GAAT (SOCKEYE SALMON, ONCORHYNCHUS NERKA) MIGRATION UP THE
GAAT HÉENI (SILVER SALMON RIVER): INFLUENCE OF ATMOSPHERIC
RIVERS ON HYDROLOGIC VARIABILITY
by
Devin Wittig
B.Kin., University of Alberta, 2016
B.Ed., Yukon University, 2022

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
NATURAL RESOURCES AND ENVIRONMENTAL STUDIES

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
July 2025
© Devin Wittig, 2025

Committee Members
Supervisor: Stephen Déry. PhD
Department of Geography, Earth and Environmental Sciences
University of Northern British Columbia
Prince George, Canada

Committee Member: Peter Jackson. PhD
Department of Geography, Earth and Environmental Sciences
University of Northern British Columbia
Prince George, Canada

Committee Member: Jonathan Moore. PhD
Biological Science (Primary) and Resource and Environmental Management
Simon Fraser University
Vancouver, Canada

ii

Abstract
Atmospheric rivers (ARs) drive hydrometeorological variability, influencing precipitation, river
discharge and water temperature. This study quantifies how ARs contribute to precipitation and
hydrology in the Gaat Héeni (Silver Salmon River) Watershed, a key migration corridor for
Sockeye Salmon. I integrate ERA5-Land reanalysis and the SIO-R1 AR catalog with in-situ
hydrometric measurements and biological monitoring data, including video observations at a
waterfall barrier and escapement counts from a weir. This approach enables me to evaluate how
AR-driven changes in hydrologic conditions subsequently influence Sockeye Salmon migration
success.
Sockeye Salmon jump success, modelled using logistic regression, was primarily influenced by
river discharge, with peak success occurring at 10–12 m³ s⁻¹ discharge levels. Jump success at
SR3-3 declined at discharge levels above 16 m³ s⁻¹ or below 9 m³ s⁻¹. Although water
temperatures of 13–15 °C coincided with optimal jump success, water temperature was not a
statistically significant predictor, reducing confidence in its influence relative to discharge.
Results show that AR events contributed between 15.9% and 39.1% of seasonal precipitation
from 1991 to 2023, with the highest contributions in fall (37.9%) and winter (24.1%). During the
2024 monitoring season, an AR event on 22–24 July triggered a discharge surge from 8.7 to
30.0 m³ s⁻¹ within two days, reflecting the watershed’s rapid hydrologic response. Centroid lag
analysis revealed a median discharge lag of 1.6 days following AR-driven precipitation,
underscoring the sensitivity of discharge timing to ARs.

iii

Table of Contents
Committee ......................................................................................................................................ii
Abstract..........................................................................................................................................iii
Table of Contents...........................................................................................................................iv
List of Tables.................................................................................................................................vi
List of Figures...............................................................................................................................vii
Acknowledgments..........................................................................................................................ix
Glossary...........................................................................................................................................x
Chapter 1: INTRODUCTION......................................................................................................... 1
1.1 Territorial Acknowledgement ................................................................................................... 1
1.2 Motivation ................................................................................................................................. 1
1.3 Gaat Héeni Watershed Physiography ....................................................................................... 3
1.4 Objectives ................................................................................................................................. 6
1.5 Thesis Structure ........................................................................................................................ 7
Chapter 2: LITERATURE REVIEW.............................................................................................. 8
2.1 Changing Climate ..................................................................................................................... 8
2.2 Meteorological Phenomena .................................................................................................... 10
2.3 Reanalysis Products ................................................................................................................ 13
2.4 Gaat Physiology ...................................................................................................................... 14
2.5 Effects of Hydrologic Variability on Gaat .............................................................................. 17
2.6 Effects of Changing Water Temperature ................................................................................ 20
Chapter 3: DATA .......................................................................................................................... 22
3.1 Reanalysis Data ....................................................................................................................... 22
3.2 Meteorological Data ................................................................................................................ 22
3.3 Hydrologic Data ...................................................................................................................... 24
3.4 SIO-R1 Atmospheric River Catalog ....................................................................................... 24
3.5 Escapement Data ..................................................................................................................... 25

iv

Chapter 4: METHODS ................................................................................................................. 26
4.1 Installation of Meteorological Station ..................................................................................... 27
4.2 Extraction and Validation of Reanalysis Data ........................................................................ 28
4.3 Climatology of the Gaat Héeni Watershed ............................................................................. 31
4.4 Atmospheric River Climatology ............................................................................................. 32
4.5 Impact of ARs on the Study Region ....................................................................................... 33
4.6 Discharge Sensitivity to Daily Precipitation and AR Events.................................................. 35
4.7 Quantifying Optimal Hydrological Conditions for Successful Gaat Barrier Passage ............ 38
4.8 Influence of Jump Success on Gaat arrival at Gaat Áayi........................................................ 42
Chapter 5: RESULTS ................................................................................................................... 44
5.1 Meteorological Data................................................................................................................ 44
5.2 Extraction and Validation of Reanalysis Data ........................................................................ 47
5.3 Climatology of Study Region ................................................................................................. 51
5.4 Atmospheric River Climatology ............................................................................................. 56
5.5 Impact of ARs on the study region ......................................................................................... 58
5.6 Discharge Sensitivity to Daily Precipitation and AR Events.................................................. 62
5.7 Quantifying Optimal Hydrological Conditions for Successful Gaat Barrier Passage ............ 64
5.8 Influence of Jump Success on daily Gaat arrival at Gaat Áayi............................................... 67
Chapter 6: DISCUSSION ............................................................................................................. 69
6.1 Contribution of ARs to Seasonal and Annual Precipitation ................................................... 69
6.2 The Impact of Changing Atmospheric River Patterns on Gaat .............................................. 70
6.3 Hydrological thresholds for Gaat Arrival at Gaat Áayi .......................................................... 72
6.4 Evaluating the Effectiveness of Video Monitoring for Assessing Fish Passage Success ...... 75
Chapter 7: CONCLUSION ........................................................................................................... 76
7.1 Summary ................................................................................................................................. 76
7.2 Study Limitations .................................................................................................................... 80
7.3 Future Direction ...................................................................................................................... 82
REFERENCES ............................................................................................................................. 85

v

List of Tables
Table 3.1: Meteorological Station Metadata. ................................................................................ 23
Table 3.2: Hydrologic Station Metadata ....................................................................................... 24
Table 5.1: Wald Z-Test Results for Hydrologic Predictors of Gaat Jump Success at SR3-3 ........ 67

vi

List of Figures
Figure 1.1 Map of Gaat Héeni Watershed and Regional Context .................................................. 2
Figure 1.2: Annual Gaat Áayi Escapement Totals, 1980–2022 ...................................................... 3
Figure 1.3: Photograph of SR3-3 ....................................................................................................5
Figure 1.4: Taku River Daily Discharge, January 2018–December 2020 ...................................... 6
Figure 2.1: Schematic of Atmospheric River Structure and Intensity .......................................... 10
Figure 2.2: Lower Llewellyn Daily Rainfall 2013-2023 ............................................................. 12
Figure 2.3: Snow Water Equivalent Across Yukon and Northwestern BC .................................. 13
Figure 4.1: Gaat Héeni Meteorological Station Installation ......................................................... 28
Figure 4.2: Flowchart showing methods for Section 4.7 .............................................................. 38
Figure 4.3: Laminar Flow Zone Within the SR3-3 Channel ......................................................... 39
Figure 5.1: Daily Meteorological and Hydrological Observations at Gaat Héeni ....................... 46
Figure 5.2: Validation of ERA5-Land Temperature Data ............................................................. 47
Figure 5.3: Validation of ERA5-Land Snow Water Equivalent Data ........................................... 48
Figure 5.4: Validation of ERA5-Land Snow Depth Data ............................................................. 49
Figure 5.5: Validation of ERA5-Land Daily Precipitation Data .................................................. 50
Figure 5.6: Validation of ERA5-Land Monthly Precipitation Data ............................................. 51
Figure 5.7: Monthly Temperature Distribution in the Gaat Héeni Watershed ............................. 52
Figure 5.8: Monthly Snow Water Equivalent in the Gaat Héeni Watershed................................. 53
Figure 5.9: Monthly Snow Depth for the Gaat Héeni Watershed ................................................. 53
Figure 5.10: Annual Precipitation Distribution in the Gaat Héeni Watershed ............................. 54
Figure 5.11: Monthly Distribution of Total Precipitation, Snowfall, and Rainfall ...................... 55
Figure 5.12: Monthly Atmospheric River Frequency and Net IVT ............................................. 56

vii

Figure 5.13: Atmospheric River Monthly Distribution and Net IVT .......................................... 57
Figure 5.14: Monthly Distribution of Atmospheric River Separated by Decade ......................... 58
Figure 5.15: Seasonal Precipitation Attributed to ARs in the Gaat Héeni Watershed .................. 59
Figure 5.16: Annual Precipitation Attributed to ARs in the Gaat Héeni Watershed ..................... 60
Figure 5.17: Monthly Precipitation Attributed to ARs in the Gaat Héeni Watershed ................... 61
Figure 5.18: Mann-Kendall Trend Analysis: Precipitation in the Gaat Héeni Watershed ............ 61
Figure 5.19: Monthly Contribution of ARs to Total Precipitation, June-October 2024 ............... 63
Figure 5.20: Daily Discharge and Precipitation, June-October, 2025 ......................................... 64
Figure 5.21: Daily Number of Jump Attempts at SR3-3, July 2024 ............................................ 65
Figure 5.22: Predicted Jump Success Dependent on Discharge and Water Temperature ............ 66

viii

Acknowledgements
Gunalchéesh, thank you (Tlingit Language, 2024), to the Taku River Tlingit First Nation
(TRTFN) for supporting this research on their traditional and unceded territory. TRTFN has
stewarded this land for time immemorial, and I hope my research contributes to its continued
stewardship. I would also like to thank Mark Connor, Dr. Stephen Déry, Dr. Peter Jackson,
Dr. Jonathan Moore, Dr. Eran Hood, Lisa Rickard, Robin Brand, Dr. Christopher Sergeant,
Brittany Milner, Dr. Bruno Sobral, Tawni Drinnan, and any reviewers I may have missed for
their time and efforts. I would also like to thank funders, including the Natural Sciences and
Engineering Research Council of Canada, through a Discovery Grant, Polar Research Canada,
and the Council of Yukon First Nations, for their support.

ix

Glossary
Atmospheric River (AR): A synoptic-scale corridor of intense moisture transport (>1500
km), linked to extratropical cyclones and strong low-level winds.
ERA5-Land: A high-resolution (0.1°) reanalysis dataset from the European Centre for
Medium-Range Weather Forecasts (ECMWF), reconstructing historical land-atmosphere
interactions.
Escapement: The proportion of Salmon that successfully migrate upstream to spawn is
provided at an annual timescale.
Gaat (Sockeye Salmon; Oncorhynchus nerka): A keystone anadromous salmonid in the
Taku River Basin, highly sensitive to hydrological shifts.
Gaat Héeni (Silver Salmon River): The river addressed in this research serves as a critical
migration corridor for Gaat.
Gaat Áayi (Silver Salmon Lake): A primary spawning habitat within the Taku River Basin
and the spawning location of Gaat migrating up the Gaat Héeni. It holds high ecological
and cultural significance as the only lake-spawning habitat accessible by the Taku River
Tlingit First Nation without aircraft.
Integrated Water Vapour Transport (IVT): A standardized metric quantifying atmospheric
moisture flux (kg m⁻¹ s⁻¹) to classify AR strength and assess precipitation potential.

x

Optimal Passage Conditions: The hydrometric thresholds defining the discharge and water
temperature ranges that maximize Gaat migration success.
SIO-R1 Atmospheric River Catalog: A dataset from Scripps Institution of Oceanography
tracking AR landfalls along the North American West Coast.
SR3-3: A hydrodynamic barrier within the Gaat Héeni canyon, featuring obstructions and
high turbulence that hinder Gaat passage.

xi

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 1: INTRODUCTION
1.1 Territorial Acknowledgement
I begin by acknowledging that the land on which this research occurred is the traditional and
unceded territory of the Taku River Tlingit First Nation (TRTFN).
1.2 Motivation
The motivation for this project is rooted in the unique characteristics of the Gaat Héeni River
(Silver Salmon River; Figure 1.1). The Gaat Héeni River, characterized by its steep canyon
constriction, multiple natural barriers, continuous meteorological and hydrometric monitoring,
and a multi-decadal record of culturally significant Gaat (Sockeye Salmon; Oncorhynchus nerka)
escapement, provides an opportunity to quantify how atmospheric rivers (ARs) drive hydrologic
variability and influence migration success.
Gaat Áayi (Kuthai Lake), one of the four major Gaat spawning lakes in the Taku River
Watershed, is critical to TRTFN and is the location of one of their focal Gaat monitoring
programs. As the only spawning lake TRTFN can access without aircraft, Gaat Áayi has high
cultural and food security value. From 2006 to 2023, the number of Gaat reaching Gaat Áayi
remained low, with partial run collapses in 2018 and 2021. In contrast, 2024 exhibited a marked
resurgence, with a record escapement of 13,375 Gaat (Figure 1.2). Field work by the TRTFN
Fisheries Department has determined that the canyon in the lower Gaat Héeni presents a
migration barrier to Gaat under both low and high discharge (Taku River Tlingit Fisheries
Department, 2017). Additionally, observations from TRTFN fisheries staff have shown an
increase in the frequency with which the Gaat Héeni Watershed experiences these low and high
discharge conditions (Taku River Tlingit Fisheries Department, 2019).
1

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

Figure 1.1: Map illustrating the extent of the Gaat Héeni Watershed (grey polygon) in northwestern British Columbia, Canada.
Key monitoring infrastructure is highlighted, including meteorological (orange triangles), snow survey (blue circles), and video
monitoring (green squares) stations. The inset map (top right) shows the study area's location within the broader context of
British Columbia and the Yukon border region.

2

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

Figure 1.2: Annual escapement of Gaat (Oncorhynchus nerka) into Gaat Áayi from 1980 to 2024. Gaat Áayi is the only major
spawning lake in the Taku River Watershed accessible to the Taku River Tlingit First Nation without aircraft. Escapement has
remained consistently low from 1991-2023, with near collapses in 2018 (13) and 2021 (26). In contrast, 2024 recorded the
highest escapement on record, with 13,375 Gaat returning to spawn. No escapement data were collected from 1982 to 1991.

The selection of this region for my work is driven by the limited research coverage on ARs and
impacts on meteorological variables in the area, the potential impact of ARs to the hydrology of
the Gaat Héeni, and the effect of river discharge and water temperature on Gaat. The presence of
the Gaat Héeni canyon as a significant migration obstacle for Gaat (Taku River Tlingit Fisheries
Department, 2022), further underscores the importance of investigating the influence of ARs on
hydrologic conditions in the region. By examining the impacts of ARs on Gaat migration and
hydrologic variability, the study aims to contribute to the conservation of culturally important
Gaat populations and inform management strategies in the face of changing environmental
conditions.
1.3 Gaat Héeni Watershed Physiography
The Gaat Héeni Watershed lies on the lee side of the Coast Mountains, resulting in lower
precipitation and runoff compared to more western watersheds (Taku River Tlingit Fisheries
Department, 2017). Gaat Áayi, shaped by subglacial processes, has substrates of chert,
3

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

limestone, and argillite (Aitken, 1958) favorable for Gaat spawning (Eiler et al., 1992).
Downstream of Gaat Áayi, Gaat Héeni meanders through an alluvial deposit, creating large
groundwater aquifers that enhance conditions for Gaat spawning and rearing (Taku River Tlingit
Fisheries Department, 2017). For the final 700 m before its confluence with the Nakina River,
Gaat Héeni descends steeply through a canyon characterized by multiple geomorphic barriers,
including the obstruction designated as SR3-3 (Taku River Tlingit Fisheries Department, 2017;
Figure 1.3).
The Gaat Héeni canyon is divided into three sub-reaches (Taku River Tlingit Fisheries
Department, 2017) based on geomorphic features and channel obstructions that influence water
flow and fish movement. The first segment, Sub-reach 1 (SR1), extends approximately 150 m
upstream from the Nakina River confluence. It is a boulder channel comprising a series of pools
with a gradient of about 11%, tightly confined by bedrock and large boulders (Taku River Tlingit
Fisheries Department, 2017). Sub-reach 2 (SR2) has a length of around 420 m with a gentler
gradient of 4% and does not present any significant migration barriers (Taku River Tlingit
Fisheries Department, 2017). The final segment, Sub-reach 3 (SR3), represents the upper canyon
portion, extending for approximately 110 m with an 11% slope (Taku River Tlingit Fisheries
Department, 2017). Segment SR3 contains five rated obstacles; SR3-3 is the most pronounced,
serving as the principal geomorphic constraint. SR3-3 consists of a protruding drop with the
vertical height between pools being approximately 2.5 m (Taku River Tlingit Fisheries
Department, 2017).

4

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

Figure 1.3: Photograph of SR3-3, the primary hydraulic barrier in the Gaat Héeni canyon. SR3-3 consists of a cascading drop
with a vertical separation of approximately 2.5 m between pools (Taku River Tlingit Fisheries Department, 2022).

Except for its alpine ridges and scattered wetlands, the Gaat Héeni Watershed is largely forested
and spans an elevation gradient of 303 m to 1766 m. Lacking long-term hydrometric stations,
hydrologic parameters are calculated by utilizing reanalysis data (Section 3.1). United States
Geological Survey (2024) data for the Taku River, a station near Juneau, Alaska, highlights
typical spring snowmelt runoff and fall storm-influenced high flows, excluding summer glacial
outbursts (Figure 1.4).

5

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

Figure 1.4: Hourly discharge (m³ s⁻¹) from the Taku River near Juneau, Alaska, recorded by the United States Geological Survey
(USGS) between 2018 and 2020. The hydrograph reveals a consistent seasonal pattern dominated by spring snowmelt and late
summer to fall rainfall events. Discharge exceeds 1000 m³ s⁻¹ multiple times yearly, reflecting contributions from snowmelt and
intense precipitation events. (United States Geological Survey, 2024)

1.4 Objectives
The objectives of this research are:
1.

To assess the contribution of ARs to meteorological variables within the Gaat Héeni
Watershed.

2.

To identify the water temperature and river discharge conditions that allow for the
successful transit of Gaat to Gaat Áayi.

3.

To provide fisheries managers with tools to identify impactful flow and/or water
temperature values.

6

1. INTRODUCTION

GAAT MIGRATION UP THE GAAT HÉENI

1.5 Thesis Structure
This thesis is structured into seven chapters, each contributing to the discussion of how ARs
influence hydrological variability and Gaat migration. Chapter 2 provides a literature review
covering AR dynamics, climate change, and hydrological variability, emphasizing the role of
ARs in shaping freshwater ecosystems and Gaat migration. It also examines the strengths and
limitations of reanalysis datasets, including the ERA5-Land reanalysis product. Chapter 3
outlines the datasets used within this work, including ERA5-Land (Muñoz-Sabater et al., 2021),
in-situ meteorological and hydrometric data from stations within the Gaat Héeni Watershed, the
SIO-R1 AR catalog (Gershunov et al., 2017) and escapement records from the Taku River Tlingit
First Nation Fisheries Department. Chapter 4 validates ERA5-Land reanalysis data against insitu observations, outlines the analytical methods used to identify AR events, quantifies their
effects on discharge and precipitation, and models how discharge and water temperature
influence Gaat migration success. Chapter 5 presents the results, including validation of ERA5Land precipitation and temperature data, characterization of AR climatology and impacts in the
study region, analysis of AR-driven changes in daily precipitation and discharge, and statistical
relationships between AR-driven hydrologic variability and Gaat migration. Chapter 6 discusses
these findings in the context of hydrological change, fisheries management, and Gaat
conservation. Finally, Chapter 7 synthesizes the thesis’s key conclusions, limitations, and future
direction.

7

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 2: LITERATURE REVIEW
This literature review examines hydrometeorological factors shaping hydrological variability and
Gaat escapement. The review begins by exploring the impacts of climate change, including
shifting temperature regimes, altered precipitation patterns, and evolving snowmelt dynamics. It
subsequently analyzes ARs, discussing climatology, moisture transport mechanisms, and
influence on precipitation and river discharge. The review then explores the impacts of climate
change, including shifting temperature regimes, altered precipitation patterns, and evolving
snowmelt dynamics. It also evaluates the applicability and limitations of reanalysis products,
such as ERA5-Land, in modelling hydrometeorological trends. Additionally, it investigates Gaat
physiology, focusing on metabolic constraints, behavioural adaptations, and environmental
stressors affecting migration. Finally, the literature review synthesizes research on the effect of
river discharge and water temperature on Gaat.
2.1 Changing Climate
Climate change is a long-term alteration in climate patterns, identifiable through statistical
analysis of changes in mean and variability of climate properties, persisting for decades or longer
(IPCC, 2025). It can result from natural processes, such as volcanic eruptions (Gerlach, 2011), or
anthropogenic influences, notably increasing greenhouse gas concentrations and land use
changes (Wood et al., 2023). These changes manifest as rising global temperatures, altered
precipitation patterns, and more frequent extreme weather events (Frandsen & Hasselbalch,
2024).
Climate change is significantly affecting British Columbia, Southeast Alaska, and Yukon
(BCSAKY), particularly through shifts in temperature, snowfall, and precipitation patterns

8

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

(Fraser & Jackson, 2018; Hawcroft et al., 2012). Rising temperatures are causing earlier spring
thaws and shifting vegetation dynamics, with projections indicating a 3.9–6.9 °C increase in
mean air temperature from 1961–1990 by the end of the twenty-first century (Reid et al., 2022;
SNAP-EWHALE, 2012). Warming also alters snow accumulation and melt timing (Foster et al.,
2016). The proportion of rain and rain-on-snow events is accelerating, leading to reduced spring
snowpack and earlier melts (Arora et al., 2025; Thériault et al., 2023), which heightens spring
flood risks and reduces water availability in drier months (Hay & McCabe, 2010; Foster et al.,
2016). Additionally, increased rainfall intensity is expected to exacerbate flooding and disrupt
ecological systems reliant on stable precipitation patterns (Hay & McCabe, 2010).
The Yukon River, whose headwaters lie adjacent to the Gaat Héeni Watershed, is experiencing
rapid climate change, with rising precipitation and air temperatures (Hay & McCabe, 2010).
Higher water temperatures are inducing heat stress in Chinook Salmon (Oncorhynchus
tshawytscha) during spawning migrations (Cunningham et al., 2018; Howard & Von Biela,
2023). These changes are impacting both freshwater ecosystems and marine habitats in the
Pacific Ocean bordering the BCSAKY region (Feddern et al., 2023).

