Skip to main content
SLU publication database (SLUpub)

Abstract

Large-scale assessments of plant condition are needed in restored plant communities to evaluate if ecosystems are functioning similarly to natural reference systems. Remote sensing can yield information on plant condition in a more efficient way than current on-ground physiological methods, which often require destructive sampling and are difficult to perform at large scales. In this study, we explore the use of near-surface remote sensing approaches-leaf-level thermography and hyperspectral reflectance-to estimate key physiological parameters and track seasonal change in a diverse group of plant species growing in a restored mine site of Western Australia. In this system, deviations from expected plant condition are mainly driven by water availability, which translates in various degrees of drought stress. We investigated the relationships between traditionally measured physiological variables and remotely sensed proxy indicators, with the following goals: (1) to assess the extent to which remotely sensed proxy indicators can reliably estimate key physiological variables in functionally diverse plant species, and (2) to explore a multivariate approach to track seasonal change in different species using a combination of remotely sensed proxy indicators. Our results indicate that univariate relationships between physiological variables and their remotely sensed proxy indicator for leaf water content and total chlorophyll concentration were accurately predicted for all species at the leaf level. This was not the case for fluorescence and stomatal conductance. Through multivariate analysis, we found that, when grouping samples through seasonal data centroids, three different data sets (physiological variables, all remotely sensed proxy indicators and remotely sensed proxy indicators from hyperspectral reflectance only) outlined very similar transitions across seasons in four out of the five studied species. This indicates that, when combined, remotely-sensed proxy indicators generated through leaf hyperspectral reflectance have the potential to detect deviations in plant condition in a wide range of plant functional types. Using this approach to detect changes in levels of plant stress could enable early detection of potential issues in restoration and therefore may be used to inform adaptive management to improve long-term restoration strategy and outcomes. Practical implication: through the multivariate approach based on remotely sensed proxy indicators that we present here, plant condition assessments can be conducted more efficiently in the context of ecological restoration.

Keywords

ecological restoration; ecophysiology; hyperspectral reflectance; near-surface remote sensing; proxy indicators of plant condition; thermal imaging

Published in

Ecological Solutions and Evidence
2026, volume: 7, number: 1, article number: e70195

SLU Authors

UKÄ Subject classification

Environmental Sciences
Horticulture
Ecology

Publication identifier

  • DOI: https://doi.org/10.1002/2688-8319.70195

Permanent link to this page (URI)

https://res.slu.se/id/publ/146181