1 / 75

水文气象推动工程水文计算核心课题研究的发展

水文气象推动工程水文计算核心课题研究的发展. Prof. Bingzhang Lin (林炳章教授) lbz@nuist.edu.cn 应用水文气象研究院 (AHMRI) 、水文气象学院 (COHM) 南京信息工程大学 (NUIST) 2012 年全国水文学术研讨会 南京 , 中国 10-11 December, 2012. 工程水文计算核心研究课题 是什么?. 美国某雨量站 100 年日雨量 AMS (011901) 28-4229 ANMAX 100

mariah
Download Presentation

水文气象推动工程水文计算核心课题研究的发展

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. 水文气象推动工程水文计算核心课题研究的发展水文气象推动工程水文计算核心课题研究的发展 Prof. Bingzhang Lin (林炳章教授) lbz@nuist.edu.cn 应用水文气象研究院(AHMRI)、水文气象学院(COHM) 南京信息工程大学 (NUIST) 2012年全国水文学术研讨会 南京, 中国 10-11 December, 2012

  2. 工程水文计算核心研究课题 是什么?

  3. 美国某雨量站100年日雨量AMS • (011901) 28-4229 ANMAX • 100 • 3.46 2.60 5.65 4.90 3.02 7.15 2.97 2.86 2.70 3.18 • 3.39 2.13 2.34 2.61 4.39 2.04 3.06 3.92 3.00 3.67 • 1.97 1.45 2.78 2.73 3.98 1.98 2.17 2.00 1.80 2.08 • 1.72 2.61 3.10 3.73 2.82 2.17 2.10 6.78 6.05 4.55 • 1.92 3.20 3.50 3.45 2.78 3.33 1.42 2.62 3.23 4.20 • 2.42 3.23 2.22 4.25 3.21 4.02 1.37 4.16 1.81 5.05 • 2.44 3.84 2.80 2.29 3.35 3.18 5.06 2.41 3.58 2.10 • 4.29 3.80 3.78 6.63 2.77 3.42 2.25 2.71 2.84 2.35 • 2.83 1.90 2.63 2.46 4.68 4.10 2.51 1.95 2.59 4.71 • 4.94 3.88 2.81 2.51 2.85 3.72 2.91 2.26 4.88 2.63

  4. 美国某雨量站100年日雨量AMS (Sorted + Grouped) (011901) 28-4229 ANMAX 100 1.55 1.61 1.64 1.952.04 2.05 2.15 2.18 2.21 2.23 2.25 2.27 2.31 2.36 2.38 2.38 2.42 2.46 2.46 2.52 2.55 2.56 2.60 2.65 2.66 2.73 2.74 2.77 2.79 2.85 2.85 2.94 2.95 2.96 2.96 2.97 2.98 2.983.06 3.07 3.10 3.14 3.15 3.15 3.18 3.19 3.20 3.21 3.22 3.23 3.24 3.30 3.37 3.40 3.42 3.473.52 3.61 3.61 3.63 3.64 3.66 3.66 3.78 3.80 3.84 3.88 3.91 3.92 3.97 4.06 4.16 4.22 4.23 4.29 4.31 4.35 4.40 4.454.51 4.56 4.65 4.72 4.76 4.82 4.86 4.985.16 5.31 5.34 5.53 5.56 5.60 5.73 5.746.416.867.52 7.698.11

  5. 经验频率曲线 Histogram Increment = 0.5 in.

  6. f (x) 概率密度函数曲线 a x b

  7. 什么是频率计算? Frequency Analysis -- is a statistical approach using sample or samples to estimate the population probability distribution. 频率计算是一种利用一个或数个样本来推求总体概率分布的统计方法。

  8. 水文频率分析有两大难题 Theoretical true value of frequencies is unknown forever. (100-year ?) -- 真值永远不知道! 2. There is no analytical way to derive a theoretical distribution to best fit the data. (GLO or GEV or PE3 ?) -- 无法从理论上推导出分布函数!

