Managing woody plant expansion in native grasslands

Worldwide, natural grasslands are threatened by the expansion of unwanted woody plant species. This also holds for Canada; a Saskatchewan rancher who wrote to us said: “When my grandfather came over 100 years ago, there was not a tree that could stop a plow. This piece of land today is 70% covered in trees and bushes”. Woody plant encroachment (WPE) has become the second most important process that leads to grassland loss in the Great Plains Biome (Working Lands for Wildlife, 2022), affecting the food industry, the economy, and the environment. For grassland management practices to be effective, accurate monitoring of grassland health is important. The advantages of using GIS and remote sensing data to achieve this goal are many: large-scale coverage, near-real time monitoring, cost-efficiency, consistency, and visualization attributes (Soubry et al., 2021).

In a literature review that we conducted, we found that there is no universal remotely sensed WPE monitoring framework available (Soubry & Guo, 2022). Therefore, the objectives of this study are:

1. to explore remote sensing and GIS approaches for appropriate detection of WPE in grasslands,
2. to investigate factors that are influencing WPE using spatial analysis, and
3. to identify areas most vulnerable to WPE to enable better land management practices.

Our study areas include native prairie regions in the Moist Mixed, Mixed grassland, and Cypress Upland ecoregions of Saskatchewan (SK) (Figure 1).

The steps followed for each objective are summarized in the flow chart below (Figure 2).

For my second objective, I used 30 cm aerial imagery from 2018 to quantify the distribution of shrub cover with an object-based approach in Cypress Hills Interprovincial Park (CHIPP), West Block, Saskatchewan. A shrub cover map was generated, and GIS was used to investigate the relationship between shrub cover and topo-edaphic and anthropogenic variables (Table 1). More details can be found in Soubry et al. (2022).

To examine the relationship between the variables in Table 1 and shrub cover in the park, I developed multiple models that combined variables related to topography and soil, anthropogenic influences, and a mixture of both. A simple ordinary least square regression was used to examine these relationships, and spatial autocorrelation in the residuals of each model were accounted for.

For objective 3, eight priority parameters related to high shrub cover in the park were selected based on the models’ results. I conducted feature overlay analysis of these priority parameters in ArcMap to get a feature overlay count, which was then translated into management priority classes, and a management priority map was created.

From objective 1, I found that spring was the best season to distinguish shrubs from grass (using the blue and red wavelength regions) while each season had a different spectral region that was more correlated to shrub cover. Summer was the best season to spectrally discriminate western snowberry from wolfwillow; with red and blue wavelength regions being more important.

The methods from objective 3 resulted in a management priority map (Figure 3). These results were then incorporated by the Ministry of Parks, Culture, and Sport of Saskatchewan in a management plan to address woody plant expansion in CHIPP. More details can be found in Government of Saskatchewan (2022).

This research serves as the base for achieving long-term resilience of native grassland species and their habitats by better understanding the interaction of local factors on WPE expansion. Future work will focus on the integration of satellite and climate data to look at historical trends of WPE.

I am deeply grateful to Dr. Xulin Guo, my supervisor, Mr. Merek Wigness for inspiring us into this research topic, to Dr. Thuan Chu and Dr. Eric Lamb for provision of study areas and field data, to Larissa Robinov for assistance with publication writing and editing, to the rangeland managers at CHIPP, West Block, and to the numerous people that facilitated fieldwork. Lastly, I am thankful for funding from the University of Saskatchewan, the Government of Saskatchewan, and NSERC.

Government of Saskatchewan. (2022). Management Plan to Address Woody Plant Expansion in Cypress Hills Interprovincial Park (Issued March).

Guo, X., & Soubry, I. (2022). Final Report for Monitoring of Shrub Component to Support for Ecosystem Management in Saskatchewan Cypress Hills Interprovincial Park.

Soubry, I., Doan, T., Chu, T., & Guo, X. (2021). A Systematic Review on the Integration of Remote Sensing and GIS to Forest and Grassland Ecosystem Health Attributes , Indicators , and Measures. Remote Sensing, 13(3262), 1–30.

Soubry, I., & Guo, X. (2021a). Identification of the Optimal Season and Spectral Regions for Shrub Cover Estimation in Grasslands. Sensors, 21(3098), 1–26. https://doi.org/https://doi.org/ 10.3390/s21093098

Soubry, I., & Guo, X. (2021b). Seasonal Spectral Separation of Western Snowberry and Wolfwillow in Grasslands with Field Spectroradiometer and Simulated Multispectral Bands. Environments, 8(7), 60. https://doi.org/10.3390/environments8070060

Soubry, I., & Guo, X. (2022). Quantifying Woody Plant Encroachment in Grasslands: A Review on Remote Sensing Approaches. Canadian Journal of Remote Sensing, 1–42. https://doi.org/10.1080/07038992.2022.2039060

Soubry, I., Robinov, L., Chu, T., & Xulin, G. (2022). Mapping shrub cover in grasslands with an object-based approach and investigating the connection to topo-edaphic factors. Geocarto International, .

Working Lands for Wildlife. (2022). Dr. Dirac Twidwell: Saving the Last Grasslands. https://www.wlfw.org/ask-an-expert-dr-dirac-twidwell-saving-the-last-grasslands/ (last accessed on 23 October 2023)