Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part...
Transcript of Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part...
![Page 1: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/1.jpg)
Uncertainties in planetary boundary layer schemes and current status of urban boundary layer simulations at OU
Xiaoming Hu
September 16th @ 3:00 PM, NWC 5600
Contributors:
Fuqing Zhang, Pennsylvania State Univ
John W. Nielsen-Gammon, Texas A&M Univ
![Page 2: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/2.jpg)
• Part 1: Evaluate three PBL schemes in the WRF model
• Part 2: Improve a PBL scheme through EnKF parameter estimation
• Part 3: Current status of urban BL simulations at OU
![Page 3: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/3.jpg)
Part 1: Evaluate three PBL schemes in the WRF model
Importance of PBL schemes
• The accuracy of the PBL scheme is critical for forecasts of local thermally driven flows and air quality while it also affects forecasts of larger-scale meteorological phenomena (Hacker and Snyder, 2005)
![Page 4: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/4.jpg)
Three PBL schemes in WRF MYJ, YSU, ACM2
• MYJ: local, down gradient
• YSU, ACM2: local+non-local
(YSU implicit,
ACM2 explicit)
Non-local Local
YSU: the Yonsei University scheme
MYJ: the Mellor–Yamada–Janjic scheme
ACM2: the asymmetric convective model scheme, v2
![Page 5: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/5.jpg)
Configurations Episode & Resolution Period: July – Sept., 2005 Resolution: 108km, 36km, 12km, 4km Grids: 53×43, 97×76, 145×100,
166×184
Model Configurations
YSU, ACM2, MYJ PBL schemes
WSM 6-class graupel scheme
NOAH land-surface model (LSM)
Dudhia short wave radiation RRTM long wave radiation Grell-Devenyi ensemble cumulus
scheme
Domains and TCEQ, NWS/FAA sites
![Page 6: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/6.jpg)
Mean T2 and dew point over 211 NWS/FAA sites
MYJ gives the coldest and moistest biases near the surface
![Page 7: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/7.jpg)
Mean T2 at 15 and 00 CST
The model captures the spatial variation of temperature,
but MYJ predicts the lowest temperature near the surface
![Page 8: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/8.jpg)
Mean PBL Height
MYJ underpredicts PBL height over most sites
![Page 9: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/9.jpg)
Difference of T2 and HFX between simulations with YSU and MYJ
Difference of sensible heat flux (HFX) cannot explain difference of T2
![Page 10: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/10.jpg)
Mean profiles of T and moisture
MYJ doesn’t mix as high as YSU and ACM2 during daytime
![Page 11: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/11.jpg)
Mean temperature profile difference from 9 CST at 11 CST
MYJ underestimates entrainment fluxes.
![Page 12: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/12.jpg)
Normalized Kz profile due to different p
p controls the local vertical mixing coefficient in ACM2 PBL scheme
![Page 13: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/13.jpg)
Mean profile of T and QVAPOR from runs with altered p
The similarity between the sensitivity of WRF to varied mixing strength and the
sensitivity of WRF to different PBL schemes confirms that much of the sensitivity of
WRF to different PBL schemes is attributable to their different vertical mixing strengths.
![Page 14: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/14.jpg)
Mean profiles of T and moisture
MYJ doesn’t mix as high as YSU and ACM2 during daytime
![Page 15: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/15.jpg)
Conclusions
1.The YSU and ACM2 schemes both tend to predict
higher T and lower moisture, and thus smaller
biases, than the MYJ scheme in the lower
atmosphere during daytime because of their
stronger vertical mixing.
2.The above conclusion is verified by the
experiments with the WRF model with altered
vertical mixing strength.
![Page 16: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/16.jpg)
• Part 2: Improve the performance of a PBL scheme through EnKF parameter estimation
![Page 17: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/17.jpg)
WSP sensitivity to 10 parameters in ACM2
WSP is mostly sensitive to p, Rc.
![Page 18: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/18.jpg)
Correlation between parameters & WSP
WSP shows the largest correlation with p, Rc. Thus p, Rc have the largest identifiability
![Page 19: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/19.jpg)
Sensitivity to p
Lower p => stronger vertical mixing => higher PBL height.
![Page 20: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/20.jpg)
Use EnKF to update p, Rc
• Deterministic simulation (NoDA)
• Regular EnKF (NoPE)
• Parameter estimation EnKF (SSPE)
– Update p, Rc simultaneously as updating regular states
– Assimilate wind profiler data only every 6-hour between Aug. 30-Sept. 2, 2006 over Texas
• Deterministic simulation with estimated parameters (NoDAnew)
![Page 21: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/21.jpg)
Wind vectors at Sept 1, 10 CST
(a) observations
(b) NoDA
(c) NoPE
(d) SSPE
SSPE shows the best agreement for surface wind.
