集合海浪预报的发展与应用回顾

总结了国内外集合海浪预报的发展及应用现状,现有集合海浪预报的方法及优缺点。以NOAA/NCEP机构为例,给出了集合海浪主流预报产品的种类,通过集合预报产品的分析可以看出,集合海浪预报能够将传统的确定性预报扩展至概率预报领域,可给出更多可能出现的未来状态,能提供单纯确定性预报所不能提供的额外信息,已成为国际上业务化海洋学未来发展的重要方向之一。     

两次暴雨过程的主观及集合预报检验

对各级预报员、集合预报成员在两次暴雨天气中的24 h和48 h降水、温度预报结果分别进行检验。结果表明:降水预报中预报员分级降水TS评分随降水级别增大而减小,所有预报对小雨均有较好的预报。暴雨预报漏报明显,漏报率几乎为100%,然而集合成员中的最大值,对暴雨预报有一定的启示作用,业务工作中应得到重视。针对降水和温度预报总体而言,集合成员中集合平均值和控制预报表现相对优秀,可在业务中作为重要参考。     

基于20—30d振荡的长江下游地区夏季低频降水延伸期预报方法研究

用长江下游降水低频分量和环流低频主成分,构造多变量时滞回归模型(MLR)和主成分复数自回归模型(PC-CAR)的混合预报模型(MLR/PC-CAR),对长江下游降水低频分量进行延伸期逐日变化预报,延长预报时效。通过2011年6—8月预测试验表明,20—30 d时间尺度的长江下游低频降水预测时效可达50 d左右,采用南半球中高纬度地区850 hPa低频经向风的主成分作为预测因子的模型的预测精度明显高于东亚地区低频经向风作为预测因子的模型。这表明在20—30 d时间尺度上,长江下游降水与南半球中纬度绕球遥相关(SCGT)型有关的主分量的时滞相关更加密切。进一步对于较强20—30 d振荡的多年资料构建的MLR/PC-CAR混合模型预测试验表明,SCGT是预测夏季长江下游低频降水未来50 d变化的显著信号。基于SCGT的发展和演变,对于把握类似长江下游地区2011年6月初旱涝急转和7月中旬持续降水和强降水过程异常变化过程很有帮助,SCGT可以作为夏季长江下游20—30 d低频降水和强降水过程进行延伸期预报的主要可预报性来源之一。     

An Analysis of Historical and Future Temperature Fluctuations over China Based on CMIP5 Simulations简

The trends and fluctuations of observed and CMIP5-simulated yearly mean surface air temperature over China were analyzed.In general,the historical simulations replicate the observed increase of temperature,but the multi-model ensemble （MME） mean does not accurately reproduce the drastic interannual fluctuations.The correlation coefficient of the MME mean with the observations over all runs and all models was 0.77,which was larger than the largest value （0.65） from any single model ensemble.The results showed that winter temperatures are increasing at a higher rate than summer temperatures,and that winter temperatures exhibit stronger interannual variations.It was also found that the models underestimate the differences between winter and summer rates.The ensemble empirical mode decomposition technique was used to obtain six intrinsic mode functions （IMFs） for the modeled temperature and observations.The periods of the first two IMFs of the MME mean were 3.2 and 7.2,which represented the cycle of 2-7-yr oscillations.The periods of the third and fourth IMFs were 14.7 and 35.2,which reflected a multi-decadal oscillation of climate change.The corresponding periods of the first four IMFs were 2.69,7.24,16.15 and 52.5 in the observed data.The models overestimate the period of low frequency oscillation of temperature,but underestimate the period of high frequency variation.The warming rates from different representative concentration pathways （RCPs） were calculated,and the results showed that the temperature will increase by approximately 0.9℃,2.4℃,3.2℃ and 6.1℃ in the next century under the RCP2.6,RCP4.5,RCP6.0 and RCP8.5 scenarios,respectively.     

Bias-Corrected Short-Range Ensemble Forecasts for Near-Surface Variables during the Summer Season of 2010 in Northern China

A running mean bias （RMB） correction ap- proach was applied to the forecasts of near-surface variables in a seasonal short-range ensemble forecasting experiment with 57 consecutive cases during summer 2010 in the northern China region. To determine a proper training window length for calculating RMB, window lengths from 2 to 20 days were evaluated, and 16 days was taken as an optimal window length, since it receives most of the benefit from extending the window length. The raw and 16-day RMB corrected ensembles were then evaluated for their ensemble mean forecast skills. The results show that the raw ensemble has obvious bias in all near-surface variables. The RMB correction can remove the bias reasonably well, and generate an unbiased ensemble. The bias correction not only reduces the ensemble mean forecast error, but also results in a better spreaderror relationship. Moreover, two methods for computing calibrated probabilistic forecast （PF） were also evaluated through the 57 case dates： 1） using the relative frequency from the RMB-eorrected ensemble; 2） computing the forecasting probabilities based on a historical rank histogram. The first method outperforms the second one, as it can improve both the reliability and the resolution of the PFs, while the second method only has a small effect on the reliability, indicating the necessity and importance of removing the systematic errors from the ensemble.     

