Improved representation of convection, tropical clouds and precipitation in climate models

Anvil-cloud. Photo by Traute Crueger ©2017

Anvil-cloud. Photo by Traute Crueger ©2017

In a new study, Dr Karsten Peters and Dr Traute Crueger, both scientists in the department “The Atmosphere in the Earth System” at the Max Planck Institue for Meteorology (MPI-M),  together with their colleagues Prof Christian Jakob and Dr Benjamin Möbis from Monash University, Australia, show for the first time that a more realistic simulation of tropical weather on timescales of weeks in a general circulation model (GCM) critically depends on the representation of tropical precipitation on much shorter (daily) timescales.

A constant balancing of radiant heat loss and convective energy input takes place in the atmosphere. Warm air rises condensing water and releasing latent heat to balance the energy the atmosphere looses through radiation to space.  The term convection describes this process. The aim to develop a more accurate representation of small-scale atmospheric vertical motions, or convection, and the surface rainfall resulting from it in GCMs is a longstanding issue in the atmospheric science community.

In their study, the authors showed that by tracking the area of convection, which mostly sets the vertical convective mass flux, it is possible to much more accurately mimic what is observed in nature on the short time-scales of tropical weather. Unexpectedly, the approach led also to improvement in the behavior of the model on longer timescales, from weeks to months. For the first time it shows that processes that govern the behavior of tropical weather also play a role in shaping its longer-term variability.

The new approach is enabled by methods developed in a previous study (Peters et al. 2013), where a stochastic multi-cloud model was introduced and its behavior trained to represent observed convection at two tropical locations.  The work is the fruit of a long-standing cooperation in the development of new convective schemes involving the MPI-M, the Australian Research Council’s Centre of Excellence for Climate System Science and the German Weather Service through its Hans Ertel Research Centers.

Original publication:
Peters, K., T. Crueger, C. Jakob, and B. Möbis (2017), Improved MJO-simulation in ECHAM6.3 by coupling a Stochastic Multicloud Model to the convection scheme, J. Adv. Model. Earth Syst., 9, Opens external link in current windowdoi: 10.1002/2016MS000809.

Previous study:
Peters, K., C. Jakob, L. Davies, B. Khouider, and A. Majda, 2013: Stochastic Behavior of Tropical Convection in Observations and a Multicloud Model. J. Atmos. Sci., 70, 3556–3575, Opens external link in current windowdoi: 10.1175/JAS-D-13-031.1.


Dr Karsten Peters
Max Planck Institute for Meteorology
Phone: +49 40 - 41173 - 424
Email: Opens window for sending emailkarsten.peters(at)

Dr Traute Crueger
Max Planck Institute for Meteorology
Phone: +49 40 - 41173 - 105
Email: Opens window for sending emailtraute.crueger(at)