Mesoscale Modelling of Optical Turbulence in the ......Mesoscale Modelling of Optical Turbulence in...

Mesoscale Modelling of Optical Turbulence in the Atmosphere:

Quantifying the Impact of Ultra-High Vertical Resolution

COAT – December 2nd, 2019

Sukanta Basu

Associate Professor

Faculty of Civil Engineering and Geosciences

s.basu@tudelft.nl

Acknowledgement: Ping He

Optical Turbulence Forecasting

2 Source: Trinquet and Vernin (2009)

All Roads Lead to Rome

Cn2 Parameterization

Higher-order Closure Approach Regression (ANN) Approach

DNS/LES Approach Scaling Approach

Estimation

(e.g., Radiosonde)

Forecasting

(Mesoscale Model)

3

Radiation

Planetary

Boundary

Layer

Land Surface

Microphysics

Convection

Mesoscale Modeling

4

Modeling Framework

Mesoscale Atmospheric

Modeling

Global Reanalysis

(~30 km Resolution)

Pressure(lat,lon,z,t)

Temperature(lat,lon,z,t)

Moisture(lat,lon,z,t)

Turbulent Diagnostics

Refractive

Index(lat,lon,z,t)

Static Geographic Data

(e.g., Terrain, Landuse)

Geometric Optics-

Based Ray Tracing

Realistic Long-Range

Optical Ray Trajectories

Ray Origin(lat,lon,z,t)

Ray Propagation

Azimuth

Ray Elevation Angle

Parameterized

Cn2 (lat,lon,z,t)

5

Objective

Intermittent Optical Turbulence in Free Atmosphere

7

May 26, 2006; Masciadri et al. (2008)

ORM: Roque de los Muchachos on the island of La Palma

OT: El Teide on the island of Tenerife

Both observatories are on the Canary Islands and about 160 km from each other

Time-Height Plot of Cn2 from G-SCIDARs

Intermittent Optical Turbulence in Free Atmosphere (Cont.)

8 Dome C, Antarctic Cerro Tololo, Chile

Challenge

Vertical Grids in Mesoscale Models

10 Source: Alligo et al. (2009)

Relevant Literature

11

Effects of Vertical Resolution on Eddy Viscosity

12

Source: Skamarock et al. (2019)

Methodology

Higher-order Closure (HOC) Approach

14

HOC Approach (Cont.)

15

HOC Approach (Cont.)

16

Case Study:

Hawaii 2002 Thermosonde Campaign

Hawaii 2002

Large-Scale Behavior of Ascent Rate

19 Source: McHugh et al. (2008)

Computational Domain (9/3/1 km)

20

Vertical Grids in WRF Runs

21

Dec 11-12, 2002

Mean Temperature Profiles

23

Mean Wind Speed Profiles

24

𝐶𝑛2 Profiles

25

Simulated 𝐶𝑛2 over Hawaii and Pacific Ocean

26

# vertical grid points: 51

Simulated 𝐶𝑛2 over Hawaii and Pacific Ocean (Cont.)

27

# vertical grid points: 101

Simulated 𝐶𝑛2 over Hawaii and Pacific Ocean (Cont.)

28

# vertical grid points: 286 (uniform grid spacing of 100 m)

Simulated “Seeing” (Total)

30

Source: McHugh et al. (2008)

Simulated “Seeing” (Ground Layer)

31

Simulated “Seeing” (Upper Layer)

32

Dec 12-13, 2002

Mean Temperature Profiles

34

Mean Wind Speed Profiles

35

𝐶𝑛2 Profiles

36

Simulated 𝐶𝑛2 over Hawaii and Pacific Ocean

37

# vertical grid points: 51

Simulated 𝐶𝑛2 over Hawaii and Pacific Ocean (Cont.)

38

# vertical grid points: 101

Simulated “Seeing” (Total)

39

Source: McHugh et al. (2008)

Horizontal Cross-Sections

Island Wakes

41 Source: NASA

Flow Structures around Big Island

42

Source: Smith and Grubišic (1993)

Spatial Distribution of 𝐶𝑛2 (# Vertical Grid Points = 286; 1 km MSL)

43 2 UTC, Dec 12 7 UTC, Dec 12

Spatial Distribution of 𝐶𝑛2 (# Vertical Grid Points = 286; 10 km MSL)

44 2 UTC, Dec 12 7 UTC, Dec 12

Case Study:

Paranal 2017

Observed 𝐶𝑛2 over Paranal

46 Courtesy: James Osborn

Simulated 𝐶𝑛2 over Paranal

47

# vertical grid points: 201

Simulated 𝐶𝑛2 over Paranal (Cont.)

48

# vertical grid points: 286 (uniform grid spacing of 100 m)

To be continued…

Richardson (1922)

50

“Perhaps some day in the dim future it will be possible to advance the computations faster than the

weather advances and at a cost less than the saving to mankind due to the information gained. But

that is a dream.”

Richardson’s

forecast factory

Source: Bengtsson