Stochastic modeling of hydroclimatic variability

气候变化将对加拿大社会产生重大影响。全球气候模式（GCM）是用于预测温度和降水等变量变化的主要工具。遗憾的是，目前的大气环流模型的分辨率非常粗糙，使得直接的模型输出不适用于许多地方或区域影响研究。各种降尺度技术已被提出来增加区域气候变化情景的细节水平。本研究计划的总体目标是为气候变化研究的统计降尺度技术的发展做出贡献。研究的重点是随机天气生成技术的发展。目前的天气生成器有一定的局限性，包括它们不能保持一个区域内多个地点之间的空间相关性，也不能再现年际变化。此外，也有一些问题，天气发电机应如何调节GCM模拟的气候变化。该研究计划将解决目前的随机天气发电机的缺点，并建议开发一个计算机模型，集成了一系列的天气生成技术，用于气候变化研究。这将对加拿大具有理论和实际价值，因为气候变化预计将对农业和水电等关键部门产生重大影响。

Climate change will have significant impact on the Canadian society. Global climate models (GCMs) are the main tool used to predict changes in variables such as temperature and precipitation. Current GCMs unfortunately have very coarse resolution, making direct model output unsuitable for many local or regional impact studies. Various downscaling techniques have been proposed to increase the level of detail of regional climate change scenarios.The general objective of this research program is to contribute to the development of statistical downscaling techniques for climate change studies. The particular focus of the research is the development of techniques for stochastic weather generation. Current weather generators have certain limitations, including their inability to preserve the spatial correlation between multiple sites in a region and their inability to reproduce interannual variability. In addition, there are also some questions as to how weather generators should be conditioned on GCM simulated climate change.The research program will address the shortcomings of current stochastic weather generators and proposes the development of a computer model that integrates a range of weather generation techniques for use in climate change studies. This will be of both theoretical and practical value to Canada where climate change is expected to have significant impact on key sectors such as agriculture and hydropower.