The Open Collections website will be undergoing maintenance on Wednesday December 7th from 9pm to 11pm PST. The site may be temporarily unavailable during this time.

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Impacts of environmental change on tree productivity in temperate-maritime forest ecosystems Hember, Robbie Andrew

Abstract

This thesis investigated observed responses of forest productivity to environmental change and their predictability using semi-empirical carbon (C) cycle models in temperate-maritime conifer forests in coastal British Columbia, Canada. Effects of environmental stress and historical responses to environmental trends were constrained using observations of gross primary production (Pg) from eddy-covariance flux towers and stemwood growth (Gsw) and mortality (Msw) from permanent forest inventory plots. Observations suggested a long-term increasing trend in Gsw extending back to the Little Ice Age, with decadal fluctuations in association with several 20th century drought episodes. Statistical models driven with climate variability, alone, could not reproduce the observed trend in Gsw, while climate variability and sensitivity to carbon dioxide (CO₂), combined, expressed a moderately strong capacity to reproduce past trends and variability. Observations also indicated substantial wave-like fluctuations in Msw that could not be explained by stand density-dependent processes, alone, while additional functions of drought sensitivity via linear-threshold functions of evapotranspiration (ET) and precipitation (P) improved model predictions. The capacity to predict tree productivity was explored within a more mechanistic modelling framework, focusing on evaluation of physical principles used to simulate Pg in production efficiency models (PEMs) and subsequent application within the established forest productivity model, 3-PG, to simulate Pg, Gsw, and Msw. Comparison with observations highlighted several deficiencies in the representation of environmental stress in PEMs that restrict the capacity to accurately simulate transient responses to environmental change, some of which arise from the model reduction and scaling techniques employed by PEMs, while others reflect unsettled physiological understanding. Consistent with regression model simulations, absence of CO₂ fertilization in 3-PG led to inability to reproduce observed trends in Gsw. This research demonstrated that representation of environmental sensitivity in models of Gsw and Msw does not lead to appreciable increases in model precision, yet is absolutely necessary to achieve temporally-unbiased simulations at the regional scale. Findings also demonstrate the critical role of observation networks, including permanent forest inventories and longterm continuous meteorological and hydrological measurements as a necessary means of advancing and implementing model representation of environmental controls on forest productivity.

Item Media

Item Citations and Data

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International