Remote sensing data is enabling the analysis of functional diversity at different scales

August 16, 2021

NOMIS researcher and UZH President Michael Schaepman and colleagues have demonstrated that their approach to deriving plant functional traits, phylogenies and genetics works on a regional level. Their research — an interdisciplinary effort combining geography, physics, mathematics, computational science, ecology, biodiversity and genetics — was published in Ecology and Evolution on July 22. Excerpts from the introduction follow:

Michael Schaepman (Photo: Frank Brüderli)

Plant functional traits have specific spatial distributions as a result of different abiotic and biotic factors interacting at different spatial and temporal scales (Funk et al., 2017). For instance, environmental and local heterogeneity, phylogenetic distance, and plant–plant interactions such as competition and facilitation can act as important drivers of trait variation and affect coexistence mechanisms at different organizational levels from individuals to communities (Gross et al., 2009; Hart et al., 2016; Valladares et al., 2015). Understanding how these different drivers differ across spatial and temporal scales could help to gain insights into the possible responses of species, communities, and ecosystems to environmental change.

Here, we combine the RS approach with an explicit delineation of individual-tree crowns in a temperate forest (≈5.5 ha), allowing us to use general linear modeling as commonly done in ecology to partition trait variation among individuals into variation due to taxonomy (e.g., class, species) and environment. Our goal is to assess the potential of RS-derived traits to identify taxonomic units. Traits with high taxonomic and low environmental variation between individuals would be promising tools to separate species and by extension even genotypes within species in the future (Cavender-Bares et al., 2016). Furthermore, we aim to describe environmental influences on trait variation between and within species for a complete forest stand. With the mapping of individual-tree traits, we want to bridge the gap between field and remote sensing measurements, and to open doors to the possibility of studying biodiversity at a very high spatial resolution under natural conditions. Through our measurements, we specifically ask: (a) Can plant traits be assessed by remote sensing at the level of individual-tree crowns? (b) To which extent can different traits be used to identify and map taxonomic units? (c) How does environmental variation influence variation in these traits and does this vary among taxonomic units?

Read the Ecology and Evolution publication: Remotely sensed between-individual functional trait variation in a temperate forest