9

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

2.2 Meteorological Phenomena
The meteorological phenomenon examined in this thesis is the atmospheric river (Figure 2.1).
ARs are defined as elongated (>1500 km) bands of atmospheric moisture transport, typically
associated with strong low-level winds ahead of the cold front of an extratropical cyclone
(Gimeno et al., 2012; Gomis-Cebolla et al., 2023; Ralph et al., 2020; Sharma & Déry, 2020). A
key characteristic of ARs is their high integrated water vapour transport (IVT), which quantifies
the total water vapour flux through a vertical column of the atmosphere. Studies have established
an IVT threshold of ≥ 250 kg m⁻¹ s⁻¹ for AR classification (Espinoza et al., 2018; Gershunov et
al., 2017; Ralph et al., 2019).

Figure 2.1: Schematic summary of the structure and strength of an atmospheric river, illustrating integrated water vapour
transport (IVT) and integrated water vapour (IWV; Glossary of Meteorology, 2022). ARs are narrow corridors of intense moisture
flux, typically >1500 km in length, linked to extratropical cyclones and low-level jets.

Strong low-level winds in ARs often precede the cold fronts of extratropical cyclones,
influencing weather patterns and precipitation (Payne et al., 2020). These winds enhance
moisture transport in the lower troposphere, where ARs interact with the low-level jet stream
(Park et al., 2021). When moist air is lifted over topography or converges with frontal systems,
10

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

extreme precipitation can result (Lavers et al., 2020). Additionally, extratropical cyclones
facilitate the development of low-level jets, which transport warm, moist air from the tropics into
mid-latitudes (Guo et al., 2020). This interaction drives heavy precipitation and intensifies
cyclogenesis, reinforcing storm development (Eiras-Barca et al., 2018). Understanding these
processes is essential for understanding the formation of ARs (Lavers et al., 2020).
ARs provide significant contributions to total precipitation (Sharma & Déry, 2020) and
fluctuations in river discharge (Dettinger, 2011; Ralph et al., 2020). Landfalling ARs trigger
increases in river discharge, often producing rapid runoff and flooding in affected basins (Ralph
et al., 2019). Within the British Columbia and southeastern Alaska (BCSAK) region, ARs
contribute between 11% to 22% of river runoff (Sharma & Déry, 2020). Moreover, these events
can elevate runoff levels by 12.5 times compared to non-AR-related runoff (Zhou et al., 2025).
The heightened runoff accelerates water transport into river systems, resulting in elevated
discharge levels (Davie & Quinn, 2019). ARs also contribute between 5% and 33% of total
precipitation in the BCSAK region (Sharma & Déry, 2020). As the frequency of ARs increases,
particularly in the fall season, their associated precipitation is expected to rise through 2100
(Radić et al., 2015). These shifts are likely to heighten the risk of flooding and other hydrological
hazards, with significant environmental and societal consequences (Rahman et al., 2023). This
finding is corroborated by Sobral & Déry (2024), who reported that ARs significantly influenced
the water budget in the Nechako River Watershed, situated in northern British Columbia. Their
study revealed that ARs account for 21% of the basin's annual precipitation, with contributions
reaching up to 45% of seasonal rainfall and 24% of seasonal snowfall.
Data provided by Sagar (2023) and supported by the SIO-R1 AR Catalog (Gershunov et al.,

11

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

2017) identified an AR event on 1 December 2020 at the Llewellyn Glacier weather station
(Figure 2.2). This event resulted in 106.6 mm of rain at Llewellyn Glacier (Sagar, 2023). This
observed AR may have contributed to the 209% of seasonal Snow Water Equivalent (SWE;
Figure 2.3) seen in Atlin during March 2021 (Kolot et al., 2021).
TRTFN Fisheries observed elevated flows through the Gaat Héeni canyon during the 2021 Gaat
migration period, which may have resulted from the earlier AR event and subsequent snowpack
conditions.

Figure 2.2: Time series of daily rainfall (mm) recorded at the Llewellyn Glacier meteorological station. An atmospheric river
event on 1 December 2020 delivered 106.6 mm of rainfall (Sagar, 2023).

12

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

Atlin
Figure 2.3: Spatial distribution of snow water equivalent (SWE) as a percentage of the historical median across Yukon
Territory. SWE values in the southern and central Yukon were substantially above average, with widespread measurements
exceeding 130%. Elevated SWE in Atlin, 209% of seasonal, indicates deep snowpack storage potential preceding the spring
freshet (Kolot et al., 2021).

2.3 Reanalysis Products
Reanalysis products are comprehensive datasets that provide a detailed record of various
meteorological and surface variables over a historical period. These products integrate model
data with in-situ observations, including ground-based and satellite measurements, to produce a
consistent global dataset (Hersbach et al., 2020). Reanalysis products such as ERA5 (Hersbach et
al., 2020), ERA5-Land (Muñoz-Sabater et al., 2021), CaSPAr (Mai et al., 2020), Daymet
(Thornton et al., 2022), and ClimateNA (Wang et al., 2016) provide comprehensive datasets for
meteorological and surface variables over historical periods.
Given its centrality to my research, I will provide a more detailed discussion of the ERA5
(Hersbach et al., 2020) and ERA5-Land products (Muñoz-Sabater et al., 2021). The fifth
generation of the ECMWF Reanalysis (ERA5), developed by the European Centre for Medium13

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

Range Weather Forecasts (ECMWF) as part of the Copernicus Climate Change Service, is a
prominent reanalysis product (Hersbach et al., 2020). ERA5 assimilates a broad range of satellite
and in-situ measurements to develop a numerical model. This methodology generates hourly
global datasets with a horizontal resolution of 0.25° × 0.25° (~27.75 km × 27.75 km) for the
atmosphere and 0.5° × 0.5° (~55 km × 55 km) for ocean waves (Hersbach et al., 2020). ERA5
provides a high-resolution record of the global atmosphere, land surface, and ocean waves from
1950 to the present, with data updated in near real-time, offering an invaluable resource for
climate and weather analysis (Hersbach et al., 2020). ERA5 exhibits improved agreement with
observations in terms of temporal and spatial correlations compared to other reanalysis products,
making it a valuable resource for climatological and meteorological studies (Tarek et al., 2020).
ERA5-Land enhances the capabilities of the ERA5 dataset by providing a higher spatial
resolution of 0.1° (~9 km) and focusing on land surface variables, offering up to 53 parameters
related to the water and energy cycles (Muñoz-Sabater et al., 2021). This refined dataset is useful
in mountainous regions and winter months due to its ability to distinguish between rain and
snow, improving the representation of precipitation dynamics (Muñoz-Sabater et al., 2021).
2.4 Gaat Physiology
Gaat, Sockeye Salmon (Oncorhynchus nerka), is one of the smaller species of Pacific Salmon,
typically measuring ~50 to ~80 cm in length and weighing between ~2 to ~7 kg (Beacham et al.,
2004; Fisheries, 2025). A 2009 study of Fraser River (British Columbia) Gaat reported a mean
fork length of ~59 cm and an average weight of ~2.5 kg for adult fish sampled (Thomas, 2010).
Gaat follow a complex life cycle, requiring physiological adaptations throughout its three to fiveyear lifespan (Fisheries, 2025; Hruska et al., 2010). Most Gaat populations exhibit a "lake-type"
14

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

life history, where the fry emerge from river gravel, migrate to a lake for 1-3 years before
becoming smolts and migrating to the Pacific Ocean to grow (Beacham et al., 2004; Fisheries,
2025). After two to five years at sea, they return to freshwater to spawn and die (Beamish et al.,
2016; Fisheries, 2025; Quinn, 2005; Ruggerone & Irvine, 2018). This spawning typically occurs
over the summer or fall (Beacham et al., 2014; English et al., 2005).
Anatomically, Gaat have a broad head, large eyes, and a terminal mouth lined with sharp teeth
(Fisheries, 2025). Gaat have large eyes positioned to the sides of their heads that allow them to
spot prey, but are also adapted to see a wide range of light, including ultraviolet. Spawning phase
males develop an exaggerated snout as well as a humped back referred to as a kype (Fisheries,
2025; Morton, 1965). At the juvenile stages, Gaat's diet consists primarily of zooplankton and
shifts to encompass a variety of organisms after they migrate to the ocean (Farley et al., 2007;
Fisheries, 2025). These organisms include squid, decapod larvae, Pacific sandlace (Ammodytes
hexapterus; Tyler et al., 2001), and micronektonic organisms (Radchenko & Mathisen, 2004).
Upon beginning their upriver migration to spawn, Gaat cease feeding (Rand et al., 2006) and rely
on endogenous energy stores (Eliason et al., 2011).
Gaat exhibit a striking transformation in coloration during their spawning phase, transitioning
from a silvery-blue oceanic appearance to vibrant red bodies and greenish heads (Khrustaleva,
2024). The bodies of male Gaat turn bright red, and their heads take on a greenish hue,
enhancing their visibility and appeal to females (Foote et al., 2004). Females also undergo a
colour change, becoming redder, though their coloration is typically less intense than that of
males (Hanson & Smith, 1967).
The swimming performance of Gaat influences their growth and reproductive success (Burt et
15

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

al., 2012; Kocan et al., 2006). Migration over long distances imposes significant energetic
demands, requiring Gaat to maintain sustained aerobic activity and exert bursts of anaerobic
power to navigate turbulent river currents and natural obstacles (Birnie-Gauvin et al., 2023).
Numerous studies have documented the adverse effects of environmental stressors such as rising
water temperatures, infectious diseases, and pollutants on swimming ability, which ultimately
reduce migration success and increase physiological strain (Burt et al., 2012; Kocan et al., 2006;
Wagner et al., 2005).
During their upriver migration, Gaat rely on lipid reserves accumulated during their oceanic
phase, which are metabolized to support the sustained effort required for migration (Crossin et
al., 2009; Eliason et al., 2011; Magnoni et al., 2006). Migration speed varies considerably among
Gaat populations, driven by environmental conditions and population-specific physiological
traits (Crossin et al., 2004; Hinch & Bratty, 2000; Quinn, 1988). In the Fraser River, the average
swimming speed of migrating Gaat has been estimated at 2.4 km h-1 (Quinn, 1988). This
supports work showing that the most efficient swimming speeds for mature Gaat are estimated at
1.8 km h-1, with efficient swimming occurring within the range of 1.0–2.6 km h-1 (Hinch &
Bratty, 2000). Factors such as water temperature and river discharge exert a significant influence
on energy expenditure during migration, with elevated temperatures accelerating metabolic rates
and hastening depletion of energy stores, thus increasing the risk of premature exhaustion
(Carey et al., 2019; Groot, 1995; Prystay et al., 2020; Rand et al., 2006). Furthermore, migration
conditions such as elevated river discharge conditions further strain these energy reserves,
contributing to elevated mortality rates in this critical upriver migration phase (Crossin et al.,
2009; Rand et al., 2006).

16

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

As ectothermic fish, Gaat’s body temperature is in direct thermal equilibrium with the water they
inhabit, meaning their metabolic rate is tightly coupled to water temperature (Crossin et al.,
2008). This thermal equilibrium directly influences the metabolic and physiological processes of
Gaat, including: metabolic rate (Crossin et al., 2008; Young et al., 2006), migration speed
(Hanson et al., 2008), and swimming performance (Eliason et al., 2011; Farrell, 2009). The
relationship between environmental temperature and metabolic rate in salmon is wellestablished, with warmer water leading to higher energy demands (Hanson et al., 2008), reduced
oxygen uptake (Brett, 1971; Steinhausen et al., 2008) and increased physiological stress (Eliason
et al., 2011). Supporting this, Eliason et al. (2011) found that at least 50% of the aerobic scope is
needed for the Coastal Weaver Gaat populations to reach their spawning area. These ectothermic
characteristics limit Gaat's ability to maintain optimal physiological function when
environmental conditions deviate from historical norms (Crossin et al., 2008). In response to
rising temperatures, specific populations of Gaat have adjusted their migration timing or sought
out cooler water refugia, showing behaviour modification to environmental stressors (Goniea et
al., 2006; Mathes et al., 2010). For example, Coastal Weaver Gaat have shifted their migration
period up to six weeks earlier from 1995 to 2008 (Eliason et al., 2011). Additionally, modelling
has shown that with a projected 2 °C increase in river temperature by 2100, Fraser River Gaat
will migrate ∼10 days earlier than Gaat in 2010 (Reed et al., 2011). Finally, Gaat in the
Columbia River system now initiate their migration 10.3 days earlier than in the 1940s, likely as
a direct response to warming water temperatures (Carey et al., 2017; Crozier et al., 2011).
2.5 Effects of Hydrologic Variability on Gaat
Variation in river discharge influences Gaat migration success, timing, and survival to spawning
grounds (Henderson et al., 1992; Quinn et al., 1997; Rand et al., 2006). The magnitude and
17

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

seasonality of these discharge patterns are affected by hydrometric inputs including changes in
snowfall (Déry et al., 2016; Kang et al., 2014), rainfall (Milliman et al., 2008; Modarres &
Ouarda, 2013), air temperature (Blöschl, 1991), and anthropogenic disturbances (Bathurst et al.,
2017; Graf, 2006; Huang et al., 2015). Moreover, as discussed in Section 2.1, meteorological
phenomena including ARs can further amplify runoff, increasing it by up to 12.5 times (Zhou et
al., 2025).
Building on this, watershed characteristics further modulate how hydrometric inputs affect river
discharge. In boreal forests, the seasonal freeze-thaw cycle increases runoff and infiltration when
soils are frozen, while also limiting groundwater recharge (Shanley & Chalmers, 1999).
Meanwhile, the structure and composition of forest stands influence soil infiltration capacity,
with mature forest cover enhancing water retention and reducing surface runoff (Rahmat et al.,
2018). However, differences in forest age, soil type, and vegetation characteristics affect the
extent to which watersheds can absorb and store water (Rahmat et al., 2018). Clearcutting in
Finland’s boreal forests led to elevated runoff rates, gradually returning to baseline levels as
forest cover regenerated (Ide et al., 2013). Climate change further complicates watershed-scale
hydrology, with warmer temperatures leading to increased snow accumulation and earlier
snowmelt, increasing spring discharge in high-latitude areas (Adam et al., 2009; Fallot et al.,
1997).
The timing, frequency, magnitude, and duration of discharge events (Poff et al., 1997) shape a
range of fluvial habitat characteristics. These include channel morphology (Hauer et al., 2011;
Mosley, 1982), sediment transport (Lenzi et al., 2006; Turcotte et al., 2011), flow refugia
(Mathers et al., 2022; Sueyoshi et al., 2014), thermal refugia (Mathers et al., 2022), bank erosion

18

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

(Hooke, 1979), extensive woody debris recruitment (Comiti et al., 2016), water quality
(Mattikalli et al., 1996), hydrologic connectivity (Grantham, 2013), and the thermal regime
(Dickson et al., 2012; White et al., 2017). In the Tagliamento River, a large gravel-bed system
draining the Italian Alps, floods approaching 200% of bankfull discharge mobilise the entire
braid-belt, entirely reworking high bars and islands (Surian et al., 2009). These elevated
discharge levels drive bank retreats up to 6 m, forcing channel realignment and redistribution of
coarse bed material (Surian et al., 2009). Additionally, maintaining hydrologic connectivity
requires minimum discharge thresholds that vary for each section of a river. Grantham (2013)
found that flow requirements for sustaining connectivity varied among different sections of
Sausal Creek, a tributary of California’s Russian River Basin.
Variation in river discharge influences the bioenergetic feasibility of Gaat migration and
subsequent spawning success (Hinch & Rand, 1998; Rand et al., 2006). Elevated Fraser River
discharges in 1997 (7500 m3 s-1, ~50 % above the 1956 low-flow benchmark) led to simulated
energy depletion in Gaat 1000 km upstream (Rand et al., 2006). In 2001, under lower discharge
levels (5400 m³ s-1), Gaat reached the spawning grounds with energy reserves within normal
ranges (Hinch & Rand, 1998). Yet it is prolonged sub-bankfull flows that most acutely constrain
migrations. In British Columbia's Koeye River, mature Gaat avoided freshwater during periods
of low water, delaying until rainfall restored river discharge (Atlas et al., 2021). Similarly, low
discharge contributed to the near-complete failure of the Snake River Gaat run in 2015 (Crozier
et al., 2020).
Within the same watershed, the Taku River Tlingit Fisheries Department has observed that both
high and low discharge on the Gaat Héeni River can obstruct migration, documenting cases of

19

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

fish holding downstream during extreme flood and drought events (TRTFN Fisheries
Department, 2022).
2.6 Effects of Changing Water Temperature
Water temperature influences the migration of Gaat by affecting metabolic rates (Atlas et al.,
2021; Islam et al., 2019), aerobic scope (Eliason et al., 2011), stress responses (Mathes et al.,
2010), and behavioural decisions (Jeffries et al., 2012; Martins et al., 2012). Field and modelling
studies have demonstrated thresholds and dose-response relationships between temperature and
migration success in natural watersheds (Atlas et al., 2021; Eliason et al., 2011).
River discharge influences water temperature, with high discharge buffering rivers against
heating by adding volume and accelerating downstream transport. Low river discharge can result
in shallow, slow-moving water that warms rapidly. In the Koeye River, discharge levels below
4 m³ s⁻¹ elevated water temperatures by 2–3 °C, resulting in the mortality of half the Gaat
population (Atlas et al., 2021). Modelling in the Fraser River showed a 20 % summer-flow
decline yields ~0.9 °C warming and doubles days >20 °C (Islam et al., 2019). Thus, elevated
summer discharge tempers peak water temperatures, while drought-level lows intensify thermal
stress on migrating Gaat.
Across decadal time frames, persistent elevation of water temperature in the Fraser River has
intensified metabolic stress on migrating Gaat (Islam et al., 2019). Sustained exposure of Gaat to
water temperature ≥19 °C has been causally linked to prolonged en route holding, delayed arrival
at spawning grounds, and heightened pre-spawn mortality (Jeffries et al., 2012; Martins et al.,
2012).

20

2. LITERATURE REVIEW

GAAT MIGRATION UP THE GAAT HÉENI

Telemetry and escapement records for the Columbia–Snake River corridor reveal contrasts in
successful Gaat migration. During the hot, low discharge summer of 2015, only 8–12 % of adult
Snake River Gaat reached their headwater spawning grounds after passing Bonneville Dam,
compared with the typical 60–80 % survival (Crozier et al., 2020). River temperatures held at
22–23 °C for prolonged periods, forming a thermal barrier that halted migrations and triggered
mass stranding and disease outbreaks (Crozier et al., 2020).
Behavioural studies confirm that Gaat actively avoid elevated water temperatures by seeking
thermal refugia. Telemetry studies of early-run Gaat in the Harrison River reveal a marked
thermal constraint on migration success. When river temperatures reached highs of 21.5 °C,
16% of tagged Gaat reached their natal grounds. In contrast, when entry temperatures remained
below 18 °C, 72% of migrating Gaat completed the journey without resorting to thermal refuge
(Mathes et al., 2010).
Emerging evidence suggests Gaat are adapting to changes in water temperature
(Armstrong et al., 2016; Eliason et al., 2011; Reed et al., 2011). In the Fraser River, Gaat
populations traversing long, high-energy corridors possess greater aerobic scope and higher
thermal optima than coastal Gaat from shorter, cooler routes (Eliason et al., 2011). Fraser River
field studies likewise show that warm migration years favour individuals with superior
cardiorespiratory performance, strengthening selection for these traits (Armstrong et al., 2016).
In contrast, eco-evolutionary simulations for Columbia River Gaat (Snake River stock) project
that the first upstream entry could advance by ~10 days under a +2 °C scenario, cutting quasiextinction risk by 83 % relative to non-evolving populations (Reed et al., 2011).

21

3. DATA

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 3: DATA
Chapter 3 presents the datasets used in this thesis. For additional details on the region's
physiography, refer to Section 1.3 and Figure 1.1. The datasets introduced below include:
reanalysis datasets (Section 3.1), meteorological data (Section 3.2), hydrological data
(Section 3.3), the SIO-R1 Atmospheric River Catalog (Section 3.4), and escapement data from
Gaat Áayi (Section 3.5).
3.1 Reanalysis Data
ERA5-Land (Muñoz-Sabater et al., 2021) data were extracted over the Gaat Héeni Watershed
domain (55°N–58.5°N, 132°W–134°W) to assess long-term meteorological variability. The
dataset, with a 0.1° (~9 km) spatial resolution, was extracted at an hourly timescale for 1991–
2024 and aggregated to daily values for comparison with in-situ meteorological and hydrometric
records.
The extracted variables were air temperature (°C), snow depth (cm), SWE (mm), total
precipitation (mm), geopotential (m2 s−2), and snowfall (m of water equivalent). These variables
were used to analyze seasonal and interannual variability in precipitation phase partitioning,
snow accumulation, and temperature trends.
3.2 Meteorological Data
Meteorological data were obtained from four stations in and around the Gaat Héeni Watershed to
characterize temperature, precipitation, snow depth, and SWE. These stations, listed in Table 3.1,
include: Llewellyn Glacier, Atlin (ECCC Station 1200560), Gaat Héeni, and the Atlin Lake
Snow Survey site (Government of British Columbia Station 4E02B), each providing hourly
meteorological observations.
22

3. DATA

GAAT MIGRATION UP THE GAAT HÉENI

The Llewellyn Glacier Station (2013–2024) records total rainfall (mm), whereas the Atlin Station
(1981–2021; ECCC Station 1200560) provides a long-term record of air temperature (°C), snow
depth (cm), and total rainfall (mm). The Gaat Héeni Station, installed in 2024 (Section 4.1),
records total precipitation (mm), wind speed (km h⁻¹), air temperature (°C), relative humidity (%)
and atmospheric pressure (hPa). The Atlin Lake Snow Survey Station (2004–2024; Government
of British Columbia Station 4E02B) provides SWE (mm) and snow depth (cm) measurements.
Table 3.1: Meteorological station metadata (Environment and Climate Change Canada, 2024; Government of British Columbia,
2024).

