Combining open NEON and Landsat data to connect biodiversity and disturbance regime

Jasper Van doninck, Jonathan Knott and Phoebe L. Zarnetske, Department of Integrative Biology, Michigan State University, East Lansing, MI, Jasper Van doninck, Jonathan Knott and Phoebe L. Zarnetske, Ecology, Evolution, and Behavior Program, Michigan State University, East Lansing, MI, Annie C. Smith, Washington State Department of Natural Resources, WA, Jonathan Knott, Forest Inventory & Analysis, USFS Northern Research Station, St. Paul, MN, Patrick Bills, Michigan State University, Sydne Record, Biology, Bryn Mawr College, Bryn Mawr, PA, Benjamin Baiser, Wildlife Ecology and Conservation, University of Florida, Gainesville, FL, Angela L. Strecker, Institute for Watershed Studies, Western Washington University, Bellingham, WA

Background/Question/Methods
The support of the FAIR data principles by the National Ecological Observatory Network (NEON) offers opportunities to quantify the influences of biodiversity drivers at a continental scale. Unfortunately, although NEON measures many abiotic drivers of biodiversity (e.g., climate, soil attributes), it lacks spatially explicit data on past land use, as well as anthropogenic and natural disturbances. Without these key data, it is challenging to interpret biodiversity patterns and capture important cross-scale interactions resulting from anthropogenic processes. For example, the disturbance history of a location has profound influences on biodiversity, ecosystem processes and structure, species distributions and abundances, and biotic interactions.
In 2008, the USGS announced a free and open Landsat data policy, making a massive amount of high resolution satellite images acquired over several decades available to the scientific community. Combined with the increased availability of high-performance computing facilities and advances in algorithms, this has led to development of land use, landcover and change layers at high resolution and continental to global scales. Yet, thematically detailed layers, describing disturbance detection and attributing disturbance agent for different landcover types, are unavailable for the continental scale covered by NEON.
In this study, we process Landsat data acquired between 1984 and 2020 in Google Earth Engine through the LandTrendr temporal segmentation algorithm. We then used a secondary classification that incorporates spatial context to obtain continental-scale layers of disturbance detection and attribution. From these we then derived disturbance metrics for the different NEON sites to connect dimensions of biodiversity and their drivers for several taxonomic groups across spatial scales.
Results/Conclusions
Landsat-derived layers of disturbance detection and attribution for the contiguous United States covering the period 1984-2020 will be shared through the Environmental Data Initiative (EDI) repository and/or as Google Earth Engine assets to encourage reuse. Data and scripts related to calibration/validation of these disturbance layers will be available through the project’s GitHub (https://github.com/NEON-biodiversity).
Preliminary results show that species richness at NEON sites is co-defined by disturbance regime. The spatial and temporal scales at which biodiversity influences species richness varies between taxa, as does the direction of this relationship. The availability of free and open data products through NEON and the Landsat program is critical to our efforts to quantify these relationships.