数值天气预报中分析同化基本方法的历史发展脉络和评述

数值天气预报中分析同化的基本方法先后经历了多项式函数拟合方法、逐步订正方法、最优插值方法、变分方法和集合卡尔曼滤波方法。本文首先根据相关的经典文献力求本色地介绍这些方法的基本思想和实施的具体要点;然后,着重于它们的上下承接关系,试图阐述同化的历史发展脉络,评述这些方法的显著特征和创新性,以期清晰地理解资料同化的循序渐进的内在发展逻辑。此外,从起源上阐明“主观分析”与“客观分析”、“初猜场”与“背景场”、“分析”与“同化”、以及“更新”、“新息”等基本概念,以期准确地理解和把握“大气资料同化”的由来和内涵。     

北京7·21暴雨暖区中尺度对流系统的数值模拟

许多业务模式对北京2012年7月21日特大暴雨的预报均是以锋面降水为主。在冷锋过境前,实际北京西南部强降水主要以暖区降水为主。本文利用30个成员的中尺度非静力数值模式（WRF）,通过3次集合卡尔曼滤波(EnKF)同化地面和探空资料,对这次暴雨过程进行了集合数值模拟。通过对比分析模拟降水较好与较差成员发现,好的成员能够模拟出河北中西部及北京西南地区触发出的暖区中尺度对流系统（MCS）以及相对稳定的系统配置,使得MCS在北京上空充分发展,从而较好地模拟了这次特大暴雨过程中的暖区降水;而较差的成员则没能模拟出暖区降水过程,降水以锋面降水过程为主,并且雨带位置偏南、出现时间滞后。集合成员间模拟结果出现的较大差异,和初始场中低值系统位置的较大差异有直接关系,因此通过EnKF提高模式成员初始场的准确率,从而准确模拟后期主要影响系统的移动和发展,是成功模拟暖区对流系统触发和维持的关键所在。     

混合集合预报法在华南暴雨短期预报中的试验

-WRF多模式集合3组试验,对比分析混合集合预报法与传统方法的降水预报效果。结果表明:ARPS模式集合改善了广东省南部局地强降水预报,该方法在中雨、大雨、暴雨量级改进效果显著。WRF模式集合对广东省北部强降水预报优于ARPS模式集合,但空报、漏报率较大,该方法有一定局限性。ARPS-WRF多模式集合在降水落区和量级预报上均优于传统方法。混合集合预报法利用低分辨率 (36 km) 集合预报和高分辨率 (12 km) 控制预报实现了高分辨率 (12 km) 集合预报,改善了降水预报效果,该方法可为业务高分辨率集合预报提供参考。     

基于TIGGE资料的集合成员优选方法

利用2007年6月8日—8月31日东亚地区TIGGE集合预报资料中欧洲中期天气预报中心(European Centre for Medium-range Weather Forecasts,ECMWF)和英国气象局(United Kingdom M et Office,UKM O)两个中心的地面2 m气温资料进行集合成员优选研究。结果表明,对于24~96 h预报,集合成员优选方法能够较好地选出预报技巧较高和预报技巧较低的集合成员。个例分析表明,在极端天气出现的地区,优选集合平均的预报优势较为明显。对比ECMWF和UKMO的集合成员优选结果发现,ECMWF的预报效果优于UKMO的预报效果。     

Comparative simulation study of effects of eddy-topography interaction in the East/Japan Sea deep circulation

In this study the structure and seasonal variations of deep mean circulation in the East/Japan Sea (EJS) were numerically simulated using a mid-resolution ocean general circulation model with two diffe...     

Siphonostomatoid copepods infecting Squalus acutipinnis Regan, 1908 off South Africa

Squalus acutipinnis (previously included in S. megalops) individuals caught off the east, south and west coasts of South Africa were examined for siphonostomatoid copepod infection. Collected siphonostomatoids were fixed and preserved in 70% ethanol and studied under microscopes using both transmitted and incidental light. The copepods were identified and the host-parasite associations estimated by calculating their prevalence, mean intensity and mean abundance. The siphonostomatoids represented four families and six species, and included new host records for Nemesis sp., Lernaeopoda sp. and Eudactylina acanthii, and a new geographical record for E. acanthii. Squalus acutipinnis individuals were infected by relatively low loads of siphonostomatoids. The most commonly found species was Achtheinus pinguis, the only species found off all three coasts. There was no difference in the prevalence of A. pinguis in sharks sampled off the three coasts but the highest mean intensity was found on the East Coast followed by the West Coast. The overall prevalence of A. pinguis was 33.6%, and was an order of magnitude higher than that of the other siphonostomatoids recorded. At such low levels of infection it will thus be necessary to examine a large number of hosts to determine the biodiversity of siphonostomatoids and geographical differences in prevalence in South African waters.     

