system.