Public Lab Research note

UN-SPIDER argues for multispectral humanitarian satellite imagery

by warren | December 05, 2011 14:44 05 Dec 14:44 | #611 | #611

Social Impact through Satellite Remote Sensing – Visualizing Acute and Chronic Crises beyond the Visible Spectrum

by Global Pulse,

November 30th 2011

For decades, satellite remote sensing has provided fundamental insights in countless physical science fields such as ecology, geosciences, atmospheric physics, and chemistry. However, as it relates to human and socioeconomic processes, satellite remote sensing is an incredibly powerful tool that is underutilized. Human behavior and socioeconomic parameters have been successfully studied via proxy through remote sensing of the physical environment by measuring the growth of city boundaries and transportation networks, crop health, soil moisture, and slum development from visible and multispectral imagery.

The NASA/ NOAA image of Earth’s “Lights at Night” is routinely used to estimate economic development and population density. There are many examples of the conventional uses of remote sensing in humanitarian-related projects including the Famine Early Warning System Network (FEWS NET) and the UNITAR's Operational Satellite Applications Programme (UNOSAT), which provides remote sensing for humanitarian and disaster relief. Yet even with these successful applications, we've just begun to scratch the surface of what remotely sensed data can provide for prevention, mitigation and response to acute and chronic human crises.

Many successful remote sensing projects have focused exclusively on the visible spectrum - what one would see of they looked down from an airplane at the ground surface. Yet in order to discern objects or patterns of interest (buildings, markets, roads, vehicles, etc.), high spatial resolution remote sensing data are necessary. It's important to note up front, though, that two other types of data resolution are also critical in remotely sensing the Earth's surface: spectral resolution and temporal resolution.High spatial resolution remote sensing data have been utilized successfully in a number of recent disasters to rapidly and accurately map the developing situation on the ground during crises, such as earthquake, flood, landslide, and civil unrest events.

After the 2010 Haiti earthquake, volunteers used released GeoEye imagery to digitize roads into OpenStreetMap. High resolution imagery was key to assessing the situation and for the first time, released widely to the public through Google Earth and other outlets. The Satellite Sentinel Project is goes beyond imaging natural disasters and utilizes DigitalGlobe and other commercial imagery to serve as witness to potential humanitarian crises and human rights crimes in near real-time. Grassroots takes a participatory, public domain approach to monitoring crises with balloon and kite photography. They are using systems that involve attaching digital cameras and infrared sensors to weather balloons. has been able to acquire imagery for monitoring the 2010 oil spill in the Gulf of Mexico and has developed a community around their DIY airborne environmental sensors.

Again, however, these projects (as well as many others in the humanitarian space) rely on high spatial resolution with limited utilization of higher spectral or temporal resolution. While high spatial resolution is necessary to "see" what is happening on the ground in the visible spectrum of light that our eyes detect, there are other kinds of data that can be obtained by using remote sensing to "see" in other parts of the electromagnetic spectrum (spectral resolution). Also, one must take into account how frequently these remotely sensed data are gathered (temporal resolution).

Spectral resolution can be the most difficult to understand since many lay users have only used imagery that looks like what their eye sees, such as the GeoEye imagery in Google Earth. There are many sensors that allow us to "see" the Earth's surface in ways other than the visible part of the electromagnetic spectrum. All passive sensors, whether they are in your digital camera or on a satellite, are measuring energy reflected or emitted from the surface of the Earth. This energy either comes from the sun or from the heat energy generated at the molecular level of materials on the ground. Active sensors, such as radar, send down energy (radio waves) to the Earth's surface and measure the returned signal. Spectral resolution refers to the coverage of the electromagnetic spectrum that a sensor can measure.

Many sensors that can detect visible light energy that also have a "band" that measures energy in the near-infrared part of the electromagnetic spectrum. These sensors can be used to measure the abundance of vegetation on the ground. Some have simple thermal detectors that can measure temperature. For more detailed analysis of the Earth's surface, the thermal infrared can be used to say what the composition of geologic and urban materials are on the ground and how those materials hold heat. This requires a higher spectral resolution than most sensors have. Some instruments like the NASA satellite Landsat have a "band" in the thermal infrared which can provide a proxy for surface temperature. The NASA satellite ASTER can has a higher spectral resolution than Landsat since it measure five bands or "slices" of the thermal infrared and so can be used to identify the composition of materials on the ground and measure surface temperature within 1.5 degrees C.

...article continues ...

Original Page:


Login to comment.