几种格点化温度滚动订正预报方案对比研究

为了快速获得更为精准的格点温度预报产品,使用国家信息中心高分辨率、高频次的温度格点多元融合产品和欧洲中期天气预报中心全球模式2 m温度预报场资料,采用8种误差订正方案进行滚动订正预报试验。选择2017年1月1日至2月28日和6月1日至7月31日两个时间段进行两次回报模拟试验,并对订正前后的预报结果进行格点和站点检验分析,结果表明:8种方案对模式直接输出的预报场有正技巧订作用,全格点滑动误差回归模型订正和全格点滑动双因子回归模型订正效果最优,两种方案都能使订正场的格点平均绝对误差在2℃以下,3、6和9 h的格点准确率均在0.9以上。全格点滑动误差回归模型的检验评分略微好于全格点滑动双因子回归模型,表明作为预报模型因子的起报时刻误差场比数值模式因子在短期订正中扮演着更为重要的角色。     

长江上游流域数值预报产品评价分析

使用宜昌市气象局研制的数值预报产品评价分析和客观订正系统,对ECMWF、T213L31模式分别进行预报误差分析和降水主要影响系统预报能力分析。结果表明:ECMWF模式总体误差水平好于T213L31模式,T213L31模式24 h的总体误差水平接近ECMWF 72 h预报;ECMWF、T213L31对影响长江上游流域暴雨的主要天气系统有不同的预报能力,使用宜昌市气象局所研制的订正系统可从一定程度上改善模式预报的输出结果。     

Another Simple Method of Spectral Correction to Obtain Robust Eddy-Covariance Results

The eddy-covariance method is the state-of-the-art tool to measure mass and energy fluxes, though many measuring systems (particularly closed-path systems) show strong spectral attenuation that causes significant underestimates of actual mass fluxes. The standard way to correct these underestimates is to use theoretically derived transfer functions. Practical implementation has shown that the attenuation, especially of closed-path systems, is underestimated by this method, even when the additional tube damping is considered. This paper introduces and tests three alternative site-specific and hardware specific correction methods based on spectral analysis, which typically enhance—additionally to the classical correction—the calculated mass fluxes based on closed-path analysers, as their inevitable tubing and the filters used cause additional flow attenuation. Two of the three methods are based on a site-specific and hardware specific transfer functions, the third uses direct comparison of cospectra. Primarily the methods based on transfer functions proved to be easy-to-handle once established for the specific set-up and measurement site. They represent practical and robust methods to correct for spectral attenuation.     

微波湿度计资料在GRAPES模式中偏差订正方法研究

根据微波湿度计MHS(Microwave Humidity Sounder)辐射率资料及GRAPES(Global/Regional Assimilation and Pr Ediction System)模式的特点,建立适用于MHS资料的偏差订正系统,该系统包括扫描和气团偏差订正,其中气团偏差订正考虑水汽资料的特性,采用三种不同预报因子组合的方案。偏差订正结果表明:MHS各个通道的扫描偏差表现出不同特征;偏差订正后观测残差基本服从均值为零的高斯分布,且观测残差的均值有所降低并随时间变化平稳;三种气团偏差订正方案都有明显的订正效果,其中方案三的订正效果最佳。     

Grapes模式预报西南地区夏季2m温度的检验评估

利用西南地区2004、2005年夏季的实况温度对G rapes模式输出的2m温度预报产品进行检验,结果表明,不管是单日预报误差还是月平均预报误差,都没有随预报时效的延长而增大,对四川盆地大部分地区,模式预报温度偏高,对重庆、云南、贵州及川西高原南部的温度预报,基本上表现为14时偏高,02时偏低,对川西高原北部和西藏则表现为预报温度偏低。分析表明,模式中对高原地形处理的不真实是温度预报产生误差的重要原因之一,利用回归分析方法能对模式温度预报进行有效的订正。检验分析结果为进一步改进模式,提高要素预报产品质量提供了一定的依据。     

一个ENSO动力-相似误差订正模式及其后报初检验

为有效利用历史资料中的相似信息,减小模式误差对ENSO这类跨季节-年际尺度预测问题的影响提高动力模式的预测水平.作者利用一种基于统计相似的模式误差订正方法,以国家气候中心简化海气耦合模式为平台建立了相应的动力-相似误差订正(DAEC)模式,并着重探讨了系统相似程度(全相似或部分相似)、误差重估周期以及相似样本个数等因素对预报效果的影响.结果表明,利用该方法可以有效地改善原有模式的预报性能,其中 "全相似" 比 "部分相似" 更能反映海气耦合系统的相似程度,从而对模式误差做出更为准确的估计,使预报误差明显减小.海洋和大气的误差重估周期对结果也有较大影响,在不同相似程度下分别存在着某种最优配置使得预报效果达到最佳.另外,在对相似样本存在状况及影响的研究中则发现在当前资料长度内整体上只存在着有限个相似样本,在此范围内随着样本取样数目的增加DAEC模式的预报性能逐渐提高.     

天气雷达回波衰减订正算法的研究(Ⅱ):数值模拟与个例实验

通过数值模拟研究雷达反射率因子呈均匀分布情况下的“可订正厚度”，计算了“无误差”情况下的订正效果以及“观测误差”、“k的误差”、库分辨等因素对订正效果的影响，还用实际观测个例进行衰减订正实验，并与解析算法和迭代算法的订正结果进行比较。结果表明，逐库算法，尤其是逐库近似算法，比迭代法计算效率高，比解析法有更大的稳定范围。     

几种流域面雨量计算方法的比较

简单介绍了计算流域面雨量的泰森多边形法和逐步订正格点法，并运用此两种方法及气象站资料的算术平均法与气象站和水文站资料的算术平均法计算了1998年(5～9月)、1999年(5～9月)及2000年(5～8月)湖南5个小流域的实况面雨量。从整体上看，几种方法的计算结果均无显著差异，但不同方法所得面雨量的逐日差异有的还是相当明显的。     

An improved method for correction of air temperature measured using different radiation shields

The variation of air temperature measurement errors using two different radiation shields （DTR502B Vaisala,Finland,and HYTFZ01,Huayun Tongda Satcom,China） was studied.Datasets were collected in the field at the Daxing weather station in Beijing from June 2011 to May 2012.Most air temperature values obtained with these two commonly used radiation shields were lower than the reference records obtained with the new Fiber Reinforced Polymers （FRP） Stevenson screen.In most cases,the air temperature errors when using the two devices were smaller on overcast and rainy days than on sunny days; and smaller when using the imported rather than the Chinese shield.The measured errors changed sharply at sunrise and sunset,and reached maxima at noon.Their diurnal variation characteristics were,naturally,related to changes in solar radiation.The relationships between the record errors,global radiation,and wind speed were nonlinear.An improved correction method was proposed based on the approach described by Nakamura and Mahrt （2005） （NM05）,in which the impact of the solar zenith angle （SZA） on the temperature error is considered and extreme errors due to changes in SZA can be corrected effectively.Measurement errors were reduced significantly after correction by either method for both shields.The error reduction rate using the improved correction method for the Chinese and imported shields were 3.3％ and 40.4％ higher than those using the NM05 method,respectively.