The issues of present climate and future climate change are widely regarded as hot spots in climate impact study. It is widely recognized that the changes in the frequency and intensity of extreme events are likely to have more significantly negative impacts than changes in mean climate on many vulnerable aspects of human health, social organization and natural systems. Meanwhile, the need for regional-scale projections of climate variables and thresholds that are directly relevant to impact researchers and stakeholders has been strongly advised.

Under the support of the IPCC AR4, over 20 modeling groups around the world conducted climate change simulation by different coupled general circulation models (CGCMs). Ten of these models provide estimation for extreme indices/indicators in the present and future climates. This offers researchers opportunities to conduct the multi-model ensemble analysis of the simulation and projection of extreme events. Bayesian model averaging (BMA), which provides away to combine different models, is a rather promising method for calibrating ensemble modeling and forecasts in climate impact research.

Therefore, the CGCM multi-model ensemble projections with BMA approach was used in order to offer the most comprehensive analysis of global temperature and precipitation extremes in the Tibetan Plateau. The extremes of temperature and precipitation are described by seven indices, namely, the frost day (FD), percentage of nights when T_min>90th percentile (TN90), consecutive dry days (CDD), annual count of days when precipitation ≥10 mm (R10), maximum 5-day precipitation total (R5D), simple daily intensity index (SDII), and annual total precipitation when precipitation >95th percentile (R95T). Results indicate that frost days decrease over the Tibetan Plateau in the 21st century. More frequent warm nights are also projected in the plateau. The increases of these temperature extremes under A2 and A1B scenarios are more pronounced than under B1. Heavy precipitation events for single days and pentads are projected to increase in their intensity over most parts of Tibetan Plateau. CDD, R10, R5D, R95T and SDII collectively suggest more extreme precipitation in the region (2011–2020). In addition, impacts of climate extremes changes on local water resources and fragile ecosystem are discussed as an extension of this article. The findings will be beneficial to project regional responses in this unique region to global climate change, and then to formulate regional strategies against the potential menaces of climate change.

The main finding has been published on Communications in Global and Planetary Change, 2012, 80-81: 1-13. The paper is also archived at http://www.sciencedirect.com/science/article/pii/S0921818111001408.

Spatial pattern of climate extremes in 1951–1999: (a) FD of observation (HadEX); (b) FD of multi-model ensemble.