Station

Climate ID

Latitude (oN),
Longitude (oW)

Variables

Elevation (m)

Period of
Record

Llewellyn Glacier N/A

Rainfall (mm)

59.1054, 134.0379

890

2013-2024

Atlin

Snow Depth (cm)

59.5780, 133.6895

696

1981-2021

59.1538, 133.0428

533

2024

59.5886, 133.7102

730

2004-2024

1200560

Air Temperature (°C)
Total Precipitation (mm)
Snowfall (cm)
Rainfall (mm)
Gaat Héeni

N/A

Total Precipitation (mm)
Wind Speed (km h-1)
Air Temperature (°C)
Relative Humidity (%)
Atmospheric Pressure (hPa)

Atlin Lake

4E02B

SWE (mm)
Snow Depth (cm)

23

3. DATA

GAAT MIGRATION UP THE GAAT HÉENI

3.3 Hydrologic Data
Hydrologic data from two stations in the Gaat Héeni Watershed monitor water temperature, river
discharge, and river level (Table 3.2). Each station records data at an hourly resolution, allowing
for detailed analysis of hydrological variability. The Lower Gaat Héeni Station records hourly
water temperature (°C) from 2021 to 2024 whereas the Gaat Héeni Station, installed in 2024,
measures hourly river discharge (m³ s⁻¹ ), level (m), and velocity (m s⁻¹).
Table 3.2: Hydrologic station metadata (C. Sergeant, personal communication, 15 January 2024).

Station
Variables
Lower Gaat Héeni Water Temperature (oC)
Gaat Héeni
Discharge (m3 s-1)
Level (m)

Latitude (oN), Longitude (oW)
59.1152, 133.0025
59.1538, 133.0428

Elevation (m) Period of Record
315
2021-2024
533
2024

Velocity (m s⁻¹)

3.4 SIO-R1 Atmospheric River Catalog
Although multiple AR identification tools exist at both global and regional scales (Gershunov et
al., 2017; Guan & Waliser, 2019; Vallejo-Bernal et al., 2023), this thesis uses the Scripps
Institution of Oceanography's (SIO)-R1 AR Catalog (Gershunov et al., 2017;
http://cw3e.ucsd.edu/Publications/SIO-R1-Catalog/, accessed 5 March 2025) to characterize AR
activity and IVT. The SIO-R1 catalog was constructed using 6-hourly specific humidity and
wind data at a 2.5° × 2.5° spatial resolution, derived from the NCEP–NCAR (National Centers
for Environmental Prediction–National Center for Atmospheric Research) reanalysis dataset
(Kalnay et al., 1996). AR events are identified using an IVT threshold ≥ 250 kg m⁻¹ s⁻¹, ensuring
a consistent framework for detecting moisture transport events. The catalog provides information
on AR frequency, duration, and landfall locations, with data covering the North American West
Coast (20°N to 60°N) from January 1948 to December 2024. Validation by Gershunov et al.
24

3. DATA

GAAT MIGRATION UP THE GAAT HÉENI

(2017) against Neiman et al. (2008) employed Special Sensor Microwave Imager (SSMI)
observations and demonstrated strong agreement between the datasets, reinforcing its reliability
for AR analysis.
For this study, I focus on 1991 to 2024, aligning with extracted ERA5-Land data (Section 3.1) to
facilitate a robust assessment of AR-driven hydroclimatic impacts. AR landfall locations were
extracted from 55°N to 60°N, aligning with the study region.
3.5 Escapement Data
Escapement data for Gaat are monitored at a weir located downstream of Gaat Áayi, managed by
TRTFN. TRTFN Fisheries (2024) collects and shares these data with the Department of Fisheries
and Oceans Canada. Escapement counts are obtained through a weir structure, with the sole
passage point continuously monitored by both a motion capture camera system and field
technicians. Data from 1980-1981 and 1992-2024 have been collected, with observations
occurring annually between July and September (Figure 1.2). For 2024, Fisheries and Oceans
Canada provided daily arrival counts of Gaat at Gaat Áayi.

25

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 4: METHODS
This section outlines the methodologies that will be used to investigate hydroclimatic variability,
reanalysis data validation, AR climatology, and passage dynamics. A meteorological station
(Section 4.1) has collected meteorological and hydrological measurements, including air
temperature, atmospheric pressure, wind speed, total precipitation, relative humidity, discharge,
river level, and river velocity. ERA5-Land reanalysis data will be extracted and validated against
in-situ observations (Section 4.2) using bias, root mean square error (RMSE), and the coefficient
of determination (R²). The climatology of the Gaat Héeni Watershed will be characterized
(Section 4.3) by analyzing long-term trends in air temperature, precipitation, and snowpack
conditions. AR climatology will be examined using the SIO-R1 AR Catalog (Section 4.4),
applying spatial and temporal classification criteria to identify and track AR landfall events. The
impact of ARs on watershed scale hydroclimatic conditions will be assessed (Section 4.5) by
quantifying AR contributions to total precipitation and evaluating seasonal and interannual
variability. Section 4.6 outlines the methods used to quantify the timing and sensitivity of daily
discharge responses to precipitation events in the Gaat Héeni Watershed during the 2024
monitoring period. The analysis applies event-based classification of AR and non-AR
precipitation, constructs dual-axis hydrographs, and calculates precipitation centroid lags to peak
discharge to characterize event-scale hydrological responses. Finally, video monitoring and a
logistic regression with a logit function will assess how discharge and water temperature
influence Gaat passage success through SR3-3 (Section 4.7). Section 4.8 outlines the methods
used to quantify the influence of daily jump success at SR3-3 on Gaat arrival anomalies at Gaat
Áayi using linear regression models at lag intervals of one, two, and three days.

26

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

4.1 Installation of Meteorological Station
An initial review of meteorological and hydrometric stations within and adjacent to the Gaat
Héeni Watershed revealed a lack of verifiable data. To address this gap, the first phase of this
research involved installing a hydrometric station on the Gaat Héeni (Figure 4.1). Located on the
west bank of the Gaat Héeni River at 533 m above sea level, this station records air temperature
(°C), atmospheric pressure (hPa), wind speed (m s⁻¹), total precipitation (mm), relative humidity
(%), discharge (m³ s⁻¹), river level (m), and river velocity (m s⁻¹). Instrumentation is anchored on
a 3 m stainless‐steel tripod, which supports the R.M. Young Wind Monitor 05103 (wind speed
and direction) and a Meter Group ATMOS-14 sensor (air temperature, relative humidity, and
atmospheric pressure). Data are recorded by an XLINK500-IR-1 datalogger with integrated
Iridium telemetry, powered by a 50 W solar panel and a 120 Ah AGM battery in a weatherproof
enclosure. A separate scaffold, erected 4 m from the tripod, supports the OTT Pluvio2 total
precipitation gauge to ensure unobstructed catchment and minimal flow disturbance. A noncontact radar flow-meter system (Sommer RQ-30a) is deployed on an adjacent structure to
measure river discharge continuously. All sensors log data at 15-minute intervals and transmit
hourly to the Hydromet Cloud network. Data collection began on 17 June 2024 and continues to
the present, with discharge measurements suspended for the winter starting on 15 October 2024.

27

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

Figure 4.1: This year-round meteorological station, photographed on 9 October 2024, monitors meteorological and hydrological
conditions in the Gaat Héeni Watershed. It records total precipitation (mm), wind speed (m s⁻¹), air temperature (°C), relative
humidity (%), and atmospheric pressure (hPa), with data collected by an OTT Pluvio², RM Young wind monitor, and ATMOS-14
sensor. A 50-watt solar panel supplies power, and an XLINK500 datalogger with Iridium telemetry transmits data remotely. A
nearby OTT RQ-30a radar sensor provides measurements of river discharge (m³ s⁻¹), level (m), and velocity (m s⁻¹).

4.2 Extraction and Validation of Reanalysis Data
Reanalysis data for the Gaat Héeni Watershed and adjacent regions (Figure 1.1) were extracted
from the ERA5‐Land reanalysis product (Muñoz-Sabater et al., 2021) for the period 1991-2023.
The extracted variables were air temperature at 2 m (K), snow depth (cm), snow depth water
equivalent (m of water equivalent), total precipitation (mm), geopotential (m2 s−2), and snowfall
(m of water equivalent). ERA5-Land data, available on a 0.1° (octahedral) reduced Gaussian grid
with hourly resolution (Copernicus Climate Change Service, 2019), were extracted for grid cells
encompassing in-situ validation sites and aggregated to a daily timescale. To ensure consistency,
in-situ observations were converted from Pacific Standard Time (PST) to Universal Coordinated
Time (UTC) prior to validation (Clelland et al., 2024).

28

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

Daily ERA5-Land values for air temperature, snow depth, and snow depth water equivalent were
calculated by averaging the extracted hourly values over each 24-hour period.
Air temperature at 2 m (in degrees Kelvin [K]) was converted to degrees Celsius (°C) by
applying the linear relation (Hall et al., 2024; Preston-Thomas, 1990) shown in Equation 1:
(1)

𝑇°C = 𝑇𝐾 − 273.15
Daily total precipitation and snowfall values were computed using Equation 2, where tp
represents total precipitation in millimeters and d is the day of interest (Copernicus Climate
Change Service, 2019). The value of tp at d+1 00 UTC accounted for the total precipitation
accumulated during the preceding 24-hour period.
𝑡𝑝 [𝑚𝑚] = 𝑡𝑝𝑑+1 00𝑈𝑇𝐶 [𝑚] × 1000

(2)

Snow depth water equivalent (SWE) in ERA5-Land is expressed in meters of water equivalent
and represents the depth of water that would result across the grid cell if the entire snowpack
melted (Kouki et al., 2023; Muñoz-Sabater et al., 2021). To convert SWE to from meters (m)
millimeters (mm), the SWE values are multiplied by 1000.
ERA5-Land performance was assessed at two meteorological stations: Atlin (ECCC Station
1200560) and Atlin Lake (Government of British Columbia Station 4E02B). A date-matched
dataset was created to ensure a robust comparison by aligning daily in-situ observations with
daily aggregated ERA5-Land outputs. Accuracy was quantified using bias (mean difference),
root mean square error (RMSE), coefficient of determination (R²), and the p-value to assess
statistical significance.

29

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

Surface elevation for each ERA5-Land grid cell was derived from the geopotential (m2 s−2),
which represents gravitational potential energy per unit mass (Gershunov et al., 2017).
Geopotential elevation (m) was obtained by dividing geopotential by standard gravitational
acceleration (g = 9.80665 m s⁻²; Copernicus Climate Change Service, 2019).
To account for elevation differences between ERA5-Land grid cells and in-situ meteorological
stations during air temperature validation, a standard lapse rate correction of −6.5 °C km⁻¹ was
applied (Kattel & Yao, 2018; Li et al., 2015). This correction adjusts ERA5-Land air temperature
outputs to better reflect in-situ elevations by compensating for the typical decrease in air
temperature with altitude.
The elevation difference, ΔElev (m), was calculated as the in-situ elevation minus the ERA5Land grid cell elevation. For example, at Atlin (ECCC station 1200560), the station elevation
was 697 m, while the ERA5-Land elevation was 673.6 m, resulting in a ΔElev of 23.4 m. The
unadjusted ERA5-Land air temperature is TERA5 (°C), and the resulting elevation-adjusted air
temperature is Tadj (°C). The lapse rate–adjusted air temperature was calculated as:
−6.5

𝑇𝑎𝑑𝑗 = 𝑇𝐸𝑅𝐴5 + (1000) × 𝛥𝐸𝑙𝑒𝑣

(3)

SWE and snow depth were validated at a daily temporal scale using in-situ data from Atlin Lake
(Government of British Columbia Station 4E02B) collected between 2004 and 2023. To improve
agreement between ERA5-Land outputs and observed values, bias correction was applied
following the approach outlined by Chen et al. (2013).
Bias was calculated as the difference between the ERA5-Land values and the corresponding
in-situ observations (Equation 4a). This metric quantifies the over- or underestimation in the
30

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

ERA5-Land dataset. The number of paired observations is denoted by N, ERA5Land refers to the
modelled values, and InSitu refers to the observed values. Bias was calculated as:
1

(4a)

𝐵𝑖𝑎𝑠 = 𝑁 ∑(𝐸𝑅𝐴5𝐿𝑎𝑛𝑑 − 𝐼𝑛𝑆𝑖𝑡𝑢)
The bias-corrected value, BiasCorr, was then derived by subtracting the Bias calculated in 4a
from each ERA5-Land value. The bias-corrected ERA5-Land value was calculated as:
𝐵𝑖𝑎𝑠𝐶𝑜𝑟𝑟 = 𝐸𝑅𝐴5𝐿𝑎𝑛𝑑 − 𝐵𝑖𝑎𝑠

(4b)

Time series plots were used to analyze seasonality and long-term variability in snow depth (cm)
and SWE (mm).
ERA5-Land precipitation and snowfall data were validated against in-situ measurements from
the Atlin (ECCC Station 1200560) meteorological station at daily and monthly timescales.
Scatter plots incorporating a 1:1 line were used to visualize the relationship between modelled
and observed values.
4.3 Climatology of the Gaat Héeni Watershed
The ERA5-Land data validated in Section 4.2 were used to characterize the climatology of the
Gaat Héeni Watershed. ERA5-Land reanalysis data, including daily temperature, snowfall, SWE,
total precipitation, snow depth, and geopotential, were extracted for 1991–2023. Data were
spatially averaged across all ERA5-Land grid cells whose centroids intersected the Gaat Héeni
Watershed boundary to produce a representative climatological overview. Spatially averaged
data were computed across grid cells at the native hourly resolution (UTC scale) before any
temporal aggregation. For air temperature, an elevation correction was applied (Equation 2) to

31

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

account for the difference between the mean elevation of the ERA5-Land extracted grid cells
(915 m) and the Gaat Héeni Watershed’s mean elevation (955 m).
Monthly air temperature data were visualized using a box and whisker plot, while SWE and
snow depth were visualized through time series plots to assess interannual variability. Rainfall
was derived by calculating the difference between total precipitation and snowfall (Espinoza et
al., 2018).
Total precipitation and snowfall extracted from ERA5-Land were visualized using a stacked bar
chart to illustrate the annual distribution of rainfall and snowfall. Rainfall was derived by
subtracting snowfall from total precipitation (Jeoung et al., 2022; Li et al., 2025). Monthly
distributions of total precipitation, snowfall, and rainfall were also represented using box and
whisker plots to highlight seasonal patterns, interquartile ranges (IQRs), and outliers.
Annual and monthly precipitation trends were analyzed to assess long-term variability in
moisture input and phase partitioning between rain and snow. Total precipitation and snowfall
were plotted as a stacked bar chart to illustrate the annual distribution of rainfall and snowfall.
Monthly distributions of total precipitation, snowfall, and rainfall were visualized using box and
whisker plots to illustrate seasonal patterns, IQRs, and outliers.
4.4 Atmospheric River Climatology
Section 4.4 investigates the spatial distribution, frequency, and intensity of ARs making landfall
between 55°N and 60°N from 1 January 1991 to 31 December 2023. This period corresponds to
meteorological and hydrometric data availability, as outlined in Sections 3.3 and 3.4. The
SIO-R1 AR Catalog (Gershunov et al., 2017) was adapted for daily resolution to identify and

32

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

classify AR landfall events. The methodological framework follows the approach outlined by
Gershunov et al. (2019), incorporating a set of specific criteria. Each daily time step was
analyzed for spatial patterns of contiguous grid cells with IVT ≥ 250 kg m⁻¹ s⁻¹ to identify ARs.
Only events extending at least 1500 km in length and intersecting the coastline were classified as
ARs. The coastal grid cell exhibiting the maximum IVT was designated as the landfalling
location for each event. To maintain temporal continuity, the landfalling location of a single AR
event could not shift more than 1000 km (~8.9° latitude) between consecutive days.
Additionally, two ARs making landfall in the same general coastal region were classified as
distinct events if separated by at least 24 hours. The SIO-R1 AR catalog was used to analyze
seasonal variability, interannual trends, and changes in AR intensity across the study period.
To assess the seasonal distribution of ARs, I calculated the monthly frequency of AR landfall
events and the associated mean net IVT for each month from 1991 to 2023. The net IVT for each
month was determined by summing the IVT values of all AR landfall events occurring within
that month across the 33 years, providing a cumulative measure of moisture transport associated
with ARs. Monthly statistical distributions, including mean, median, IQR, and total range, were
computed, with outliers defined as values exceeding 1.5 times the IQR (Gautam & Singh, 2022).
To further explore potential decadal-scale shifts in AR activity, the study period was divided into
three sub-periods (1991–2001, 2002–2012, and 2013–2023), with each sub-period analyzed
separately using the same statistical metrics applied to the full dataset.
4.5 Impact of ARs on the Study Region
Building on the methodology outlined in Sections 4.3 and 4.4, Section 4.5 quantifies AR
contributions to seasonal and annual precipitation in the Gaat Héeni Watershed. The analysis

33

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

covers January 1991 to December 2023, aligning with the SIO-R1 AR Catalog and the ERA5Land reanalysis dataset extracted in Section 4.2 (Gershunov et al., 2017; Sharma & Déry, 2020).
Following the methodology outlined in Section 4.4, ARs are defined as contiguous grid cells
with an IVT ≥ 250 kg m⁻¹ s⁻¹, extending at least 1500 km and intersecting the coastline between
55°N and 60°N (Gershunov et al., 2017). To quantify the portion of precipitation attributed to
ARs, I identify the initial AR landfalling day and the following day (Gershunov et al., 2017).
This accounts for the lag time as ARs move inland from the coast (Sobral & Déry, 2023) and
mitigates some uncertainty related to UTC-to-local time differences in precipitation datasets
(Sharma & Déry, 2020).
I integrate AR landfalling events from the SIO-R1 AR Catalog with ERA5-Land precipitation,
snowfall data, and the derived rainfall calculated in Section 4.3. To ensure spatiotemporal
consistency, ERA5-Land data are averaged across the Gaat Héeni Watershed (Sharma and Déry,
2020) with precipitation classified as AR-driven when the IVT criterion is met (Equation 5). The
fraction of precipitation attributed to ARs (FAR) is then calculated as the ratio of total
precipitation during AR landfall days (PAR, mm) to the total seasonal precipitation (PTotal, mm).

FAR =

𝑃𝐴𝑅
𝑃𝑇𝑜𝑡𝑎𝑙

(5)

The seasonal variability of AR-driven precipitation is calculated by computing the fraction of
total precipitation attributed to ARs for each season, revealing seasonal dependencies in AR
contributions. Seasons are delineated as spring (March–May), summer (June–August), fall
(September–November), and winter (December–February) (Sobral & Déry, 2023). To evaluate
long-term variability, the study period is stratified into three sub-periods (1991–2001, 2002–

34

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

2012, and 2013–2023), ensuring consistency with the AR climatology trend analysis outlined in
Section 4.2.
I constructed a stacked bar chart spanning the 33-year study period (1991–2023) to visualize
interannual variability in AR-driven and non-AR-driven precipitation. This plot quantifies annual
precipitation totals attributable to AR events (blue bars) and non-AR precipitation (grey bars).
I also calculated decadal means for 1991–2001, 2002–2012, 2013–2023, and the whole 1991–
2023 period to contextualize longer-term patterns.
Furthermore, a Mann-Kendall (MK) test (Kendall, 1948; Mann, 1945) is employed at both
annual and seasonal scales to detect monotonic trends in AR-driven precipitation, with statistical
significance assessed at the 95% confidence level. A significance threshold of 0.05 indicates that
results with p-values of 0.05 or below are deemed statistically significant (Ravi & Odorico,
2005). This approach enables the identification of potential shifts in AR frequency and their
contribution to total precipitation over time.
4.6 Discharge Sensitivity to Daily Precipitation and AR Events
Section 4.6 describes the methodology used to examine how daily discharge and precipitation in
the Gaat Héeni Watershed are influenced by AR events, with a specific focus on event-scale
hydrological responses. The objective is to determine the timing and sensitivity of discharge to
AR-driven precipitation during the 2024 monitoring season.
Identification of AR events followed the IVT threshold approach outlined in Section 4.4, using
landfall data from the updated SIO-R1 AR Catalog (Gershunov et al., 2017). Events were
classified as ARs when IVT ≥ 250 kg m⁻¹ s⁻¹ along a contiguous corridor of at least 1500 km in

35

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

length. The AR also had to land on the North American west coast between 55°N and 60°N to be
included in the analysis. Following the approach in Section 4.5, precipitation was attributed to
AR activity if it occurred on the AR landfall day or the following day.
Daily precipitation and discharge data were aggregated from hourly data collected at the Gaat
Héeni meteorological and hydrometric stations (Sections 3.2 and 3.3). Daily data were used to
align with the daily resolution of the SIO-R1 AR catalog. Each day in the dataset from 17 June to
10 October 2024 was classified as either AR-driven or non-AR-driven, enabling the partitioning
of precipitation by source. To summarize precipitation distribution, grouped bar charts were used
to compare cumulative AR and non-AR contributions, consistent with the seasonal AR
precipitation quantification methods described in Section 4.5. These classifications were used to
construct a dual-axis hydrograph (Mathai & Mujumdar, 2022), in which daily discharge is
overlaid with AR-driven and non-AR-driven precipitation.
Defining the hydrological response in the Gaat Héeni Watershed involved quantifying the lag
between the precipitation centroid and peak discharge based on daily precipitation and discharge
records. Precipitation events comprised one or more consecutive days with ≥1 mm of
precipitation, separated by at least one precipitation-free day (Li et al., 2023). For each identified
event, the precipitation centroid time (tWC) was computed to represent the effective timing of
water input into the system. The precipitation centroid time (tWC) is the weighted mean timing of
precipitation during an event, representing when the majority of the event’s precipitation
occurred. Precipitation centroid time was calculated using Equation 6, where Pi is the
precipitation (mm) recorded during the ith hour, ti is the timestamp (in days), and n is the total
number of daily observations within the precipitation event (Dingman, 2015).

