River network structuring of climate and landscape effects in salmon watersheds

Author: 
Date created: 
2019-01-09
Identifier: 
etd20069
Keywords: 
Climate change
Fluvial landscape
Fraser River
Thompson River
Unsupervised learning
Abstract: 

Climate change is altering historic patterns of temperature and precipitation worldwide with significant implications for the abiotic and biotic dynamics of river ecosystems. Flowing downhill, precipitation aggregates into creeks, streams and eventually rivers, forming a branching network over the landscape architecturally similar to the branches, limbs and trunk of a tree. Linking disparate locations, these river networks integrate the varied expressions of climate within a watershed. Thus, the river network offers a framework for understanding how spatial patterns of climate are organized and become manifest in rivers. By considering the river network's structuring of climate and landscape interactions, we might better understand how climate and land-use change impact river ecosystems and more clearly identify particularly vulnerable biota. In chapter 2, I examine how river networks dampen signals of climate change in hydrologic flow by integrating varied flow trends from upstream. I demonstrate that by integrating a diverse climate portfolio, the network accumulates changing flow regimes of different volatility, direction and magnitude, such that on average downstream climate change trends are moderated. In chapter 3, I consider the match-mismatch potential of juvenile salmon migrating towards the springtime zooplankton resource pulse in the estuary. I show that populations further from, and whose climate is more dissimilar to, the estuary, are more likely to miss the peak zooplankton bloom. These findings suggest migratory distance influences phenological mismatch risk among populations. My fourth chapter develops an unsupervised machine learning method for cleaning stream temperature data to facilitate big data studies. In chapter 5, I gathered temperature data at over 100 locations throughout a watershed the size of Ireland, over 4 years at 2-hour intervals resulting in over 1 million data points. These data informed a spatial stream network model that quantified how landscape features and river connectivity control seasonal temperature dynamics. These temperature dynamics across space and time revealed that different adult salmon migrations have very different exposures to warm temperatures. Collectively, these findings illustrate that river networks: 1. integrate and dampen signals of climate change, 2. structure phenological match-mismatch patterns and 3. organize thermal exposure potential of biota.

Document type: 
Thesis
Rights: 
This thesis may be printed or downloaded for non-commercial research and scholarly purposes. Copyright remains with the author.
File(s): 
Senior supervisor: 
Jonathan Moore
Department: 
Science: Biological Sciences Department
Thesis type: 
(Thesis) Ph.D.
Statistics: