Linking LiDAR with streamwater biogeochemistry in coastal temperate rainforest watersheds

The goal of this study was to use watershed characteristics derived from light detection and ranging (LiDAR) data to predict stream biogeochemistry in Perhumid Coastal Temperate Rainforest (PCTR) watersheds. Over a 2-day period, we sampled 37 streams for concentrations of dissolved C, N, P, major cations, and measures of dissolved organic matter quality (specific ultraviolet absorbance, SUVA254) and bioavailability. Random forest - classification and regression tree analysis showed that aboveground biomass and structure and watershed characteristics, inclusive of mean watershed slope and elevation, watershed size, and topographic wetness, explained more than 60% of the variation in concentration for most measured constituents. These results indicate this approach may be particularly useful for predicting stream biogeochemistry in small forested watersheds where fine resolution is needed to resolve subtle differences in forest biomass, structure, and topography. Overall, we suggest that the use of LiDAR in many of the small and remote watersheds across the Southeast Alaskan PCTR as well as other forested regions could help inform land management decisions that have the potential to alter ecosystems services related to watershed biogeochemical fluxes.