Social–ecological landscape patterns predict woody encroachment from native tree plantings in a temperate grassland

Authors

Victoria M. Donovan, University of Nebraska-LincolnFollow
Jessica L. Burnett, University of Nebraska-LincolnFollow
Christine H. Bielski, University of Nebraska-LincolnFollow
Hannah E. Birge, University of Nebraska-Lincoln
Rebecca Bevans, University of Nebraska-Lincoln
Dirac L. Twidwell Jr, University of Nebraska‐LincolnFollow
Craig R. Allen, University of Nebraska-LincolnFollow

ORCID IDs

Victoria Donovan

Jessica Burnett

Christine Bielski

Dirac Twidwell

Craig R. Allen

Document Type

Article

Date of this Version

2018

Citation

Ecology and Evolution. 2018;8:9624–9632.

Comments

© 2018 The Authors.

Open access

DOI: 10.1002/ece3.4340

Abstract

Afforestation is often viewed as the purposeful planting of trees in historically non-forested grasslands, but an unintended consequence is woody encroachment, which should be considered part of the afforestation process. In North America’s temperate grassland biome, Eastern redcedar (Juniperus virginiana L.) is a native species used in tree plantings that aggressively invades in the absence of controlling processes. Cedar is a well-studied woody encroacher, but little is known about the degree to which cedar windbreaks, which are advocated for in agroforestry programs, are contributing to woody encroachment, what factors are associated with cedar spread from windbreaks, nor where encroachment from windbreaks is occurring in contemporary social–ecological landscapes. We used remotely sensed imagery to identify the presence and pattern of woody encroachment from windbreaks in the Nebraska Sandhills. We used multimodel inference to compare three classes of models representing three hypotheses about factors that could influence cedar spread: (a) windbreak models based on windbreak structure and design elements; (b) abiotic models focused on local environmental conditions; and (c) landscape models characterizing coupled human-natural features within the broader matrix. Woody encroachment was evident for 23% of sampled windbreaks in the Nebraska Sandhills. Of our candidate models, our inclusive landscape model carried 92% of the model weight. This model indicated that encroachment from windbreaks was more likely near roadways and less likely near farmsteads, other cedar plantings, and waterbodies, highlighting strong social ties to the distribution of woody encroachment from tree plantings across contemporary landscapes. Our model findings indicate where additional investments into cedar control can be prioritized to prevent cedar spread from windbreaks. This approach can serve as a model in other temperate regions to identify where woody encroachment resulting from temperate agroforestry programs is emerging.