庐山2008和2009年两次冰冻天气对比分析

利用高空与地面观测资料,主要从形势场、温度层结、地理形势及地面气象要素等方面,对庐山2008年和2009年两次冰冻天气过程进行对比分析。结果表明,庐山2009年冰冻灾害与2008年初中国南方出现的大范围冰冻灾害有些不同特点,2009年冰冻灾害更具有明显的庐山局部性特征。认为庐山冰冻灾害的出现主要决定于925—850hPa层次的温度层结,当层结温度降为0℃以下并伴有较明显降水时,并不一定需要太长时间和明显逆温层的存在庐山就可以出现冰冻灾害。     

粒子群算法与支持向量机相结合的热带气旋强度预报试验

支持向量机(SVM)的惩罚参数及核参数的选择直接影响到模型效果,通过粒子群算法(PSO)解决支持向量机的参数选择问题,实现了参数选择的自动化。将该方法应用于热带气旋强度预报,利用气候持续性因子,挑选了1990年的100个左右样本进行预报检验,预报时效为12 h、24 h、36 h、48 h的强度平均绝对误差分别为3.00、4.35、4.93和6.68 m/s。另外,还与国外预报结果及采用最小二乘回归法的预报结果进行了效果的比较,SVM方法显示了更好的预报能力。     

玉米螟越冬死亡率的气象条件及其预报模型研究

根据1994-2004年黑山气象资料和玉米螟越冬期资料,采用通径分析方法,对玉米螟越冬期的气象条件与死亡率的关系进行分析,建立了4月和5月死亡率的预报模型。结果表明：玉米螟受3-5月的气象因子影响较大,死亡率明显高于越冬前的死亡率,建立的模型能够反映玉米螟越冬期气象条件与死亡率的响应关系,可在实际预报业务中作为参考依据。考虑越冬前死亡率的预报拟合率明显高于不考虑越冬前死亡率的模拟结果,说明在对玉米螟进行预报时,只考虑气象因子,不考虑虫源基数是不全面的。     

多特征差分核支持向量机模型的湿地变化检测

文章通过对不同特征空间、不同时相信息和不同类别引入多核函数组合和类别加权方式,构建了基于多特征差分核的加权支持向量机变化检测模型,能够获得平衡样本大小、综合多特征空间的变化检测结果.以上海崇明岛东端地区为例,利用4期ETM+/TM影像数据,基于构建的SVM变化检测模型,最终得到研究区湿地资源的空间变化格局,结果表明检测模型能够有效、准确地提取多种湿地变化类别的空间分布信息;并从景观水平上对检测结果进行格局分析,为湿地保护和管理提供科学参考.     

Assessment of global meteorological,hydrological and agricultural drought under future warming based on CMIP6

在全球变暖背景下,分析和预测干旱的变化趋势和传播规律对于区域生态环境安全和灾害管理具有重要意义.本文基于第六次国际耦合模式比较计划(CMIP6),分析了SSP2-4.5和SSP5-8.5两种变暖情景下的气象(标准化降水指数SPI和标准化降水蒸发指数SPED,水文(标准化径流指数SRI)和农业(标准化土壤水分指数SSI)...     

Automated Prediction of CMEs Using Machine Learning of CME – Flare Associations

Machine-learning algorithms are applied to explore the relation between significant flares and their associated CMEs. The NGDC flares catalogue and the SOHO/LASCO CME catalogue are processed to associate X and M-class flares with CMEs based on timing information. Automated systems are created to process and associate years of flare and CME data, which are later arranged in numerical-training vectors and fed to machine-learning algorithms to extract the embedded knowledge and provide learning rules that can be used for the automated prediction of CMEs. Properties representing the intensity, flare duration, and duration of decline and duration of growth are extracted from all the associated (A) and not-associated (NA) flares and converted to a numerical format that is suitable for machine-learning use. The machine-learning algorithms Cascade Correlation Neural Networks (CCNN) and Support Vector Machines (SVM) are used and compared in our work. The machine-learning systems predict, from the input of a flare’s properties, if the flare is likely to initiate a CME. Intensive experiments using Jack-knife techniques are carried out and the relationships between flare properties and CMEs are investigated using the results. The predictive performance of SVM and CCNN is analysed and recommendations for enhancing the performance are provided.     

