Wojciech Grabowski — National Center for Atmospheric Research (NCAR)
Hugh Clifton Morrison — University Corporation for Atmospheric Research
Xiaoqing Wu — Iowa State University
Ann M. Fridlind — NASA - Goddard Institute for Space Studies
Andrew Ackerman — NASA - Goddard Institute for Space Studies
Adrian Hill — UK Meteorological Office
Jiwen Fan — Pacific Northwest National Laboratory
Todd R. Jones — Colorado State University
—
Jean-Pierre Chaboureau — University of Toulouse, France/CNRS
Jean-Pierre Pinty — University of Toulouse, France/CNRS
Category
Cloud Properties
Description
Precipitation rate statistics at the surface and at 2.5 km elevation from eight cloud-resolving model simulations and four sensitivity tests are compared with the large-scale forcing data set, which is based on the scanning C-POL radar at Darwin, and with ECMWF re-analysis at three grid cells considered representative of local conditions. Mean value and standard deviations are listed, followed by percentage difference between forcing and C-POL data.
An idealized cloud-resolving model (CRM) intercomparison specification was based on observations obtained under active and suppressed monsoon conditions over 16 days during the Tropical Warm Pool-International Cloud Experiment (TWP-ICE). Results of eight CRM simulations were submitted for the baseline case described in the specification. Sensitivity test results were also submitted for half of the eight models, wherein nudging to observed thermodynamic fields with a six-hour timescale was included. Results from the eight different models generally exhibit a wider spread than between baseline and sensitivity results from any single model. Considering all of the simulations together, predicted values of precipitation rate and liquid and ice water path, cloud optical thickness, and the ratio of upwelling over downwelling shortwave radiative flux all vary by about a factor of two. As a function of total predicted condensate path, liquid and ice water path and surface precipitation rate all tend to increase with a similar trend. While cloud optical thickness also exhibits the same general trend, scatter is much greater. By contrast, the ratio of upwelling to downwelling shortwave radiative flux at top of atmosphere is both widely scattered and poorly correlated with total condensate path, and instead closely correlated with the ratio of liquid water path to total condensate path, consistent with the dominant contribution of liquid to cloud optical thickness. We investigate the liquid water fraction and radiative flux differences in the context of simulated cloud macrophysical and microphysical structure. We present evidence that uncertainties in large-scale forcing fields are likely to produce differences in model results comparable to the spread in most reported values (e.g., precipitation rate), but can still identify various likely outlying simulation results through comparison of a wide range of reported diagnostics with disparate datastreams, such as cloud optical thickness range observed at the surface, upwelling longwave radiative flux range measured by satellite, and reflectivity fields measured by scanning precipitation radar.