  9. 就像打靶时不知靶心在何处! The true value of frequencies such as 100-year is unknown forever. Something? like bulls eye unknown while shooting. ﹡ ﹡

  10. 全球气候变化加剧使得问题更加复杂

  11. (1900 – 2000年) Ohio流域1797站仅有16%检测出有线性趋势 Map of stations that exhibit linear trend in mean for AMP over 20th century

  12. (1958 年前后半个世纪比较) Ohio流域531站仅有18%检测出跳跃性 Map of stations that exhibit shift in mean for AMP over the 20th century

  13. 中国2436站仅有17.5%检测出有线性趋势 (1960 – 2012年)

  14. 中国1649站仅有10.8%检测出跳跃性 (1985年前后30年比较)

  15. Findings 总的来说,中美两国地区(流域)年最大日雨量均值在过去的近一百年中没有明显的趋势性和跳跃性; 但是,在受检测的美国Ohio 流域却有将近50%站点的年最大日雨量呈现出明显的离差增大,其SD平均增大了23%。--- 这意味着什么呢?

  16. 这意味着: We may observe more and more extreme hydrometeorological events (droughts or floods) in the Ohio River Basin area in the near future than before. So the world. --(将有越来越多的干旱和洪涝灾害,尽管其均值保持不变)

  17. Faced with the acceleration of climate change, What is our job ?

  18. 我们的使命之一 Exploration of a Robust (稳健) and Reliable (可靠) Approach to Precipitation Frequency Analysis of Extreme Events。

  19. 我们该做(能做)什么? L-Moments Method – focusing on the issue of precision in terms of parameter estimation 线性矩解决精度 Regional Analysis – focusing on the issue of accuracy in terms of uncertainties of quantiles 地区分析解决准确度

  20. Methods of Parameter Estimation 1. Conventional Moments Method (CMM) 常规矩 2. L-Moments Method (LMM) 线性矩

  21. L-Moments Definition: L-moments are expectations of certain linear combinations of order statistics (Hosking, 1989)

  22. Application of RLM to PrecipitationFrequency Analysis in the U.S.(1) Fig. 6 Sketch of regionalization of the U.S.

  23. 比较常规矩和线性矩的偏态性能 常规矩Cs偏态很明显 线性矩L-CS几乎是不偏

  24. 比较常规矩和线性矩的稳健性 CMM plays poorly to outlier (10.37”/day on 7/22/1947 at #2682, PA) Cs cannot model this outlier even for N = 500 yr. 常规矩模拟特大值很差线性矩模拟特大值性能很好

  25. Regional L-moments Analysis in China(A pilot study – Application to Lake Tai Basin)

  26. Location of 96 raingauges in Taihu Lake

  27. 太湖流域及其周边141雨量站点被划分为8个水文气象一致区太湖流域及其周边141雨量站点被划分为8个水文气象一致区

  28. Taihu 96 daily max, fit GLO Taihu 96 daily max, fit GLO Fig.1 Biasedness on skewness of CMM Fig.2 Unbiasedness on skewness of LMM Comparison for Taihu Lake Data (1) o on Biasedness

  29. Taihu 96 daily max, fit GLO Taihu 96 daily max, fit GLO Fig.4 Robustness to outlier by LMM Fig.3 Difficult to model the outlier by CMM Comparison for Taihu Lake Data(2)oon Robustness

  30. 地区分析法 – 构造水文气象一致区 Assume:A rainfall could be decomposed into the regional component reflecting the common properties in the region and the local component reflecting the individual local properties.

