Estimates of per capita consumption in four African countries. Stanford researchers used machine learning to extract information from high-resolution satellite imagery to identify impoverished regions in Africa. (Image credit: Neal Jean et al.)Estimates of per capita consumption in four African countries. Stanford researchers used machine learning to extract information from high-resolution satellite imagery to identify impoverished regions in Africa. (Image credit: Neal Jean et al.)

Using Machine Learning to Map Poverty from Satellite Imagery

Satellite images are now being used to map poverty levels around the world using machine learning used to analyze specific poverty data using a convolutional neural network.

Sudden Landslide Identification Product (SLIP) developed by NASA detects landslide potential by analyzing satellite imagery for changes in soil moisture, muddiness, and other surface features. The Landsat 8 satellite capture the left and middle images on September 15, 2013, and September 18, 2014—before and after the Jure landslide in Nepal on August 2, 2014. The processed image on the right shows areas in red indicating a probable landslide and areas in yellow showing a possible landslide. Source: NASA.Sudden Landslide Identification Product (SLIP) developed by NASA detects landslide potential by analyzing satellite imagery for changes in soil moisture, muddiness, and other surface features. The Landsat 8 satellite capture the left and middle images on September 15, 2013, and September 18, 2014—before and after the Jure landslide in Nepal on August 2, 2014. The processed image on the right shows areas in red indicating a probable landslide and areas in yellow showing a possible landslide. Source: NASA.

Using Remote Sensing to Automate the Detection of Landslides

The Sudden Landslide Identification Product (SLIP) developed by NASA detects landslide potential by analyzing satellite imagery for changes in soil moisture, muddiness, and other surface features.

Average number of reporting weather stations in Rwanda during 1981 to 2013. Note the drastic drop in 1994. (Data source: Dinku et. al, 2016)Average number of reporting weather stations in Rwanda during 1981 to 2013. Note the drastic drop in 1994. (Data source: Dinku et. al, 2016)

Building Missing Weather Data

Called the ENACTS (Enhancing National Climate Services) initiative, scientists are using satellite data in order to estimate rainfall, temperature, and other information to fill in a 15 year gap in climate data collection for Rwanda.

Map showing the predicted distribution model of Ashokan edicts on the basis of geology, population, climate and topography.Map showing the predicted distribution model of Ashokan edicts on the basis of geology, population, climate and topography. Source: Gillespie, et al., 2016.

UCLA Researchers are Using Geospatial Technologies to Predict Potential Ancient Buddhist Sites

Archaeologist Monica Smith and geographer Thomas Gillespie identified 121 locations that they hope will reveal some of Ashoka’s edicts using GIS analys