OOS 51-4 - Consequences of multiple global change factors for leaf-level spectra of tree species from the boreal-temperate ecotone.

By the late 21st century, global temperatures are projected to rise significantly, with changes to the amount and timing of rainfall. These changes will likely affect plant physiological processes and consequently the growth, survival, and competitive ability of plants. Previous research shows that a suite of plant chemical, morphological, and physiological traits can be detected from leaf-level spectral reflectance, which provide a means to scale to ecosystem processes through spectroscopic remote sensing. How global change factors such as warming and drought affect spectral signatures is currently not clear. Here, we assessed how experimental warming and rainfall reduction affect spectral reflectance across temperate and boreal tree species.

This research uses the Boreal Forest Warming at an Ecotone in Danger (B4WarmED) climate change research facility located at two sites in the boreal-temperate ecotone of northern Minnesota. At each site, 72 research plots have been established with factorially replicated climate manipulations, namely: i) below and aboveground warming (control, +1.7°C, +3.3°C), ii) environmental light conditions (closed and open canopy), and iii), rainfall removal (control and -40% of summer rainfall), but only in the open canopy. In the 2021 and 2022 growing seasons, we measured full-range leaf-level spectral reflectance (350-2500nm) on nine tree species from the boreal-temperate ecotone, including both gymnosperms and angiosperms and three distinct populations of Quercus macrocarpa (bur oak) Oklahoma, Illinois, and Minnesota. In total, we measured spectra on ~2400 fresh and ~1500 pressed dry leaves.

Using these data, we conducted a series of partial least square discriminant analyses to determine whether: i) tree species can be distinguished both within and among different environments, and ii) the climate treatment in which plants were grown can be identified from green and dry-pressed leaf spectra. Overall, we found that regardless of treatment, species could be distinguished using either green and pressed dry leaf spectra with high accuracy and that trees grown in different climate treatments display some differences in spectral reflectance primarily in visible and shortwave infrared regions. We also found that angiosperms grown in the highest level of warming were moderately distinguishable from those grown in ambient conditions in both closed and open canopy conditions. However, these climate treatments could not be spectrally distinguished with a good confidence for gymnosperms or when intermediate warming and/or rainfall reduction treatments were considered. Our findings show that global changes leave distinct imprints on reflectance spectra with the strongest effects evident through the visible and shortwave infrared.