Evaluating Changes in Visible to Short-Wave Infrared Spectral Reflectance of Arctic Mosses in Response to Experimental Drying to Find the Best Predictors of Moisture Content

Mosses are a dominant understory component in the Arctic and because of sparse canopy cover, contribute to spectral signals used in remote sensing estimates of various ecologically important characteristics such as productivity, phenology, and vegetation mapping. However, little is known about their contributions to community level spectra or how moisture content influences those spectral signals. Unlike vascular plants, mosses cannot actively regulate moisture content and are highly susceptible to desiccation. Previous research has shown that moss reflectance is sensitive to tissue moisture content. Here, a lab-controlled drying experiment was conducted to identify the best spectral predictors of moisture content of moss as well as distinguishing characteristics of their spectral profile compared to vascular plants. Additionally, a pilot study tested whether moss could drive community-level reflectance in situ in response to short-term moisture changes. Significant changes in the near infrared and short-wave infrared regions of moss spectra were observed in response to moisture content fluctuations and could be used to determine moisture content. Moisture indices derived from spectral reflectance were able to predict moisture content with a high degree of certainty. The red edge inflection point and slope obtained from derivative spectra were found to be good distinguishing characteristics between moss and vascular plant spectra for the purpose of classification. Lastly, moisture content of moss was shown to significantly drive community-level spectra where moss and vascular plants were interspersed. These findings demonstrate the need to consider whether mosses are present in a spectrally mixed signal and to be aware of moisture content and its effects on overall spectra. Given the influence that both mosses and moss moisture have on overall spectra, incorporating this once semi-forgotten understory component will be critical for better predictions and modeling of the changing Arctic ecosystem