Helping Climate Models Rain at the Right Time

Researchers use ARM observations to improve the timing of rainfall in climate models.

The Science

The Southern Great Plains (SGP) region exhibits a unique rainfall peak at night due to interactions among large-scale circulation, local convection, and mesoscale weather systems. This complexity poses a major challenge for climate models, which commonly predict peak rainfall over SGP to occur at noon rather than at night. Scientists used Atmospheric Radiation Measurement (ARM) observations and a cloud scheme from a weather model to identify and correct the causes of this bias in a commonly used climate model, the Community Atmosphere Model.

The Impact

Scientists identified and applied two changes to the convective cloud formulation commonly used in the Community Atmosphere Model, thereby significantly improving the timing of peak rainfall over the SGP region simulated by the model.


In this study, researchers explored the triggering mechanisms of diurnal rainfall events due to multiscale processes and how to best represent them in climate models. Rainfall events for seven summers (2002-2008) were simulated with a single-column version of the Community Atmosphere Model using both the default cumulus parameterization (ZM) for this climate model and a cumulus parameterization more widely used in weather models and able to produce the correct rainfall timing. By comparing the model output obtained using the two schemes to observations from the Department of Energy’s (DOE) ARM Climate Research Facility, they identified the critical triggering mechanisms for producing the appropriate timing of convective rainfall in the climate model. One key issue was that observations showed that strong temperature inversions often developed in the nighttime and were sustained until morning, so the most convective air parcels were often about the planetary boundary layer; however, the ZM convective scheme required all convective air parcels to originate in the boundary layer. Relaxing this constraint and allowing convective parcels that originate above the boundary layer was the key to simulating nighttime rainfall. A second issue was that the ZM scheme produced unrealistic frequent weak rainfall events; adding a convective inhibition constraint to prevent the occurrence of convection when the energy barrier of dry layers is too large for the air parcels to be lifted to the level of free convection eliminated these events. When both changes were included, the ZM scheme produced a significantly improved daily rainfall cycle.

Principal Investigator(s)

Yi-Chi Wang
Academia Sinica, Taipei, Taiwan


Data used in this study was provided by the ARM program supported by DOE’s Office of Science, Office of Biological Research. Research funding was provided by the Ministry of Science and Technology, Taiwan under grant NSC100-2119-M-001-029-MY5.


Wang, Y.-C., H.-L. Pan, and H.-H. Hsu. 2015. “Impacts of the Triggering Function of Cumulus Parameterization on Warm-Season Diurnal Rainfall Cycles at the Atmospheric Radiation Measurement Southern Great Plains Site,” Journal of Geophysical Research Atmospheres 120(20), 10681-702. DOI: 10.1002/2015JD023337.6.1.