Food Security
Paragraphs

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.

All Publications button
1
Publication Type
Journal Articles
Publication Date
Journal Publisher
Remote Sensing MDPI
Authors
David Lobell
Stefania Di Tommaso
Marshall Burke
Authors
Beth Duff-Brown
News Type
News
Date
Paragraphs

The U.S. government's global hunger and food security initiative, Feed the Future, has prevented 2.2 million children from experiencing malnutrition in sub-Saharan Africa, according to new research led by Stanford Health Policy's PhD candidate Tess Ryckman.

The researchers compared children’s health in 33 low- and middle-income countries in sub-Saharan Africa. In 12 of those countries, Feed the Future provided services such as agricultural assistance and financial services for farmers, as well as direct nutrition support, such as nutrient supplementation. 

The study, published online Dec. 11 in The BMJ, found a 3.9 percentage point decrease in chronic malnutrition among children served by Feed the Future, leading to 2.2 million fewer children whose development has been harmed by malnourishment.

“What we see with stunting rates is striking,” Ryckman said. “I would argue that 2 million fewer children stunted over seven years is major progress and puts a substantial dent in total stunting levels. And that’s 2 million children who will now have the levels of physical and cognitive development to allow them to reach their full potential.”

Stunting, or having a low height for a particular age, is a key indicator of child malnutrition. Children who aren’t properly nourished in their first 1,000 days are more likely to get sick more often, to perform poorly in school, grow up to be economically disadvantaged and suffer from chronic diseases, according to the World Health Organization.

A Controlled Study

Feed the Future is thought to be the world’s largest agricultural and nutrition program, with around $6 billion in funding from USAID (plus more from other federal agencies) between 2010 and 2015. Despite its size, much remains unknown about the effectiveness of the program.

The researchers analyzed survey data on almost 900,000 children younger than 5 in sub-Saharan Africa from 2000 to 2017. They compared children from the Feed the Future countries with those in countries that are not participants in the program, both before and after the program’s implementation in 2011.

The researchers found the results were even more pronounced — a 4.6 percentage point decline in stunting — when they restricted their sample to populations most likely to have been reached by program. These included children who were younger when the program began, rural areas where Feed the Future operated more intensively, and in countries where the program had greater geographic coverage.

“Our findings are certainly encouraging because it has been difficult for other programs and interventions to demonstrate impact on stunting, and this program has received a lot of funding, so it’s good to see that it’s having an impact,” Ryckman said.

Multifaceted Approach to Nutrition

Experts are divided about the best way to help the world’s 149 million malnourished children: Is assistance that directly targets nutrition, such as breastfeeding promotion or nutrient supplementation, more effective? Or is it also beneficial to tackle the problem at its root by supporting agriculture and confronting household poverty?

The authors, including Stanford Health Policy’s Eran Bendavid, MD, associate professor of medicine, and Jay Bhattacharya, MD, PhD, professor of medicine, a senior fellow (by courtesy) at the Freeman Spogli Institute of International Studies and a senior fellow senior fellow at the Stanford Institute for Economic Policy Research, said their analysis supports the value of a multifaceted approach to combating malnutrition among children, namely leveraging agriculture and food security interventions.

“Independent evaluations of large health policy programs such as Feed the Future help build the evidence base needed to tackle persistent patterns of undernutrition,” said Bendavid, an epidemiologist. “The widespread prevalence of stunting and chronic undernutrition is among the most common and yet most stubborn cause of underdevelopment in the world, and learning what works in this space is sorely needed.”

The researchers, including Stanford medical students Margot Robinson and Courtney Pederson, speculated that possible drivers of the program’s effectiveness include three features of Feed the Future’s design: its country-tailored approach; its focus on underlying drivers of nutrition, such as empowering female farmers; and its large scale and adequate funding.

The authors hope their independent evaluation of the program might lead to more funding and support for it. At the very least, they said, it should demonstrate to people working on Feed the Future and the broader global nutrition program community that programs focused mostly on agriculture and food security — indirect contributors to malnutrition — can lead to success.

Value Unknown

Feed the Future has been scaled back in recent years — it once served 19 countries and now reaches only 12. The program’s budget also remains somewhat murky.

“While there isn’t much data on the program’s funding under the Trump administration, the program appears to have been scaled back, at least in terms of the countries where it operates,” Ryckman said. “It’s possible that some of these gains could be lost, absent longer-term intervention from Feed the Future.”

Image
eran

The researchers also did not look at whether the program provided high value for the money spent.

“While we find that it has been effective, it hasn’t led to drastic declines in stunting and it is unclear whether it is good value for money,” she said.

Ryckman also noted that USAID’s own evaluation of its program is tenuous because it looked only at before-and-after stunting levels in Feed the Future countries without comparing the results to a control group or adjusting for other sources of bias, which is problematic because stunting is slowly declining in most countries.

