CONVECTIVE NOWCASTING USING IMPROVED OPTICAL FLOW METHOD AND SATELLITE TBB DATA
-
摘要: 为了在缺少雷达观测的地区开展对流临近预报，利用光流法和半拉格朗日外推法进行了卫星云图外推实验，同时利用无辐散约束改进光流矢量来避免云图辐散失真。(1) 光流法反演的风场能够准确反映出台风涡旋环流结构，采用半拉格朗日方案进行外推，可以保持云系的旋转特性，具有良好的稳定性和精度，但随着外推时间的增加光流矢量中的噪声会导致云图辐散失真。(2) 无辐散约束减少了风矢量杂乱无序现象，弥补了缺失的光流，还能对风速进行平滑，使风速空间梯度更平滑。(3) 用改进后的风场进行外推，避免了云系辐散失真，在保持其形态不被破坏的同时，还能减少云团TBB虚假增长。(4) 检验表明改进后的外推预报方案，具有更小的平均绝对误差，MAE提高了4%，临界成功指数提升了9%。Abstract: To carry out convective nowcasting in areas lacking radar observations, the present study uses the methods of optical flow and semi-Lagrangian extrapolation to conduct satellite cloud image extrapolation experiments. Moreover, the optical flow vector is improved by using non-divergence constraint to avoid divergence and distortion of the cloud image. Case studies show that: (1) The wind field retrieved by using the optical flow method can accurately reflect the typhoon vortex circulation structure. The semi-Lagrangian scheme is employed for extrapolation as it can maintain the rotation characteristics of the cloud system, and has good stability and accuracy; however, as the extrapolation time increases, the noise in the optical flow vector will lead to divergence and distortion of the cloud image. (2) The non-divergence constraint reduces the disorder of the wind vector, makes up for the missing optical flow, and can also smooth the wind speed to make the wind speed spatial gradient smoother. (3) The improved wind field is used for extrapolation to avoid cloud divergence and distortion. The cloud shape will not be damaged, and the false growth of cloud cluster TBB will be reduced. (4) The test shows that the improved extrapolation forecasting scheme has a smaller average absolute error, with MAE increased by 4%, and critical success index increased by 9%.
-
Key words:
- optical flow method /
- semi-Lagrangian extrapolation /
- COTREC /
- FY-4A /
- TBB
-
图 2 采用半拉格朗日方法对台风“山竹”的6 h外推预报和对应的观测
a 1~a4为9月15日19时、20时、21时、16日00时卫星TBB观测；b1~b4为与观测时间对应的采用原始风场外推预报结果；c1~c4为采用无辐散约束处理后的风场外推预报结果，时间与观测一致。
图 3 预报评分
a. MAE；b.-70 ℃阈值预报技巧评分；c.-60 ℃阈值预报技巧评分；d.-50 ℃阈值预报技巧评分。黑色线条为原始风场外推预报，红色线条为无辐散约束改进后的风场外推预报。
表 1 评分统计表
预报技巧评分 MAE POD FAR CSI POS 13.5 0.73 0.73 0.25 NPOS 14.05 0.78 0.76 0.23 Rate -0.04 -0.07 -0.04 0.09 
下载: 导出CSV
-
[1] WILK K J, GRAY K G. Processing and analysis techniques used with the NSSL weather radar system[C]. Preprints 14th Conf On Radar Meteor, 1970: 367-374. [2] 陈明轩, 王迎春, 俞小鼎. 交叉相关外推算法的改进及其在对流临近预报中的应用[J]. 应用气象学报, 2007, 18(5): 690-701. doi: 10.3969/j.issn.1001-7313.2007.05.014 [3] TUTTLE J D, FOOTE G B. Determination of the Boundary Layer Airflow from a Single Doppler Radar[J]. J Atmos Oceanic Technol, 1990, 7(2): 218-232. doi: 10.1175/1520-0426(1990)007<0218:DOTBLA>2.0.CO;2 [4] RINEHART R E. A pattern recognition technique for use with conventional weather radar to determine internal storm motions[J]. Atmos Technol, 1981, 13(1): 119-134. [5] LI L, SCHMID W, JOSS J. Nowcasting of motion and growth of precipitation with radar over a complex orography[J]. J Appl Meteor, 1995, 34(6): 1 286-1 300. doi: 10.1175/1520-0450(1995)034<1286:NOMAGO>2.0.CO;2 [6] 陈雷, 戴建华, 陶岚. 一种改进后的交叉相关法(COTREC)在降水临近预报中的应用[J]. 热带气象学报, 2009, 25(1): 117-122. https://rdqxxb.itmm.org.cn/article/id/200901015 [7] SHERMAN C A. A mass-consistent model for wind fields over complex terrain[J]. J Appl Meteor, 1978, 17(4): 312-319. [8] 黄旋旋, 朱科锋, 赵坤. 改进后TREC外推方法在台风临近降雨预报中的应用[J]. 气象科学, 2017, 37(5): 610-618. https://www.cnki.com.cn/Article/CJFDTOTAL-QXKX201705005.htm [9] 邹海波, 章毅之, 解震华, 等. COTREC方法在江西地区的应用和检验[J]. 