![Page 22: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/22.jpg)
Profiles of WSP and T
SSPE predicts higher PBLH to match profiler data.
![Page 23: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/23.jpg)
Evolution of p
During most of time, SSPE predicts p value lower than 2.0 (default).
![Page 24: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/24.jpg)
Bias and error of T2
SSPE predicts the least cold bias.
![Page 25: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/25.jpg)
Conclusions
1.PBL schemes remain one of the primary sources
of inaccuracies in model simulation. Vertical
mixing strength plays an important role in
performance of PBL schemes
2.Real-data experiments show that simultaneous
state and parameter estimation with EnKF
performs better than deterministic simulation and
regular EnKF by providing optimized flow
dependent parameters in the PBL scheme
![Page 26: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/26.jpg)
• Part 3: Current status of urban boundary layer simulations at OU
![Page 27: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/27.jpg)
MODIS observed UHI, daytime
March 2005
![Page 28: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/28.jpg)
MODIS observed UHI, nighttime
July 2006
![Page 29: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/29.jpg)
UHI splits precipitation?
![Page 30: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/30.jpg)
Configurations of WRF-UCM Episode & Resolution Period: July 13., 2005 Resolution: 40.5, 13.5, 4.5, 1.5, 0.5km
Model Configurations
YSU PBL schemes
WSM 6-class graupel scheme
Dudhia short wave radiation RRTM long wave radiation
NOAH land-surface model (LSM)
+ an urban canopy model (UCM)
𝐻𝐹𝑋 = 𝐹𝑢𝑟𝑏 × 𝐻𝐹𝑋𝑢𝑐𝑚 + (1 − 𝐹𝑢𝑟𝑏) × 𝐻𝐹𝑋𝑁𝑂𝐴𝐻
![Page 31: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/31.jpg)
Schematic of urban canopy model
Slab model like NOAH Urban canopy model (UCM)
𝐻𝐹𝑋 = 𝐹𝑢𝑟𝑏 × 𝐻𝐹𝑋𝑢𝑐𝑚 + (1 − 𝐹𝑢𝑟𝑏) × 𝐻𝐹𝑋𝑁𝑂𝐴𝐻
![Page 32: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/32.jpg)
Detailed land use data USGS 1994 NLCD 2006
Urban expansion since 1994
![Page 33: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/33.jpg)
Simulated urban heat island
![Page 34: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/34.jpg)
Challenges in initializing WRF-UCM
Default soil moisture cannot represent urban
![Page 35: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/35.jpg)
Solution for initializing soil moisture?
• Run an uncoupled (offline) LSM constrained by observed forcing conditions long enough to develop an equilibrium soil properties?
Chen et al. (2011) run NOAH/UCM in an offline mode for 18 months.
Run NASA Land Information System (LIS) in an offline mode?
![Page 36: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/36.jpg)
References
1. Hu, X.-M., J. W. Nielsen-Gammon, and F. Zhang (2010), Evaluation of
Three Planetary Boundary Layer Schemes in the WRF Model, J. Appl.
Meteor. Climatol., 49, 1831–1844.
2. Nielsen-Gammon, J. W., X.-M. Hu, F. Zhang, and J. E. Pleim (2010),
Evaluation of Planetary Boundary Layer Scheme Sensitivities for the
Purpose of Parameter Estimation, Mon. Wea. Rev., 138, 3400–3417.
3. Hu, X.-M. , F. Zhang, and J. W. Nielsen-Gammon (2010), Ensemble-
based simultaneous state and parameter estimation for treatment of
mesoscale model error: A real-data study, Geophys. Res. Lett., 37, L08802,
doi:10.1029/2010GL043017.
![Page 37: Uncertainties in planetary boundary layer schemes and ...xhu/files/XHu2011SeptNWCimprove... · Part 1: Evaluate three PBL schemes in the WRF model Importance of PBL schemes • The](https://reader034.fdocuments.us/reader034/viewer/2022051801/5ad980877f8b9ae1768bf034/html5/thumbnails/37.jpg)
Links
1. http://faculty-staff.ou.edu/H/Xiaoming.Hu-1/
2. http://journals.ametsoc.org/doi/abs/10.1175/2010JAMC2432.1
3. http://journals.ametsoc.org/doi/abs/10.1175/2010MWR3292.1
4. http://www.agu.org/pubs/crossref/2010/2010GL043017.shtml
Simulations of WRF-UCM
5. http://www.caps.ou.edu/micronet/WRF-UCM.html