36

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

𝑡𝑤𝑐 =

∑𝑛
𝑖=1 𝑃𝑖 ×𝑡𝑖
∑𝑛
𝑖=1 𝑃𝑖

(6)

To evaluate the hydrologic response of the Gaat Héeni Watershed to each precipitation event, the
time of peak discharge, tₚₖ, was identified as the timing of the maximum observed streamflow
following the precipitation event. The timing of the precipitation input was characterized using
the precipitation centroid time (tWC). The lag between the precipitation centroid and the resulting
discharge peak, referred to as the centroid lag to peak discharge (tLPS), measures the watershed’s
response time. This lag, expressed in days, is expressed in Equation 7:

tLPS=tPK−tWC

37

(7)

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

4.7 Quantifying Optimal Hydrological Conditions for Successful Gaat Barrier Passage
Section 4.7 discusses how water temperature (T, °C) and river discharge (Q, m³ s⁻¹) influence
Gaat jump success at obstruction SR3-3. The methodological workflow is summarized in Figure
4.2.
Video monitoring: 5-30 July 2024
12:00-18:00, 15-min intervals

Extract river discharge (Q) and water
temperature (T) aligned to video monitoring
blocks

Calculate proportion of successful jumps per
monitoring block

𝑝

Logistic Regression with logit function: log 1−𝑝𝑖 = 𝛽0 + 𝛽1 𝑇𝑖 + 𝛽2 𝑇𝑖 2 + 𝛽3 𝑄𝑖 + 𝛽4 𝑄𝑖 2
𝑖

Generate heatmap across T and Q ranges

Perform Wald Z-Tests
Figure 4.2: A flowchart outlining methods shown to determine optimal hydrologic conditions for the successful jump of Gaat at
SR3-3.

Previous assessments by the Taku River Tlingit Fisheries Department (2017) had identified
SR3-3 as a key migratory barrier (Section 1.4; Figure 4.3). Video monitoring of SR3-3 was
conducted from 5 to 30 July 2024, with observations carried out from 1200-1800 local time
(UTC−7) on each monitoring day. During the first 15 minutes of every hour, a Nikon Zf fullframe camera fitted with a Tamron 150–500 mm f/5-6.7 zoom lens captured 4 K UHD
38

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

(3840 × 2160 px) footage at 60 fps from a fixed tripod. Focal length was adjusted to 200 mm to
frame the wetted jump face. Although daily filming was planned, field constraints limited data
collection to 110 15-minute observation blocks distributed over 19 monitoring days (six blocks
on each of 17 days and two blocks on 23 July). During this period, a total of 92,687 jump
attempts was recorded. No footage was obtained on 15 July, 18 July, or 24-27 July owing to crew
fatigue and hazardous river conditions. Each jump attempt recorded on video was scored as a
binary outcome: 1 for a successful jump or 0 for a failure. A successful jump was defined as
Gaat landing with forward momentum (Lauritzen et al., 2005) in a laminar flow region (Mueller
et al., 2008).

July 5 - 14.93 m³ s⁻¹

July 17 – 10.69 m³ s⁻¹

Figure 4.3: Photographs on 5 and 17 July 2024 comparing discharge conditions at SR3-3, a canyon constriction that is a
migration barrier for Gaat (Sockeye Salmon; Oncorhynchus nerka). Discharge at 1200 hours on 5 July 2024 was 14.93 m³ s⁻¹,
while discharge decreased to 10.69 m³ s⁻¹ by 1500 hours on 17 July 2024. Red polygons in each image delineate areas of laminar
surface flow.

To analyze water temperature and discharge trends, I extracted averaged hourly discharge
(Q, m³ s⁻¹) data from the Gaat Héeni hydrometric station and utilized hourly water temperature
(T, °C) data from the lower Gaat Héeni (C. Sergeant, personal communication, 2024). Data were
extracted between 5 and 30 July 2024 to align with the video monitoring periods. A composite
figure was produced in which the daily jump‐attempt count is shown as a vertical bar chart, and

39

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

the daily mean river discharge (m³ s⁻¹) and daily mean water temperature (°C) are each overlaid
as a point joined by a continuous line. Hourly discharge and water temperature measurements
recorded during each day’s observation window were first averaged to yield a single mean value
per day.
To assess how hydrologic conditions affected Gaat jump success, a logistic-regression model
(Dockery et al., 2017) with a logit link was fitted to the jump data. The logit link function
transforms the probability of a binary outcome into log-odds, allowing for a linear relationship
between predictors and the transformed response, while constraining predicted probabilities to
values between 0 and 1 (Hosmer et al., 2013; Peng et al., 2002). In this model, p represents the
probability of a successful jump (Hosmer et al., 2013). The probability of jump success was
modelled as a function of water temperature (T, °C) and river discharge (Q, m³ s⁻¹), including
both linear and quadratic terms for each predictor (Maynard et al., 2017). Quadratic terms were
retained because passage success typically peaks at intermediate flows and water temperatures
and declines at extremes (Cruz-Jesus et al., 2017; Haro & Castro-Santos, 2012). To mitigate
observer bias and pseudo-replication arising from repeated jump attempts within 15-minute
monitoring blocks (Bolker et al., 2009; Harrison et al., 2018), all jump attempts recorded in each
15-minute monitoring interval were aggregated into a single binomial observation (Lazic et al.,
2020). Each aggregated data point thus represents the proportion of successful jumps (Si) relative
to total jump attempts (Ni) observed in that interval, explicitly defining the probability (pi) used
in the logistic regression model (Equation 8);
𝑆

𝑝𝑖 = 𝑁𝑖
𝑖

Equation 9 is a logistic regression with a logit function and incorporates four hydrologic
40

(8)

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

predictors: water temperature (T, °C), river discharge (Q, m³ s⁻¹), and their respective quadratic
terms (T2 and Q2). In this formula, the logit of the jump-success probability is expressed as a
linear combination of five regression coefficients: the intercept β0, which provides the baseline
log-odds of success when all predictors equal zero, the linear terms β1 (water temperature) and
β3 (discharge), which quantify first-order effects, and the quadratic terms β2 (water temperature
curvature) and β4 (discharge curvature).
𝑝

log 1−𝑝𝑖 = 𝛽0 + 𝛽1 (𝑇𝑖 ) + 𝛽2 (𝑇𝑖 2 ) + 𝛽3 (𝑄𝑖 ) + 𝛽4 (𝑄𝑖 2 )
𝑖

(9)

To visualize the model’s behaviour over the range of observed temperature (T) and discharge (Q)
conditions, I generated a two-dimensional prediction grid using the ggplot2 package in R
(R Core Team, 2024; Wickham, 2016). Water temperature (T, °C) and river discharge (Q, m³ s⁻¹)
were separated into 250 equally spaced values spanning their empirical minima and maxima. For
each node on this grid, the linear (T, Q) and quadratic (T², Q²) predictors described in Equation 8
were computed. These values were combined with the predicted probabilities (pᵢ) to predict jump
success. The resulting heatmap illustrates the relationship between environmental variables and
jump success probability and utilises a uniform and colourblind friendly palette.
In R (R Core Team, 2024), the maximum and minimum predicted probabilities on the prediction
grid were identified using which.max() and which.min(), extracting the corresponding water
temperature and discharge values (Bolker et al., 2009; Harrison et al., 2018). The 110 fitted
block‐level probabilities were then summarized with the summary() function to obtain the
minimum, first quartile, median, mean, third quartile, and maximum.
Wald z-tests were used to evaluate the statistical significance of each coefficient βⱼ in the model

41

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

(McArdle, 1988). The Wald z-test is defined in equation 10 where βⱼ is the maximum likelihood
estimate of the coefficient and SE(βⱼ) is its standard error (Hosmer et al., 2013). Predictors with
p-values > 0.05 are considered nonsignificant within this analysis.
𝛽ⱼ

𝑧ⱼ = 𝑆𝐸(𝛽ⱼ)

(10)

4.8 Influence of Jump Success on Gaat Arrival at Gaat Áayi
This section aims to determine how the daily jump success at SR3-3 influences the arrival of
Gaat at Gaat Áayi. Daily jump success (pd) was calculated by aggregating total successful jumps
and jump attempts from all 15-minute observation intervals recorded each monitoring day
(Equation 11). Sdi represents successful jumps in block i on day d, Ndi indicates total jump
attempts in block i on day d.
𝑝𝑑 = ∑(𝑆𝑑𝑖 )/ ∑(𝑁𝑑𝑖 )

(11)

To isolate daily variations in the arrival of Gaat at Gaat Áayi, daily arrival anomalies (ΔA) at the
Gaat Áayi weir were computed as daily arrival counts (Ad) minus the mean daily arrival count
(A) recorded over the observation period (Equation 12):
𝛥𝐴 = 𝐴d − 𝐴

(12)

A linear regression model was fitted, which tested relationships between observed daily jump
success (pd) and arrival anomalies 𝛥𝐴 at lag intervals (n) of one, two, and three days (Equation
13). Equation 14 was derived by algebraically rearranging the original linear regression
(Equation 13) to isolate the regression coefficient (β). Due to the absence of empirical migration
time data, it was assumed that Gaat would take one, two, or three days (n) to swim 22.5 km from
SR3-3 to Gaat Áayi. This assumption aligns with previously discussed Gaat migration speeds
(Hinch & Bratty, 2000; Quinn, 1988), but introduces inherent uncertainty, as actual swim speeds
42

4. METHODS

GAAT MIGRATION UP THE GAAT HÉENI

may vary considerably due to hydrological conditions or individual fish physiology. The error
term (ε) accounts for residual variation that the model does not explain.
𝛥𝐴𝑛 = 𝛽 × 𝑝𝑑 + 𝜀
𝛥𝐴

𝜀

𝑑

𝑑

𝛽 = 𝑝n+𝑝

(13)
(14)

R² and statistical significance (p-values) of the regression coefficient β were examined at lag
intervals of one, two, and three days to determine the lag at which jump success most strongly
and significantly predicted upstream arrival anomalies.

43

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 5: RESULTS
The results presented in this chapter synthesize the meteorological, hydrological, and climatic
factors shaping the Gaat Héeni Watershed from 1991 to 2023. Particular emphasis is placed on
precipitation, air temperature, river discharge, and ARs. Section 5.1 initiates the analysis by
presenting data collected from the meteorological station located in the upper Gaat Héeni
Watershed during the period of 17 June to 10 October 2024. Section 5.2 validates ERA5-Land
reanalysis data through comparison with in-situ measurements, providing a foundation for
subsequent analyses. Building upon this, Section 5.3 explores the climatological characteristics
of the region, elucidating long-term trends and seasonal variations. Section 5.4 delves into the
patterns and dynamics of ARs, highlighting their frequency, intensity, and seasonal distribution
in the Gaat Héeni Watershed. In Section 5.5, the influence of ARs on key meteorological
variables is quantified, revealing their significant role in modulating regional precipitation.
Section 5.6 analyzes daily precipitation and discharge during the 2024 monitoring period to
quantify the short-term hydrological response of the Gaat Héeni Watershed to AR and non-AR
events. Section 5.7 applies the aforementioned findings to assess the optimal hydrometric
conditions for Gaat, identifying key thresholds and environmental factors influencing migration
success. Finally, Section 5.8 quantifies the relationship between daily jump success at SR3-3 and
daily arrival anomalies at Gaat Áayi, determining the lag intervals at which jump success most
significantly predicts arrival variability.
5.1 Meteorological Data
To illustrate the short-term dynamics of meteorological and hydrological conditions in the Gaat
Héeni Watershed, Figure 5.1 presents hourly observations recorded at the Gaat Héeni station
from 18 June to 10 October 2024. It shows temporal variations across meteorological and
44

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

hydrological parameters including air temperature (°C), atmospheric pressure (hPa), wind speed
(m s⁻¹), total precipitation (mm), relative humidity (%), discharge (m³ s⁻¹), river level (m) and
river velocity (m s-1). Recorded air temperatures ranged from –5.6 °C to 27.4 °C, with
atmospheric pressure values between 912.4 hPa and 945.3 hPa.
Hydrological variability was evident in both precipitation and discharge data. A pronounced
precipitation event occurred between 22 and 24 July 2024, during which 47 mm of rainfall was
recorded. This precipitation event resulted in a rapid hydrological response, with river discharge
increasing from 8.7 m3 s-1 to 30.0 m3 s-1 over the same period and river level and velocity
showing similar increases.

45

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Air Temperature (°C)

Atmospheric Pressure (hPa)

Wind Speed (m s⁻¹)

Total Precipitation (mm)

Relative Humidity (%)

Discharge (m³ s⁻¹)

River Level (m)

River Velocity (m³ s⁻¹)

Date

Figure 5.1: Hourly meteorological and hydrometric data from the Gaat Héeni Station, collected between 18 June and 10 October
2024, provide a detailed record of summer hydroclimatic conditions. The figure presents air temperature (°C), atmospheric
pressure (hPa), wind speed (m s⁻¹), total precipitation (mm), relative humidity (%), discharge (m³ s⁻¹), river level (m) and river
velocity (m s-1).

46

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

5.2 Extraction and Validation of Reanalysis Data
The validation of ERA5-Land reanalysis data for air temperature, SWE, and snow depth was
conducted using in-situ data from Atlin (1991-2021; ECCC Station 1200560) and Atlin Lake
(2004-2023; Government of British Columbia Station 4E02B).
At Atlin (1991-2021; ECCC Station 1200560), daily ERA5-Land air temperature exhibits a bias
of -2.1 °C relative to observed values, with a root mean square error (RMSE) of 3.5 °C, an R² of
0.9, and a p-value of 0.039 (Figure 5.2). After adjusting for elevation at the same daily scale, the
bias is reduced to -1.9 °C, RMSE improves to 3.4 °C, and the R2 remains at 0.9, with a p-value
of 0.035.

a)

b)

Figure 5.2: Hourly ERA5-Land reanalysis air temperatures are compared with air temperatures from the Atlin station (ECCC
Station 1200560) from 1991–2001. Panel a) shows unadjusted ERA5-Land values, while panel b) displays air temperatures
corrected for elevation using a standard lapse rate of −6.5 °C km⁻¹. Elevation correction reduced the mean bias from –2.1 °C to
−1.9 °C and decreased the RMSE from 3.5 °C to 3.4 °C. In both cases, the strength of the linear relationship remained high (R² =
0.9) with a p-value of 0.035.

47

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

The validation of SWE extracted from ERA5-Land highlights underestimation in the ERA5Land dataset (Figure 5.3). At Atlin Lake (2004-2023; Government of British Columbia Station
4E02B), ERA5-Land has a bias of 283.1 mm, RMSE of 290.9 mm, and an R² of 0.33 with a
p-value of 0.52. After bias correction (Equations 3a and 3b), the bias diminishes to 0.0 mm, and
RMSE improves to 67.0 mm. R² remains at 0.331 with a p-value of 0.52.

Figure 5.3: Comparison of the observed SWE measurements from the Atlin Lake Snow Survey Station (Government of British
Columbia Station 4E02B) with ERA5-Land reanalysis data for the exact dates. Three datasets are shown: observed SWE
(orange), original ERA5-Land SWE (green), and bias-corrected ERA5-Land SWE (blue). The original ERA5-Land product
substantially overestimated SWE, with a mean bias of 283.1 mm and a root mean square error (RMSE) of 290.9 mm. Following
bias correction, the mean bias was eliminated and the RMSE was reduced to 67.0 mm, while the coefficient of determination (R²
= 0.33) remained unchanged.

The snow depth data from ERA5-Land demonstrate a consistent overestimation at Atlin Lake
(2004-2023; Government of British Columbia Station 4E02B), as illustrated in Figure 5.4.
Initially, the ERA5-Land dataset shows a bias of 82.5 cm, an RMSE of 83.6 cm, and an R² of
0.70 with a p-value of 0.34. After applying bias correction (Equations 3a and 3b), the bias is
reduced to 0.0 cm, and the RMSE improves to 13.5 cm. The R² remains unchanged at 0.70, with
48

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

a p-value of 0.34.

Figure 5.4: Comparison of ERA5-Land reanalysis snow depth with manually collected in-situ observations from the Atlin Lake
Snow Survey Station (Government of British Columbia Station 4E02B) from 2004–2023. Before bias correction, ERA5-Land
exhibits a mean bias of 82.5 cm and a root mean square error (RMSE) of 83.6 cm. The coefficient of determination (R²) was 0.70,
with a p-value of 0.34, reflecting moderate agreement. After applying bias correction, the mean bias became 0 and RMSE
improved significantly to 13.5 cm, while R² remained unchanged.

Figure 5.5 compares daily precipitation from Atlin (1991-2021; ECCC Station 1200560) and
ERA5-Land. The analysis shows a poor correlation: R2 of 0.002, p-value of 0.92, and an
RMSE of 4.58 mm. This relationship suggests that ERA5-Land poorly captures daily
precipitation at Atlin, consistent with previous work (Goswami et al., 2024; Wu et al., 2023).

49

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Figure 5.5: Total daily precipitation from ERA5-Land reanalysis and in-situ measurements at the Atlin meteorological station
(ECCC Station 1200560) from 1991–2020. ERA5-Land data were extracted for the grid cell overlapping the station and
temporally aligned with the in-situ observations. ERA5-Land demonstrates limited skill in capturing daily precipitation
variability, with a weak correlation (R² = 0.002, p = 0.92), a root mean square error (RMSE) of 4.58 mm, and a mean bias of 1.14
mm.

Despite the poor agreement at a daily scale, at a monthly scale, Atlin (1991-2021; ECCC Station
1200560) and ERA5-Land at a monthly scale (Figure 5.6) show a moderate correlation with an
R2 of 0.42 and a p-value of 0.04. This suggests that while ERA5-Land does not capture daily
precipitation, it represents long-term variability when daily errors are aggregated over time
(Wu et al., 2023).

50

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Figure 5.6: Comparison of total monthly precipitation from ERA5-Land reanalysis with monthly aggregated in-situ observations
at the Atlin meteorological station (ECCC Station 1200560) for 1991–2020. ERA5-Land data were spatially and temporally
aligned with station records. Monthly aggregation improves agreement compared to the daily scale, with a stronger correlation
(R² = 0.42, p < 0.05), a reduced RMSE of 22.3 mm, and a mean bias of 9.7 mm. These results indicate that ERA5-Land is more
reliable for capturing precipitation variability at coarser temporal resolutions.

5.3 Climatology of the Study Region
Section 5.3 analyzes the climatological conditions in the Gaat Héeni Watershed from 1991 to
2023. During this period, the mean annual air temperature averaged –2.3 °C, with the coldest
year (1993) reaching –3.5 °C and the mildest year (2022) reaching –1.2 °C. Total mean annual
precipitation was 481.9 mm, varying from a minimum of 335.9 mm in 1995 to a maximum of
606.1 mm in 2004. The mean annual rainfall was 165.3 mm, with yearly totals ranging from
93.2 mm in 1992 to 235.6 mm in 2017.
For air temperature (°C), an elevation correction was applied using Equation 2 to account for the
51

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

40 m elevation difference between the extracted ERA5-Land grid cells (915 m) and the Gaat
Héeni Watershed mean elevation (955 m). This resulted in a 0.26 °C decrease in the extracted
ERA5-Land air temperature. The temperature cycle in the Gaat Héeni Watershed follows a
seasonal pattern (Figure 5.7) with cold winters, rapid spring warming, warm summers, and a
cooling autumn transition. Looking at monthly air temperature distributions, winter (December–
February) remains the coldest period, with January recording the lowest mean temperature
(-14.8 °C), reaching a minimum of -26.8 °C. February remains cold (-12.1 °C avg.), with lows of
-21.7 °C, while March (-8.5 °C avg.) signals the beginning of seasonal warming. Spring (April–
May) sees a sharp increase in temperatures, with April averaging -1.9 °C and May transitioning
to positive temperatures (3.3 °C avg.), reaching highs of 7.4 °C. Summer (June–August) reaches
peak warmth, with July recording the highest mean temperature (10.9 °C) and maximum highs
of 14.3 °C.

Figure 5.7: Distribution of monthly air temperatures extracted from ERA5-Land in the Gaat Héeni Watershed (1991 to 2023).
Box and whisker plots capture each month's interquartile range, minimum/maximum values, median, and outliers, highlighting
the seasonal variability and spread of daily air temperatures.

The seasonal SWE (mm) and snow depth (cm) extracted from ERA5-Land (Figures 5.8 and 5.9)
is characterized by a consistent cycle of accumulation and ablation. Peak SWE occurs in
February to March, with an annual mean of 416.0 mm and observed annual peaks ranging from
52

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

300.7 mm to 606.1 mm. Similarly, the mean annual snow depth is 175.3 cm with a maximum
snow depth of 260.2 cm in 2005 and a minimum snow depth of 100.1 cm in 1995. Complete
ablation typically occurs by May– June.

Figure 5.8: Monthly Mean Snow Water Equivalent (SWE) in the Gaat Héeni Watershed (1991–2023) extracted from ERA5-Land.
SWE accumulation begins in October and steadily increases through winter, peaking in March at a mean value of 416.0 mm. The
maximum SWE recorded was 606.1 mm in 2006, while the minimum was 300.7 mm in 1997.

Figure 5.9: Monthly Mean Snow Depth across the Gaat Héeni Watershed (1991–2023). Extracted from ERA5-Land reanalysis
data, the mean annual snow depth is 175.3 cm, with a maximum of 260.2 cm in 2005 and a minimum of 100.1 cm in 1995.

The seasonal precipitation distribution follows a well-defined pattern. Snowfall dominates winter
months, and rainfall becomes the primary precipitation phase during the warmer seasons (Figure
5.10 and Figure 5.11). Winter (November–March) is characterized by high snowfall

53

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

accumulation, with December (52.7 mm), January (50.1 mm), and February (35.2 mm) recording
the highest monthly mean snowfall totals. Spring (April–May) marks a transition from snowfall
to rainfall, with April reporting 25.1 mm of mean monthly precipitation, predominantly as snow
(0.6 mm rainfall), while May sees an increased rainfall fraction (13.8 mm) out of a mean total
precipitation of 26.4 mm. Summer (June–August) is entirely rainfall-driven, with mean monthly
precipitation increasing progressively from June (34.4 mm) to August (39.4 mm), reflecting
enhanced convective activity and storm-driven precipitation variability. Autumn (September–
October) represents a secondary wet season, with September (58.3 mm) and October (58.6 mm)
recording the highest mean monthly precipitation.

Figure 5.10: Total annual precipitation is partitioned into rainfall and snowfall within the Gaat Héeni Watershed. Data are
extracted from ERA5-Land reanalysis data (1991-2023) and are spatially averaged across grid cells. Precipitation shows
interannual variability, with snowfall consistently contributing the majority of total precipitation.

54

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

a)

b)

c)

Figure 5.11: Monthly distribution of total precipitation (top), snowfall (middle), and rainfall (bottom) in the Gaat Héeni
Watershed from 1991 to 2023. Total precipitation and snowfall are extracted from ERA5-Land, while rainfall is the difference
between total precipitation and snowfall. Total precipitation shows peaks during autumn and early winter (October–January) and
a secondary rise in late summer (September–October). Snowfall dominates from October to April, peaking in January, and is
nearly absent by summer. In contrast, rainfall is minimal in winter but increases sharply from May to September, peaking in late
summer.

