Spatial Patterns In Plant Structural Diversity Across A Shrub Encroachment–Native Grassland Landscape Mosaic In The Northern Chihuahuan Desert

Drylands are heterogeneous landscapes that provide important ecosystem services, but they are threatened by global change factors that may influence aspects of their diversity and function. Environmental factors may influence the three-dimensional physical characteristics of vegetation structure (structural diversity) across spatial scales in arid ecosystems. This study investigated how environmental factors, including shrub cover, landform, and elevation, are related to vegetation structural diversity metrics across different spatial scales and scopes across the Jornada Experimental Range, a long-term ecological research site located in the northern Chihuahuan Desert. A canopy height model generated by Airborne Observation Platform (AOP) LiDAR data and collected by the National Ecological Observatory Network in 2022 were used to derive structural diversity metrics across the dryland landscape: mean height, maximum height, and top rugosity. Structural diversity metrics and environmental predictor rasters were aggregated to 10 m, 20 m, and 40 m spatial grain sizes and cut to spatial extents of 50 m, 100 m, and 200 m, respectively which produced two spatial scopes (extent-to-grain ratios) with two to three sets of absolute scale datasets (fixed grain size with different extent sizes). A total of 56 non-overlapping plots were randomly selected across the study area. Non-parametric tests, including Spearman's Rank Correlation and the Kruskal-Wallis test, were used to investigate how environmental variables -shrub cover, elevation, and landform type - were associated with structural diversity at different scales, i.e., within similar scope (extent-to-grain ratios) or absolute scales (different grain and extent combinations). The proportion of shrub cover was a significant negative correlate of vegetation structure metrics in the Jornada dryland landscape, showing that denser shrub cover is associated with a decrease in structural diversity. However, models with shrub cover as a predictor of structural diversity only occurred at one scope value (Scope value = 100), while the other scope value of 25, did not display a similar significant trend. In contrast, elevation and landforms were not significantly associated with structural diversity, which was consistent across all spatial datasets (scope or absolute scale). My findings emphasize that ecological patterns do not act uniformly across spatial scales in the landscape and that the absolute spatial scale and scope value require careful consideration when selecting the appropriate values for ecological studies in drylands.