COS 311-2 - Using remote sensing to determine the impact of invasive species on plant characteristics, soil properties, and primary productivity

Invasive plants are threats to biodiversity, ecosystem functioning, and services. Previous studies have reported conflicting results on the impacts of biological invasions on ecosystem characteristics and have revealed that such impacts are scale- and ecosystem-dependent. With its synoptic view, remote sensing can contribute to improving our understanding of the impact of biological invasions across large spatial scales. Our goals were to (1) determine the impact of invasive species on plant characteristics and primary productivity, here expressed as aboveground biomass, (2) identify the underlying mechanisms by which species invasion impacts soil properties, and (3) assess the capability of remotely-sensed data to capture the impacts of species invasion.
To address our goals, we focused on an invasive legume known as Lespedeza cuneata (L. cuneata) at the Joseph H. Williams Tallgrass Prairie Preserve in Oklahoma, USA. We measured percent cover of L. cuneata at 100 30m by 30m plots, collected in situ top-of-canopy foliage samples to quantify plant characteristics, including community-weighted mean nitrogen, phosphorus, and potassium concentration, and quantified aboveground biomass. We also collected soil samples to quantify soil nitrogen and carbon. Additionally, we collected airborne hyperspectral imagery with spatial resolution of 1m. Using our vegetation trait measurements and remotely-sensed spectral data, we developed partial least square regression models to estimate plant characteristics and aboveground biomass from remote sensing data. We then assessed the impact of L. cuneata invasion on plant characteristics and aboveground biomass using generalized additive models. Finally, we identified the mechanisms through which L. cuneata impacted soil properties using structural equation models.
Results showed that L. cuneata invasion shifted aboveground characteristics towards those of L. cuneata by significantly increasing community-weighted mean nitrogen and phosphorus concentration (R2=0.39 and 0.19, respectively). Moreover, L. cuneata significantly increased primary productivity (R2=0.30). The associations between L. cuneata percent cover and soil carbon and nitrogen were positive and significant. Our study also showed that remote sensing can capture the impacts of species invasion on plant characteristics and primary productivity. Moreover, when we used remotely-sensed plant characteristics and biomass to assess the aboveground-belowground linkages, our results were comparable to those obtained from in situ measurements. These findings can advance our understanding of the impact of species invasion on above- and belowground characteristics across large-scales using remote sensing which, in turn, can improve management of biological invasions.