Precipitation-generated oscillations in open cellular cloud fields

Journal name:
Nature
Volume:
466,
Pages:
849–852
Date published:
DOI:
doi:10.1038/nature09314
Received
Accepted

Cloud fields adopt many different patterns that can have a profound effect on the amount of sunlight reflected back to space, with important implications for the Earth’s climate. These cloud patterns can be observed in satellite images of the Earth and often exhibit distinct cell-like structures associated with organized convection at scales of tens of kilometres1, 2, 3. Recent evidence has shown that atmospheric aerosol particles—through their influence on precipitation formation—help to determine whether cloud fields take on closed (more reflective) or open (less reflective) cellular patterns4, 5. The physical mechanisms controlling the formation and evolution of these cells, however, are still poorly understood6, limiting our ability to simulate realistically the effects of clouds on global reflectance. Here we use satellite imagery and numerical models to show how precipitating clouds produce an open cellular cloud pattern that oscillates between different, weakly stable states. The oscillations are a result of precipitation causing downward motion and outflow from clouds that were previously positively buoyant. The evaporating precipitation drives air down to the Earth’s surface, where it diverges and collides with the outflows of neighbouring precipitating cells. These colliding outflows form surface convergence zones and new cloud formation. In turn, the newly formed clouds produce precipitation and new colliding outflow patterns that are displaced from the previous ones. As successive cycles of this kind unfold, convergence zones alternate with divergence zones and new cloud patterns emerge to replace old ones. The result is an oscillating, self-organized system with a characteristic cell size and precipitation frequency.

At a glance

Figures

  1. Cloud albedo calculated from large eddy simulation of closed and open cellular structures.
    Figure 1: Cloud albedo calculated from large eddy simulation of closed and open cellular structures.

    The simulations differ only in the magnitude of the initial aerosol concentrations. High aerosol concentrations favour non-precipitating closed cells (a) while low aerosol concentrations favour precipitating open cells (b). The brightest regions are associated with the thickest clouds and (in the case of open cells) the precipitation-generating zones.

  2. Updraught and downdraught patterns illustrating surface convergence and divergence zones in open cells.
    Figure 2: Updraught and downdraught patterns illustrating surface convergence and divergence zones in open cells.

    ac, Plan views of near-surface updraught (purple) and downdraught (green) patterns, each separated by one hour. w is vertical velocity; positive values are updraughts and negative values are downdraughts. Updraught regions correspond to surface convergence, and downdraughts to surface divergence zones. Four Y-shaped updraught patterns in a are labelled 1, 2, 3 and 4 to illustrate the evolution of the open cellular structures from one time (times relative to the start of the simulation) to the next (6:40 to 7:40 to 8:40). They occur at strong convergence zones and are favoured for the strongest convection. As time progresses (b and c), precipitation at these points changes the surface flow to a divergent one. New cellular structures emerge from the old planform as the cold-pool outflows interact with one another and generate new convergence zones that in turn generate new precipitation zones. d, Vertical cross-section of a precipitation-generated outflow and creation of a convergence zone as observed by ship-based radar and Doppler lidar (located at range 0km and height 0km). Radar reflectivity (>20dBZ; dark red) is indicative of significant rain. Lidar data show air flow towards (green) and away from (yellow) the lidar. Arrows indicate the direction of flow.

  3. A simple two-dimensional model of Rayleigh-Benard convection and oscillating Rayleigh-Benard convection.
    Figure 3: A simple two-dimensional model of Rayleigh–Bénard convection and oscillating Rayleigh–Bénard convection.

    Results are at normalized height z = 0.5 (0z1) and for normalized time τ. The colour scheme is a rendering of the normalized temperature. The period 0<τ<100 shows the transition from conduction to convection. The period 100<τ<300 shows a steady-state convective pattern developing. The period 300<τ<600 shows relatively small negative temperature perturbations, of one-quarter of the temperature gradient, applied to the locations of maximum temperature for a period of one-quarter of the relaxation time of the system. The system is resilient to the perturbations. The period 600<τ<800 shows larger temperature perturbations of one-third of the temperature gradient applied for a period of one-third of the relaxation time. The system oscillates back and forth between steady states as the simulated effects of precipitation modify the open cellular structure.

