High-resolution Modeling Studies of the Changing Risks of Damage from Extratropical Cyclones

1. Motivation

Extratropical cyclones that generate damaging windstorms are among the costliest of natural disasters, with Western Europe and the British Isles being particularly hard hit. These storms often generate flooding rains, strong winds, and can cause storm surges that impact coastal regions. There are potential competing effects of climate change on extratropical cyclones, making it unclear how these storms will be affected as Earth warms. In the lower atmosphere, polar amplification is expected to lead to a weaker temperature gradient, which would reduce the baroclinicity and likely reduce the strength and frequency of storms. With a warmer climate, however, it is expected there will be more water vapor in the atmosphere, which could lead to stronger storms through diabatic processes.

2. Methods

2.1 Climate Change

To analyze possible changes in these extreme events, climate and case study simulations are used to compare events in a current and projected future climate. The future climate projection is generated using a “pseudo global-warming” process similar to previous studies (Frei et al., 1998, Schär et al. 1996). Five IPCC AR4 models using the A2 scenario are used to generate an average ensemble change. These global climate models (GCMs) vary in resolution, so their output is interpolated to a common 1˚ x 1˚ grid to match operational analysis data that are used as boundary and initial conditions in the limited area simulations. The data are also linearly interpolated in the vertical to make all GCM levels match those of the operational analysis. The 100-year thermodynamic climate changes are taken as the difference between the decadal averages for future (2090-2099) and present (1990-1999) climates.

Boundary conditions for the control runs use unaltered operational analyses. The 100-year climate change is added to these to create future boundary conditions for the climate change simulations. A second climate change scenario includes doubled CO2 as well as the thermodynamic changes to examine how sensitive the climate change results are to increased greenhouse gases.

2.2 Simulations

We focus on the winter months when conditions are most favorable for the generation of damaging storms. Ten seasons of simulations for January February and March are performed on a 120 km grid, comparable to current GCMs, and on a high-resolution, 20 km, grid. The higher resolution is expected to provide a more detailed look at the storm climate, especially by resolving diabatic processes that contribute to storm development. A case study of Xynthia, a damaging windstorm in late February 2010, is also analyzed with a focus on the dynamical changes of the storm resulting from climate change. This case study has one-way nests with a parent domain of 60 km and inner nests of 20 km and 6.6 km. The boundary conditions for the control simulation and the climate change simulation will be done the same way as the climate simulations.

All simulations are performed with the Weather Research and Forecasting (WRF) model, version 3.2.1 and cover the area of Europe and the North Atlantic storm track. Changes in storm intensity are quantified using extreme value theory, in order to focus on the damaging tail of the distributions of winds and precipitation.

3. Preliminary results

Storm tracks for present and future climates are analyzed to see if there are frequency changes or track shifts. Significant changes in the climatological stormtrack and its responses to climate change are expected going from 120 km to 20 km grid spacing, based on our earlier results from a case study of Xynthia. For Xynthia, there are only small changes in the minimum sea-level pressure in the future climate, despite the added moisture, while the winds and precipitation are stronger in the future simulation. It is hypothesized that the diabatic effects are responsible for the stronger winds, since present and future storms do not differ notably in their intensities. A physics ensemble of this storm is examined to see if these results are robust and to explore the dynamics driving these changes.