“These types of evaluations are misleading,” Ryckman said. “The U.S. government really needs to prioritize having their programs independently evaluated using more robust methods. That was part of our motivation for doing this study.”

Support for the study was provided by the National Institutes of Health (grant P20-AG17253), the National Science Foundation and the Doris Duke Charitable Foundation.

 

All News button
1
Paragraphs

The economic costs of Indonesia’s 2015 forest fires are estimated to exceed US $16 billion, with more than 100,000 premature deaths. On several days the fires emitted more carbon dioxide than the entire United States economy. Here, we combine detailed geospatial data on fire and local climatic conditions with rich administrative data to assess the underlying causes of Indonesia’s forest fires at district and village scales. We find that El Niño events explain most of the year-on-year variation in fire. The creation of new districts increases fire and exacerbates the El Niño impacts on fire. We also find that regional economic growth has gone hand-in-hand with the use of fire in rural districts. We proceed with a 30,000-village case study of the 2015 fire season on Sumatra and Kalimantan and ask which villages, for a given level of spatial fire risk, are more likely to have fire. Villages more likely to burn tend to be more remote, to be considerably less developed, and to have a history of using fire for agriculture. Although central and district level policies and regional economic development have generally contributed to voracious environmental degradation, the close link between poverty and fire at the village level suggests that the current policy push for village development might offer opportunities to reverse this trend.


  •  
All Publications button
1
Publication Type
Journal Articles
Publication Date
Journal Publisher
World Development Journal
Authors
Rosamond L. Naylor
Paragraphs

As the global population and people’s incomes rise, the demand for ocean-derived food will continue to grow. At the same time, hunger and malnutrition continues to be a challenge in many countries, particularly in rural or developing areas. Looking to the ocean as a source of protein produced using low-carbon methodologies will be critical for food security, nutrition and economic stability, especially in coastal countries where hunger and malnutrition are a challenge. Yet these advances in ocean production can only be achieved with a concurrent focus on addressing threats to ocean health, such as climate change and overfishing.

All Publications button
1
Publication Type
Conference Memos
Publication Date
Journal Publisher
High Level Panel for a Sustainable Ocean Economy
Authors
Rosamond L. Naylor
Paragraphs

Understanding the determinants of agricultural productivity requires accurate measurement of crop output and yield. In smallholder production systems across low- and middle-income countries, crop yields have traditionally been assessed based on farmer-reported production and land areas in household/farm surveys, occasionally by objective crop cuts for a sub-section of a farmer’s plot, and rarely using full-plot harvests. In parallel, satellite data continue to improve in terms of spatial, temporal, and spectral resolution needed to discern performance on smallholder plots. This study evaluates ground- and satellite-based approaches to estimating crop yields and yield responsiveness to inputs, using data on maize from Eastern Uganda. Using unique, simultaneous ground data on yields based on farmer reporting, sub-plot crop cutting, and full-plot harvests across hundreds of smallholder plots, we document large discrepancies among the ground-based measures, particularly among yields based on farmer-reporting versus sub-plot or full-plot crop cutting. Compared to yield measures based on either farmer-reporting or sub-plot crop cutting, satellite-based yield measures explain as much or more variation in yields based on (gold-standard) full-plot crop cuts. Further, estimates of the association between maize yield and various production factors (e.g., fertilizer, soil quality) are similar across crop cut- and satellite-based yield measures, with the use of the latter at times leading to more significant results due to larger sample sizes. Overall, the results suggest a substantial role for satellite-based yield estimation in measuring and understanding agricultural productivity in the developing world.

All Publications button
1
Publication Type
Journal Articles
Publication Date
Journal Publisher
American Journal of Agricultural Economics
Authors
David Lobell
Marshall Burke
George Azzari, Sydney Gourlay, Zhenong Jin, Talip Kilic, Siobhan Murray
Paragraphs

Feeding a growing population while reducing negative environmental impacts is one of the greatest challenges of the coming decades. We show that microsatellite data can be used to detect the impact of sustainable intensification interventions at large scales and to target the fields that would benefit the most, thereby doubling yield gains. Our work reveals that satellite data provide a scalable approach to sustainably increase food production.

All Publications button
1
Publication Type
Journal Articles
Publication Date
Journal Publisher
Nature Sustainability
Authors
Meha Jain, Balwinder-Singh, Preeti Rao, Amit K. Srivastava, Shishpal Poonia, Jennifer Blesh, George Azzari, Andrew J. McDonald
David Lobell
-

Fighting to End Hunger at Home & Abroad:  Ambassador Ertharin Cousin shares her journey & lessons learned

A Conversation in Global Health with Ertharin Cousin

FSI Payne Distinguished Lecturer | Former Executive Director of the World Food Programme | TIME's 100 Most Influential People

RSVP for conversation & lunch: www.tinyurl.com/CIGHErtharinCousin (please arrive at 11:45 am for lunch)

Professor Ertharin Cousin has been fighting to end global hunger for decades. As executive director of the World Food Programme from 2012 until 2017, she led the world’s largest humanitarian organization with 14,000 staff serving 80 million vulnerable people across 75 countries. As the US ambassador to the UN Agencies for Food and Agriculture, she served as the US representative for all food, agriculture, and nutrition related issues.

