Mountain pine beetle (Dendroctonus ponderosae) and spruce beetle (Dendroctonus rufipennis) epidemics have led to extensive mortality in lodgepole pine (Pinus contorta) and Engelmann spruce (Picea engelmannii) forests in the Rocky Mountains of the western US. The impacts of these disturbances on ecosystem-scale water fluxes are complex, owing to their variable and transient nature. Further, anticipated responses in watershed hydrological processes, including significant decreases in evapotranspiration (ET), have not been observed. In this work, we are using the Terrestrial Regional Ecosystem Exchange Simulator (TREES) model to mechanistically simulate ecosystem-scale ET for two forests impacted by bark beetles, one dominated by lodgepole pine and the other by Englemann spruce. Each site was instrumented with an eddy covariance tower, and ET observations recorded during several consecutive growing seasons of bark beetle epidemics were used to evaluate simulated ET. Four model cohorts, each parameterized individually and scaled spatially within the eddy covariance footprint, were used to simulate live overstory, beetle-attacked overstory, dead overstory, and regenerating understory during each growing season. For the beetle-attacked cohorts, conducting xylem area scalars that reduced whole-tree conductance were developed using temperature-driven models of bark beetle life cycle and blue stain fungal growth, and were calibrated using sap flux data.
Results/Conclusions
In the lodgepole pine forest, simulated ecosystem-scale, half-hourly ET matched observed ET well (R2 = 0.69, slope = 0.97, RMSE = 0.067 mm hr-1). At the same site, over a five-year period, the overstory mortality increased by over 40%, however simulated and observed growing season ET did not show a concomitant decline. By the fifth year of the study, the regenerating understory cohort represented 39% of total simulated growing season ET. Simulation work for the Englemann spruce site is currently ongoing, but the results from the lodgepole pine site highlight the importance of incorporating vegetation-driven compensatory effects in models that simulate ET in beetle-impacted forests. Ignoring such effects could lead to significantly underestimating ET at the ecosystem-scale. These results account for recently reported observations showing increasing understory growth in subsequent years following overstory mortality and provide a mechanistic explanation for the compensatory effects of such growth on ecosystem-scale ET.