  4. Oscillations in precipitation rate.
    Figure 4: Oscillations in precipitation rate.

    a, Time series of domain-averaged surface precipitation rate P associated with simulations in Fig. 2 (cell size ~15km; red line), P from an Atlantic Ocean sounding with smaller cell size (~4km; black line16) and P from a southeast Pacific Ocean sounding with larger cell size (~20km; blue line)30. We note the distinct periodicity in surface P and the relationship between cell size and precipitation frequency. b, Time series of column-maximum precipitation Pm for a subset of the Fig. 2 domain, further illustrating oscillatory behaviour. c, Pm as a function of x (horizontal) distance over a limited range of y (horizontal) distance at a single model time step. d, Contours of Pm in x–time space showing a ‘gridded’ structure of Pm with distinct spatial/temporal periodicity. (That is, combinations of b and c, but with averaging over the entire y distance.) The column-maximum vertical velocity (solid contours; averaged over all y as in Pm) is well-correlated with Pm. Precipitation in cloud systems that are not organized exhibits time series characterized by intermittent, irregular events.

References

  1. Krueger, A. F. & Fritz, S. Cellular cloud patterns revealed by TIROS I. Tellus 13, 17 (1961)
  2. Agee, E. M. Observations from space and thermal convection: a historical perspective. Bull. Am. Meteorol. Soc. 65, 938949 (1984)
  3. Garay, M. J., Davies, R., Averill, C. & Westphal, J. A. Actinoform clouds: overlooked examples of cloud self-organization at the mesoscale. Bull. Am. Meteorol. Soc. 85, 15851594 (2004)
  4. Stevens, B. et al. Pockets of open cells and drizzle in marine stratocumulus. Bull. Am. Meteorol. Soc. 86, 5157 (2005)
  5. Sharon, T. M. et al. Aerosol and cloud microphysical characteristics of rifts and gradients in maritime stratocumulus clouds. J. Atmos. Sci. 63, 983997 (2006)
  6. Wood, R. et al. Open cellular structure in marine stratocumulus sheets. J. Geophys. Res. 113 doi:10.1029/2007JD009371 (2008)
  7. Rayleigh, L. On convection currents in a horizontal layer of fluid, when the higher temperature is on the under side. Phil. Mag. 32, 529546 (1916)
  8. Twomey, S. The influence of pollution on the shortwave albedo of clouds. J. Atmos. Sci. 34, 11491152 (1977)
  9. Atkinson, B. W. & Zhang, J.-W. Mesoscale shallow convection in the atmosphere. Rev. Geophys. 34, 403431 (1996)
  10. Krishnamurti, R. On cellular cloud patterns—Part 1: Mathematical model. J. Atmos. Sci. 32, 13531363 (1975)
  11. Getling, A. V. Rayleigh-Bénard Convection: Structures and Dynamics. In Advanced Series in Nonlinear Dynamics Vol. 11 (World Scientific, 2001)
  12. Graham, A. Shear patterns in an unstable layer of air. Phil. Trans. R. Soc. A232, 285296 (1933)
  13. Helfand, M. & Kalnay, E. A mechanism for open or closed cellular convection. J. Atmos. Sci. 40, 631650 (1983)
  14. Rosenfeld, D., Kaufman, Y. J. & Koren, I. Switching cloud cover and dynamical regimes from open to closed Benard cells in response to the suppression of precipitation by aerosols. Atmos. Chem. Phys. 6, 25032511 (2006)
  15. Savic-Jovcic, V. & Stevens, B. The structure and mesoscale organization of precipitating stratocumulus. J. Atmos. Sci. 65, 15871605 (2008)
  16. Xue, H., Feingold, G. & Stevens, B. Aerosol effects on clouds, precipitation, and the organization of shallow cumulus convection. J. Atmos. Sci. 65, 392406 (2008)
  17. Wang, H. & Feingold, G. Modeling mesoscale cellular structures and drizzle in marine stratocumulus. Part I: Impact of drizzle on the formation and evolution of open cells. J. Atmos. Sci. 66, 32373256 (2009)
  18. Baker, M. & Charlson, R. J. Bistability of CCN concentrations and thermodynamics in the cloud-topped boundary layer. Nature 345, 142145 (1990)
  19. Willis, G. E. & Deardorff, J. W. Laboratory observations of turbulent penetrative-convection planforms. J. Geophys. Res. 84, 295302 (1979)
  20. Strogatz, S. H. Exploring complex networks. Nature 410, 268276 (2001)
  21. Arenas, A. et al. Synchronization in complex networks. Phys. Rep. 469, 93153 (2008)
  22. Yair, Y., Aviv, R. & Ravid, G. Clustering and synchronization of lightning flashes in adjacent thunderstorm cells from lightning location networks data. J. Geophys. Res. 114 D09210 doi:10.1029/2008JD010738 (2009)
  23. Wood, R. et al. Preliminary confrontation of the VOCALS hypotheses with observations and modeling. CLIVAR 7, 59 (2009)
  24. Guo, Z., Shi, B. & Zheng, C. A coupled lattice BGK model for the Boussinesq equations. Int. J. Numer. Methods Fluids 39, 325342 (2002)
  25. Parodi, A., Emmanuel, K. A. & Provenzale, A. Plume patterns in radiative convective flow. New J. Phys. 5, 106.1106.17 (2003)
  26. Heylighen, F. The science of self-organization and adaptivity. In Encyclopedia of Life Support Systems (EOLSS, 2001)
  27. Wang, H. & Feingold, G. Modeling mesoscale cellular structures and drizzle in marine stratocumulus. Part II: The microphysics and dynamics of the boundary region between open and closed cells. J. Atmos. Sci. 66, 32573275 (2009)
  28. Barabási, A.-L. & Albert, R. Emergence of scaling in random networks. Science 286, 509512 (1999)
  29. Albrecht, B. A. Aerosols, cloud microphysics and fractional cloudiness. Science 245, 12271230 (1989)
  30. Wang, H., Feingold, G., Wood, R. & Kazil, J. Modelling microphysical and meteorological controls on precipitation and cloud cellular structures in southeast Pacific stratocumulus. Atmos. Chem. Phys. Discuss. 10, 63476362 (2010)