面实体匹配的集成学习CatBoost方法

现有的面实体多指标几何匹配方法在计算综合相似度和确定最终匹配实体时面临着指标权重和阈值难以科学量化的难题,集成学习算法通过构建并结合多个机器学习器来完成学习任务,在解决分类问题时体现出了较为明显的性能优势。为此,本文提出了一种基于集成学习算法CatBoost的面实体匹配方法,将匹配问题转化为分类问题。选取形状、面积、方向和位置4个几何特征作为模型分类特征;利用过采样与欠采样相结合的混合重采样技术减轻原始训练样本的类别不平衡度;借助贝叶斯优化算法确定CatBoost模型的最优超参数;引入可解释人工智能领域的SHAP解释框架从全局和局部两个角度解释各输入特征对匹配结果的影响。在青藏高原的面状湖泊数据上对本文提出的方法进行了验证,实验结果表明：对模型预测影响最大的特征是位置,然后依次是面积、形状,影响最小的特征是方向。CatBoost匹配方法在实验数据集上的查准率、查全率和F1-score分别达到0.9937、0.9753和0.9844,相比于直接使用样本不均衡的原始样本进行模型训练,分别提高了约5.8%、0.6%和3.3%。与传统的面实体多指标双向匹配方法和逻辑回归、K近邻、决策树、神经网络等常规机器学习分类算法相比,集成学习算法CatBoost性能表现更加优异,在避免指标权重和阈值设置难题的同时取得了较好的匹配结果。     

中国近地表气温直减率及其季节和类型差异

近地表气温直减率是推测近地表气温空间分布的重要参数。中国幅员辽阔,气候和地形地貌条件复杂,直接使用反映对流层平均状况的单一气温直减率(0.65℃/100 m)很难表征中国近地表气温直减率的季节和类型差异。本文利用中国839个国家气象站点2000-2013年的近地表气温数据,分别在国家尺度和综合自然区划尺度上使用多元回归分析方法计算各个季节的平均气温直减率(lr_meanT)、平均最低气温直减率(lr_minT)和平均最高气温直减率(lr_maxT),并借助空间插值算法对气温直减率的可靠性进行了验证,最后分析了其季节和类型的差异。结果表明：①在国家尺度上,3种气温直减率均小于0.65℃/100 m;lr_minT、lr_meanT和lr_maxT的季节差异分别为0.05、0.13和0.24℃/100 m,且一般有夏季最大、冬季最小的季节规律;春、夏、秋、冬季气温直减率的类型差异分别为0.12、0.05、0.11和0.26℃/100m,且有lr_minT>lr_meanT>lr_maxT的规律。②在综合自然区划尺度上,气温直减率大多低于0.65℃/100 m,且存在明显的地域差异;夏季气温直减率一般大于冬季气温直减率,季节差异大多超过0.10℃/100 m;气温直减率类型差异半数区域超过或等于0.10℃/100 m,在春、夏、秋季,半数左右的区域lr_maxT >lr_minT,在冬季,多数区域的lr_minT >lr_maxT,lr_meanT一般处于lr_maxT和lr_minT之间。     

LOW FREQUENCY VARIABILITY OF INTERANNUAL CHANGE PATTERNS FOR GLOBAL MEAN TEMPERATURE DURING THE RECENT 100 YEARS

The TEEOF method that expands temporally is used to conduct a diagnostic study of the variation patterns of 1, 3, 6 and 10 years with regard to mean air temperature over the globe and Southern and Northern Hemispheres over the course of 100 years. The results show that the first mode of TEEOF takes up more than 50% in the total variance, with each of the first mode in the interannual oscillations generally standing for annually varying patterns which are related with climate and reflecting long-term tendency of change in air temperature. It is particularly true for the first mode on the 10-year scale, which shows an obvious ascending trend concerning the temperature in winter and consistently the primary component of time goes in a way that is very close to the sequence of actual temperature. Apart from the first mode of all time sections of TEEOF for the globe and the two hemispheres and the second mode of the 1-year TEEOF, interannual variation described by other characteristic vectors are showing various patterns, with corresponding primary components having relation with long-term variability of specific interannual quasi-periodic oscillation structures. A 2 T test applied to the annual variation pattern shows that the abrupt changes for the Southern Hemisphere and the globe come closer to the result of a uni-element t test for mean temperature than those for the Northern Hemisphere do. It indicates that the T 2 test, when carried out with patterns of multiple variables, seems more reasonable than the t test with single elements.     

B08RDP区域集合预报温度场质量评估与综合偏差订正

针对B08RDP(The Beijing 2008 Olympics Research and Development Project)5套区域集合预报资料,系统分析了各套集合预报温度场的预报质量。在此基础上运用集合预报的综合偏差订正方法对温度场进行偏差订正,并对其效果进行了分析讨论。结果显示:5套B08RDP区域集合预报中,美国国家环境预报中心(NCEP)区域集合预报温度场的整体预报质量最高,平均预报误差最小,离散度也最为合理,预报可信度和可辨识度均较优;而中国气象科学研究院(CAMS)的温度预报误差过大,预报质量最差。整体上看,除NCEP之外的4套集合预报的温度场均存在集合离散度偏小的问题;综合偏差订正能有效减小各集合预报温度场的集合平均均方根误差,改善集合离散度的质量,显示出综合偏差订正方案对集合预报温度场偏差订正的良好能力。