Montane conifer water use under earlier and more episodic snowmelt
AdvisorHarpold, Adrian A.
AltmetricsView Usage Statistics
Snowmelt is the primary source for soil water that supplies summer tree water use in the Mediterranean climate of the Sierra Nevada, California. Inter-annual differences in climate combined with topography (e.g. elevation, slope, and aspect) control the timing and amount of snow water inputs. Yet, few studies have explored how differences in the timing and magnitude of snowmelt affect tree water use in Mediterranean conifer forests. Thus, existing land surface model representation of these processes are not well-verified with observations. In Chapter 2, we used sap flux, meteorological, and hydrological measurements on north and south aspects in the Sagehen Creek watershed (Sierra Nevada, California) to investigate snowmelt controls on the timing of tree water use. We found that tree water use prior to its seasonal peak was controlled largely by air temperature and spring weather patterns caused cooler, cloudier days, that reduced tree water use. Snowmelt-driven differences in the timing of soil moisture limitations led to an earlier peak and subsequent decline of tree water use. Overall, our results suggest that earlier snow disappearance and shifts from snow to rain will potentially shift the peak of tree water use earlier and increase the days that trees are water limited. These observations provided process insights that were then used as the basis for the objective functions to test model skill in Chapter 3. Land surface models summarize our understanding of thermodynamic and hydrologic processes. We used a multi-model framework called the Structure for Unifying Multiple Modeling Alternatives (SUMMA) to test the skill of model decisions (i.e. equations) and parameters used to calculate transpiration. We found that stomatal resistance models of different complexities were able to capture the early season behavior of tree water use that was modulated by spring weather patterns. Models that used the Ball-Berry stomatal resistance equation (most complex representation of stomatal resistance) comprised the majority of the 100 best models in terms of timing of peak and rate of decline of transpiration. We found an unexpected split between shallow and deep root depths in the top-performing models that highlights a need for further testing for potential compensatory effects between different parameters and model decisions. Overall, this work increases understanding of the effects of climate variability and change on tree water use and increases model fidelity for predicting future water use that could support forest and water management.