Daniel Lachat: Wind Trends in the Far North: Model-based Climate Projection

2015-02-25T15:03:07+00:00February 25th, 2015|2014 Conference, Climate Change|

Daniel Lachat (York University)

Title: Wind Trends in the Far North: Model-based Climate Projection

The North American Regional Reanalysis (NARR) data have shown increases in wind speed in Ontario of up to 60% for the past thirty years. Since measurement stations are scarce in the far-north due to challenges in accessibility and maintenance, the NARR model offers the possibility of assessing climate trends in the coastal regions of Hudson and James Bay Lowlands where these increases are strongest. Reduced sea-ice cover has been recognized as the dominant factor for accelerated warming in polar regions due to the ice-albedo feedback and enhanced release of heat from the newly ice-free water bodies. The resulting reduction of atmospheric stability is accompanied by a downward transfer of horizontal momentum from aloft, inducing increases in surface wind speeds in marine environments. There are multiple impacts of increasing wind speed in the northern region, e.g. coast erosion due to wave energy increases, threat to power supply infrastructure etc., reflecting its high vulnerability. To ensure adaptation of the northern communities to these changes, climate projection to the near future is essential. The Regional Climate Model (RCM) PRECIS (Wang et al. 2012, 2014) driven by the boundary conditions of the HadCM3 ensemble of coupled ocean-atmosphere global climate model, allows one to project climate variables such as wind, temperature and precipitation up to 2095 with a high spatial resolution of 25kmx25km grid cells. In order to quantify the effects of modeling uncertainties, the initial parameters of the HadCM3 GCM were perturbed in the Quantifying Uncertainty in Model Predictions (QUMP) research project, which results in 17 different GCM’s referred to as HadCM3Q0-16. For the purpose of validation, we compare the PRECIS RCM output with the North American Regional Climate Change Assessment Program (NARCCAP) model data, which is generated by running a variety of combinations of GCM’s and RCM’s, for the same study domain. Aside from estimating climate parameters in the future, climate models also offer the possibility of analyzing climate processes in remote regions where actual measurements are scarce. This is particularly evident for assessing climate variables over water bodies such as Hudson and James Bay. Furthermore, with the usage of RCM outputs, we have conducted a detailed examination, much more than possible using a mere spatial interpolation between measurement stations, of the impact of climate variability on spatially limited phenomena, such as the exposure of power transmission lines linking the coast of James Bay with the provincial power grid. The abundance of climate parameters derived from RCM’s do not only allow the quantification of future climate impacts, but also a comprehensive statistical analysis in the quest for causal relationships of various underlying processes. In the context of threat to the indigenous communities arising from climate change impact on northern environments, we examine the balance between model uncertainties associated with future climate projections and their usefulness in developing adaptation strategies.