Robust relationship between air quality and infant mortality in Africa
We know that breathing dirty air is bad for human health, and recognition of this fact has caused many countries in the developed world to adopt policies to clean up their air. Many developing countries, though, still face very high levels of air pollution, and we still know surprisingly little about what these high exposure levels mean for health. Relative to these other threats, how important is the harm caused by air pollution? Answering this question is important for figuring out policy priorities, however, in most poor countries, air quality monitors are unavailable, and traditional methods for monitoring air quality don’t adequately account for the burning of biomass – things like wood and animal waste – common in these countries.
Satellite detection of rising maize yield heterogeneity in the U.S. Midwest
The future trajectory of crop yields in the United States will influence food supply and land use worldwide. We examine maize and soybean yields for 2000–2015 in the Midwestern U.S. using a new satellite-based dataset on crop yields at 30m resolution. We quantify heterogeneity both within and between fields, and find that the difference between average and top yielding fields is typically below 30% for both maize and soybean, as expected in advanced agricultural regions. In most counties, within-field heterogeneity is at least half as large as overall heterogeneity, illustrating the importance of non-management factors such as soil and landscape position. Surprisingly, we find that yield heterogeneity is rising in maize, both between and within fields, with average yield differences between the best and worst soils more than doubling since 2000.
Using remotely sensed temperature to estimate climate response functions
Temperature data are commonly used to estimate the sensitivity of societally relevant outcomes to ongoing climate changes. In many tropical regions, however, temperature measures are often very sparse and unreliable, limiting our ability to understand climate change impacts. For this project we evaluated satellite measures of near-surface temperature (Ts) as an alternative to traditional air temperatures (Ta) from weather stations, and in particular their ability to replace Ta in econometric estimation of climate response functions. We find that satellite data performs as well as ground station data in areas of high station density and better than ground stations in areas of low station density.
death per 1,000 child-years
Understanding Spatial Patterns of Child Mortality in Sub-Saharan Africa
Detailed spatial understanding of levels and trends in under-5 mortality is needed to improve the targeting of interventions to the areas of highest need, and to understand the sources of variation in mortality. To improve this understanding, we developed data to map and analyze under-5 mortality in sub-Saharan Africa in the 1980s, 1990s, and 2000s.
Combining Satellite Imagery and Machine Learning to Predict Poverty
The elimination of poverty worldwide is a global priority, but the data to measure progress toward this goal is lacking in many countries. In this project, we developed an approach that combines machine learning with high-resolution satellite imagery to provide local-level estimates of wealth and consumption in African countries.
Global Non-Linear Effect of Temperature on Economic Production
For this project we analyzed changes in temperature and changes in economic output (as measured by per capita gross domestic product) for 166 countries for the years 1960-2010. We then combined these historical estimates with projections of future climate change from global climate models, and projections of how countries’ economies might develop absent climate change. Our findings demonstrate that changes in temperature have substantially shaped economic growth in both rich and poor countries over the last half century, and that future warming is likely to reduce global economic output, relative to a world without climate change.