Using imaging spectroscopy to assess the impacts of invasive plants on aboveground and belowground characteristics

Abstract

ABSTRACT Invasive plants are threats to biodiversity, ecosystem functioning, and services. Previous studies have reported that the impacts of biological invasions on ecosystem characteristics can be scale- and ecosystem-dependent. Current methods to assess the impacts of biological invasions have mainly focused on traditional field observations, limiting the scale at which biological invasions can be studied. With its synoptic view, remote sensing can contribute to improving our understanding of the impacts of biological invasions across large spatial scales. However, the application of remote sensing to determine the impacts of invasive plants on ecosystem characteristics, including aboveground and belowground, has not yet been explored. Therefore, our goals were to (1) determine the impacts of invasive plants on aboveground functional traits and productivity, (2) assess the underlying mechanisms through which species invasion impacts belowground characteristics, and (3) determine the capability of remotely-sensed data to capture the impacts of species invasion on aboveground and belowground characteristics. To address our goals, we focused on Lespedeza cuneata (L. cuneata), an invasive legume at the Joseph H. Williams Tallgrass Prairie Preserve in Oklahoma, U.S. We measured percent cover of L. cuneata, quantified aboveground biomass, collected top-of-canopy foliage samples and soil samples in the field, and collected airborne imaging spectroscopy. We used remotely-sensed spectral data and in situ-measured traits to estimate plant functional traits and aboveground biomass. We then assessed the impacts of L. cuneata invasion on aboveground functional traits and biomass using generalized additive models. We also identified the mechanisms through which L. cuneata impacted belowground characteristics using structural equation models. We developed generalized joint attribute models using in situ aboveground and belowground characteristics and predicted belowground characteristics throughout our study site by applying the developed model to remotely-sensed aboveground characteristics. Our findings showed that L. cuneata invasion shifted aboveground functional traits toward those of L. cuneata by significantly increasing community-weighted mean (CWM) foliar nitrogen and phosphorus concentrations. Moreover, L. cuneata significantly increased aboveground biomass, a proxy for aboveground productivity. We also showed that imaging spectroscopy captured the impacts of species invasion on aboveground functional traits and productivity. More importantly, we provided substantial evidence suggesting that imaging spectroscopy can be used to predict belowground characteristics through the aboveground-belowground linkages. These findings can significantly advance our understanding of the impacts of biological invasions on belowground characteristics across large scales which is often challenging to quantify using field methods.