How large is large enough? Large-eddy simulation of clear and cloudy convective boundary layers Stephan R. de Roode Peter G. Duynkerke Harm J. J. Jonker (1) Institute for Marine and Atmospheric Research Utrecht (IMAU), Utrecht, The Netherlands (1) Thermofluids Section, Department of Applied Physics, Delft University of Technology, Delft, The Netherlands The investigations were supported by the Netherlands Organization for Scientific Research (NWO). This work was sponsored by the National Computing Facilities Foundation (NCF) for the use of supercomputer facilities.
Simulation of the clear convective boundary layer 4 different horizontal domain sizes L L = 3200 m L = 6400 m L = 12800 m L = 25600 m 1200 1200 L = 3200 m 1000 1000 L = 6400 m L = 3200 m L = 12800 m L = 6400 m L = 25600 m L =12800 m 800 800 L = 25600 m height [m] height [m] 600 600 400 400 200 200 0 0 0 0.125 0.25 0.375 0.5 -0.02 0 0.02 0.04 0.06 w' � v ' [mK/s] w'w' [mK/s]
Variance spectra in the middle of the boundary layer at t=8h 100 km 10 km 1 km 0.1 km 0.4 0.3 L = 3200 m L = 6400 m -2 ] L = 12800 m 2 s L = 25600 m 0.2 vertical velocity kS [m 0.1 0 -5 10 0.0001 0.001 0.01 L=3.2 km BL height -3 1 10 -4 8 10 -4 6 10 2 ] virtual potential kS [K -4 4 10 temperature -4 2 10 0 0 10 -5 10 0.0001 0.001 0.01 -1 ] wavenumber k [m
(Co)-variance spectra in the middle of the boundary layer 100 km 10 km 1 km 0.1 km -4 1 10 -5 8 10 -1 ] L = 3200 m L = 6400 m -1 )ms L = 12800 m -5 6 10 L = 25600 m kS [(g kg -5 moisture flux 4 10 -5 2 10 0 0 10 -5 10 0.0001 0.001 0.01 L=3.2 km BL height -7 2 10 -7 1.5 10 2 ] -1 ) moisture variance kS [(g kg -7 1 10 -8 5 10 0 0 10 -5 10 0.0001 0.001 0.01 -1 ] wavenumber k [m
Structures in the middle of the clear convective BL
Liquid water path evolution in stratocumulus simulation during the night (kg/m 2 )
(Co)-variance spectra in the middle of the stratocumulus layer 100 km 10 km 1 km 0.1 km 0.08 L = 3200 m 0.06 L = 6400 m -2 ] L = 12800 m 2 s vertical velocity L = 25600 m ww [m 0.04 variance k S 0.02 0 -5 -4 -3 -2 10 10 10 10 L=3.2 km BL height -9 8 10 -9 6 10 2 ] -1 ) moisture variance kS [(g kg -9 4 10 -9 2 10 0 0 10 -5 -4 -3 -2 10 10 10 10 -1 ] wavenumber k [m
(Co)-variance spectra in the middle of the stratocumulus layer 100 km 10 km 1 km 0.1 km 0.12 L = 3200 m 0.1 L = 6400 m L = 12800 m 0.08 -2 ] L = 25600 m uu-variance 2 s 0.06 kS [m 0.04 0.02 0 -5 10 0.0001 0.001 0.01 L=3.2 km BL height 0.12 0.1 0.08 -2 ] 2 s vv-variance 0.06 kS [m 0.04 0.02 0 -5 10 0.0001 0.001 0.01 -1 ] wavenumber k [m
Production of variance in the clear CBL Žw'w' = P w 2 = 2 g � 0 w' � v ' Žt P Ž � v ' � v ' 2 = - 2 w' � v ' Ž � v = P � v Žt Žz P 1 1 2 > 0 P w 2 < 0 , P � v 0.8 0.8 scaled height scaled height 0.6 0.6 2 < 0 P w 2 > 0 , P � v 0.4 0.4 2 > 0 P w 2 > 0 , P � v 0.2 0.2 0 0 -0.02 0 0.02 0.04 0.06 304.6 304.8 305 -1 ) v (K) w' � v ' (mKs �
Length scale in stratocumulus A = 1 , B = 0.61 � ≈ 180 (subcloud) w' � v ' = A w' � l ' + B w'q t ' A ≈ 0.6 , B ≈ 1200 (cloud) 1.2 1 0.8 i z/z 0.6 w' � v ' w' � l ' 0.4 0.2 0 -0.04 -0.02 0 0.02 -1 ) (Kms
How large is large enough? Conclusions • larger w-fluctuations at mesoscales, more production of mesoscale scalar variance • Scalar variance at mesoscales may grow with time -> longer duration of simulation requires larger horizontal domain This talk: http://www.phys.uu.nl/~roode/presentations.html Paper: http://www.phys.uu.nl/~roode/publications.html S. R. de Roode, P. G. Duynkerke and H. J. J. Jonker, 2002: Large Eddy Simulation: How large is large enough? Submitted to the J. Atmos. Sci.
Recommend
More recommend