相似集合预报技术在粤西风场预报中的应用

利用ERA5再分析资料与常规站点观测资料,采用相似集合法(AnEn)对粤西地面风的日平均风速与日最大风速释用进行订正.结果表明:日平均风速与日最大风速订正后的准确性得到明显提高,且在地形复杂的山区也有较好的订正效果.日平均风速的逐日均方根误差总体小于日最大风速,且更为稳定.对比相似成员集合性能,日平均风速各成员也比日最...     

时间序列分析在电离层短期预报中的应用

根据IGS提供的2012年TEC数据,在得到TEC值残差序列的基础上,利用谱分析去掉周期项和Matlab工具箱去掉趋势项并采用平滑算法去噪后,对随机项进行时间序列分析。目前,电离层格网点的预报均建立在原始TEC上,阐述了采用差分后的TEC值进行预报的方法,以减少日变化周期项的影响。根据AR（p）模型预报的结果加上周期项和趋势项后,再加上前一天对应时段的TEC值与IGS发布的数据比较,结果表明,该方法利用短期IGS发布的TEC值进行电离层预报能取得较高的精度。     

太阳辐射短时临近预报R模型的构建及验证

提高太阳辐射短时临近预报（<6 h）的准确率是确保电网调度的重要举措,也是极具挑战性的技术瓶颈之一。基于云-辐射关系,利用地面观测的太阳辐照度反演的云相对辐射强迫比值,构建了太阳辐射短时临近预报模型（R模型）,并用美国南部大平原中心站16 a的辐照度观测数据,对R模型的预报性能进行了评估。结果表明：（1）有云存在的个例中,R模型较传统的简单持续性模型（Simple模型）的预报性能有很大提升,相比于预报性能较高的智能持续性模型（Smart模型或RCRF模型）仍有2%～25%的改进。（2）在16 a包含2.9×105个8类云状个例的总体检验中,当预报时效超过1 h时,R模型的预报性能显著优于Simple模型和RCRF模型。相对于RCRF模型,R模型在6 h预报时效下,对总辐射和直接辐射的预报性能可分别提高25%和19%,预报时效分别延长了1.5 h和1 h。（3）R模型为太阳辐射短时临近预报提供了准确率更高的基准模型。同时,该模型可仅依靠地面短期的辐照度观测资料即可预报,为缺少同期气象要素观测的光伏电厂的辐射预报提供了新的途径或新的可能。     

高分辨率遥感影像分类的多示例集成学习

精选示例特征嵌入多示例学习(MILES)算法在对噪声较强的训练样本进行学习时表现出良好的性能,但其判断规则可能带来遥感影像分类结果的不确定性。针对这一问题,提出用Bagging和AdaBoost集成MILES的多示例集成学习算法,使用粗包细分、多样性密度和最大似然分类相结合抑制分类不确定性的方法,实现了高分辨率遥感影像分类中多示例学习与集成学习的组合。采用Quick Bird、IKONOS等高分辨率遥感影像进行试验,结果表明多示例集成学习能有效控制遥感影像分类结果的不确定性,具有良好的应用前景。     

Data fusion and classifier ensemble techniques for vegetation mapping in the coastal Everglades

This study examined the applicability of data fusion and classifier ensemble techniques for vegetation mapping in the coastal Everglades. A framework was designed to combine these two techniques. In the framework, 20-m hyperspectral imagery collected from Airborne Visible/Infrared Imaging Spectrometer was first merged with 1-m Digital Orthophoto Quarter Quads using a proposed pixel/feature-level fusion strategy. The fused data set was then classified with an ensemble approach based on two contemporary machine learning algorithms: Random Forest and Support Vector Machine. The framework was applied to classify nine vegetation types in a portion of the coastal Everglades. An object-based vegetation map was produced with an overall accuracy of 90% and Kappa value of 0.86. Per-class classification accuracy varied from 61% for identifying buttonwood forest to 100% for identifying red mangrove scrub. The result shows that the framework is promising for automated vegetation mapping in the Everglades.