Hyperspectral indices of Sphagnum moss water content and photosynthetic capacity reflect CO2 and CH4 exchange in a northern peatland ecosystem

Colin Tucker, Northern Research Station, US Forest Service, Madison, WI, Ally O'Neill, College of Forest Resources and Environmental Sciences, Michigan Technological University, Houghton, MI, Karl Meingast, School of Forestry, Michigan Technological University, Houghton, MI, Laura Bourgeau-Chavez, Michigan Tech Research Institute, Michigan Technological University, Ann Arbor, MI, Erik A. Lilleskov, Climate, Fire and Carbon Cycle Sciences, US Forest Service, Northern Research Station, Houghton, MI and Evan S. Kane, School of Forest Resources and Environmental Science, Michigan Technological University, Houghton, MI

Background/Question/Methods
Sphagnum mosses are a dominant driver of carbon cycling in northern ombrotrophic and oligotrophic peatland ecosystems. They can tolerate, and even create, saturated conditions that favor their own growth, suppress that of other plants, and enhance CH4 production. They dominate northern peatland NPP, and have intrinsically recalcitrant tissues, with the net result that Sphagnum-dominated peatlands store a large fraction of the world’s C. Because Sphagna are non-vascular, their water uptake depends on water table position and changes in capillarity. In two mesocosm studies, we evaluated the response of ecosystem CO2 and CH4 cycling to Sphagnum community composition, and altered water tables and rainfall. One goal of the larger study was to evaluate the use of hyperspectral data to detect changes in the hydrology and C cycling in these ecosystems, which we report on here. To do so, we used a field portable spectroradiometer to derive multiple hyperspectral indices of water content and vegetation function. We also measured Sphagnum growth, and ecosystem CO2 and CH4 exchange using static chamber techniques.
Results/Conclusions
Hyperspectral indices were strongly correlated with Sphagnum water content, likely due to the sponge-like water storage capacity of the large, empty hyaline cells that make up the bulk of volume of these mosses. These indices had significant and yet weaker relationships with water table position. Sphagnum growth was strongly related to both water table depth and multiple spectral indices reflecting near surface water content. Net ecosystem production (NEP) reduced by altered rainfall and lowered water tables, with responses that differed among hummock and hollow moss species. NEP and Sphagnum growth showed highly positively correlated responses to near surface moisture indices, with patterns that differed among Sphagnum treatments. CH4 effluxes were highly sensitive to water table, and were lower in hummock compare to hollow mosses, and were strongly correlated with hyperspectral indices of water content. From this, we infer that these hyperspectral indices are promising for detecting changes in the ecosystem C cycle driven by hydrological patterns reflected in the water content and greenness of Sphagnum mosses.