Measuring biological heterogeneity in the northern mixed prairie: a remote sensing approach

Abstract

Biological heterogeneity, defined as the degree of dissimilarity between biological variables (e.g., biomass, green vegetation and Leaf Area Index [LAI]), is one of the most important and widely applicable concepts in ecology due to its close link with biodiversity. To investigate grassland biological heterogeneity, we selected three transects extending from upland to valley grasslands at Grasslands National Park (GNP), Canada, representing the northern mixed grassland. For the purposes of our analysis, three types of data were collected: remote sensing ground level hyperspectral data, biological data (LAI, biomass, and vegetation cover) and environmental data (soil moisture, organic content, and bulk density). Methodologically, field‐level remote sensing data were used to calculate spectral vegetation indices. These indices, plus the biological variables, were then used in regression analyses with the goal of assessing the feasibility of using remote sensing data to study biological heterogeneity. The results indicate that it is feasible to use ground‐level remote sensing data to represent biological variables. These indices can explain about 40–60 percent of the biological variation. Semivariogram analyses were further applied on these data to investigate their range of spatial variation. Spatial variations in the