  31. 水文气象一致区 • 每个雨量都可以分解成共性分量和个性分量; • 一致区内所有站点的共性分量是同分布的、并按资料长度加权配合一条最优的无量纲概率分布曲线; • 此地区无量纲分布曲线与各个站点的本地分量“叠加”,组成该站的概率分布曲线; • 推求各个站点的频率估计值。

  32. 地区分析的优点:大大降低估计值的不确定性 Reduced uncertainties in terms of confidence interval (Regional analysis provides more stable estimates) Fig. 5 Regional analysis substantially reduce the uncertainties of frequency estimates

  33. 我国现有技术与新技术的差距 我国 – 主观性强、资料信息有限 常规矩法 + (单点、单时段、指定单一线型)+目估适线 新技术– 客观性强、充分利用资料信息 线性矩 + (地区分析、多时 段、多线型比较)+准则判断

  34. 我们的使命之二 为重大工程(大型水坝、核电站)防洪设计 和大城市防洪规划服务的PMP估算 Probable Maximum Precipitation -- 水文气象成因分析

  35. 20世纪30年代– 美国天气局(NWS)开始运用水文气象途径,尝试从成因分析上回答:某地某时段能降下最大的雨量是多少?这就是: Maximum Possible Precipitation (MPP)  Probable Maximum Precipitation (PMP)

  36. Definition of PMP (1) Probable maximum precipitation (PMP) is defined as the greatest depth of precipitation for a given duration meteorologically possible for a design watershed or a given storm area at a particular location at a particular at a particular location at a particular time of year, with no allowance made for long-term climate trends. (WMO No.-1045, Manual on Estimation of PMP, Geneva, 2009) 可能最大降水定义为一年的特定时间中、在特定地点和给 定时段内、在某一设计流域上或者给定暴雨面积下,气象 上所可能降下的最大雨量;这个降水量是不考虑气侯的长 期变化趋势。(WMO “PMP估算手册”,第三版)

  37. PMP Estimation Methodology-- International Practice In general, mainly two types of approaches in design practice of PMP studies: Hydrometeorological (HYDROME) & Statistical (STAT) (a) Moisture maximization Maximum 12-hr persisting dew point (HYDROME) (b) Storm transposition Storm Separation + Adjustments (HYDROME) (c) Use of D-A-D curves Envelopment (HYDROME) (d) Statistical approach Modified frequency analysis (STAT)

  38. 最主要方法 -- 暴雨移置 最难的是山区暴雨移置 -- 核心:暴雨分割(Storm separation) 辐合雨(Convergence component) 地形雨(Orographic component)

  39. 天气分析+雨量统计特性+地形、地貌 For a storm rainfall, the rainfall intensity for a given point P(x,y) in a drainage at any time can be defined by 6-2. Hence, the area-averaged rainfall R for the whole drainage area of A during the period of is given below:

  40. 莫拉克台风24-hr等雨量线图 1,583mm / 24-hr 2,372mm / 48-hr 2,682mm / 72-hr

  41. 莫拉克台风水汽入流 Development of the SDOIF for the Target Area (b) Major Moisture Flux during Morakot (left) Power Spectrum of the WNPSM Index (right) PMP Workshop No. 1 August 31, 2014 Page 41

  42. 台湾代表性雨量站分布图

  43. 格点化的台湾地形增强因子图 (10kmx10km)

  44. 莫拉克暴雨分割前后 Before After

  45. 概化的台湾地区辐合雨分量模式 300 450 600 700 800 928

  46. 辐合雨分量可以扩大移置范围

  47. Application example: Transposed to HK Moisture flux Typhooncyclone Meet ! Westerly moisture air

  48. 构造移入地地形增强因子图 (5kmx5km)

  49. Application of Results Estimation of Embryonic PMP in Hong Kong (c) Step 3. get a preliminary view of how they look like of the estimated PMP and its spatial distribution as a comparable way to the Step (2) – (i) Convert the convergence component into a gridded frame like the SDOIF 4-hr sub-mesh showing here is for the 4-hr Morakot pattern at resolution of 5kmx5km PMP Workshop No. 1 August 31, 2014 Page 49

  50. Application of Results Estimation of Embryonic PMP in Hong Kong (e) Step 3. get a preliminary view of how they look like of the estimated PMP and its spatial distribution as a comparable way to the Step (2) – (iii) Distribution of Gridded-average Embryonic PMP in HK PMP Workshop No. 1 August 31, 2014 Page 50

More Related