Application of PSO-LIBSVM in Recognition of Potassium Salt Deposits

The shortage of potassium salt seriously restricts the development of China's agriculture. Increasing the exploration and development of potash will help improve the self-sufficiency of potassium in China. With rich potassium salt resources, Sichuan basin is one of the most important research areas for potash exploration and development in China. Polyhalite is an important solid potassium salt mineral in Sichuan basin, often intercalated in rock minerals such as anhydrite, rock salt and dolomite. Aiming at the problem that conventional log interpretation methods are difficult to accurately identify polyhalites, this paper proposed a new Support Vector Machine (SVM) recognition method based on Particle Swarm Optimization (PSO) to classify polyhalites in Sichuan basin. Based on particle swarm optimization and support vector machine theory, combined with logging interpretation theory, the effective data sensitive to polyhalite logging response were selected as input samples to generate training sets and test sets randomly. The Radial Basis Function (RBF) parameters were optimized by particle swarm optimization, and the classification and prediction model of polyhalite was established. Compared with mud logging results, the recognition accuracy of SVM model based on particle swarm optimization reached 97.5758%, which is obviously better than that of SVM model optimized by cross validation method in recognition accuracy and speed. The results show that the model has broad application prospects in potash exploration in Sichuan basin.     

基于遗传算法优化的支持向量机在岩性识别中的应用

为提高测井岩性识别的精度,本文结合乌夏地区岩芯资料和测井数据,总结该地区砂砾岩测井响应特征,优选出声波、自然伽马、密度、中子孔隙度和电阻率等5条测井曲线参数作为训练和测试样本,通过遗传算法挑选出最佳的支持向量机核函数参数σ和惩罚因子C,建立支持向量机岩性识别模型。结果表明该模型实际数据预测总体符合率为81.6%,在识别准确率上与传统测井识别砂砾岩岩性方法相比都有明显提升。     

耦合PCA-SVM算法的金矿矿床规模预测分析研究

判断矿床(点)的类型是矿床勘探中的重要内容,传统预测金矿成矿规模的方法不仅耗时耗力,而且所需的经济成本较大。为提高矿床规模的勘探效率和准确度,揭示元素与金矿成矿规模的潜在联系,文中提出了耦合主成分分析(principal component analysis,PCA)和支持向量机(support vector machine,SVM)算法的预测分析PCA-SVM(principal component analysis-support vector machine)方法。该方法先通过主成分分析提取数据中的主要特征,再将主要特征带入支持向量机算法,从而训练出最优分类器以预测金矿成矿规模。文中共使用了3 812个金矿样本数据用于学习训练和预测分析,训练准确率为92.3%,测试准确率为88.7%,分别比直接使用支持向量机算法高出14.3%和17.1%。基于PCA-SVM的预测模型,不仅消除了人为主观因素的影响,而且有效提高了勘探过程中矿床预测的准确率和矿床勘探的效率,为地质勘查工作提供依据。     

中国公众对气候变化和气象灾害认知的社会性别差异

对2015年全国公众气象服务评价调查结果的分析显示,我国公众中男性对气候变化认识程度以及对气象灾害预警的认知水平和满意度都比女性高。城镇女性选择环保产品、环保出行及愿意对生活或工作方式做出改变来应对气候变化的比例高于城镇男性,而农村男性选择调整或改变种植或养殖方式、改变种植或养殖品种及转换谋生方式的比例高于女性。男性选择购买相关气象或气候保险、参加培训或辅导的比例比女性高,而女性选择学习气象灾害和气候变化相关的专业知识比例高于男性。女性对气象服务信息的需求比例比男性高。建议：1)积极推进气候变化及其灾害认知的社会性别研究,建立灾害认知性别数据库,将社会性别融入到气象灾害风险管理的整个过程;2)在气象服务中引入社会性别视角,加强针对女性的宣传和指导,开发适合女性的服务产品和信息传播渠道,提升她们应对灾害的意识和能力。