Download references

Author information

Affiliations

  1. NOAA Earth System Research Laboratory (ESRL), Chemical Sciences Division, Boulder, Colorado 80305, USA

    • Graham Feingold &
    • Wm. Alan Brewer
  2. Department of Environmental Sciences, Weizmann Institute, Rehovot 76100, Israel

    • Ilan Koren
  3. Pacific Northwest National Laboratory, Richland, Washington 99352, USA

    • Hailong Wang
  4. Department of Atmospheric Sciences, School of Physics, Peking University, 5 Yiheyuan Road, Beijing 100871, China

    • Huiwen Xue

Contributions

The principal idea was conceived jointly by G.F. and I.K. The large eddy simulations were designed and performed by H.W., H. X. and G.F. I.K. performed the two-dimensional model simulations and the satellite image analysis. W.A.B. performed the analysis of the lidar data. All authors discussed the results and commented on the manuscript.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

PDF files

  1. Supplementary Information (849K)

    This file contains Supplementary Information comprising: Large Eddy Simulation; Animation of evolving open cellular structure; Animation of coupled oscillator; MSG SEVIRI satellite imagery; Lidar and radar data; Rayleigh-Bénard oscillations resulting from buoyancy reversals using a 2-D model, Legends for Supplementary Movies 1-3, Supplementary Figure S3 with legend and References.

Image files

  1. Supplementary Movie 1 - Supplementary Figure 1 (2.6M)

    This movie contains an animation of open cellular structures in Fig. 2 (for full legend see Figure 1 in Supplementary Information file page 8).

Movies

  1. Supplementary Movie 2 - Supplementary Figure (17.3M)

    This movie contains an animation of the coupled oscillator (for full legend see Figure 2 in Supplementary Information file page 8).

  2. Supplementary Movie 3 - Supplementary Figure 4 (6.4M)

    This movie shows satellite imagery of oscillating open cells with animation of images at 30 min intervals (for full legend see Figure 4 in Supplementary Information file page 8).

Additional data