Comparing stability in random forest models to map Northern Great Plains plant communities in pastures occupied by prairie dogs using Pleiades imagery

Abstract

Abstract. Black-tailed prairie dogs (Cynomys ludovicianus) have been described as a keystone species and are important for grassland conservation, yet many concerns exist over the impact of prairie dogs on plant biomass production and consequently livestock production. The ability to map plant communities in pastures colonized by prairie dogs can provide land managers with an opportunity to optimize rangeland production while balancing conservation goals. The aim of this study was to test the ability of random forest (RF) models to classify five plant communities located on and off prairie dog towns in mixed-grass prairie landscapes of north central South Dakota, assess the stability of RF models among different years, and determine the utility of utilizing remote sensing techniques to identify prairie dog colony extent. During 2015 and 2016, Pleiades satellites were tasked to image the study site for a total of five monthly collections each summer (June–October). Training polygons were mapped in 2016 for the five plant communities and used to train RF models. Both the 2015 and 2016 RF models had low (1 %) out-of-bag error rates. However, comparisons between the predicted plant community maps using the 2015 imagery and one created with the 2016 imagery indicate over 32.9 % of pixels changed plant community class between 2015 and 2016. The results show that while RF models may predict with a high degree of accuracy, overlap of plant communities and interannual differences in rainfall may cause instability in fitted models. A final RF model combining both 2015 and 2016 data yielded the lowest error rates and was also highly accurate in determining prairie dog colony boundaries.

Long‐Term Effects of Black‐Tailed Prairie Dogs on Livestock Grazing Distribution and Mass Gain

“…In contrast, areas occupied continuously for 7–10 years exhibited significant reductions in perennial grass cover and increased bare soil exposure for multiple years after plague‐induced prairie dog removal (Augustine et al 2014). Similarly, in northern mixed prairie, core areas of colonies that are occupied continuously for decades show increased bare soil exposure, increased dominance by unpalatable subshrubs and forbs, and loss of perennial grass cover, whereas the more recently colonized edges often have significant cover of closely cropped grasses (Coppock et al 1983, Brennan et al 2020). These colony edges can provide valuable foraging areas for livestock during the growing season (Brennan 2019), and in some cases, even during dormant season grazing (Sierra‐Corona et al 2015).…”

“…At the same time, some lesser reduction in stocking rate may be warranted if the goal is to prevent a decline in individual cattle performance. Estimates of the extent of core colony areas with elevated bare soil exposure and unpalatable forb dominance (Brennan et al 2020) could potentially provide a more appropriate estimate of the necessary magnitude of reduction in stocking rate. Economic analyses are also needed to compare strategies of reducing stocking rate versus maintaining stocking rate with reduced animal gains, and to compare both to the costs of controlling prairie dogs.…”

“…In our study, we did not map spatial variation in vegetation communities or cover within the colonies. Recent advances in remote sensing are enabling more accurate mapping of areas within colonies with elevated bare soil exposure and cover of unpalatable forbs (Brennan et al 2020). Knowledge of the location and extent of such areas could potentially provide an improved estimate of the effect of prairie dogs on livestock performance, and perhaps assist with more targeted adaptive management approaches to minimize the effects.…”

“…Our finding that cattle mass gains were suppressed to a limited degree and only when colonies were extensive (e.g., by 4–8% when colonies occupied 30–60% of pastures) is consistent with the idea that negative effects of prairie dogs may primarily arise from the loss of perennial grasses on core colony areas that are occupied continuously in between plague epizootics. Because of recurrent plague epizootics on time scales of roughly 5–15 years across the western Great Plains, these core areas typically comprise a small portion of the landscape, even when colonies occupy a majority of a given pasture (Augustine et al 2008, Hartley et al 2009, Brennan et al 2020). In our study, we did not map spatial variation in vegetation communities or cover within the colonies.…”

Evaluating the Impact of Grazing Cessation and Reintroduction in Mixed Prairie Using Raster Time Series Analysis of Landsat Data

“…Satellite remote sensing, with various temporal and spatial resolution, is a promising approach for monitoring biophysical parameters as indicators of prairie responses. Remote sensing has been used to measure productivity [26][27][28][29], community composition [30], phenology [31][32][33], ground cover [34], leaf area index [35], vegetation greenness [36], canopy height [37], canopy structure [38], species distribution [39,40], woody encroachment, and exotic species invasion [41]. Long-term experiments are particularly necessary to tease apart inconsistent responses of native prairie ecosystems to management practices that vary in frequency, intensity, and duration [24].…”