Basin Scale Snowpack and Soil Moisture Variability Inference from Inter-Annual Tree-Ring Chronologies in the Sierra Nevada

Our study objective is to explore whether tree ring chronologies are sufficiently sensitive to key out information about the snow and soil water content spatial variability regime in mountain ranges dominated by seasonal snowpack. The analysis is carried through the development of a principal component regression model that simulates tree-ring inter-annual radial growth chronologies using site specific climate indices such as the duration of the dormancy season (winter), the duration of the growth season (spring), the duration of the dry season (summer) and the available seasonal soil moisture at the root zone. A highly skilled model can be used to understand the dominant land surface processes that affect tree ring growth in a given location and thus reconstruct these indices from the tree ring chronologies. For the development of this model, we relied on annual chronologies of earlywood and latewood ring width and latewood density that were developed for five conifer species (Tsuga mertensiana, Abies magnifica, Abies concolor, Juniperus occidentalis, and Pinus ponderosa) from five sites within a 60 km radius at the American River watershed and its vicinity. The climate indices were derived for each sampling site by implementing a sub-daily spatially congruent land surface model that simulates snow, soil water content, and actual evapotranspiration for 1915- 2016.

The skill of the principal component regression model is compared to the results from a simple process based model and another principal components regression model that uses monthly time series of temperature and precipitation. In addition, the ability of this model to capture the spatial and temporal variability among the tree sampling sites is examined.