| 英文摘要: | Extratropical cyclones are the main generators of the strong winds that cause large ocean waves in temperate regions of the world. The severity of the winds associated with these storms is poorly represented by the coarse resolution of current global climate models (GCMs), making it challenging to produce projections of the future climate of large waves. Wind data from GCMs can be downscaled in resolution using dynamical methods, resulting in a successful reproduction of the mean wave climate, but a suboptimal reproduction of the storm wave climate1. Projections of large wave occurrence can also be produced using statistical downscaling methods, although such methods have previously been applied only to three or less GCMs2, 3, preventing a robust assessment of confidence in projections based on variation between models. Consequently, considerable uncertainty remains in projections of the future storm wave climate. Here we apply a statistical diagnostic of large wave occurrence in eastern Australia to 18 different GCMs, allowing model variations to be examined in greater detail than previously possible. Results are remarkably consistent between different GCMs, allowing anthropogenic influences to be clearly demonstrated, with fewer days with large waves expected to occur in eastern Australia due to increasing greenhouse gas concentrations. There is growing interest in understanding the climatology of surface ocean waves, partly due to their role in coastal erosion and inundation when coupled with rising sea levels4, as well as their potential for renewable energy generation5. Although tropical cyclones can have some influence on the occurrence of large waves in subtropical regions, the largest waves along the central east coast of Australia are most commonly attributable to extratropical cyclones6. The large waves caused by these storms can have severe impacts on coastal regions, such as being a major contributor to elevated water levels due to wave set-up7. Large waves can also have desirable benefits for coastal areas including recreational pursuits such as surfing, as well as influencing biodiversity within ocean ecosystems8. Any projected change in the future wave height spectrum could therefore be expected to have both desirable and undesirable impacts on coastal regions. The diagnostic method used here to produce projections of large wave occurrence is based on geopotential height in the upper troposphere. Previous studies have examined contour maps of geopotential height for eastern Australia, finding that a strong curvature of the contours provides a good indication of the likelihood of extratropical cyclone occurrence9, 10, 11. To examine whether or not this is also the case for large wave occurrence, Fig. 1 shows contour maps of geopotential height (at the 500 hPa pressure level obtained from the European Centre for Medium-Range Weather Forecasts ERA-Interim reanalyses12) for four different wave height ranges: 6 m or larger, 4–6 m, 2–4 m and 2 m or smaller. Wave height is calculated as the largest wave height observed on a given day at any one of five ocean buoys13 along the central east coast of Australia. The contour maps represent the average of all days when the wave height was within the specified range, calculated for the period of available wave observations (from 1992 to 2010).
| http://www.nature.com/nclimate/journal/v4/n4/full/nclimate2142.html
|