55

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

5.4 Atmospheric River Climatology
ARs exhibit a distinct seasonal cycle across the study region with the highest frequency of ARs
occurring in summer and fall, specifically August, September, and October (Figure 5.12).
September recorded the highest frequency, averaging approximately four ARs per month,
followed closely by August. In contrast, the lowest AR activity was observed in March and
April, where the average number of AR landfalls remained below one event per month.

Figure 5.12: Average number of atmospheric river (AR) landfalls per month between 55°N and 60°N from 1991 to 2023, based
on the SIO-R1 AR Catalog. Bar height represents the monthly AR frequency, while shading reflects the total monthly net IVT
(kg m⁻¹ s⁻¹), with darker bars indicating higher cumulative vapour transport. AR activity peaks in late summer and early fall, with
September and August exhibiting the highest frequency of ARs.

IVT followed a similar seasonal pattern, with the highest values occurring in late summer and
early autumn. The net IVT peaked in September, exceeding 10,000 kg m⁻¹ s⁻¹, with August and
October also exhibiting elevated values (Figure 5.13).

56

GAAT MIGRATION UP THE GAAT HÉENI

Net IVT kg m⁻¹ s⁻¹

5. RESULTS

Number of Atmospheric Rivers per Month

Figure 5.13: Climatological relationship between the mean monthly frequency of atmospheric river (AR) landfalls and the
corresponding net integrated water vapour transport (IVT, kg m⁻¹ s⁻¹) for the 55°N–60°N coastal sector from 1991 to 2023, using
data from the SIO-R1 AR Catalog. Each point represents a month, with its position indicating the mean number of AR events and
cumulative IVT, while colour denotes IVT magnitude.

The distribution of AR landfalls exhibited distinct seasonal and interannual variability, with
notable increases in frequency and dispersion occurring in late summer and early fall (August,
September, and October; Figure 5.14). The four-panel box and whisker plot illustrates the
interquartile range (IQR), median, and occurrence of outliers for ARs across four distinct
periods: 1991–2001, 2002–2012, 2013–2023, and 1991–2023.
Across all periods, the median number of ARs remained low from January through June before
increasing sharply in July and peaking between August and October. The highest median values
were recorded in September and October, consistent with climatological expectations of
heightened AR activity during late summer and early fall (Radić et al., 2015). Outliers during
these months highlight the episodic nature of high-activity years, contributing to overall
interannual variability.

57

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

a)

b)

A comparative
analysis of the three subperiods indicates a
progressive increase in
LAR frequency over time, particularly during peak months. During 1991–
2001 from August, to October, the LAR distribution exhibited relatively
low variability, with no extreme outliers, and a median three LARs in
September and October. In contrast, the 2002–2012 period displayed an
expanded distribution, with a median of three LARs in August and
October and four in
September. From 2002-2012 there was also
c)
a notable increase
in outliers, indicating heightened year-to- d)
year variability.
The most recent period (2013–2023)
reflects a modest rise in LAR frequency, with median values of four over
the fall, representing a slight increase compared to earlier decades.

Figure 5.14: Monthly distribution of atmospheric river (AR) landfalls for (a) 1991–2001, (b) 2002–2012, (c) 2013–2023, and (d) 1991–
2023. Box and whisker plots show AR counts with medians (black lines), interquartile range (grey boxes), and outliers (>1.5×IQR, black
dots). A consistent late-summer to early-autumn (August–October) peak is evident, with higher medians and greater variability in recent
periods (2002–2023) compared to 1991–2001.

5.5 Impact of ARs on the Study Region
The contribution of ARs to total precipitation within the study region from 1991 to 2023 exhibits
seasonal and interannual variability (Figure 5.15). On average, ARs account for 15.9% to 39.1%
of seasonal precipitation, with the most significant contributions occurring in fall and winter. Fall
experiences the highest AR-driven precipitation fraction, with 37.9% of total seasonal
precipitation attributed to ARs from 1991 to 2023. This proportion has varied over the past three
decades, with the most recent period (2013–2023) showing an AR contribution of 39.1% in fall,
compared to 36.3% (2002–2012) and 38.3% (1991–2001). Seasonality is further evident by ARs
delivering 24.1 % of winter precipitation (1991-2023), 15.9 % in spring (1991–2023) and 22.9 %
in summer (1991–2012). These seasonal contrasts reflect the synoptic conditions governing AR

58

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

occurrence and landfall intensity.

Figure 5.15: Seasonal contribution of atmospheric rivers (ARs) to total precipitation across the Gaat Héeni Watershed, aggregated
by decade and over the whole study period using bias-corrected ERA5-Land precipitation and the SIO-R1 AR Catalog. Panel a)
(top left) displays data for 1991–2001, with ARs contributing 20.5% of winter, 22.3% of spring, 24.1% of summer, and 38.3% of
fall precipitation. Panel b) (top right) represents 2002–2012, showing increased winter AR influence (25.0%) but reduced spring
and summer contributions (12.2% and 18.7%, respectively). Panel c) (bottom left) illustrates the most recent period, 2013–2023,
where ARs account for 26.7% of winter, 13.5% of spring, 26.0% of summer, and 39.1% of fall precipitation. Panel d) (bottom
right) shows the overall climatology for 1991–2023, highlighting that ARs contribute the largest share in fall (37.9%), followed
by winter (24.1%), summer (22.9%), and spring (15.9%).

On an annual scale, AR events contributed between 16.4% and 37.8% of total annual
precipitation in the Gaat Héeni Watershed from 1991 to 2023 (Figure 5.16). Annual AR-driven
precipitation ranged from a minimum of 55.0 mm in 1995 to a maximum of 230.0 mm in 2004,
while total annual precipitation varied from 330.0 mm (1996) to 610.0 mm (2004). When
considering decadal means, AR-driven precipitation remains stable: 132 mm yr⁻¹ (27 %) in 1991–
2001, 129 mm yr⁻¹ (25 %) in 2002–2012, and 137 mm yr⁻¹ (27 %) in 2013–2023. Combined,

59

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

these yield a mean of 133 mm yr⁻¹, representing 26.4 % of total precipitation for 1991–2023.

Figure 5.16: Annual total precipitation partitioned into AR-driven and non-AR-driven components for the Gaat Héeni Watershed
from 1991 to 2023. Blue bars represent precipitation totals (mm) attributed to atmospheric river (AR) landfall events.

Heatmaps of monthly AR contributions over the 1991–2023 period illustrate the variability in
AR-driven precipitation (Figure 5.17). Consistent with seasonal trends, AR activity is
concentrated in the late summer to fall (August–November).

60

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

a)

b)

c)

Figure 5.17: Monthly percentage contribution of atmospheric rivers (ARs) to total precipitation across three latitudinal bands
within the study region, from 1991 to 2023. Panel a) (left) corresponds to the 55°–57.5°N band, panel b) (centre) to 55.0°N, and
panel c) (right) to 57.5°N. Each heatmap cell represents a month's AR contribution as a percentage of total monthly precipitation,
with colour intensity ranging from 0% (purple) to 50% (yellow). Clear seasonal patterns emerge in all panels, with the highest
AR contributions typically occurring in fall and early winter (September–December).

To assess long-term trends, a Mann-Kendall (MK) test was applied to annual and seasonal ARdriven precipitation (Figure 5.18). No significant trends were found at the annual scale
(p = 0.765), suggesting that while AR contributions fluctuate, they do not exhibit a statistically
significant increase or decrease over the entire period.
At the seasonal scale, while winter (p = 0.202), summer (p = 0.126), and fall (p = 0.415) show
trends of increasing precipitation, none approach significance. In contrast, spring precipitation
exhibits a decreasing trend (p = 0.158), though it remains statistically insignificant. Collectively,
these non-significant trends imply that interannual variability governs AR precipitation,
precluding the emergence of a coherent long-term trend.

61

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Figure 5.18: Mann-Kendall trend test results for atmospheric river (AR) related precipitation across four time intervals (1991–
2001, 2002–2012, 2013–2023, and 1991–2023), disaggregated by season and annually. Each cell displays the p-value of the test,
with triangle markers indicating the direction of the trend: blue upward-facing triangles denote increasing trends, and red
downward-facing triangles indicate decreasing trends. No trends across any time scale or season reach the conventional threshold
for statistical significance (p < 0.05).

5.6 Discharge Sensitivity to Daily Precipitation and AR Events
To investigate the short-term hydrological impacts of ARs, daily precipitation and discharge data
were analyzed for the 2024 monitoring season (17 June–10 October). This approach enabled
evaluation of discharge sensitivity to AR-induced precipitation events.
Figure 5.19 presents the net precipitation (mm) for five temporal periods separated into
AR-driven and non-AR-driven precipitation. AR events contributed to total precipitation in July
(87.7%), early October (75.5%), and September (40.3%), indicating that a minority of days
accounted for a substantial proportion of rainfall. In contrast, the late June period (17–30 June)
showed minimal AR influence (19.3%).

62

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Total Precipitation
AR-Driven Precipitation

Figure 5.19: Monthly partitioning of precipitation into total precipitation and atmospheric river (AR) driven precipitation at the
Gaat Héeni meteorological station for the 2024 observation period. Light gray bars represent total precipitation and black bars
show the AR-driven subset, defined as precipitation occurring on the day of AR landfall and the subsequent day. Percentages
indicate the proportion of monthly precipitation attributable to AR activity. The June and October labels reflect partial months of
data coverage (17–30 June and 1–10 October).

Figure 5.20 illustrates daily discharge (m³ s⁻¹) overlain with AR-driven and non-AR-driven
precipitation. Distinct discharge responses were observed following high-intensity AR events,
such as the 22–24 July event, when AR-driven precipitation triggered a sharp increase in
discharge from 8.7 m³ s⁻¹ to 30.0 m³ s⁻¹.

63

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Figure 5.20: Daily discharge (m³ s⁻¹; navy line) and precipitation (mm; inverted bars) at the Gaat Héeni meteorological station
from 17 June to 10 October 2024. Precipitation is partitioned into AR-driven (dark blue) and non-AR-driven (blue) totals, with
each bar representing cumulative daily rainfall. Discharge reflects mean daily values derived from continuous measurements.
Following snowmelt-driven attenuation in June, baseflows stabilize at low levels throughout late summer and early autumn,
punctuated by a major AR event in mid-July and more minor rainfall-driven fluctuations in late September and early October.

Centroid lag to peak discharge was calculated for precipitation events across the 2024
observation period to characterize the timing between precipitation inputs and resultant river
discharge response. Time between the centroid of precipitation and resulting discharge peak
ranged from 0.2 to 5.7 days, with a mean of 2.1 days and a median of 1.6 days. These values
highlight the rapid and variable response of the Gaat Héeni Watershed to precipitation inputs.
5.7 Quantifying Optimal Hydrological Conditions for Successful Gaat Barrier Passage
Between 5 and 30 July 2024, I conducted daily video monitoring at obstruction SR3-3 on the
Gaat Héeni to quantify jump success (Figure 5.21). Hourly discharge and water temperature
measurements recorded during each day’s observation window were averaged to yield daily
means, ranging from 10.57 °C to 14.11 °C and 8.85 m³ s⁻¹ to 28.02 m³ s⁻¹. These hydrological
conditions provide the framework for interpreting temporal patterns in passage behaviour under
baseline and elevated discharge events. Observational coverage was curtailed during the peak
discharge interval (24–27 July) owing to safety constraints imposed by an AR event (Figure

64

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

5.21), resulting in data gaps precisely when discharge and temperature reached their maximum
values.

Figure 5.21: Composite daily time series of Gaat passage success at obstruction SR3-3 on the Gaat Héeni (5–30 July 2024).
Vertical grey bars show total daily jump attempt counts recorded by video monitoring, over which daily mean river discharge
(blue circles and solid line) and daily mean water temperature (orange square and dashed line) are overlaid. Six 15-minute
monitoring blocks were conducted each day except on 23 July, when only two blocks were run. The percentages above each bar
indicate the daily percentage of successful jumps rounded to the nearest percent. Monitoring was suspended on 15 and 18 July
due to crew fatigue and on 24–27 July under unsafe river conditions. The 22–24 July interval corresponds to an atmospheric river
event during which both discharge and temperature peaked.

To quantify how water temperature and river discharge jointly influence Gaat jump‐success
probability, I fitted a logistic‐regression model that includes linear and quadratic terms for
temperature (T, °C) and discharge (Q, m³ s⁻¹), as detailed in Section 4.7. Figure 5.22 presents the
model’s predicted probability surface as a two‐dimensional heatmap. The colour gradient, from
deep purple through magenta to yellow, indicates the probability of successful jumps. On the
prediction grid, the model yields a maximum probability of 0.157 at a water temperature of 15.22
°C and river discharge of 12.15 m³ s⁻¹. Across the 110 observed 15-minute blocks, fitted
probabilities range from 0.016 (minimum) to 0.15 (maximum), with a mean of 0.06, a median of
0.06, and interquartile values of 0.04 (1st quartile) and 0.07 (3rd quartile).
Predicted success peaks (~0.13–0.16) at temperatures of 14–15 °C combined with moderate
65

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

discharges of 10–12 m³ s⁻¹, whereas probabilities fall below 0.05 under extremes of high
discharge (≥ 16 m³ s⁻¹) or low temperature (≤ 11 °C).

Figure 5.22: Heatmap of predicted jump success probability from a logistic-regression model (with linear and quadratic terms for
water temperature T and discharge Q) fitted to 15-minute block observations collected at obstruction SR3-3 (5–30 July 2024).
Colour intensity (deep purple to yellow) reflects the modelled values, which range from a minimum of 0.01 at T = 9.66 °C and
Q = 12.85 m³ s¹ to a maximum of 0.16 at T = 15.22 °C and Q = 12.15 m³ s¹. White circles represent the water temperature and
discharge conditions observed during each video monitoring block.

To assess the individual contributions of each predictor within the fitted logistic regression
model, I evaluated coefficient estimates using Wald z-tests (Table 5.1).

66

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

Table 5.1: Wald Z-Test Results for Hydrologic Predictors of Gaat Jump Success at SR3-3

Predictor

Estimate (β)

Std. Error

z value

p-value

(Intercept)

−10.61

2.09

−5.07

< 0.001

T

+0.38

0.28

+1.33

0.182

T²

−0.0028

0.011

−0.26

0.796

Q

+0.66

0.099

+6.69

< 0.001

Q²

−0.029

0.0043

−6.71

< 0.001

Among the predictor variables, discharge (Q) showed a significant positive linear effect on jump
success (β = 0.66, p < 0.001). This suggests that increasing discharge increases the likelihood of
a successful jump, likely by improving hydraulic depth and discharge. Furthermore, the quadratic
term for discharge (Q²) was also significant and negative (β = −0.029, p < 0.001), confirming the
presence of a non-linear discharge relationship. Together, these terms indicate that passage
success increases with discharge up to a peak, after which excessive discharge reduces success.
In contrast, temperature (T) and temperature squared (T²) were not statistically significant
(p = 0.182 and p = 0.796, respectively). This suggests that within the observed summer water
temperature range, water temperature had no strong linear or curvilinear effect on jump success.
5.8 Influence of Jump Success on Gaat Arrival at Gaat Áayi
The relationship between daily jump success (pd) at SR3-3 and arrival anomalies (ΔA) at Gaat
Áayi was investigated using linear regression models with lag intervals of one, two, and three
days.
The regression results revealed a significant, positive relationship between daily jump success
and Gaat arrival anomalies at all lag intervals. The model with a 1-day lag exhibited an R² value

67

5. RESULTS

GAAT MIGRATION UP THE GAAT HÉENI

of 0.640 and a p-value of < 0.01, indicating that daily jump success accounted for approximately
64% of the variation in arrival anomalies at the 1-day lag. The model with a 2-day lag showed an
even stronger relationship, with an R² of 0.709 and a p-value of < 0.01, suggesting that jump
success was a more reliable predictor of arrival anomalies after a 2-day delay. Finally, the 3-day
lag model had an R² of 0.431 and a p-value of < 0.01, which, although weaker than the 2-day
model, still demonstrated a statistically significant association between jump success and arrival
anomalies at longer lags.

68

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 6: DISCUSSION
AR-driven precipitation influences hydrological variability in the Gaat Héeni Watershed, shaping
seasonal and annual precipitation patterns and influencing river discharge (Section 6.1). As AR
frequency, intensity, and seasonal distribution shift, these changes influence Gaat migration
(Section 6.2). High and low AR-driven discharge have been linked to run collapses, underscoring
the need to establish hydrological thresholds for successful arrival of Gaat at Gaat Áayi (Section
6.3). Fluctuations in river discharge, partly driven by ARs, together with variations in water
temperature, determine the feasibility of passage through critical geomorphic barriers. To better
understand these impacts, video monitoring was employed at SR3-3 to assess Gaat passage
success (Section 6.4).
6.1 Contribution of ARs to Seasonal and Annual Precipitation
ARs contribute to seasonal and annual precipitation variability, particularly in coastal and
mountainous regions (Ralph et al., 2019; Sharma & Déry, 2020). In the Gaat Héeni Watershed,
ARs account for 15.9% to 37.9% of seasonal precipitation from 1991-2023 (Figure 5.15), with
the highest contributions occurring in fall (37.9%) and winter (24.1%). Notably, the fraction of
fall precipitation attributed to ARs (Figure 5.15) has varied only slightly across decades with no
statistically significant trend. On an annual scale, AR-driven precipitation in the Gaat Héeni
Watershed ranged from 55.0 mm in 1995 to 230.0 mm in 2004, with a mean annual contribution
of 139.6 mm over the 1991–2023 period (Figure 5.16). These events accounted for 16.4% to
37.8% of total annual precipitation, highlighting their influence on the hydrology of the Gaat
Héeni Watershed. If fall AR precipitation increases (Radić et al., 2015), it could enhance SWE
accumulation (Figure 5.11), elevating spring discharge volumes. Additionally, elevated AR
activity during summer can generate abrupt river discharge and water temperature increases
69

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

during the Gaat migration period. For example, during the 2024 observation period, an AR made
landfall from 22 to 24 July, accounting for 88% of that month’s precipitation (Figure 5.19) and
increasing discharge from 8.7 to 30.0 m³ s⁻¹ (Figures 5.20 and 5.21).
Interannual variability in AR frequency and intensity further influences precipitation patterns.
Although no statistically significant trend (p = 0.765) in total AR-driven precipitation is present
at an annual scale (Figure 5.18), seasonal shifts are evident. The Mann-Kendall test (Figure 5.18)
suggests a possible increase in summer (p = 0.126) and fall (p = 0.415) AR contributions,
though neither trend is statistically significant. These findings align with research showing
enhanced AR activity in late summer and fall along the North Pacific Coast (Gershunov et al.,
2019; Sobral & Déry, 2023).
6.2 The Impact of Changing Atmospheric River Patterns on Gaat
Changes in the distribution and intensity of ARs shape the hydrological conditions of the Gaat
Héeni Watershed, with potential consequences for the daily arrival of Gaat at Gaat Áayi and
annual escapement. The frequency, duration, and intensity of ARs modulate precipitation
patterns, river discharge, and water temperature. These are key variables governing the ability of
Gaat to navigate critical migration barriers (Cannon et al., 2017; Ralph et al., 2019). Section 6.2
examines how these hydrological factors may influence escapement trends by integrating
escapement and hydrological records from 1991 to 2023. Notably, escapement data reveal partial
run collapses in 2018 and 2021, with only 13 and 26 Gaat reaching Gaat Áayi, respectively,
compared to the long-term average of 3031 Gaat (Figure 1.2).
River discharge during the Gaat migration period is influenced by snow accumulated in the fall
(August to November) and precipitation from the current year. In 2017, the Gaat Héeni
70

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

Watershed received 117 mm of AR-driven precipitation, representing 22% of that year’s total,
and in 2018, 115 mm of AR-driven precipitation, accounting for 30% of annual precipitation
(Figure 5.16). The two-year average of 116 mm yr⁻¹ (26%) falls below the decadal mean of
137 mm yr⁻¹ (27%) for the period 2012–2023 (Figure 5.16). This depressed AR-driven
precipitation input occurred alongside subdued summer discharge conditions (Taku River Tlingit
Fisheries Department, 2019), which were associated with higher jump heights at key barriers and
reduced hydraulic connectivity. These conditions likely impeded upstream movement by
elevating fallback rates and prolonging migration durations, consistent with observations from
other watersheds (Padilla et al., 2015; Mathes et al., 2010).
In 2020, the Gaat Héeni Watershed recorded 214.8 mm of AR-driven precipitation, representing
36.4% of annual precipitation, while in 2021, AR-driven precipitation totaled 118.6 mm,
accounting for 22.5% of annual precipitation (Figure 5.16). The two-year average of
166.7 mm yr⁻¹ (29.4%) exceeds the decadal mean of 137 mm yr⁻¹ (27%) for the period 2013–
2023 (Figure 5.16). This elevated AR-driven precipitation contributed to elevated snowpack
accumulation during the 2020–2021 winter (Kolot et al., 2021), leading to elevated summer
discharge (Taku River Tlingit Fisheries Department, 2022). These high discharge conditions
disrupted migration success and resulted in higher-than-normal numbers of Gaat holding below
SR3-3 (Taku River Tlingit Fisheries Department, 2022). Similar to findings in other watersheds,
elevated discharge and turbulence reduce passage success, particularly at steep-gradient canyon
sites where hydraulic complexity is elevated (MacNutt et al., 2006).
Analysis from Section 5.8 demonstrates a temporal linkage between jump success at SR3-3 and
the arrival of Gaat at Gaat Áayi, with the strongest correlation (R² = 0.709; p < 0.001) observed

