Presentation is loading. Please wait.

Presentation is loading. Please wait.

利用循环Hybrid ETKF-3DVAR 改进黄海海雾数值模拟的初始场

Similar presentations


Presentation on theme: "利用循环Hybrid ETKF-3DVAR 改进黄海海雾数值模拟的初始场"— Presentation transcript:

1 利用循环Hybrid ETKF-3DVAR 改进黄海海雾数值模拟的初始场
高山红 感谢 Xuguang WANG ! MODIS, MTSAT and FY satellite images

2 引子 2005年3月9日的一次平流冷却海雾过程(Gao et al.,2007) 1:天气系统(锋面) 2:湍流与长波辐射 3:云微物理

3 三个方面对海雾模拟很重要! ● 初始场 (高山红等, 2010a, b) ● 云微物理方案 辐射方案 ● 边界层方案 model obs
sea fog sea fog

4 内容 1. 何谓hybrid ETKF-3DVAR? 海雾个例与同化试验 … 黄海海雾数值预报之展望!

5 1. 何谓Hybrid ETKF-3DVAR? (a) 3DVAR (3 dimensional varational) 同化结果 背景场
(分析场) 背景场 (初估场) 观测 背景误差 观测误差

6 1. 何谓Hybrid ETKF-3DVAR? (b) cycling 3DVAR 3DVAR 06 MAR 08 MAR
UTC 00 UTC 06 MAR 08 MAR 3DVAR 3DVAR 3DVAR Cycling 3DVAR UTC 00 UTC 05 MAR 06 MAR 08 MAR 3DVAR interval is 6h Start Point of Forecast

7 1. 何谓Hybrid ETKF-3DVAR? (c) Hybrid 3DVAR 3DVAR所用的背景误差场是
静态!非流依赖(flow-dependent)! 时间平均 时间平均 非流依赖 流依赖 (Hamill et al., 2006)

8 1. 何谓Hybrid ETKF-3DVAR? (c) Hybrid 3DVAR (Ensemble+3DVAR) 如何获取“流依赖”信息?
集合预报 (X1, 2, …, N)! 上标含义: a: 分析场 f: 预报场

9 Ensemble Transform Kalman filter (ETKF, Wang,2008a,b)
1. 何谓Hybrid ETKF-3DVAR? Ensemble Transform Kalman filter (ETKF, Wang,2008a,b) (d) cycling hybrid ETKF-3DVAR

10 1. 何谓Hybrid ETKF-3DVAR? (e) 如何得到初始集合个体? 方法1:time-lagged forecast
方法2:random perturbation

11 2. 海雾个例与同化试验 (a) 观测事实(2006年3月)
20 LST 06 Mar 02 LST 07 Mar 08 LST 07 Mar 12 LST 07 Mar 14 LST 07 Mar 20 LST 07 Mar 02 LST 08 Mar 05 LST 08 Mar MTSAT IR (Gao et al,2009) TBB of IR1

12 (b) 同化试验设计 Exp-A: No 3DVAR 06 MAR 08 MAR Exp-B: 3DVAR 06 MAR 08 MAR
00 UTC 00 UTC 06 MAR 08 MAR 3DVAR Exp-B: 3DVAR UTC 00 UTC 06 MAR 08 MAR 3DVAR 3DVAR Exp-C: Cycling 3DVAR (intv=3h) UTC 3DVAR 00 UTC 05 MAR 06 MAR 08 MAR Exp-D: Cycling Hybrid ETKF-3DVAR (time-lagged) member number = 12 Exp-E: Cycling Hybrid ETKF-3DVAR (random-perturbation) Start Point of Forecast

13 Specifications of WRF run
Domains Back-ground FNL Data (1.0°x 1.0°) NEAR-GOOS Daily SST (0.25°x 0.25°) Resolution 30km, 10km; 44 levels PBL YSU Cumulus Kain-Fritsch Moisture Lin et al. Radiation LW: RRTM SW: Dudhia Surface Noah land-surface model Simulation Period 06_ _00 UTC Mar ( 48 h )

14 (d) 所同化的观测数据类型 sound synop ships airs gps QuikSCAT

15 (e) 模拟结果之比较 OBS Exp-A FNL only Single 3DVAR Exp-B Cycling 3DVAR Exp-C
H-E-3DVAR time-lagged Exp-D Cycling H-E-3DVAR random Exp-E

16 (f) 结果分析 地面气压场与风场的改进 a-d分别为Exp-B —Exp-E与Exp-A的之间的差异
2006年3月6日08LST 1000hPa位势高度场(m)

17 3. 黄海海雾数值预报之展望 1. 目前研究结果令人鼓舞! 2. Cycling Hybrid ETKF-3DVAR的业务可行性?
3. 为将来集合预报奠定基础! 1:多初始场 2:多参数化 3:多模式


Download ppt "利用循环Hybrid ETKF-3DVAR 改进黄海海雾数值模拟的初始场"

Similar presentations


Ads by Google