This article is part of the Fall 2021 issue of the Earth Science Matters Newsletter.
Modeling past and future land cover in the Delaware River Basin
Landscape change has a critical impact on a wide variety of processes related to the status and health of both human and ecological systems. Remote sensing, such as information gathered from satellite imagery, provides a vital tool for mapping and monitoring changes on the Earth’s surface. However, consistent remote sensing data is limited in availability, with even the longest active observation system, Landsat, only collecting data back to 1972. In order to extend the record of land change data, scientists must create models that simulate what conditions may have been in the past and that can forecast what conditions may be in the future based off of different variables. Then with the availability of these even longer-term modelled landscape change datasets, assessments of the impacts of past change on human and ecological processes can be determined, and planning and mitigation activities for potential future changes to the landscape can be anticipated and visualized.
The U.S. Geological Survey has developed a modeling framework capable of both reconstructing historical landscapes, and modeling scenarios of future landscape change. The Forecasting Scenarios of land use (FORE-SCE) model was designed to use the historical satellite record to inform long-term modeling of landscape change. Information on the rates, types, and spatial patterns of landscape change is used to develop key model parameters that allow it to anticipate future rates of change, as well as create realistic patterns of landscape change. The model uses real parcels (boundaries) of land ownership and land management to capture the units at which real land-change decisions are made.
The FORE-SCE model was recently applied in the Delaware River Basin, an area with substantial change in agricultural and urban patterns over the last few decades. Major issues in the region include changes in water use and water availability. Researchers reconstructed the historical landscape in the basin back to 1700 at 10-year increments, representing what the landscape used to look like prior to European settlement. They also modeled landscape change forward through 2100, using several different scenarios to represent uncertainties in future conditions. The model is responsive to changes in both water availability and climate, with the occurrence or spatial patterns of both human land uses and natural vegetation, shifting in response to future climate change and water scenarios.
The outcome of these analyses is a long-term database of landscape change for the Delaware River Basin, at 10-year increments from 1700 through 2100, with seven different scenarios of future change from 2016 to 2100. This research is being used to address impacts on water resources in the region, including the potential impacts if a drought similar to the famous 1960s drought occurred today. The low flows associated with that drought resulted in saltwater intrusion into the estuary, threatening freshwater drinking supplies for Philadelphia, Trenton, and other cities in the region. Low flows also impacted energy generation on hydroelectric dams, diminished water quality, and caused ecological disruptions such as changes in stream temperature. With modeled land cover in the region, the impacts of a possible similar drought in today’s landscape that has a higher population and greater water use demands can be assessed. The Delaware River Basin data will also be used to assess biodiversity and conservation priorities in relationship with the Administration’s America the Beautiful initiative, a focus to conserve 30% of our Nation’s lands and waters by 2030.
The paper, "Prototyping a methodology for long-term (1680-2100) historical-to-future landscape modeling for the conterminous United States" was recently published in Land.