TWP-ICE/ACTIVE MONSOON CASE STUDY

Liquid water path (LWP) measurements and ECMWF model results (white lines and symbols) versus example case study simulations (colored lines and symbols):



Comments:

• Forcing data set derived from variational analysis over the TWP-ICE domain bounded by sounding sites, provided by Shaocheng Xie and Minghua Zhang (summary with download link), reported as 3-hr mean values over the area of approximately 25,000 km². Because this is primarily derived from MWR retrievals, which are not available under precipitating conditions, we compare it here to LWP in model columns with PR < 0.02 mm/h.
• Microwave radiometer (MWR) retrievals are supplied by Dave Turner. Retrievals are instantaneous point measurements at 40-s frequency provided at two locations, the DOE ARM Darwin site and the Southern Surveyor ship (both plotted at reported frequency in middle panel).
• Tropical Cloud Properties and Heating Rate Profiles (TCPRHP) retrievals at Darwin are supplied by Sally McFarlane, Jim Mather, and Sally McFarlane. Values represent instantaneous profiles over the DOE ARM site reported at 10-s and 45-m resolution.
• ECMWF supplies hourly mean data at 0.56x0.56-degree resolution. We calculate LWP by integrating liquid condensate over all layers in each of the two grid boxes centered nearest to and immediately west of the Darwin ARM site at (130.43,-12.82) and (129.73,-12.47), both shown. The mean value is negligibly changed if a third grid box to the southeast at (131.13,-13.52) is also included (these three grid points are stored in the ARM archive for the Darwin site), bringing the total area to approximately 11,500 km².
• All measurement data have been downloaded from the DOE ARM data archive (www.arm.gov) and use in publications is subject to the policies described there. Any errors in plotting or representation shown here should be blamed on Ann Fridlind.
• Model results are instantaneous values reported at 3-hr frequency and 1-km horizontal resolution (and averaged over 55 km) over a domain of approximately 37,000 km². LWP is reported as either "total" (including both cloud liquid and liquid-phase precipitation) or "non-precipitating" (including only LWP where PR < 0.02 mm/h).

Preliminary summary:

• When model mean LWP is therefore calculated from columns where PR < 0.02 mm/h, predicted LWP is still more than a factor of two lower than retrieved. At the same time, a simple comparison of every MWR retrieval with domain-maximum LWP where PR < 0.02 mm/h indicates that peak predicted LWP generally exceeds any maximum measured during the campaign. Model predicted total LWP is also about a factor of three higher than ECMWF results. Considering the relatively good agreement between modeled and observed PR and precipitating area, agreement seems very poor here.