Primary Research Topics
Land subsidence measured with D-InSAR
Land subsidence in river deltas can heighten storm surges, salinate groundwater, intensify river flooding, destabilize infrastructure, and accelerate shoreline retreat. Using satellite-based D-InSAR (Differential Interferometric Synthetic Aperture Radar), we measured land motion along the coast of the Yellow River Delta in China. We found that groundwater pumping at fish farms has caused subsidence rates as high as 250 mm/y, which exceeds local and global average sea level rise by nearly two orders of magnitude. These results suggest a major hazard for Asian megadeltas, where fish farming dominates land-use and groundwater pumping is common to regulate salinity in ponds. To learn more about this problem, see the original publication in Geophysical Research Letters [Higgins et al., 2013], or read these associated features on the websites of Nature or the UN-sponsored Future Earth program. Sprechen Sie Deutsch? Check out this recent interview with German Public Radio!
Urban subsidence measured with SBAS-InSAR
Groundwater extraction, hydrocarbon extraction, tectonics, isostatic adjustment and natural compaction of sediments can all cause cities to sink, but without high-resolution measurements it can be impossible to tell which is having the greatest impact in any given city. Interferometric Synthetic Aperture Radar (InSAR) is a satellite-based technique that can produce maps of ground deformation with mm-scale vertical accuracy. In Higgins et al. , we used a technique called Small Baseline Subset (SBAS) InSAR to search for controls on subsidence in major cities such as Dhaka, Bangladesh, where dense populations exert rapidly-changing influences on the surface and subsurface.
Visible and infrared remote sensing
In addition to its many other applications, visible and infrared remote sensing can be used to map inundation, identify new reservoirs, track coastline changes, and measure suspended sediment concentrations in rivers. From time to time I team up with the Flood Observatory (formerly Dartmouth Flood Observatory/DFO) at CU-Boulder to look at these applications. In 2012, I joined the DFO on a publication examining the potential for global mapping of storm surges using a number of sensors [Brakenridge et al., 2012]. In 2016, we are collaborating on an analysis of the flood history of Myanmar, using DFO products to examine the flood events during the period when the country was largely closed to international press. With IPython Notebooks from Google Earth Engine developers, I also mapped and classified irrigation inundation at 30-m resolution in Bangladesh using the MCD12 (MODIS) Landcover classification as a training dataset for LANDSAT-7 observations.
Delta sustainability and risk assessment modeling
I am part of the Belmont Forum’s DELTAS project, an international collaboration of government, university, and NGO researchers focused on vulnerability and sustainability of coastal river deltas. In this project I am working to co-develop an open-access, science-based, integrative modeling framework called the Delta Risk Assessment and Decision Support (RADS) Tool, which can provide a quantitative basis for investigating and comparing scenarios and trade-offs for decision making. Towards this end, I work with global precipitation and climate data, census data, Digital Elevation Models, Open Street Map extracts, and several other datasets that are combined in a GIS framework to automatically produce model inputs and provide model parameter estimation. As a first test for RADS, we are investigating how proposed dams and water transfers in India might soon impact sediment transport to the Ganges-Brahmaputra Delta in Bangladesh.
Hurricane Rainfall, Gulf Coast, USA
Mitigating the impacts of tropical cyclones requires accurate forecasts of precipitation within the storm. A deeper understanding of tropical cyclone rainfall is also desirable because of the large role tropical cyclones can play in regional water budgets. In the Gulf Coast region, for example, between five and ten percent of total annual precipitation can come from tropical cyclones. Improving hurricane models requires samples of actual rainfall collected from inside of hurricanes, but rain gauges often cannot withstand the conditions inside of a landfalling storm. With a member of my dissertation committee, Dr. Katja Friedrich, I developed a quality-control algorithm to improve measurements of raindrop size distributions in severe weather. We applied this algorithm to data collected inside of Hurricanes Ike and Rita (2008), which made landfall along the Texas coast in 2008 [Friedrich, Higgins et al., 2013].