气象与减灾研究, 2019, 42(2): 105-112. https://www.cnki.com.cn/Article/CJFDTOTAL-HXQO201902004.htm [10] 张亚萍, 程明虎, 夏文梅, 等. 天气雷达回波运动场估测及在降水临近预报中的应用[J]. 气象学报, 2006, 64(5): 632-634. https://www.cnki.com.cn/Article/CJFDTOTAL-QXXB200605009.htm [11] 徐亚钦, 翟国庆, 黄旋旋, 等. 基于TREC法以多重动态区域反演风场[J]. 浙江大学学报(工学版), 2011, 45(10): 1 738-1 745. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC201110010.htm [12] 韩雷, 王洪庆, 林隐静. 光流法在强对流天气临近预报中的应用[J]. 北京大学学报: 自然科学版, 2008, 44(5): 751-755. https://www.cnki.com.cn/Article/CJFDTOTAL-BJDZ200805019.htm [13] 张蕾, 魏鸣, 李南, 等. 改进的光流法在回波外推预报中的应用[J]. 科学技术与工程, 2014, 14(32): 1 671-1 815. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS201432029.htm [14] SHI X J, CHEN Z R, WANG H, et al. Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting[C]. NIPS'15: Proceedings of the 28th International Conference on Neural Information Processing Systems, 2015, 1: 802-810. [15] 张艳霞, 蒙伟光, 张诚忠, 等. LAPS云分析中卫星资料的应用及对GRAPES模式短时预报的影响[J]. 热带气象学报, 2015, 31(5): 599-607. https://rdqxxb.itmm.org.cn/article/id/20150503 [16] 王洪, 王东海, 万齐林. FY2E辐射资料的直接同化试验研究[J]. 热带气象学报, 2016, 32(3): 334-345. https://rdqxxb.itmm.org.cn/article/id/20160305 [17] 董海萍, 袁炳, 罗雨. 多源资料循环同化在台风"潭美"预报中的应用[J]. 热带气象学报, 2017, 33(4): 433-441. doi: 10.16032/j.issn.1004-4965.2017.04.001 [18] 隋新秀, 王振会, 鲍艳松, 等. FY-2E晴空风矢同化对台风分析和预报的影响研究[J]. 热带气象学报, 2018, 34(6): 819-831. doi: 10.16032/j.issn.1004-4965.2018.06.010 [19] 刘甫, 明杰, 张翰, 等. 热带气旋"凤凰"(2014)的结构演变及其引起的海洋响应分析[J]. 热带气象学报, 2020, 36(4): 552-561. doi: 10.16032/j.issn.1004-4965.2020.051 [20] 张琪, 任景轩, 肖红茹, 等. 基于FY-4A卫星资料的四川盆地MCC初生和成熟阶段特征[J]. 大气科学, 2021, 45(4): 863-873. https://www.cnki.com.cn/Article/CJFDTOTAL-DQXK202104011.htm [21] 韩芙蓉, 鹿翔, 冯晓钰, 等. 台风Lekima(1909)登陆前后动热力结构变化对浙江极端降水的影响[J]. 热带气象学报, 2021, 37(1): 34-48. doi: 10.16032/j.issn.1004-4965.2021.004 [22] VADIM L, VLADIMIR I, IRINA R, et al. Precipitation nowcasting with satellite imagery[C]. In The 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '19), August 4-8, 2019, Anchorage, AK, USA. ACM, New York, NY, USA, 9 pages. [23] KUMAR A, ISLAM T, SEKIMOTO Y, et al. Convcast: An embedded convolutional LSTM based architecture for precipitation nowcasting using satellite data[J]. PLOS ONE, 2020, 15(3): e0230114. [24] OTSUKA, SHIGENORI, KOTSUKI, et al. Nowcasting with data assimilation: A case of global satellite mapping of precipitation[J]. Wea Forecasting, 2016, 31(5): 1409-1416. [25] LIU Y, XI D G, LI Z L, et al. A new methodology for pixel-quantitative precipitation nowcasting using a pyramid Lucas Kanade optical flow approach[J]. J Hydrology, 2015, 529354-364. [26] 纪丞. 基于FY-4A卫星的对流云监测预报方法研究[D]. 南京: 南京信息工程大学, 2020. [27] 江吉喜, 范梅珠. 青藏高原夏季TBB场与水汽分布关系的初步研究[J]. 高原气象, 2002, 21(1): 20-24. https://www.cnki.com.cn/Article/CJFDTOTAL-GYQX200201003.htm [28] 曹钰, 岳彩军, 寿绍文. 热带气旋(TC)环流内对流核数、TBB特征与TC强度关系统的计合成分析[J]. 热带气象学报, 2013, 29(3): 381-392. https://rdqxxb.itmm.org.cn/article/id/20130304 [29] FARNEBACK G. Polynomial expansion for orientation and motion estimation[D]. PhD thesis, inkoping University, Sweden, 2002. [30] FARNEBACK G. Very high accuracy velocity estimation using orientation tensors, parametric motion, and simultaneous segmentation of the motion field[C]. In: Proceedings of the Eighth IEEE International Conference on Computer Vision. Volume I., Vancouver, Canada (2001)171-177. [31] ADELSON E H, ANDERSON C H, BERGEN J R, et al. Pyramid methods in image processing[J]. RCA Engineer, 1984, 29(6): 33-41. [32] 王改利, 赵翠光, 刘黎平, 等. 雷达回波外推预报的误差分析[J]. 高原气象, 2013, 32(3): 874-883. https://www.cnki.com.cn/Article/CJFDTOTAL-GYQX201303026.htm -
点击查看大图
图(3) / 表(1)
计量
- 文章访问数: 12
- HTML全文浏览量: 2
- PDF下载量: 1
- 被引次数: 0
粤公网安备 4401069904700003号