Open Collections

Detecting novel climates in projections of 21st century terrestrial ecosystem change

University of British Columbia

The emergence of locally unfamiliar climates due to anthropogenic global warming is compelling scientists and resource managers to seek ecological data and management strategies from non-local climates, known as climate analogs. In this context, novel climates—emerging conditions with no analog in the observational record—represent widening gaps in the ecological knowledge base. Identification of novel climates is essential to climate change adaptation. However, methods to detect novel climates have not kept pace with this necessity. The goal of this dissertation is to advance methods for detection of novel climates in the context of ecology and forest management. I develop a multivariate metric for climatic novelty, sigma dissimilarity, that uses the local historical range of interannual climatic variability as a scale for measuring the ecological significance of climatic differences. I apply this metric at three scales—continental, jurisdictional, and local—each of which offers a distinct perspective on the implications of climatic novelty. At the continental scale, I assess the emergence of novel climates in North America, where they are an important source of extrapolation error in ecological modeling. I demonstrate the potential for novel climates to emerge throughout the continent, particularly at low topographic positions. At the jurisdictional scale, I assess the emergence of novel climates that are not represented in a structured knowledge system for forest management—the Biogeoclimatic Ecosystem Classification for British Columbia. A parallel novelty assessment using sigma dissimilarity and random forest classification indicates a robust pattern of novel climates in BC, for which analogs from outside BC must be identified. At the local scale, I demonstrate that dependencies among climate variables can produce larger and earlier departures from natural variability than is detectable in individual variables. This multivariate departure intensification effect—evident in distinct regions of the planet in global climate models—indicates adaptive challenges for ecological and human communities as their local climates become unfamiliar. The identification of locally unfamiliar and regionally novel climates is an important step in anticipating and adapting to climate change. Further, the challenges presented by novel climates are yet another basis to advocate for global emissions reductions.

Text

2018-03-22

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/64884

Forestry

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/