Biodiversity in a Flash: Monitoring Impacts of Disturbance for Grassland Biodiversity and Ecosystem Functioning Using Hyperspectral Imagery

Monitoring biodiversity across space and time is increasingly important due to the role of biodiversity in maintaining ecosystem productivity, stability and resilience. Functional diversity constitutes an important facet of biodiversity, and describes the value and abundance of functional traits (i.e., characteristics that affect performance) in a community, which is integral to understanding both how environmental change impacts communities and how communities influence functioning. However, effectively capturing community trait distributions requires comprehensive sampling of individuals across species and environmental conditions, the time and resource intensity of which prevents frequent measurements and limits our ability to detect responses to disturbance. Hyperspectral imaging presents a promising opportunity for overcoming these limitations because it can detect functional and compositional characteristics of vegetation across large spatial extents by measuring subtle differences in light reflectance. Here, we examine the ability of hyperspectral imagery to reveal community change and predict outcomes for ecosystem functioning for a grassland community subject to a multiyear drought and grazing manipulation in Boulder, Colorado, USA. Specifically, we assess 1) the impacts of drought and grazing on spectrally-derived community trait compositions, and 2) the utility of spectral diversity for predicting changes to ecosystem functioning. We combine high resolution hyperspectral imagery (2 mm spatial resolution from 400 - 1000 nm) with indicators of ecosystem functions, including soil fertility and plant productivity, to investigate the role of community trait shifts in driving changes to ecosystem multifunctionality. Our results reveal that hyperspectral imagery successfully tracks the interactive effect of drought and grazing on community functional traits and diversity, highlighting the potential for hyperspectral imagery to monitor outcomes of disturbance for vegetation communities.