Empirically-Driven Land Cover Change Scenarios in Federal Lands

Recent advances in satellite capabilities to monitor the Earth have made it possible to monitor land cover change on annual, monthly, and even daily time scales. Many researchers are using these data to analyze the rates and causes of short-term changes and long-term trends. New research by USGS scientists is linking dense land-cover map series with climate data to establish long term climate relationships and project future land cover changes under different climate scenarios.

Understanding the complexity of past land cover changes and modeling the potential consequences of climate on future vegetation is critical in federally managed lands. This study looked at land cover in the Beaty Butte Herd Management Area (Bureau of Land Management), Hart Mountain National Antelope Refuge (U.S. Fish and Wildlife Service), and Sheldon National Wildlife Refuge (U.S. Fish and Wildlife Service). Land managers there must develop current and future plans to accomplish a range of goals like habitat conservation, land use regulation, and public safety.

Sources/Usage: Public Domain. View Media Details

Sources/Usage: Public Domain. View Media Details

To measure the sensitivity of vegetation to climate over the observed period (1985-2018), USGS scientists used the National Land Cover Dataset (NLCD) Back in Time (BIT) fractional cover maps (Figure 1) to correlate annual maps of shrub cover, herbaceous cover, and bare ground to precipitation and temperature records. Historical land cover changes were used to create a spatially explicit, business-as-usual (BAU) scenario that predicts future abundance and distribution of each cover type for 2018 to 2050. These modelled results approximate what the land cover in the study area may look like if there are no changes in climate conditions compared to the past three decades. The research team also developed a modeled scenario that incorporates possible future climate changes. The model was based on the Intergovernmental Panel on Climate Change Representative Concentration Pathway (RCP) 8.5, which is a greenhouse gas concentration trajectory that assumes continued increased emissions throughout the 21^st century; this is often presented as “the worst-case” climate change scenario.

Projections indicate changes from 2018 to 2050: locations with high bare ground cover become less bare ground dominant, locations with moderate herbaceous cover contain less herbaceous cover, and locations with low shrub cover contain more shrub cover (Figure 2). Hart Mountain National Antelope Refuge is forecast to undergo the most change, with both models projecting larger declines in bare ground and larger increases in average herbaceous and shrub cover compared to Beaty Butte Herd Management Area and Sheldon National Wildlife Refuge.

General vegetation composition does not differ dramatically between the BAU and RCP 8.5 climate scenarios despite RCP 8.5 projections of mean annual minimum temperatures increasing by 1.2 °C and annual precipitation increasing by 7.6 mm by the year 2050. Small differences at the study area scale may be partly due to the short time horizon on the 2050 RCP 8.5 scenario, or to the climatic variability over the historical period. Specifically, individual years over the observed period had more total precipitation or a warmer minimum temperature than RCP 8.5 projections, and upper limit cover change rates reflect extremes during these anomalous years.

Despite subtle differences, these scenarios present plausible future outcomes intended to help federal land managers generate 20–30-year climate management strategies surrounding the condition and availability of vegetation for species of interest. If projections of increased shrub cover are realized, it may benefit shrub-dependent wildlife while also contributing to fire fuel build-up in part of the study area. Shifting fire risks should be considered in regional fire management plans, especially given the long history of sizable fires in the study area.

This paper, “Application of empirical land-cover changes to construct climate change scenarios in federally managed lands”, was published in MDPI Remote Sensing. The published model inputs, outputs, and instructions for building the simulation model with open source software can be found here: https://doi.org/10.5066/P9LJ1FI4. 

<< Back to Fall 2020 Newsletter