USAID Feed the Future Global Poverty and Malnutrition Estimation Project

We combine the latest advances in satellite vegetation remote sensing and physical measurement with other publicly available data, processed using accessible machine learning techniques, to explore the potential to provide accurate, timely, and lower cost monitoring of key Feed the Future (FtF) outcome indicators such as asset poverty and nutritional status at the sub-national community level in low-income FtF countries. The resulting indicators enable higher frequency monitoring and adaptive targeting, as well as careful impact evaluation of FtF and other interventions when combined with rigorous research design around project or program participation



We thank the United States Agency for International Development for financial support under cooperative agreement # 7200AA18CA00014, “Innovations in Feed the Future Monitoring and Evaluation - Harnessing Big Data and Machine Learning to Feed the Future”. All data and written products are solely the authors’ responsibility and do not necessarily reflect the views of USAID or the United States Government.



Research Team


logo
Leiqiu Hu
Assistant Professor
Atomspheric and Earth Science
University of Alabama-Huntsville
logo
Oz Kira
Postdoctoral Associate
School of Integrative Plant Science
Cornell University
logo
Yanyan Liu
Senior Research Fellow
International Food Policy Research Institute
logo
David Matteson
Associate Professor
Statistics and Data Science
Cornell University
logo
Linden McBride
Assistant Professor
Economics
St. Mary's College of Maryland
logo
Ying Sun
Assistant Professor
School of Integrative Plant Science
Cornell University
logo
Jiaming Wen
Graduate Student
School of Integrative Plant Science
Cornell University

Papers


A framework for harmonizing multiple satellite instruments to generate a long-term global high spatial-resolution solar-induced chlorophyll fluorescence (SIF)

Jiaming Wen, Philipp Köhler, Grégory Duveiller, Nicholas C. Parazoo, Troy S. Magney, Giles J. Hooker, L.Yu, Christine Y. Chang, Ying Sun


Remote Sensing of Environment


20 January 2020


High‐Resolution Global Contiguous SIF of OCO‐2

L. Yu, Jiaming Wen, Christine Y. Chang, Christian Frankenberg, Ying Sun


Geophysical Research Letters


11 December 2018



Improved estimates of monthly land surface temperature from MODIS using a diurnal temperature cycle (DTC) model

Leiqiu Hu, Ying Sun, Gavin Collins, Peng Fu


ISPRS Journal of Photogrammetry and Remote Sensing


19 August 2020



Extraction of sub-pixel C3/C4 emissions of solar-induced chlorophyll fluorescence (SIF) using artificial neural network

Oz Kira, Ying Sun


ISPRS Journal of Photogrammetry and Remote Sensing


21 January 2020



Predicting poverty and malnutrition for targeting, mapping, monitoring, and early warning

Linden McBride, Christopher B. Barrett, Christopher Browne, Leiqiu Hu, Yanyan Liu, David S. Matteson, Ying Sun, and Jiaming Wen


Applied Economic Perspectives and Policy


July 2021




Multivariate random forest prediction of poverty and malnutrition prevalence

Chris Browne, David S. Matteson, Linden McBride, Leiqiu Hu, Yanyan Liu, Ying Sun, Jiaming Wen, Christopher B. Barrett


PLOS ONE


June 2021, forthcoming



Presentations





Project Reports




Data Sources



Monthly LST Products over the Feed-the-Future Countries

We focus on data from eleven USAID Feed the Future (FTF) priority countries: Bangladesh, Ethiopia, Ghana, Guatemala, Honduras, Kenya, Mali, Nepal, Nigeria, Senegal, and Uganda.





Harmonized SIF from GOME-2 and SCIAMACHY

This dataset was created by fusing SIF retrievals from SCanning Imaging Absorption spectroMeter for Atmospheric CHartographY (SCIAMACHY) and Global Ozone Monitoring Experiment 2 (GOME-2) onboard MetOp-A developed at German Research Center for Geosciences (GFZ)

Poverty and Malnutrition Prevalence Rates

This dataset was created to provide researchers with a free and open source dataset for the estimation and prediction of poverty and malnutrition prevalence in developing nations.


Each datum in this dataset contains 5 outcomes (stunted, wasted, healthy, poorest, underweight_bmi) which measure poverty/malnutrition prevalence rates estimated via DHS surveys at a particular point in space and time. Estimation is done at the enumeration area level (roughly a village). Alongside these outcomes are many features which may be explanatory of these outcomes, sampled as close as possible to the enumeration area in space and time as possible, unless otherwise indicated (see paper). Data was extensively preprocessed, and used in our associated paper.


Chris Browne, David S. Matteson, Linden McBride, Leiqiu Hu, Yanyan Liu,Ying Sun, Jiaming Wen, Christopher B. Barrett "Multivariate random forest prediction of poverty and malnutrition prevalence," PLOS ONE, Forthcoming.


Code

Translational User Guide

A guide to the Python code written by Chris Browne from Chris Browne, David S. Matteson, Linden McBride, Leiqiu Hu, Yanyan Liu,Ying Sun, Jiaming Wen, Christopher B. Barrett "Multivariate random forest prediction of poverty and malnutrition prevalence"," PLOS ONE, Forthcoming. and annotated code written by Medha Bulumulla.