71

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

at a two-day lag interval. During the 2024 monitoring season, a prominent AR event (22-24 July)
increased river discharge from 8.7 m³ s⁻¹ to 30.0 m³ s⁻¹ (Figure 5.20), surpassing optimal
discharge thresholds (10–12 m³ s⁻¹) identified in Section 5.7. Before this AR event, holding
pools immediately downstream of SR3-3 were already observed to be at high capacity, indicating
significant Gaat accumulation and emphasizing the potential impact of rapidly increased
discharge in delaying migration. Given the identified two-day lag from passage success at SR3-3
to the arrival of Gaat at Gaat Áayi, it is likely that the AR-driven discharge increase on 22-24
July significantly influenced short-term outcomes. However, I cannot make a clear statistical
linkage as observations at SR3-3 did not begin until 28 July 2024, four days after the AR event.
While the timing of these escapement events aligns with notable hydrologic variation, this thesis
acknowledges the absence of a comprehensive statistical linkage between AR-driven hydrologic
extremes and longer-term escapement trends. The observational nature of these findings means
that explicit attribution of escapement solely to AR activity remains speculative and requires
further analysis.
These findings suggest high and low AR-driven precipitation extremes are associated with
escapement variation at Gaat Áayi. Given projections of increasing AR frequency and intensity
(Arora et al., 2025; Radić et al., 2015), understanding the hydrological mechanisms regulating
Gaat escapement success will be essential for sustaining populations.
6.3 Hydrological thresholds for Gaat Arrival at Gaat Áayi
The Gaat Héeni Watershed exhibits a rapid and variable hydrological response to precipitation,
reflecting the steep gradients and confined channels. Discharge sensitivity analysis from Section
5.6 found that the lag between the centroid of precipitation and the corresponding discharge peak
72

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

ranged from 0.2 to 5.7 days, with a mean of 2.1 days and a median of 1.6 days. This underscores
the watershed’s quick runoff response, meaning shifts in precipitation timing, particularly those
driven by ARs, can have rapid implications on river discharge.
Gaat escapement in the Gaat Héeni Watershed is influenced by the hydrological constraints of
water temperature and river discharge, which determine the feasibility of passage through
geomorphic barriers, including SR3-3 (Taku River Tlingit First Nation Fisheries Department,
2017). However, escapement is also shaped by extrinsic factors, including ocean survival
(Beamish et al., 2016; Crossin et al., 2008), pre-spawn mortality (Jeffries et al., 2012), and
interannual variability in run strength. Data collected from 5-30 July 2024, not including 15, 18,
24, 25, 26, and 27 July, indicate optimal jump success was seen with discharge levels between
10–12 m³ s-1 (Figure 5.22). Notably, an increase in river discharge during video monitoring from
8.7 m³ s⁻¹ to a peak of 18.05 m³ s⁻¹ from 22-24 July coincided with an AR event on 22 July
(Figure 5.20). Although river discharge rose to 30 m³ s⁻¹, jump behaviour was not recorded
during this interval due to safety constraints. These observations illustrate how AR events can
rapidly alter hydrological conditions, potentially constraining passage success.
Water temperature is a critical factor influencing the arrival of Gaat at Gaat Áayi (Atlas et al.,
2021; Hague et al., 2011). Within the range observed in July 2024, no water temperatures
exceeded 15.28 °C, and jump success was highest between 13–15 °C (Figure 5.22). However,
statistical analysis (Table 5.1) shows that water temperature was not a statistically significant
predictor of jump success at SR3-3 (linear term p = 0.182; quadratic term p = 0.796), reducing
confidence in conclusions drawn from temperature data alone. While laboratory studies and
migration data suggest that performance declines above 19 °C (Keefer et al., 2008; Quinn et al.,

73

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

1997), this study did not observe or model behaviour at higher temperatures. Projected warming
trends (Arora et al., 2025; Reid et al., 2022; SNAP-EWHALE, 2012) suggest that rising summer
temperatures may further constrain the thermal window required for the successful arrival of
Gaat at Gaat Áayi.
In contrast to water temperature data, river discharge was a statistically significant predictor of
jump success at SR3-3 (Table 5.1; linear term p < 0.001; quadratic term p < 0.001). Discharge
fluctuations strongly determine whether Gaat can successfully overcome migration barriers.
During the 2021 partial run collapse, high discharge may have increased the hydraulic
complexity of SR3-3, contributing to the low escapement observed that year (26 fish). Gaat are
forced to expend greater energy at high discharge to maintain their position and overcome
obstacles (Hinch & Rand, 1998). This effect is particularly pronounced at SR3-3 (Taku River
Tlingit Fisheries Department, 2022), where the absence of alternative migration routes prevents
bypassing turbulent flow zones and reduces passage success (Caudill et al., 2007; Enders et al.,
2009; Rand et al., 2006).
Conversely, when discharge falls below 9 m3 s-1, hydraulic connectivity between pools and
passage corridors is disrupted. This increases the vertical height of obstacles such as the SR3-3
waterfall feature, making passage more challenging. The precise threshold at which connectivity
breaks down may vary depending on site-specific conditions. For example, in 2018, escapement
decreased to 13 individuals during a low-discharge year, but in 2019, 605 individuals passed
under similar discharge conditions. This suggests that other factors, including channel
modifications at SR3-3 (Taku River Tlingit Fisheries Department, 2019) or interannual
differences in run size, may influence passage outcomes.

74

6. DISCUSSION

GAAT MIGRATION UP THE GAAT HÉENI

6.4 Evaluating the Effectiveness of Video Monitoring for Assessing Fish Passage Success
Video monitoring at SR3-3 during the 2024 migration season yielded detailed insights into Gaat
jump success by capturing fish behaviour under varying discharge and temperature regimes.
Using a fixed camera at a consistent vantage point, this non-invasive approach recorded
individual jump attempts and, when paired with concurrent measurements of river discharge and
water temperature, allowed us to identify specific hydrological thresholds linked to successful
passage. This method helped characterize the hydrological thresholds associated with observed
jump success by integrating video observation with river discharge and water temperature.
Although beneficial, video monitoring at SR3-3 had several limitations. Waterfall spray and
turbulence from SR3-3 obscured portions of the video footage, particularly at higher discharges.
This reduced the number of fully observable jump attempts and likely led to underestimating
total passage attempts and success rates. Additionally, Gaat probably ascended via routes hidden
behind SR3-3, making it challenging to quantify total passage accurately. Turbidity following
AR-driven precipitation events further reduced video clarity. Finally, the labour-intensive
manual review of footage required frame-by-frame analysis to assess passage attempts,
introducing potential observer bias and misclassification errors.

75

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

Chapter 7: CONCLUSION
7.1 Summary
This study examines the influence of ARs on hydrologic variability and its subsequent effects on
Gaat migration in the Gaat Héeni Watershed. The research quantifies AR-driven precipitation's
seasonal and interannual variability by integrating ERA5-Land reanalysis data, SIO-R1
Atmospheric River Catalog outputs, in-situ hydrometric observations, and escapement
monitoring records. It defines the hydrometric thresholds governing Gaat migration success.
Where appropriate, hydrologic and biological responses are presented as observational rather
than causal. These findings provide critical information for fisheries management in a region
increasingly impacted by hydrologic extremes.
Sections 5.2 to 5.3 establish the long-term climatological and hydrometeorological conditions of
the Gaat Héeni Watershed (1991–2023), revealing trends in precipitation, air temperature, and
seasonal hydrologic responses. Annual precipitation has increased at a rate of 0.93 mm per year,
with a distinct shift toward higher rainfall fractions after 2010. Before 2010, the mean annual
rainfall was 160.6 mm, whereas post-2010, it increased to 183.8 mm, marking a net gain of 23.2
mm per year.
The seasonal distribution of precipitation follows a predictable snowfall-to-rainfall transition,
with snowfall dominating winter months (December-February) and rainfall peaking during late
summer and early fall (August–October). The highest precipitation volumes occur between
September and November, with ARs accounting for up to 39.1% of fall precipitation (20132023). The increasing late-season precipitation intensity suggests a growing influence of ARs on
discharge variability, particularly during the critical migration window for Gaat.
76

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

Air temperature reanalysis records indicate a seasonal warming trend, with July exhibiting the
highest mean air temperature at 10.9 °C. The interaction between increased AR-driven
precipitation and rising air temperatures has resulted in more frequent high-discharge events in
the fall. However, whether these trends have translated to measurable long-term changes in
escapement remains uncertain.
Section 5.4 examines AR frequency, seasonal variability, and long-term trends using data from
the SIO-R1 Atmospheric River Catalog (1991–2023). Results confirm that AR landfalls exhibit a
distinct seasonal cycle, with peak activity occurring in late summer and early fall. The highest
AR frequency occurs in September, averaging four landfalling events per month, followed
closely by August and October. Net IVT values exceed 10,000 kg m⁻¹ s⁻¹ in September, making
this month the most hydrologically active.
Over the past three decades, the frequency of AR landfalls has increased, particularly in late
summer and fall. The most recent period (2013–2023) has shown a higher number of AR
landfalls than previous decades, with increased interannual variability. This trend suggests that
AR-driven hydrologic variability will continue to be a dominant force shaping precipitation
patterns.
Section 5.5 applied an event-based precipitation framework to quantify AR contribution to total
precipitation. Results indicate that AR-driven precipitation accounts for 15.9% to 39.1% of
seasonal precipitation, with the highest fractions occurring in fall and winter. Highlighting this,
fall precipitation is heavily influenced by ARs with contributions of ARs to total precipitation
ranging from 38.3% (1991–2001) to 39.1% (2013–2023).

77

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

A Mann-Kendall trend analysis applied to annual and seasonal AR-driven precipitation found no
statistically significant trends at the annual scale (p = 0.765). However, at the seasonal scale, AR
contributions in fall, winter, and summer show increasing trends but do not reach statistical
significance at the 0.05 level. These findings highlight the high interannual variability of ARdriven precipitation, reinforcing the need for adaptive hydrologic monitoring to anticipate
extreme discharge events.
High-resolution observations during the 2024 monitoring season revealed that the Gaat Héeni
Watershed exhibits a dynamic hydrologic response, with discharge reacting rapidly to AR and
non-AR rainfall events. As detailed in Section 5.6, the discharge response to precipitation during
the 2024 observation period was strongly modulated by AR activity. Additionally, centroid lag
analysis revealed that discharge peaks typically occurred within 0.2 to 5.7 days following
precipitation inputs, with a mean lag of 2.1 days and a median of 1.6 days, indicative of a
generally rapid but variable hydrological response. The most pronounced discharge event
occurred during the 22–24 July AR, when discharge rapidly increased from baseflow
(~8.7 m³ s⁻¹) to 30.0 m³ s⁻¹, exceeding the 16 m³ s⁻¹ hydraulic threshold for upstream fish
passage at SR3-3. These findings underscore the nonlinear nature of runoff generation in this
steep catchment and its implications for passage timing and the arrival of Gaat at Gaat Áayi.
Observational data from Section 5.7 reveal a non-linear relationship between discharge, water
temperature, and jump success rates at SR3-3. Logistic regression modelling (Table 5.1)
demonstrated that river discharge influenced jump success (linear term, p < 0.001; quadratic
term, p < 0.001), whereas water temperature was not a statistically significant predictor (linear
term, p = 0.182; quadratic term, p = 0.796), reducing confidence in water temperature-based

78

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

conclusions. Passage at SR3-3 was most successful at discharge levels between 10–12 m³ s⁻¹
(Figure 5.22), conditions that maintain laminar flow and facilitate efficient jumping. When river
discharge exceeds approximately 16 m³ s⁻¹, increased flow turbulence and velocity gradients
reduce passage success. Conversely, low discharge conditions (<9 m³ s⁻¹) increase jump distance
requirements, thereby reducing passage success due to insufficient water depth and disrupted
hydraulic connectivity (Taku River Tlingit Fisheries Department, 2017).
Analysis in Section 5.8 clarified the temporal relationship between jump success at SR3-3 and
the arrival of Gaat at Gaat Áayi, identifying a significant and strong predictive relationship at a
two-day lag (R² = 0.709, p < 0.001). Despite statistical relationships, monitoring gaps during
peak discharge periods (particularly during the 22-24 July AR event) limit the robustness of
these conclusions under extreme discharge conditions.
While river discharge emerged as the primary hydrological factor influencing jump success,
water temperature is known to affect migration efficiency and energy expenditure (Atlas et al.,
2021; Hague et al., 2011). Video monitoring at SR3-3 indicated higher passage probability
between 13–15 °C, although no statistically significant relationship was detected (Table 5.1).
Limited observational coverage at higher water temperatures (>15 °C) prevents robust
conclusions on the upper thermal limits of successful passage, where previous work has shown
impacts of elevated water temperature. Given projected regional warming trends (Reid et al.,
2022; SNAP-EWHALE, 2012), future increases in water temperature could further constrain the
already narrow discharge and water temperature conditions optimal for successful Gaat
migration.
From a fisheries management perspective, this study provides quantitative thresholds for
79

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

discharge and water temperature that can be used to guide conservation efforts. Given the rising
frequency of ARs from August through October, it is likely that Gaat migration success will be
increasingly influenced by hydrologic variability. However, this study does not assess long-term
escapement data through formal statistical linkage to hydrometric records. To mitigate run
collapses, real-time hydrometric monitoring at SR3-3 should be prioritized. Additionally,
integrating traditional Taku River Tlingit First Nation (TRTFN) knowledge with hydrologic
modelling will strengthen conservation efforts and ensure management approaches align with
empirical data and traditional ecological insights.
Ultimately, this work underscores the intricate relationship between ARs, river hydrology, and
Gaat migration success. Defining hydrometric thresholds for successful passage provides
scientific evidence to support conservation policy and fisheries management decisions in the
Gaat Héeni Watershed.
7.2 Study Limitations
While this study provides insights into the influence of ARs on hydrologic variability and the
arrival of Gaat at Gaat Áayi, limitations will be discussed below. These relate to hydrometric
data's spatial and temporal resolution, the range of observed temperature and discharge
conditions, temperature and discharge extremes, video monitoring, and climate change
considerations. Many patterns observed, particularly those involving Gaat passage, should be
interpreted as preliminary, given the absence of formal causal analyses. Addressing these
constraints in future research will improve our ability to predict and mitigate the impacts of ARdriven hydrologic extremes on Gaat migration.
One limitation concerns the discharge measurement location in relation to the video monitoring
80

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

location. Hydrometric data were collected above the confluence of Katina Creek, meaning that
the discharge conditions at SR3-3 were inferred rather than directly measured. This introduces
potential discrepancies in discharge estimates at the actual migration site. While the general
trends in discharge variability are captured, local discharge conditions at SR3-3 will differ due to
tributary inflows such as Katina Creek. Future research should deploy sensors to measure
discharge at SR3-3, allowing for real-time discharge measurements at the precise location of
Gaat passage. This would improve the accuracy of jump success models and provide a clearer
understanding of site-specific hydrodynamics.
Another limitation relates to the resolution and reliability of ERA5-Land reanalysis data,
particularly for precipitation. While ERA5-Land provides a consistent long-term dataset, it has
documented biases in complex mountainous terrain, often underestimating extreme precipitation
events and misrepresenting orographic effects. As a result, some AR-driven discharge responses
may be underrepresented in the dataset. This limitation is particularly relevant for Section 5.6,
which showed disproportionate discharge increases under high daily precipitation. Users of these
results should recognize that modelled AR contributions likely represent conservative estimates
of true hydrologic extremes.
A significant constraint in this study is the lack of conclusive results for extremely high and low
water temperatures and river discharges. While mid-range conditions (13–15 °C and 10–12 m³
s⁻¹) were identified as optimal for Gaat passage, data on passage success under more extreme
conditions (>16 m³ s⁻¹ or <9 m³ s⁻¹ discharge, and >15 °C or <9 °C water temperature) remain
insufficient. This is partly due to logistical challenges in monitoring fish passage under extreme
conditions and the migration window not aligning with suboptimal periods. This results from

81

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

monitoring gaps and natural migration timing, which limited observations during the most
physiologically challenging conditions. Future studies should extend video monitoring efforts
across a broader range of discharge and water temperature conditions, possibly using Passive
Integrated Transponder (PIT) tagging to track passage attempts over multiple years.
While valuable for quantifying passage success, the study's video monitoring approach is limited
by field-of-view constraints and sampling frequency. Cameras were deployed to observe jump
success at SR3-3, but Gaat behaviour outside the immediate jump zone was not fully captured.
Turbidity during peak discharge occasionally reduced video clarity, leading to potential
undercounting of failed passage attempts. Additionally, the inability to safely conduct
observations during periods of elevated discharge (the 22-24 July AR event) further limited data
collection under elevated hydrologic conditions. Future research should consider alternative
monitoring techniques, such as PIT tagging (Hemeon et al., 2024) or infrared imaging
(Hiebert et al., 2000), which may provide more consistent passage observations under various
hydrologic conditions.
Finally, this study does not include future climate projections, which limits long-term
conservation planning. While historical AR frequency and intensity trends were analyzed, the
study does not predict how future warming and altered precipitation patterns may affect Gaat
migration conditions. Incorporating climate scenarios would enhance the applicability of these
findings for proactive fisheries management and long-term habitat planning.
7.3 Future Direction
The recognition of changing AR intensity and spatial distribution (Gershunov et al., 2019; Payne
et al., 2020), alongside their demonstrated local hydrometric impacts, underscores the urgency of
82

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

acquiring high-resolution, year-round hydrometric and biological data in salmonid-bearing
watersheds vulnerable to extreme events (Bellmore et al., 2023). Improved monitoring capacity
is essential to anticipating, understanding, and mitigating the ecological consequences of
intensifying AR regimes under climate change.
One research priority is the expansion of long-term, individual-based tracking across variable
discharge conditions. PIT tagging enables continuous monitoring of fish passage across
migration barriers (Hemeon et al., 2024), capturing repeat attempts, fallback events, and
movement patterns with acceptable temporal resolution

(Haro & Castro-Santos, 2012).

When coupled with antenna arrays strategically installed at hydraulic constriction points such as
SR3-3, PIT tagging serves as a complementary and non-invasive method to video monitoring,
particularly under turbid, low-light, or spray-obscured conditions (Bateman et al., 2009). Future
PIT-based studies should span multiple migration seasons to capture interannual variability in
fish behaviour and barrier navigability, with antennas also deployed downstream of Katina Creek
to resolve discharge dynamics influencing passage feasibility.
To refine visual monitoring, future work should trial multi-angle video arrays to enhance
detection coverage, especially for Gaat passage events that occur behind the waterfall or under
variable visibility conditions. While thermal or infrared imagery may marginally improve
detectability in low-light or obscured conditions (Hiebert et al., 2000), their efficacy remains
uncertain due to the ectothermic nature of Gaat. Machine learning–assisted video classification
(Saberioon et al., 2017) offers significant potential to automate behavioural annotation, reduce
manual processing time, and increase consistency across reviewers. These tools should be
validated against PIT or sonar-based observations to evaluate detection accuracy and

83

7. CONCLUSION

GAAT MIGRATION UP THE GAAT HÉENI

classification precision under variable discharge and lighting regimes.
It is important to note that escapement reflects cumulative freshwater and marine influences,
including ocean survival, predation, and pre-spawn mortality (Rand et al., 2006). As such,
hydrologic factors observed here represent only one potential constraint within a broader suite of
environmental pressures. While descriptive comparisons of paired hydrologic and escapement
years highlight possible linkages, the current dataset does not support inferential testing. Future
efforts should explicitly test these relationships with multi-decadal escapement, discharge, and
water temperature datasets.
To address these limitations, future research should focus on advancing predictive, multi-scale
modelling approaches, expanding long-term fish tracking technologies, and integrating climate
change scenarios into Gaat migration assessments. These long-term datasets will be invaluable
for understanding how Gaat populations respond to shifting AR-driven hydrology over multiple
generations. To support this, a hydrometric station should be installed below the Katina Creek
confluence to account for potential discharges affecting SR3-3. Existing measurements upstream
of the confluence may underestimate the actual discharge conditions encountered by migrating
Gaat at the SR3-3. Future studies could develop predictive models using deep learning
algorithms to forecast migration success based on real-time hydrometric conditions. This would
enable fisheries managers to implement adaptive conservation strategies that optimize passage
success during critical migration windows.

