Lake trout (Salvelinus namaycush) otoliths as indicators of past climate patterns and growth in Arctic lakes

Torvinen, Eric S.

Name:

Torvinen_E_2017.pdf

Size:

9.339Mb

Format:

PDF

Download

Author

Torvinen, Eric S.

Chair

Falke, Jeffrey

Committee

Arp, Christopher
Zimmerman, Christian
Sutton, Trent

Description

Thesis (M.S.) University of Alaska Fairbanks, 2017

The effects of climate change on freshwater ecosystems are amplified in high-latitude regions, however, Alaska climate data are limited due to the remote location of the Arctic. Predictions have indicated that warming temperatures owing to climate change could increase fish growth, but the magnitude and factors influencing these changes remain uncertain. Here I investigated the relationship between Lake Trout Salvelinus namaycush growth and physical and biological characteristics, fish community structure and climate patterns. I applied biochronology techniques to predict recent climate patterns from annual growth increments recorded on Lake Trout otoliths. Growth increments were also used to perform length-at-age back-calculations and to estimate the growth coefficient K, as described by a von Bertalanffy growth model. Lake Trout were captured from 13 climate-sensitive lakes in the Fish Creek watershed in Arctic Alaska during 2014 and 2015. Individual Lake Trout (N = 53) ranged from 471--903 mm fork length (FL; mean = 589.3 mm) and their readable annuli, representative of age, ranged from 9--55 annual growth increments. I constructed a growth chronology for the period 1977--2014 and used model selection to identify the best predictive model of relative Lake Trout growth (ring width index; RWI) as a function of climate descriptors. A single covariate model was the best predictor and indicated that RWI tracked mean August air temperature recorded at a local weather station from 1998--2013 (P < 0.001; R2adj = 0.55; RMSE = 0.048). Lake Trout growth (K) was subsequently modeled as a function of physical and biological characteristics, and fish community structure, using multiple linear regression. The highest ranked model included physical (i.e., depth, distance to river and coast, connectivity class, and number of stream intersections) and biological (sex) covariates. Model averaging indicated K was higher in deeper, well connected lakes, located further from the coast and was lower with increasing distance from a large river, though the relationship with depth was found to be the single significant covariate. This study demonstrated the utility of biochronology techniques to estimate past climate patterns in remote regions, and provided valuable knowledge regarding growth-environment relationships for Lake Trout. In turn, this information can be used to better understand the effects of a changing environment in sensitive Arctic lake ecosystems.

Date

2017-05

Type

Thesis

Collections

Fisheries