UC Berkeley

Quantifying the impacts of climate change on flora is of interest to many watershed and land managers, but has been challenging to study and distinguish from other disturbances that can reshape the landscape. This dissertation has found new tools to assist watershed managers in modeling the impacts of climate change by utilizing large comprehensive spatial datasets derived from remote sensing technology. To understand the recent expansion of the Douglas-fir (Pseudotsuga menziesii) forest within the San Francisco Peninsula Watershed across spatial-temporal scales with respect to increased precipitation over the past century, GIS models were developed using Light Detection And Ranging (LiDAR). The Douglas-fir was selected to be studied within the San Francisco Peninsula Watershed primarily because of its impact on replacing oak woodland, chaparral and grassland vegetation types. Secondly, Douglas-fir has a minimum mean annual precipitation requirement of around 34 inches, which allows it to be more competitive in formerly drier parts of the watershed. Extensive daily precipitation records spanning the last 100 years were recorded by the Spring Valley Water Company and later by the San Francisco Public Utilities Commission (SFPUC), which has provided a unique opportunity to study the Douglas-fir spread pattern influenced by increased rainfall.

The SFPUC rain gauges showed an increase in mean annual precipitation of approximately 15 cm over the past century, which is consistent with independent data on climate research for the central California coast. Using LiDAR, over 12,000 individual overstory Douglas-fir trees were identified and their heights measured. Tree cores were taken of Douglas-fir inhabiting different microclimates to determine their age and rate of growth in the study area of the San Francisco Peninsula Watershed, allowing a species distribution model (SDM) of Douglas-fir to be devised. The high density of rain gauge stations encompassing the Douglas-fir forest spanning the last 100 years provided a unique opportunity to model the advancement pattern of the Douglas-fir forest with respect to historical and future climate scenarios. The results show the Douglas-fir advances more quickly with increased annual precipitation across the entire study area.