Biophysical Heterogeneity, Hydrologic Connectivity, and Productivity of a Montane Floodplain Forest

Floodplains display exceptional variation in habitat type, connectivity, and vegetation structure that make them ideal landscapes in which to address biophysical controls on primary production. However, our ability to do so requires fine-scale assessment of biophysical complexity over large spatial gradients in habitat heterogeneity, species composition, and productivity. We used LiDAR data and hydrologic modeling to quantify surface elevation, hydrologic connectivity, and a vegetation structural diversity index (VSDI) in 551 patches across a floodplain forest of a montane river corridor. We also estimated terrestrial primary production via the normalized difference vegetation index (NDVI) in each floodplain patch. Our main goal was to compare abiotic and biotic controls on terrestrial primary production using a path analysis model to estimate direct and indirect effects on NDVI values. Across the floodplain, patch inundation was predominantly low and negatively related to mean patch elevation (r = −0.434, p < 0.001) and distance to the river channel (r = −0.397, p < 0.001). Most patches exhibited high VSDI, corresponding to a total canopy cover of 25–65% and an average canopy height of 8.5 m. Path analysis revealed direct effects of inundation on canopy cover and NDVI, indicating abiotic control on both floodplain vegetation distribution and productivity. Canopy cover mediated indirect effects of inundation on vegetation structural diversity, which was in turn a strong mediator of the effects of canopy cover on forest productivity. Our results suggest that coexisting layers of vegetation in a floodplain patch provide complementary functional traits that interact with flooding regime to collectively increase aboveground productivity.