Prior to her global work, Cousin lead the domestic fight to end hunger. As chief operating officer at America’s Second Harvest (now Feeding America), she oversaw operations for a confederation of 200 food banks across America that served more than 50,000,000 meals per year.

Stanford School of Medicine Senior Communications Strategist Paul Costello will interview Professor Cousin about her experiences, unique pathway, and the way forward for ending the global hunger crisis.

cid:image002.png@01D509A2.91178F90cid:image003.png@01D509A2.91178F90cid:image004.png@01D509A2.91178F90cid:image005.jpg@01D50A42.AF28BEA0

Li Ka Shing Room 320 

Seminars
-

Massive changes in the global food sector over the next few decades – driven by climate change and other environmental stresses, growing population and income, advances in technology, and shifts in policies and trade patterns – will have profound implications for the oceans. Roz Naylor, Senior Fellow and Founding Director of Stanford’s Center on Food Security and the Environment,  will discuss the interplay between terrestrial and marine food systems, highlighting the rising role of aquaculture in helping to meet the nutritional demands of 9-10 billion people by 2050. As a platform for her talk, she will introduce a new research initiative at Stanford on “Oceans and the Future of Food”, co-led by the Center for Oceans Solutions (COS) and the Center on Food Security and the Environment (FSE).

Free Admission is by reservation only. Please call 831-655-6200 between 8:30AM – 5:00PM, Mon-Fri, or RSVP at the Friends of Hopkins web page.

Contact:
Amanda Whitmire
831-655-6200
thalassa@stanford.edu

Boat Works Lecture Hall, Hopkins Marine Station

Lectures
Paragraphs

Accurate prediction of crop yields in developing countries in advance of harvest time is central to preventing famine, improving food security, and sustainable development of agriculture. Existing techniques are expensive and difficult to scale as they require locally collected survey data. Approaches utilizing remotely sensed data, such as satellite imagery, potentially provide a cheap, equally effective alternative. Our work shows promising results in predicting soybean crop yields in Argentina using deep learning techniques. We also achieve satisfactory results with a transfer learning approach to predict Brazil soybean harvests with a smaller amount of data. The motivation for transfer learning is that the success of deep learning models is largely dependent on abundant ground truth training data. Successful crop yield prediction with deep learning in regions with little training data relies on the ability to fine-tune pre-trained models.

All Publications button
1
Publication Type
Working Papers
Publication Date
Journal Publisher
COMPASS '18 Proceedings of the 1st ACM SIGCAS Conference on Computing and Sustainable Societies
Authors
Anna Wang
Caelin Tran
Nikhil Desai
David Lobell
Stefano Ermon
Paragraphs

Crop type mapping at the field level is necessary for a variety of applications in agricultural monitoring and food security. As remote sensing imagery continues to increase in spatial and temporal resolution, it is becoming an increasingly powerful raw input from which to create crop type maps. Still, automated crop type mapping remains constrained by a lack of field-level crop labels for training supervised classification models. In this study, we explore the use of random forests transferred across geographic distance and time and unsupervised methods in conjunction with aggregate crop statistics for crop type mapping in the US Midwest, where we simulated the label-poor setting by depriving the models of labels in various states and years. We validated our methodology using available 30 m spatial resolution crop type labels from the US Department of Agriculture's Cropland Data Layer (CDL). Using Google Earth Engine, we computed Fourier transforms (or harmonic regressions) on the time series of Landsat Surface Reflectance and derived vegetation indices, and extracted the coefficients as features for machine learning models. We found that random forests trained on regions and years similar in growing degree days (GDD) transfer to the target region with accuracies consistently exceeding 80%. Accuracies decrease as differences in GDD expand. Unsupervised Gaussian mixture models (GMM) with class labels derived using county-level crop statistics classify crops less consistently but require no field-level labels for training. GMM achieves over 85% accuracy in states with low crop diversity (Illinois, Iowa, Indiana, Nebraska), but performs sometimes no better than random when high crop diversity interferes with clustering (North Dakota, South Dakota, Wisconsin, Michigan). Under the appropriate conditions, these methods offer options for field-resolution crop type mapping in regions around the world with few or no ground labels.

All Publications button
1
Publication Type
Journal Articles
Publication Date
Journal Publisher
Remote Sensing of Environment
Authors
Sherrie Wang
George Azzari
David Lobell
Subscribe to Food Security