Physiological and ecological effects of increasing temperature on fish production in lakes of Arctic Alaska

Abstract

Lake ecosystems in the Arctic are changing rapidly due to climate warming. Lakes are sensitive integrators of climate-induced changes and prominent features across the Arctic landscape, especially in lowland permafrost regions such as the Arctic Coastal Plain of Alaska. Despite many studies on the implications of climate warming, how fish populations will respond to lake changes is uncertain for Arctic ecosystems. Least Cisco (Coregonus sardinella) is a bellwether for Arctic lakes as an important consumer and prey resource. To explore the consequences of climate warming, we used a bioenergetics model to simulate changes in Least Cisco production under future climate scenarios for lakes on the Arctic Coastal Plain. First, we used current temperatures to fit Least Cisco consumption to observed annual growth. We then estimated growth, holding food availability, and then feeding rate constant, for future projections of temperature. Projected warmer water temperatures resulted in reduced Least Cisco production, especially for larger size classes, when food availability was held constant. While holding feeding rate constant, production of Least Cisco increased under all future scenarios with progressively more growth in warmer temperatures. Higher variability occurred with longer projections of time mirroring the expanding uncertainty in climate predictions further into the future. In addition to direct temperature effects on Least Cisco growth, we also considered changes in lake ice phenology and prey resources for Least Cisco. A shorter period of ice cover resulted in increased production, similar to warming temperatures. Altering prey quality had a larger effect on fish production in summer than winter and increased relative growth of younger rather than older age classes of Least Cisco. Overall, we predicted increased production of Least Cisco due to climate warming in lakes of Arctic Alaska. Understanding the implications of increased production of Least Cisco to the entire food web will be necessary to predict ecosystem responses in lakes of the Arctic.

Keywords: Arctic Coastal Plain; climate change; coregonidae; fish bioenergetics; lake food webs.

Figure 1

Least Cisco caught in Arctic Alaska (photo credit R. Brown).

Figure 2

Water temperature data plotted through the ice-free period (day of year, DOY: 152–273) for current conditions (average daily temperature of 2009, 2010, and 2011) and future projections for 2040, 2060, and 2090. Future predictions are calculated as current water temperature plus the predicted increase for air temperature in Atqasuk, AK by the Scenarios Network for Alaska and Arctic Planning (SNAP 2013). We incorporate SNAP projections based on emission scenario B1, A1B, and A2 and averaged values by day for each emission scenario.

Figure 3

Biomass accumulated for different age classes of Least Cisco and the proportion of maximum consumption (pCmax) estimated for the annual growth of Least Cisco under current temperatures.

Figure 4

Daily weight accumulation (today's mass minus yesterday's mass) of Least Cisco plotted throughout a year. Age classes of Age-2, 9, and 18 demonstrate seasonal patterns with current temperature conditions.

Figure 5

The relative growth estimates ([final biomass − initial biomass]/initial biomass) of Least Cisco are plotted across age classes for current temperatures and for the Scenarios Network for Alaska and Arctic Planning (SNAP 2013) projected temperatures in 2040, 2060, and 2090. The average value (±SD) of biomass is calculated across the three SNAP scenarios (B1, A1B, and A2). Growth estimates are made holding constant the proportion of maximum consumption (pCmax) fitted under current conditions (see Table 1) for projections in 2040, 2060, and 2090.

Figure 6

The amount of biomass accumulated annually by each age class of Least Cisco estimated by the bioenergetics model for current temperatures and for the Scenarios Network for Alaska and Arctic Planning (SNAP 2013) projected temperatures in 2040, 2060, and 2090. The average value (±SD) of biomass is calculated across the three SNAP scenarios (B1, A1B, and A2). Biomass estimates are made holding constant (A) the proportion of maximum consumption (pCmax) or (B) the annual ration (percent biomass of fish) for projections in 2040, 2060, and 2090.

Figure 7

The relative growth estimates ([final biomass − initial biomass]/initial biomass) of Least Cisco are plotted across age classes for (A) warmer water temperature (+1°C) during ice cover or warmer temperatures during open water. (B) Relative growth estimates are plotted for shorter periods of ice cover by adding 10 open-water days or adding 20 open-water days.

Figure 8

The relative growth estimates ([final biomass − initial biomass]/initial biomass) of Least Cisco are plotted across age classes for differences in prey energy during the summer and winter. Growth estimates are made holding constant the proportion of maximum consumption (pCmax) fitted under current conditions Table 1.