Mapping tree cover expansion in Montana, U.S.A. rangelands using high‐resolution historical aerial imagery

• Published in: Remote sensing in ecology and conservation Vol. 10; no. 1; pp. 91 - 105
• Main Authors: Morford, Scott L., Allred, Brady W., Jensen, Eric R., Maestas, Jeremy D., Mueller, Kristopher R., Pacholski, Catherine L., Smith, Joseph T., Tack, Jason D., Tackett, Kyle N., Naugle, David E.
• Format: Journal Article
• Language: English
• Published: Oxford John Wiley & Sons, Inc 01.02.2024; Wiley
• Subjects: Accuracy; Aerial photography; Afforestation; Agricultural ecosystems; agricultural land use; Artificial neural networks; Biodiversity; Climate change; Cloud computing; Coniferous trees; Conifers; Deep learning; Ecosystem services; Ecosystem structure; Environmental changes; Environmental impact; global change; grassland; Grasslands; Image resolution; Neural networks; Photogrammetry; rangeland conservation; Rangelands; Remote sensing; Satellite tracking; Semantics; Spatial analysis; Structure-function relationships; Trees; Woody encroachment; United States > US; Montana; Great Plains
• ISSN: 2056-3485; 2056-3485
• DOI: 10.1002/rse2.357

Worldwide, trees are colonizing rangelands with high conservation value. The introduction of trees into grasslands and shrublands causes large‐scale changes in ecosystem structure and function, which have cascading impacts on ecosystem services, biodiversity, and agricultural economies. Satellites are increasingly being used to track tree cover at continental to global scales, but these methods can only provide reliable estimates of change over recent decades. Given the slow pace of tree cover expansion, remote sensing techniques that can extend this historical record provide critical insights into the magnitude of environmental change. Here, we estimate conifer expansion in rangelands of the northern Great Plains, United States, North America, using historical aerial imagery from the mid‐20th century and modern aerial imagery. We analyzed 19.3 million hectares of rangelands in Montana, USA, using a convolutional neural network (U‐Net architecture) and cloud computing to detect tree features and tree cover change. Our bias‐corrected results estimate 3.0 ± 0.2 million hectares of conifer tree cover expansion in Montana rangelands, which accounts for 15.4% of the total study area. Overall accuracy was >91%, but the producer's accuracy was lower than the user's accuracy (0.60 vs. 0.88) for areas of tree cover expansion. Nonetheless, the omission errors were not spatially clustered, suggesting that the method is reliable for identifying the regions of Montana where substantial tree expansion has occurred. Using the model results in conjunction with historical and modern imagery allows for effective communication of the scale of tree expansion while overcoming the recency effect caused by shifting environmental baselines. Worldwide, trees are colonizing grasslands of high conservation value, and these changes have cascading impacts on ecosystem services, biodiversity, and agricultural economies. Given the slow pace of tree encroachment, quantifying tree expansion using historical imagery at scale provides critical insight into the magnitude of environmental change. Here, we analyze 19.3 million hectares of imperiled grasslands in Montana, USA, using historical and modern aerial imagery. We show that over 15% of these lands are experiencing tree encroachment and provide a mapping application using historical and modern aerial imagery to visualize this change.