Past is prologue: historic landcover patterns predict contemporary grassland loss in the U.S. Northern Great Plains

Context Grasslands of the North American Great Plains are among the world’s most imperiled ecosystems. Determining landscapes at risk of grassland loss will benefit grassland conservation programs by enabling prioritization of parcels for acquisition. Objectives We hypothesized that gradients in the...

Full description

Saved in:
Bibliographic Details
Published in:Landscape ecology 2022-12, Vol.37 (12), p.3011-3027
Main Authors: Niemuth, Neal D., Barnes, Kevin W., Tack, Jason D., Iovanna, Rich
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Context Grasslands of the North American Great Plains are among the world’s most imperiled ecosystems. Determining landscapes at risk of grassland loss will benefit grassland conservation programs by enabling prioritization of parcels for acquisition. Objectives We hypothesized that gradients in the amount of grass in local landscapes resulting from past conversion would be a useful predictor of future conversion. We developed simple, intuitive models predicting grassland conversion across the United States northern Great Plains. Methods We used the grassland/herbaceous, pasture/hay, and emergent herbaceous wetland classes from the National Land Cover Database to evaluate grassland loss from 2001 to 2016. We classified the amount of grass in 13.2-km 2 landscapes in 2001 into percentage bins and used generalized additive models to relate these values to the total and proportion of grassland lost, by percentage bin. We applied models to landcover data to create spatial surfaces predicting conversion. Results Area of grass lost was highest in 40–60% grass bins, except for heavily cropped states, where highest losses occurred in 10–20% grass bins. Percentage of grass in local landscapes was generally a strong predictor of state-level total grassland loss and proportion of grassland lost. Predicted conversion for remaining grasslands varied within and among states. Conclusion The amount of grass in local landscapes can be a useful indicator of grassland conversion. Our simple models complement species distribution models used to guide conservation in the region. Mechanistic models of conversion can be improved by including amount of grass in local landscapes.
ISSN:0921-2973
1572-9761
DOI:10.1007/s10980-022-01528-3