Development and application of high-resolution regional climate models

区域气候模式（RCMs）目前的常规运行分辨率约为每格框50公里。该应用程序的主要推力是启动RCM物理参数化领域的工作，以开发下一代RCM，这些RCM将在显着更高的分辨率（~10km）下表现最佳。在这个尺度上，区域监测模式应该开始提供规划人员和影响评估人员在气候变化和较长期（季节性）预报领域所需的详细信息。大部分工作将致力于在高模式分辨率下更好地表示大气水循环。特别是，一个统一的和基于物理的方法来参数化湍流，对流，云的形成和降水将被追求。大部分工作将使用GEM（全球环境多尺度）模式作为基本建模工具，并将其应用于区域气候模式和云分辨分辨率。后一种分辨率（每格框~1-2km）允许明确地解析许多关键的对流和湍流环流。与非常详细的云方案一起，这使得云解析模型能够提供非常精确的关键云尺度过程模拟，尽管计算成本非常高。此模型版本将作为参考，参数化方法将据此开发和评估。特别是，湿过程参数化方案中的许多项不能观测到，因此必须从这些显式模式模拟中推导出来。与此同时，高分辨率RCM将应用于撒哈拉以南非洲地区，以模拟控制该气候敏感地区降雨时间、分布和数量的关键区域气候过程。其目的将是调查全球季节预测和撒哈拉以南非洲气候变化情景的高分辨率区域缩小的可能性。我们的进一步目标是将这些区域预测与该区域的影响评估和规划机构联系起来。

Regional Climate Models (RCMs) are presently routinely run at resolutions of around 50km per grid box.  The major thrust of this application is to initiate work in the area of RCM physical parameterisations to develop the next generation of RCMs that will perform optimally at significantly higher resolution (~10km). At this scale, RCMs should begin to provide the detailed information required by planners and impact assessors in the areas of climate change and longer-range (seasonal) forecasting. Much of the work will be devoted to developing a better representation of the atmospheric water cycle at high model resolutions. In particular, a unified and physically based approach to the parameterisation of turbulence, convection, cloud formation and precipitation will be pursued. Most of this work will use the GEM (Global Environmental Multiscale) model as the basic modelling tool and apply this in both Regional Climate mode and at Cloud Resolving resolutions. The latter resolutions (~1-2km per grid box) allow many of the key convective and turbulent circulations to be explicitly resolved. Along with a very detailed cloud scheme, this allows the Cloud Resolving Model to provide very accurate simulations of key cloud scale processes, albeit at a very high computational cost. This model version will act as a reference against which parameterisation approaches will be developed and evaluated. In particular, many terms in moist process parameterisation schemes cannot be observed and therefore must be derived from these explicit model simulations. Parallel to this effort, the high-resolution RCM will be applied over sub-Saharan Africa to simulate key regional climate processes controlling the timing, distribution and amount of rainfall over this climate sensitive region. The aim will be to investigate the potential for high-resolution regional downscaling of Global seasonal predictions and climate change scenarios for sub-Saharan Africa. We further aim to link these regional predictions to impact assessment and planning bodies in the region.