84

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

REFERENCES
Adam, J. C., Hamlet, A. F., & Lettenmaier, D. P. (2009). Implications of global climate change
for snowmelt hydrology in the twenty‐first century. Hydrological Processes, 23(7), 962–972.
https://doi.org/10.1002/hyp.7201
Aitken, J. D. (1958). Atlin map-area (Memoir No. 307).
Armstrong, J. B., Ward, E. J., Schindler, D. E., & Lisi, P. J. (2016). Adaptive capacity at the
northern front: Sockeye Salmon behaviourally thermoregulate during novel exposure to warm
temperatures. Conservation Physiology, 4(1), https://doi.org/10.1093/conphys/cow039
Arora, V. K., Lima, A., & Shrestha, R. (2025). The effect of climate change on the simulated
streamflow of six Canadian rivers based on the CanRCM4 regional climate model. Hydrology
and Earth System Sciences, 29(1), 291–312. https://doi.org/10.5194/hess-29-291-2025
Atlas, W. I., Seitz, K. M., Jorgenson, J. W. N., Millard-Martin, B., Housty, W. G., RamosEspinoza, D., Burnett, N. J., Reid, M., & Moore, J. W. (2021). Thermal sensitivity and flowmediated migratory delays drive climate risk for coastal Sockeye Salmon. FACETS, 6(1), Article
1. https://doi.org/10.1139/facets-2020-0027
Bateman, D. S., Gresswell, R. E., & Berger, A. M. (2009). Passive integrated transponder tag
retention rates in headwater populations of Coastal Cutthroat Trout. North American Journal of
Fisheries Management, 29(3), 653–657. https://doi.org/10.1577/M07-169.1
Bathurst, J. C., Birkinshaw, S. J., Cisneros Espinosa, F., & Iroumé, A. (2017). Forest impact on
flood peak discharge and sediment yield in streamflow. In N. Sharma (Ed.), River System
Analysis and Management (pp. 15–29). Springer Singapore. https://doi.org/10.1007/978-981-101472-7_2
Beacham, T. D., Cox-Rogers, S., MacConnachie, C., McIntosh, B., & Wallace, C. G. (2014).
Population structure and run timing of Sockeye Salmon in the Skeena River, British Columbia.
North American Journal of Fisheries Management, 34(2), 335–348.
https://doi.org/10.1080/02755947.2014.880761
Beacham, T. D., McIntosh, B., & MacConnachie, C. (2004). Population structure of lake‐type
and river‐type sockeye salmon in transboundary rivers of northern British Columbia. Journal of
Fish Biology, 65(2), 389–402. https://doi.org/10.1111/j.0022-1112.2004.00457.x
Beamish, R. J., Neville, C. M., Sweeting, R. M., Beacham, T. D., Wade, J., & Li, L. (2016).
Early ocean life history of Harrison River Sockeye Salmon and their contribution to the
biodiversity of Sockeye Salmon in the Fraser River, British Columbia, Canada. Transactions of
the American Fisheries Society, 145(2). https://doi.org/10.1080/00028487.2015.1123182
Bellmore, J. R., Sergeant, C. J., Bellmore, R. A., Falke, J. A., & Fellman, J. B. (2023). Modeling
Coho Salmon (Oncorhynchus kisutch) population response to streamflow and water temperature
extremes. Canadian Journal of Fisheries and Aquatic Sciences, 80(2), Article 2.
https://doi.org/10.1139/cjfas-2022-0129

85

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Birnie-Gauvin, K., Patterson, D. A., Cooke, S. J., Hinch, S. G., & Eliason, E. J. (2023).
Anaerobic exercise and recovery: roles and implications for mortality in Pacific Salmon. Reviews
in Fisheries Science & Aquaculture, 31(4), 497–522.
https://doi.org/10.1080/23308249.2023.2224902
Blöschl, G. (1991). The influence of uncertainty in air temperature and albedo on snowmelt.
Hydrology Research, 22(2), 95–108. https://doi.org/10.2166/nh.1991.0007
Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R., Stevens, M. H. H., &
White, J.-S. S. (2009). Generalized linear mixed models: A practical guide for ecology and
evolution. Trends in Ecology & Evolution, 24(3), 127–135.
https://doi.org/10.1016/j.tree.2008.10.008
Brett, J. R. (1971). Energetic responses of Salmon to temperature: A study of some thermal
relations in the physiology and freshwater ecology of Sockeye Salmon (Oncorhynchus nerka).
American Zoologist, 11(1), 99–113. https://doi.org/10.1093/icb/11.1.99
Burt, J. M., Hinch, S. G., & Patterson, D. A. (2012). Developmental temperature stress and
parental identity shape offspring burst swimming performance in Sockeye Salmon
(Oncorhynchus nerka). Ecology of Freshwater Fish, 21(2), Article 2.
https://doi.org/10.1111/j.1600-0633.2011.00535.x
Carey, M. P., Keith, K. D., Schelske, M., Lean, C., McCormick, S. D., Regish, A., &
Zimmerman, C. E. (2019). Energy Depletion and Stress Levels in Sockeye Salmon Migrating at
the Northern Edge of their Distribution. Transactions of the American Fisheries Society, 148(4),
785–797. https://doi.org/10.1002/tafs.10172
Carey, M. P., Zimmerman, C. E., Keith, K. D., Schelske, M., Lean, C., & Douglas, D. C. (2017).
Migration trends of Sockeye Salmon at the northern edge of their distribution. Transactions of
the American Fisheries Society, 146(4), 791–802.
https://doi.org/10.1080/00028487.2017.1302992
Caudill, C. C., Daigle, W. R., Keefer, M. L., Boggs, C. T., Jepson, M. A., Burke, B. J., Zabel, R.
W., Bjornn, T. C., & Peery, C. A. (2007). Slow dam passage in adult Columbia River salmonids
associated with unsuccessful migration: delayed negative effects of passage obstacles or
condition-dependent mortality? Canadian Journal of Fisheries and Aquatic Sciences, 64(7),
979–995. https://doi.org/10.1139/f07-065
Chen, J., Brissette, F. P., Chaumont, D., & Braun, M. (2013). Finding appropriate bias correction
methods in downscaling precipitation for hydrologic impact studies over North America. Water
Resources Research, 49(7), 4187–4205. https://doi.org/10.1002/wrcr.20331
Clelland, A. A., Marshall, G. J., & Baxter, R. (2024). Evaluating the performance of key ERA‐
Interim, ERA5 and ERA5-Land climate variables across Siberia. International Journal of
Climatology, 44(7), 2318–2342. https://doi.org/10.1002/joc.8456
Comiti, F., Lucía, A., & Rickenmann, D. (2016). Large wood recruitment and transport during
large floods: A review. Geomorphology, 269, 23–39.
https://doi.org/10.1016/j.geomorph.2016.06.016

86

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Copernicus Climate Change Service. (2019). ERA5-Land hourly data from 1950 to present
[Dataset]. Copernicus Climate Change Service (C3S) Climate Data Store (CDS).
https://doi.org/10.24381/CDS.E2161BAC
Crossin, G. T., Hinch, S. G., Cooke, S. J., Cooperman, M. S., Patterson, D. A., Welch, D. W.,
Hanson, K. C., Olsson, I., English, K. K., & Farrell, A. P. (2009). Mechanisms influencing the
timing and success of reproductive migration in a capital breeding semelparous fish species, the
Sockeye Salmon. Physiological and Biochemical Zoology, 82(6), 635–652.
https://doi.org/10.1086/605878
Crossin, G. T., Hinch, S. G., Cooke, S. J., Welch, D. W., Patterson, D. A., Jones, S. R. M., Lotto,
A. G., Leggatt, R. A., Mathes, M. T., Shrimpton, J. M., Van Der Kraak, G., & Farrell, A. P.
(2008). Exposure to high temperature influences the behaviour, physiology, and survival of
Sockeye Salmon during spawning migration. Canadian Journal of Zoology, 86(2), 127–140.
https://doi.org/10.1139/Z07-122
Crossin, G. T., Hinch, S. G., Farrell, A. P., Higgs, D. A., Lotto, A. G., Oakes, J. D., & Healey, M.
C. (2004). Energetics and morphology of Sockeye Salmon: Effects of upriver migratory distance
and elevation. Journal of Fish Biology, 65(3), 788–810. https://doi.org/10.1111/j.00221112.2004.00486.x
Crozier, L. G., Scheuerell, M. D., & Zabel, R. W. (2011). Using time series analysis to
characterize evolutionary and plastic responses to environmental change: a case study of a shift
toward earlier migration date in Sockeye Salmon. The American Naturalist, 178(6), 755–773.
https://doi.org/10.1086/662669
Crozier, L. G., Siegel, J. E., Wiesebron, L. E., Trujillo, E. M., Burke, B. J., Sandford, B. P., &
Widener, D. L. (2020). Snake River Sockeye and Chinook salmon in a changing climate:
implications for upstream migration survival during recent extreme and future climates. PLOS
ONE, 15(9), Article 9. https://doi.org/10.1371/journal.pone.0238886
Cruz-Jesus, F., Oliveira, T., Bacao, F., & Irani, Z. (2017). Assessing the pattern between
economic and digital development of countries. Information Systems Frontiers, 19(4), 835–854.
https://doi.org/10.1007/s10796-016-9634-1
Cunningham, C. J., Westley, P. A. H., & Adkison, M. D. (2018). Signals of large scale climate
drivers, hatchery enhancement, and marine factors in Yukon River Chinook Salmon survival
revealed with a bayesian life history model. Global Change Biology, 24(9), Article 9.
https://doi.org/10.1111/gcb.14315
Davie, T., & Quinn, N. W. (2019). Fundamentals of Hydrology (Third edition). Routledge, Taylor
& Francis Group.
Déry, S. J., Stadnyk, T. A., MacDonald, M. K., & Gauli-Sharma, B. (2016). Recent trends and
variability in river discharge across northern Canada. Hydrology and Earth System Sciences,
20(12), 4801–4818. https://doi.org/10.5194/hess-20-4801-2016
Dettinger, M. (2011). Climate change, atmospheric rivers, and floods in California - a
multimodel analysis of storm frequency and magnitude changes. JAWRA Journal of the
American Water Resources Association, 47(3). https://doi.org/10.1111/j.1752-1688.2011.00546.x

87

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Dickson, N. E., Carrivick, J. L., & Brown, L. E. (2012). Flow regulation alters alpine river
thermal regimes. Journal of Hydrology, 464–465, 505–516.
https://doi.org/10.1016/j.jhydrol.2012.07.044
Dingman, S. L. (2015). Physical Hydrology (3rd ed.). Waveland Press, Long Grove, IL.
Dockery, D. R., McMahon, T. E., Kappenman, K. M., & Blank, M. (2017). Evaluation of
swimming performance for fish passage of Longnose Dace Rhinichthys Cataractae using an
experimental flume. Journal of Fish Biology, 90(3), 980–1000. https://doi.org/10.1111/jfb.13217
Eiler, J. H., Nelson, B. D., & Bradshaw, R. F. (1992). Riverine spawning by Sockeye Salmon in
the Taku River, Alaska and British Columbia. Transactions of the American Fisheries Society,
121(6), 701–708. https://doi.org/10.1577/1548-8659(1992)121<0701:RSBSSI>2.3.CO;2
Eiras-Barca, J., Ramos, A. M., Pinto, J. G., Trigo, R. M., Liberato, M. L. R., & Miguez-Macho,
G. (2018). The concurrence of atmospheric rivers and explosive cyclogenesis in the North
Atlantic and North Pacific basins. Earth System Dynamics, 9(1), 91–102.
https://doi.org/10.5194/esd-9-91-2018
Eliason, E. J., Clark, T. D., Hague, M. J., Hanson, L. M., Gallagher, Z. S., Jeffries, K. M., Gale,
M. K., Patterson, D. A., Hinch, S. G., & Farrell, A. P. (2011). Differences in thermal tolerance
among Sockeye Salmon populations. Science, 332(6025), Article 6025.
https://doi.org/10.1126/science.1199158
Enders, E. C., Gessel, M. H., & Williams, J. G. (2009). Development of successful fish passage
structures for downstream migrants requires knowledge of their behavioural response to
accelerating flow. Canadian Journal of Fisheries and Aquatic Sciences, 66(12), 2109–2117.
https://doi.org/10.1139/F09-141
English, K. K., Koski, W. R., Sliwinski, C., Blakley, A., Cass, A., & Woodey, J. C. (2005).
Migration timing and river survival of late‐run Fraser River Sockeye Salmon estimated using
radiotelemetry techniques. Transactions of the American Fisheries Society, 134(5), 1342–1365.
https://doi.org/10.1577/T04-119.1
Environment and Climate Change Canada. (2024). Atlin 1981-2010 Station Data [Dataset].
https://climate.weather.gc.ca/climate_normals/results_1981_2010_e.html?stnID=1485&a
utofwd=1
Espinoza, V., Waliser, D. E., Guan, B., Lavers, D. A., & Ralph, F. M. (2018). Global analysis of
climate change projection effects on atmospheric rivers. Geophysical Research Letters, 45(9),
4299–4308. https://doi.org/10.1029/2017GL076968
Fallot, J.-M., Barry, R. G., & Hoogstrate, D. (1997). Variations of mean cold season temperature,
precipitation and snow depths during the last 100 years in the former Soviet Union (FSU).
Hydrological Sciences Journal, 42(3), 301–327. https://doi.org/10.1080/02626669709492031
Farley, E. V., Murphy, J. M., Adkison, M., & Eisner, L. (2007). Juvenile Sockeye Salmon
distribution, size, condition and diet during years with warm and cool spring sea temperatures
along the eastern Bering Sea shelf. Journal of Fish Biology, 71(4), 1145–1158.
https://doi.org/10.1111/j.1095-8649.2007.01587.x

88

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Farrell, A. P. (2009). Environment, antecedents and climate change: Lessons from the study of
temperature physiology and river migration of salmonids. Journal of Experimental Biology,
212(23). https://doi.org/10.1242/jeb.023671
Feddern, M. L., Schoen, E. R., Shaftel, R., Cunningham, C. J., Chythlook, C., Connors, B. M.,
Murdoch, A. D., Von Biela, V. R., & Woods, B. (2023). Kings of the north: bridging disciplines
to understand the effects of changing climate on Chinook Salmon in the Arctic-YukonKuskokwim Region. Fisheries, 48(8). https://doi.org/10.1002/fsh.10923
Fisheries, N. (2025, January 4). Sockeye Salmon | NOAA Fisheries. NOAA.
https://www.fisheries.noaa.gov/species/sockeye-salmon
Foote, C. J., Brown, G. S., & Hawryshyn, C. W. (2004). Female colour and male choice in
Sockeye Salmon: Implications for the phenotypic convergence of anadromous and
nonanadromous morphs. Animal Behaviour, 67(1), 69–83.
https://doi.org/10.1016/j.anbehav.2003.02.004
Foster, L. M., Bearup, L. A., Molotch, N. P., Brooks, P. D., & Maxwell, R. M.. (2016). Energy
budget increases reduce mean streamflow more than snow–rain transitions: using integrated
modeling to isolate climate change impacts on Rocky Mountain hydrology. Environmental
Research Letters, 11(4). https://doi.org/10.1088/1748-9326/11/4/044015
Frandsen, S. L., & Hasselbalch, J. A. (2024). Who are the green transition experts? Towards a
new research agenda on climate change knowledge. WIREs Climate Change, 15(6), e917.
https://doi.org/10.1002/wcc.917
Fraser, C., & Jackson, S. (2018). Shifts in air temperature and high‐magnitude winter
precipitation events in coastal North America: implications for environmental assessment and
management. Integrated Environmental Assessment and Management, 14(2).
https://doi.org/10.1002/ieam.2012
Gautam, S. S., & Singh, V. (2022). Adaptive discretization using golden section to aid outlier
detection for software development effort estimation. IEEE Access, 10, 90369–90387.
https://doi.org/10.1109/ACCESS.2022.3200149
Gerlach, T. (2011). Volcanic versus anthropogenic carbon dioxide. Eos, Transactions American
Geophysical Union, 92(24), 201–202. https://doi.org/10.1029/2011EO240001
Gershunov, A., Shulgina, T., Clemesha, R. E. S., Guirguis, K., Pierce, D. W., Dettinger, M. D.,
Lavers, D. A., Cayan, D. R., Polade, S. D., Kalansky, J., & Ralph, F. M. (2019). Precipitation
regime change in Western North America: the role of atmospheric rivers. Scientific Reports, 9(1),
9944. https://doi.org/10.1038/s41598-019-46169-w
Gershunov, A., Shulgina, T., Ralph, F. M., Lavers, D. A., & Rutz, J. J. (2017). Assessing the
climate‐scale variability of atmospheric rivers affecting western North America. Geophysical
Research Letters, 44(15). https://doi.org/10.1002/2017GL074175
Gimeno, L., Stohl, A., Trigo, R. M., Dominguez, F., Yoshimura, K., Yu, L., Drumond, A., Durán‐
Quesada, A. M., & Nieto, R. (2012). Oceanic and terrestrial sources of continental precipitation.
Reviews of Geophysics, 50(4), Article 4. https://doi.org/10.1029/2012RG000389
Glossary of Meteorology. (2022). American Meteorological Society.
89

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Gomis-Cebolla, J., Rattayova, V., Salazar-Galán, S., & Francés, F. (2023). Evaluation of ERA5
and ERA5-Land reanalysis precipitation datasets over Spain (1951–2020). Atmospheric
Research, 284, 106606. https://doi.org/10.1016/j.atmosres.2023.106606
Goniea, T. M., Keefer, M. L., Bjornn, T. C., Peery, C. A., Bennett, D. H., & Stuehrenberg, L. C.
(2006). Behavioral thermoregulation and slowed migration by adult fall Chinook Salmon in
response to high Columbia River water temperatures. Transactions of the American Fisheries
Society, 135(2). https://doi.org/10.1577/T04-113.1
Government of British Columbia. (2024). Atlin Lake (4E02B). https://bcmoeprod.aquaticinformatics.net/Report/Show/SnowMSS.4E02B.MSS%20Report/
Graf, W. L. (2006). Downstream hydrologic and geomorphic effects of large dams on American
rivers. Geomorphology, 79(3–4), 336–360. https://doi.org/10.1016/j.geomorph.2006.06.022
Grantham, T. E. (2013). Use of hydraulic modelling to assess passage flow connectivity for
Salmon in streams. River Research and Applications, 29(2), 250–267.
https://doi.org/10.1002/rra.1591
Guan, B., & Waliser, D. E. (2019). Tracking atmospheric rivers globally: spatial distributions and
temporal evolution of life cycle characteristics. Journal of Geophysical Research: Atmospheres,
124(23), 12523–12552. https://doi.org/10.1029/2019JD031205
Guo, Y., Shinoda, T., Guan, B., Waliser, D. E., & Chang, E. K. M. (2020). Statistical relationship
between atmospheric rivers and extratropical cyclones and anticyclones. Journal of Climate,
33(18), 7817–7834. https://doi.org/10.1175/JCLI-D-19-0126.1
Hague, M. J., Ferrari, M. R., Miller, J. R., Patterson, D. A., Russell, G. L., Farrell, A. P., &
Hinch, S. G. (2011). Modelling the future hydroclimatology of the lower Fraser River and its
impacts on the spawning migration survival of sockeye salmon: Sockeye Salmon migratory
survival. Global Change Biology, 17(1), 87–98. https://doi.org/10.1111/j.13652486.2010.02225.x
Hall, B. D., Saunders, P., & White, D. R. (2024). Digital representation of scales and units for
temperature and related quantities. 020005. https://doi.org/10.1063/5.0235453
Hanson, A. J., & Smith, H. D. (1967). Mate selection in a population of Sockeye Salmon
(Oncorhynchus nerka) of mixed age-groups. Journal of the Fisheries Research Board of Canada,
24(9), 1955–1977. https://doi.org/10.1139/f67-160
Hanson, K. C., Cooke, S. J., Hinch, S. G., Crossin, G. T., Patterson, D. A., English, K. K.,
Donaldson, M. R., Shrimpton, J. M., Van Der Kraak, G., & Farrell, A. P. (2008). Individual
variation in migration speed of upriver‐migrating Sockeye Salmon in the Fraser River in relation
to their physiological and energetic status at marine approach. Physiological and Biochemical
Zoology, 81(3), 255–268. https://doi.org/10.1086/529460
Haro, A., & Castro-Santos, T. (2012). Passage of American Shad: paradigms and realities.
Marine and Coastal Fisheries, 4(1), 252–261. https://doi.org/10.1080/19425120.2012.675975

90

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Harrison, X. A., Donaldson, L., Correa-Cano, M. E., Evans, J., Fisher, D. N., Goodwin, C. E. D.,
Robinson, B. S., Hodgson, D. J., & Inger, R. (2018). A brief introduction to mixed effects
modelling and multi-model inference in ecology. PeerJ, 6, e4794.
https://doi.org/10.7717/peerj.4794
Hauer, C., Unfer, G., Tritthart, M., & Habersack, H. (2011). Effects of stream channel
morphology, transport processes and effective discharge on salmonid spawning habitats. Earth
Surface Processes and Landforms, 36(5), 672–685. https://doi.org/10.1002/esp.2087
Hawcroft, M. K., Shaffrey, L. C., Hodges, K. I., & Dacre, H. F. (2012). How much Northern
Hemisphere precipitation is associated with extratropical cyclones? Geophysical Research
Letters, 39(24), 2012GL053866. https://doi.org/10.1029/2012GL053866
Hay, L. E., & McCabe, G. J. (2010). Hydrologic effects of climate change in the Yukon River
Basin. Climatic Change, 100(3–4), 509–523. https://doi.org/10.1007/s10584-010-9805-x
Hemeon, K. M., Peterson, D. P., Twibell, R. G., Kennedy, B. M., & Piteo, M. S. (2024). Rates of
PIT tag ingestion and gastric retention in three species of hatchery-reared salmonids and
potential implications for postrelease monitoring. North American Journal of Fisheries
Management, 44(6), 1489–1511. https://doi.org/10.1002/nafm.11050
Henderson, M. A., Levy, D. A., & Stockner, J. S. (1992). Probable consequences of climate
change on freshwater production of Adams River sockeye salmon (Oncorynchus nerka).
GeoJournal, 28(1), 51–59. https://doi.org/10.1007/BF00216406
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., Nicolas, J.,
Peubey, C., Radu, R., Schepers, D., Simmons, A., Soci, C., Abdalla, S., Abellan, X., Balsamo,
G., Bechtold, P., Biavati, G., Bidlot, J., Bonavita, M., … Thépaut, J. (2020). The ERA5 global
reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730), 1999–2049.
https://doi.org/10.1002/qj.3803
Hiebert, S., Helfrich, L. A., Weigmann, D. L., & Liston, C. (2000). Anadromous Salmonid
passage and video image quality under infrared and visible light at Prosser Dam, Yakima River,
Washington. North American Journal of Fisheries Management, 20(3), 827–832.
https://doi.org/10.1577/1548-8675(2000)020<0827:ASPAVI>2.3.CO;2
Hinch, S. G., & Bratty, J. (2000). Effects of swim speed and activity pattern on success of adult
Sockeye Salmon migration through an area of difficult passage. Transactions of the American
Fisheries Society, 129(2), 598–606. https://doi.org/10.1577/15488659(2000)129<0598:EOSSAA>2.0.CO;2
Hinch, S. G., & Rand, P. S. (1998). Swim speeds and energy use of upriver-migrating Sockeye
Salmon (Oncorhynchus nerka): Role of local environment and fish characteristics. Canadian
Journal of Fisheries and Aquatic Sciences, 55(8), 1821–1831. https://doi.org/10.1139/f98-067
Hooke, J. M. (1979). An analysis of the processes of river bank erosion. Journal of Hydrology,
42(1–2), 39–62. https://doi.org/10.1016/0022-1694(79)90005-2
Hosmer, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied Logistic Regression (Third
edition). Wiley.

