@@ 1. Introduction The TIGGE (Thorpex Interactive Grand Global Ensemble) project was designed to encourage research to determine the advantages of merging ensemble forecasts from different centers. Most verification studies using the TIGGE database have so far focused on the evaluation of upper air variables with respect to gridded analyses. For example Park et al (2007) looked at the performance of the models in terms of geopotential height at 500hPa and temperature at 850hPa. Johnson and Swinbank (2009) discussed the advantages of model combinations using geopotential height at 500hPa and mean sea level pressure. They also considered 2m Temperature in their evaluation, but still with respect to analysis data. Yi He et al (2008) studied the benefits of multimodel ensembles in a hydrological context and more recently Froude (2010) compared the performance of the ensemble systems when used for tropical cyclone forecasts. In general, these studies show that improvements in probabilistic forecasts can be obtained from combined ensembles compared to single model ensembles, especially if the models used in the combination are of relatively high quality, but the magnitude of the improvement has not been found to be large.
展开▼
机译：@@ 1.简介TIGGE（索普全球互动大合奏）项目旨在鼓励研究，以确定合并来自不同中心的集合预报的优势。迄今为止，大多数使用TIGGE数据库进行的验证研究都集中在相对于网格分析的高空变量评估上。例如，Park等人（2007年）以500hPa的地势高度和850hPa的温度来考察模型的性能。 Johnson和Swinbank（2009）讨论了使用500hPa的地势高度和平均海平面压力进行模型组合的优势。他们在评估中还考虑了2m Temperature，但仍相对于分析数据。 Yi He等人（2008年）研究了水文环境中多模式合奏的好处，最近，Froude（2010年）比较了用于热带气旋预报的整体系统的性能。通常，这些研究表明，与单个模型集成相比，组合集成可以提高概率预测，特别是如果组合中使用的模型具有较高的质量，但是改进的幅度并未发现很大。
展开▼
