Simulated response of ecosystem processes to climate change in northern California and western Nevada

In order to investigate potential climate impacts on landscape-scale ecosystem processes, I implemented a dynamic general vegetation model (DGVM) over a large domain in northern California and western Nevada on a rectangular grid of ca. 800-meter spatial resolution. I used 100 years of observed, mon...

Full description

Bibliographic Details
Main Author: McGlinchy, Maureen C.
Other Authors: Neilson, Ronald P., Law, Beverly E., Bailey, John, Forest Ecosystems and Society, Oregon State University. Graduate School
Format: Master Thesis
Language:English
unknown
Published: Oregon State University
Subjects:
Online Access:https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/zk51vk74d
Description
Summary:In order to investigate potential climate impacts on landscape-scale ecosystem processes, I implemented a dynamic general vegetation model (DGVM) over a large domain in northern California and western Nevada on a rectangular grid of ca. 800-meter spatial resolution. I used 100 years of observed, monthly climate and nine future climate projections in an attempt to explore the range of possible climate futures in the region. I selected three general circulation models (MIROC3.2(medres), UKMO-HadCM3 and CSIRO-Mk3.0), incorporating a range of 2xCO₂ temperature sensitivity. Each GCM was run through three carbon dioxide emissions scenarios (SRES A2, A1B and B1). For this analysis, I focused the study on the simulated ecological impacts under the three A2 scenarios. Historical observations and future climate scenarios were interpolated to the 800-meter grid by the PRISM model. MC1, a systems-based DGVM, compared favorably to observed data for simulations of vegetation distribution and annual streamflow. MC1 slightly overestimated annual production in the Sierra Nevada and Klamath Mountains and underestimated it in the Coast Range and Eastern Cascades. MC1 displayed a low bias for annual area burned and high bias for pyrogenic emissions. Validation of simulated model output was complicated because MC1 does not consider the effects of land management on ecosystems and the study region is heavily-impacted by development, logging, fire suppression, grazing and pre-European, indigenous land-use and burning. Under all future climate projections, an increase in growing season length and temperature led to the replacement of tundra and subalpine vegetation types with temperate conifer forest. Increased winter minimum temperatures promoted the expansion of mixed needleleaf-broadleaf forest, particularly in the mid-elevations of the Sierra Nevada and in coastal forests. In the MIROC3.2 and HadCM3 scenarios, ecosystem-level net primary productivity (NPP) did not increase with enhanced CO₂ fertilization because production remained ...