91

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Howard, K. G., & Von Biela, V. (2023). Adult spawners: A critical period for subarctic Chinook
salmon in a changing climate. Global Change Biology, 29(7), 1759–1773.
https://doi.org/10.1111/gcb.16610
Hruska, K. A., Hinch, S. G., Healey, M. C., Patterson, D. A., Larsson, S., & Farrell, A. P. (2010).
Influences of sex and activity level on physiological changes in individual adult Sockeye Salmon
during rapid senescence. Physiological and Biochemical Zoology, 83(4), 663–676.
https://doi.org/10.1086/652411
Huang, S., Krysanova, V., Zhai, J., & Su, B. (2015). Impact of intensive irrigation activities on
river discharge under agricultural scenarios in the semi-arid Aksu River Basin, Northwest China.
Water Resources Management, 29(3), 945–959. https://doi.org/10.1007/s11269-014-0853-2
Ide, J., Finér, L., Laurén, A., Piirainen, S., & Launiainen, S. (2013). Effects of clear-cutting on
annual and seasonal runoff from a boreal forest catchment in eastern Finland. Forest Ecology
and Management, 304, 482–491. https://doi.org/10.1016/j.foreco.2013.05.051
IPCC. (2025). Special Report: Global Warming of 1.5 oC.
https://www.ipcc.ch/sr15/chapter/glossary/
Islam, S. U., Hay, R. W., Déry, S. J., & Booth, B. P. (2019). Modelling the impacts of climate
change on riverine thermal regimes in western Canada’s largest Pacific watershed. Scientific
Reports, 9(1), 11398. https://doi.org/10.1038/s41598-019-47804-2
Jeffries, K. M., Hinch, S. G., Sierocinski, T., Clark, T. D., Eliason, E. J., Donaldson, M. R., Li,
S., Pavlidis, P., & Miller, K. M. (2012). Consequences of high temperatures and premature
mortality on the transcriptome and blood physiology of wild adult Sockeye Salmon
(Oncorhynchus nerka). Ecology and Evolution, 2(7), 1747–1764.
https://doi.org/10.1002/ece3.274
Jeoung, H., Shi, S., & Liu, G. (2022). A novel approach to validate satellite snowfall retrievals by
ground-based point measurements. Remote Sensing, 14(3), 434.
https://doi.org/10.3390/rs14030434
Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, M., Saha, S.,
White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, R., Chelliah, M., Ebisuzaki, W., Higgins,
W., Janowiak, J., Mo, K. C., Ropelewski, C., … Joseph, D. (1996). The NCEP/NCAR 40-year
reanalysis project. Bulletin of the American Meteorological Society, 77(3), 437–471.
https://doi.org/10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2
Kang, D. H., Shi, X., Gao, H., & Déry, S. J. (2014). On the changing contribution of snow to the
hydrology of the Fraser River Basin, Canada. Journal of Hydrometeorology, 15(4), 1344–1365.
https://doi.org/10.1175/JHM-D-13-0120.1
Kattel, D. B., & Yao, T. (2018). Temperature–topographic elevation relationship for high
mountain terrain: An example from the southeastern Tibetan Plateau. International Journal of
Climatology, 38(S1). https://doi.org/10.1002/joc.5418
Kendall, M. G. (1948). Rank Correlation Methods.
https://www.cambridge.org/core/product/identifier/S0020268100013019/type/journal_article

92

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Khrustaleva, A. (2024). Spatial and ecological structure of the Sockeye Salmon population
Oncorhynchus nerka of Kuril Lake. E3S Web of Conferences, 548, 07010.
https://doi.org/10.1051/e3sconf/202454807010
Kocan, R., LaPatra, S., Gregg, J., Winton, J., & Hershberger, P. (2006). Ichthyophonus‐induced
cardiac damage: A mechanism for reduced swimming stamina in salmonids. Journal of Fish
Diseases, 29(9), 521–527. https://doi.org/10.1111/j.1365-2761.2006.00745.x
Kolot, J., Mischler, A., Bier, A., & Goulding, H. (2021). Yukon snow survey bulletin and water
supply forecast. Yukon Government.
Kouki, K., Luojus, K., & Riihelä, A. (2023). Evaluation of snow cover properties in ERA5 and
ERA5-Land with several satellite-based datasets in the Northern Hemisphere in spring 1982–
2018. The Cryosphere, 17(12), 5007–5026. https://doi.org/10.5194/tc-17-5007-2023
Lauritzen, D. V., Hertel, F., & Gordon, M. S. (2005). A kinematic examination of wild Sockeye
Salmon jumping up natural waterfalls. Journal of Fish Biology, 67(4), 1010–1020.
https://doi.org/10.1111/j.0022-1112.2005.00799.x
Lavers, D. A., Ralph, F. M., Richardson, D. S., & Pappenberger, F. (2020). Improved forecasts of
atmospheric rivers through systematic reconnaissance, better modelling, and insights on
conversion of rain to flooding. Communications Earth & Environment, 1(1), Article 1.
https://doi.org/10.1038/s43247-020-00042-1
Lazic, S. E., Mellor, J. R., Ashby, M. C., & Munafo, M. R. (2020). A Bayesian predictive
approach for dealing with pseudoreplication. Scientific Reports, 10(1), 2366.
https://doi.org/10.1038/s41598-020-59384-7
Lenzi, M. A., Mao, L., & Comiti, F. (2006). Effective discharge for sediment transport in a
mountain river: Computational approaches and geomorphic effectiveness. Journal of Hydrology,
326(1–4), 257–276. https://doi.org/10.1016/j.jhydrol.2005.10.031
Li, Y., Sun, F., Chen, Y., Fang, G., Li, Z., Duan, W., Qin, J., Zhang, X., & Li, B. (2025).
Unraveling the complexities of rain-on-snow events in High Mountain Asia. Npj Climate and
Atmospheric Science, 8(1), 118. https://doi.org/10.1038/s41612-025-00943-y
Li, Y., Zeng, Z., Zhao, L., & Piao, S. (2015). Spatial patterns of climatological temperature lapse
rate in mainland China: A multi–time scale investigation. Journal of Geophysical Research:
Atmospheres, 120(7), 2661–2675. https://doi.org/10.1002/2014JD022978
Magnoni, L. J., Patterson, D. A., Farrell, A. P., & Weber, J.-M. (2006). Effects of long-distance
migration on circulating lipids of Sockeye Salmon (Oncorhynchus nerka). Canadian Journal of
Fisheries and Aquatic Sciences, 63(8), 1822–1829. https://doi.org/10.1139/f06-083
Mai, J., Kornelsen, K. C., Tolson, B. A., Fortin, V., Gasset, N., Bouhemhem, D., Schäfer, D.,
Leahy, M., Anctil, F., & Coulibaly, P. (2020). The Canadian Surface Prediction Archive
(CaSPAr): a platform to enhance environmental modeling in Canada and globally. Bulletin of the
American Meteorological Society, 101(3), E341–E356. https://doi.org/10.1175/BAMS-D-190143.1
Mann, H. B. (1945). Nonparametric tests against trend. Econometrica, 13(245).

93

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Martins, E. G., Hinch, S. G., Patterson, D. A., Hague, M. J., Cooke, S. J., Miller, K. M.,
Robichaud, D., English, K. K., & Farrell, A. P. (2012). High river temperature reduces survival
of Sockeye Salmon (Oncorhynchus nerka) approaching spawning grounds and exacerbates
female mortality. Canadian Journal of Fisheries and Aquatic Sciences, 69(2), 330–342.
https://doi.org/10.1139/f2011-154
Mathers, K. L., Robinson, C. T., & Weber, C. (2022). Patchiness in flow refugia use by
macroinvertebrates following an artificial flood pulse. River Research and Applications, 38(4),
696–707. https://doi.org/10.1002/rra.3941
Mathes, M. T., Hinch, S. G., Cooke, S. J., Crossin, G. T., Patterson, D. A., Lotto, A. G., &
Farrell, A. P. (2010). Effect of water temperature, timing, physiological condition, and lake
thermal refugia on migrating adult Weaver Creek sockeye salmon (Oncorhynchus nerka).
Canadian Journal of Fisheries and Aquatic Sciences, 67(1), 70–84. https://doi.org/10.1139/F09158
Mattikalli, N. M., Devereux, B. J., & Richards, K. S. (1996). Prediction of river discharge and
surface water quality using an integrated geographical information system approach.
International Journal of Remote Sensing, 17(4), 683–701.
https://doi.org/10.1080/01431169608949038
Maynard, G. A., Kinnison, M. T., & Zydlewski, J. D. (2017). Size selection from fishways and
potential evolutionary responses in a threatened Atlantic Salmon population. River Research and
Applications, 33(7), 1004–1015. https://doi.org/10.1002/rra.3155
McArdle, B. H. (1988). The structural relationship: regression in biology. Canadian Journal of
Zoology, 66(11), 2329–2339. https://doi.org/10.1139/z88-348
Milliman, J. D., Farnsworth, K. L., Jones, P. D., Xu, K. H., & Smith, L. C. (2008). Climatic and
anthropogenic factors affecting river discharge to the global ocean, 1951–2000. Global and
Planetary Change, 62(3–4), 187–194. https://doi.org/10.1016/j.gloplacha.2008.03.001
Modarres, R., & Ouarda, T. B. M. J. (2013). Modeling rainfall–runoff relationship using
multivariate GARCH model. Journal of Hydrology, 499, 1–18.
https://doi.org/10.1016/j.jhydrol.2013.06.044
Morton, W. M. (1965). The taxonomic significance of the kype in American Salmonids. Copeia,
1965(1), 14. https://doi.org/10.2307/1441233
Mosley, M. P. (1982). Analysis of the effect of changing discharge on channel morphology and
instream uses in a braided river, Ohau River, New Zealand. Water Resources Research, 18(4),
800–812. https://doi.org/10.1029/WR018i004p00800
Mueller, R. P., Southard, S. S., May, C. W., Pearson, W. H., & Cullinan, V. I. (2008). Juvenile
Coho Salmon leaping ability and behavior in an experimental culvert test bed. Transactions of
the American Fisheries Society, 137(4), 941–950. https://doi.org/10.1577/T06-244.1
Muñoz, S. E., Hamilton, B., & Parazin, B. (2023). Contrasting ocean–atmosphere dynamics
mediate flood hazard across the Mississippi River Basin. Earth Interactions, 27(1).
https://doi.org/10.1175/EI-D-22-0015.1

94

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Muñoz-Sabater, J., Dutra, E., Agustí-Panareda, A., Albergel, C., Arduini, G., Balsamo, G.,
Boussetta, S., Choulga, M., Harrigan, S., Hersbach, H., Martens, B., Miralles, D. G., Piles, M.,
Rodríguez-Fernández, N. J., Zsoter, E., Buontempo, C., & Thépaut, J.-N. (2021). ERA5-Land: A
state-of-the-art global reanalysis dataset for land applications. Earth System Science Data, 13(9),
4349–4383. https://doi.org/10.5194/essd-13-4349-2021
Neiman, P. J., Ralph, F. M., Wick, G. A., Lundquist, J. D., & Dettinger, M. D. (2008).
Meteorological characteristics and overland precipitation impacts of atmospheric rivers affecting
the west coast of North America based on eight years of SSM/I satellite observations. Journal of
Hydrometeorology, 9(1), 22–47. https://doi.org/10.1175/2007JHM855.1
Park, C., Son, S., & Kim, H. (2021). Distinct features of atmospheric rivers in the early versus
late east Asian summer monsoon and their impacts on monsoon rainfall. Journal of Geophysical
Research: Atmospheres, 126(7). https://doi.org/10.1029/2020JD033537
Payne, A. E., Demory, M.E., Leung, L. R., Ramos, A. M., Shields, C. A., Rutz, J. J., Siler, N.,
Villarini, G., Hall, A., & Ralph, F. M. (2020). Responses and impacts of atmospheric rivers to
climate change. Nature Reviews Earth & Environment, 1(3). https://doi.org/10.1038/s43017-0200030-5
Peng, C.Y. J., Lee, K. L., & Ingersoll, G. M. (2002). An introduction to logistic regression
analysis and reporting. The Journal of Educational Research, 96(1), 3–14.
https://doi.org/10.1080/00220670209598786
Poff, N. L., Allan, J. D., Bain, M. B., Karr, J. R., Prestegaard, K. L., Richter, B. D., Sparks, R. E.,
& Stromberg, J. C. (1997). The natural flow regime. BioScience, 47(11), 769–784.
https://doi.org/10.2307/1313099
Preston-Thomas, H. (1990). The international temperature scale of 1990 (ITS-90). Metrologia,
27(1), 3–10. https://doi.org/10.1088/0026-1394/27/1/002
Quinn, T. P. (1988). Estimated swimming speeds of migrating adult Sockeye Salmon. Canadian
Journal of Zoology, 66(10), 2160–2163. https://doi.org/10.1139/z88-322
Quinn, T. P. (Ed.). (2005). The behavior and ecology of Pacific Salmon and Trout. American
Fisheries Society.
Quinn, T. P., Hodgson, S., & Peven, C. (1997). Temperature, flow, and the migration of adult
sockeye salmon (fmodel) in the Columbia River. Canadian Journal of Fisheries and Aquatic
Sciences, 54(6), 1349–1360. https://doi.org/10.1139/f97-038
R Core Team. (2024). R: The R project for statistical computing. https://www.r-project.org/
Radchenko, V. I., & Mathisen, O. A. (2004). Distribution, growth, and feeding of Sockeye
Salmon in the western Bering Sea. Transactions of the American Fisheries Society, 133(3), 606–
621. https://doi.org/10.1577/T03-072.1
Radić, V., Cannon, A. J., Menounos, B., & Gi, N. (2015). Future changes in autumn atmospheric
river events in British Columbia, Canada, as projected by CMIP5 global climate models. Journal
of Geophysical Research: Atmospheres, 120(18), 9279–9302.
https://doi.org/10.1002/2015JD023279

95

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Rahman, M. S., Senkbeil, J. C., & Keellings, D. J. (2023). Spatial and temporal variability of
extreme precipitation events in the Southeastern United States. Atmosphere, 14(8), 1301.
https://doi.org/10.3390/atmos14081301
Rahmat, A., Noda, K., Onishi, T., & Senge, M. (2018). Runoff characteristics of forest
watersheds under different forest managements. Reviews in Agricultural Science, 6, 119–133.
https://doi.org/10.7831/ras.6.119
Ralph, F. M., Cannon, F., Tallapragada, V., Davis, C. A., Doyle, J. D., Pappenberger, F.,
Subramanian, A., Wilson, A. M., Lavers, D. A., Reynolds, C. A., Haase, J. S., Centurioni, L.,
Ingleby, B., Rutz, J. J., Cordeira, J. M., Zheng, M., Hecht, C., Kawzenuk, B., & Delle Monache,
L. (2020). West coast forecast challenges and development of atmospheric river reconnaissance.
Bulletin of the American Meteorological Society, 101(8). https://doi.org/10.1175/BAMS-D-190183.1
Ralph, F. M., Rutz, J. J., Cordeira, J. M., Dettinger, M., Anderson, M., Reynolds, D., Schick, L.
J., & Smallcomb, C. (2019). A scale to characterize the strength and impacts of atmospheric
rivers. Bulletin of the American Meteorological Society, 100(2). https://doi.org/10.1175/BAMSD-18-0023.1
Rand, P. S., Hinch, S. G., Morrison, J., Foreman, M. G. G., MacNutt, M. J., Macdonald, J. S.,
Healey, M. C., Farrell, A. P., & Higgs, D. A. (2006). Effects of river discharge, temperature, and
future climates on energetics and mortality of adult migrating Fraser River Sockeye Salmon.
Transactions of the American Fisheries Society, 135(3), 655–667. https://doi.org/10.1577/T05023.1
Ravi, S., & D’Odorico, P. (2005). A field‐scale analysis of the dependence of wind erosion
threshold velocity on air humidity. Geophysical Research Letters, 32(21).
https://doi.org/10.1029/2005GL023675
Reed, T. E., Schindler, D. E., Hague, M. J., Patterson, D. A., Meir, E., Waples, R. S., & Hinch, S.
G. (2011). Time to evolve? Potential evolutionary responses of Fraser River Sockeye Salmon to
climate change and effects on persistence. PLoS ONE, 6(6), e20380.
https://doi.org/10.1371/journal.pone.0020380
Reid, K. A., Reid, D. G., & Brown, C. D. (2022). Patterns of vegetation change in Yukon: Recent
findings and future research in dynamic subarctic ecosystems. Environmental Reviews, 30(3),
380–401. https://doi.org/10.1139/er-2021-0110
Ruggerone, G. T., & Irvine, J. R. (2018). Numbers and biomass of natural and hatchery origin
Pink Salmon, Chum Salmon, and Sockeye Salmon in the North Pacific Ocean, 1925–2015.
Marine and Coastal Fisheries, 10(2), 152–168. https://doi.org/10.1002/mcf2.10023
Saberioon, M., Gholizadeh, A., Cisar, P., Pautsina, A., & Urban, J. (2017). Application of
machine vision systems in aquaculture with emphasis on fish: State‐of‐the‐art and key issues.
Reviews in Aquaculture, 9(4), 369–387. https://doi.org/10.1111/raq.12143
Sagar, B. (2023). Llewellyn Data [Dataset].

96

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Shanley, J. B., & Chalmers, A. (1999). The effect of frozen soil on snowmelt runoff at Sleepers
River, Vermont. Hydrological Processes, 13(12–13), 1843–1857.
https://doi.org/10.1002/(SICI)1099-1085(199909)13:12/13<1843::AID-HYP879>3.0.CO;2-G
Sharma, A. R., & Déry, S. J. (2020). Contribution of atmospheric rivers to annual, seasonal, and
extreme precipitation across British Columbia and Southeastern Alaska. Journal of Geophysical
Research: Atmospheres, 125(9). https://doi.org/10.1029/2019JD031823
SNAP-EWHALE. (2012). Predicting future potential climate-biomes for the Yukon, Northwest
Territories and Alaska. Scenarios Network for Arctic Planning, and Ecological Wildlife Habitat
Data Analysis for the Land and Seascape Laboratory. University of Alaska Fairbanks.
http://www.snap.uaf.edu/attachments/Cliomes-FINAL.pdf
Sobral, B. S., & Déry, S. J. (2023). Spatiotemporal distribution and trend analyses of atmospheric
rivers affecting British Columbia’s Nechako Watershed. International Journal of Climatology,
43(14). https://doi.org/10.1002/joc.8230
Steinhausen, M. F., Sandblom, E., Eliason, E. J., Verhille, C., & Farrell, A. P. (2008). The effect
of acute temperature increases on the cardiorespiratory performance of resting and swimming
Sockeye Salmon (Oncorhynchus nerka). Journal of Experimental Biology, 211(24), 3915–3926.
https://doi.org/10.1242/jeb.019281
Sueyoshi, M., Nakano, D., & Nakamura, F. (2014). The relative contributions of refugium types
to the persistence of benthic invertebrates in a seasonal snowmelt flood. Freshwater Biology,
59(2), 257–271. https://doi.org/10.1111/fwb.12262
Surian, N., Mao, L., Giacomin, M., & Ziliani, L. (2009). Morphological effects of different
channel‐forming discharges in a gravel‐bed river. Earth Surface Processes and Landforms, 34(8),
1093–1107. https://doi.org/10.1002/esp.1798
Taku River Tlingit Fisheries Department. (2017). Kuthai Lake Access Improvement Assessment.
Taku River Tlingit Fisheries Department. (2019). 2019 Summary Report for the Silver Salmon
River Sockeye Access Improvement Project.
Taku River Tlingit Fisheries Department. (2022). Silver Salmon Passage Remediation Report.
Tarek, M., Brissette, F. P., & Arsenault, R. (2020). Evaluation of the ERA5 reanalysis as a
potential reference dataset for hydrological modelling over North America. Hydrology and Earth
System Sciences, 24(5), 2527–2544. https://doi.org/10.5194/hess-24-2527-2020
Thériault, J. M., Leroux, N. R., Tchuem Tchuente, O., & Stewart, R. E. (2023). Characteristics of
rain-snow transitions over the Canadian Rockies and their changes in warmer climate conditions.
Atmosphere-Ocean, 61(5), 352–367. https://doi.org/10.1080/07055900.2023.2251938
Thomas, J. O. (2010). Fraser River Sockeye recreational hook and release mortality study. J.O.
Thomas and Associates Ltd. https://psf.ca/wp-content/uploads/2021/10/Download-PDF1521.pdf#:~:text=The%20mean%20fork%20length%20was,4%20cm%20for%20beach%20seined
Thornton, M. M., Shrestha, R., Wei, Y., Thornton, P. E., & Kao, S.-C. (2022). Daymet: daily
surface weather data on a 1-km grid for North America, version 4 R1. ORNL Distributed Active
Archive Center. https://doi.org/10.3334/ORNLDAAC/2129
97

REFERENCES

GAAT MIGRATION UP THE GAAT HÉENI

Turcotte, B., Morse, B., Bergeron, N. E., & Roy, A. G. (2011). Sediment transport in ice-affected
rivers. Journal of Hydrology, 409(1–2), 561–577. https://doi.org/10.1016/j.jhydrol.2011.08.009
Tyler, A. V., Swanton, C. O., & McIntosh, B. C. (2001). Feeding ecology of maturing Sockeye
Salmon (Oncorhynchus nerka) in nearshore waters of the Kodiak Archipelago.
https://www.boem.gov/sites/default/files/boem-newsroom/Library/Publications/2001/2001059.pdf
United States Geological Survey. (2024). USGS 15041200 TAKU R NR JUNEAU AK.
Vallejo-Bernal, S. M., Wolf, F., Boers, N., Traxl, D., Marwan, N., & Kurths, J. (2023). The role
of atmospheric rivers in the distribution of heavy precipitation events over North America.
Hydrology and Earth System Sciences, 27(14). https://doi.org/10.5194/hess-27-2645-2023
Wagner, G. N., Hinch, S. G., Kuchel, L. J., Lotto, A., Jones, S. R., Patterson, D. A., Macdonald,
J. S., Kraak, G. V. D., Shrimpton, M., English, K. K., Larsson, S., Cooke, S. J., Healey, M. C., &
Farrell, A. P. (2005). Metabolic rates and swimming performance of adult Fraser River Sockeye
Salmon (Oncorhynchus nerka) after a controlled infection with Parvicapsula minibicornis.
Canadian Journal of Fisheries and Aquatic Sciences, 62(9), 2124–2133.
https://doi.org/10.1139/f05-126
Wang, T., Hamann, A., Spittlehouse, D., & Carroll, C. (2016). Locally downscaled and spatially
customizable climate data for historical and future periods for North America. PLOS ONE, 11(6).
https://doi.org/10.1371/journal.pone.0156720
White, J. C., Hannah, D. M., House, A., Beatson, S. J. V., Martin, A., & Wood, P. J. (2017).
Macroinvertebrate responses to flow and stream temperature variability across regulated and
non‐regulated rivers. Ecohydrology, 10(1). https://doi.org/10.1002/eco.1773
Wickham, H. (2016). Ggplot2. Springer International Publishing. https://doi.org/10.1007/978-3319-24277-4
Wood, R. A., Crucifix, M., Lenton, T. M., Mach, K. J., Moore, C., New, M., Sharpe, S., Stocker,
T. F., & Sutton, R. T. (2023). A climate science toolkit for high impact‐low likelihood climate
risks. Earth’s Future, 11(4). https://doi.org/10.1029/2022EF003369
Zhou, Y., North, J. S., Rhoades, A. M., Tao, J., Rudisill, W., Risser, M. D., & Collins, W. D.
(2025). Atmospheric river frequency‐category characteristics shape U.S. West coast runoff.
Journal of Geophysical Research: Atmospheres, 130(2). https://doi.org/10.1029/2024JD041805

98