Application of a numerical model to predict impacts of climate change on water temperatures in two deep, oligotrophic lakes in New Zealand

We applied a numerical hydrodynamic model (DYRESM) to two large, deep New Zealand lakes that are characterised by deep thermoclines and high wind forcing, to assess their sensitivity to changes in climate. Modifications to standard model parameters were necessary for the successful application of DYRESM. Predictions from downscaled global circulation models suggest an increase in mean air temperature, rainfall, and wind speeds. Modelling the hydrodynamics of the lakes suggests that increasing air temperatures would offset the cooling influences of increased rainfall and river flows, resulting in warmer overall lake temperatures, and an earlier, longer, and shallower thermal stratification. These physical changes could affect phytoplankton production as their light limitation would decrease in duration and intensity. However, deeper mixing caused by increases in wind speed would negate this reduction of thermocline depth. While warmer air temperatures appear to be the dominant driver of changes in thermal structure, changes in other meteorological factors, especially wind speed, are important in predicting future hydrodynamics. Compared to large, deep lakes in the Northern Hemisphere, the predicted warming rates in Lakes Wanaka and Wakatipu are slower, due partly to a lower predicted rate of atmospheric warming and the absence of winter ice cover in these lakes.