TWP-ICE LAM Intercomparison

Motivation

To facilitate our understanding of the structure and evolution of tropical monsoonal deep convection, the resulting cirrus clouds, the induced convective transport, and their impact on the large-scale dynamics and thermodynamics, the ARM/GCSS/SPARC have decided to organize a joint model intercomparison study based on the Tropical Warm Pool – International Cloud Experiment (TWP-ICE). The case description and the scientific goal of the TWP-ICE intercomparison have been documented by Fridlind et al. (2009) in detail and can be found online at http://science.arm.gov/workinggroup/cpm/scm/scmic6/documentation.html. The limited area model (LAM) intercomparison described here is a unique component of this joint modeling study on tropical deep convective clouds. In addition to the scientific questions raised by Fridlind et al. (2009), the LAM intercomparison also focuses on issues that may not be appropriately addressed by the accompanying CRM, SCM, and NWP intercomparison studies. These include:

• Can LAM simulations capture the observed wide range of dynamical processes during the TWP-ICE experiment?
• How does surface heterogeneity, in particular the land/sea contrast, affect the cloud evolution during the active and suppressed monsoon periods?
• Can LAM simulations reproduce the observed diurnal cycle of the convective cloud systems initiated by the mainland and islands?
• Can LAMs realistically simulate the observed life cycle of mesoscale convective systems (MCSs)? And how do MCSs interact with the westerly monsoonal flow to regulate the cloud systems during different monsoon periods?
• Can LAMs statistically produce the similar cloud structure, anvil cirrus, and convective transport to those simulated by CRMs if LAMs are configured with a resolution compatible to that of CRMs? If not, what are the main reasons responsible for the difference?

Case Specification

Time period

• Monsoon trough (‘landfoon’): 23-25 Jan 2006 (start on 12:00Z 22 Jan)
• Suppressed monsoon: 28-30 Jan 2006 (start on 12:00Z 27 Jan)
• CRM period, 0Z 18 Jan - 0Z 3 Feb (optional)

Forcing data

• ECMWF analyses
• NCEP reanalyses (if ECMWF is not available)

Model domain and resolution

• 128.891ºE – 132.891ºE, 13.923ºS-10.925ºS with model center at 130.891ºE, 12.425ºS
• Horizontal grid spacing: 1 km
• Vertical: 76 levels (model top at least 24 km) (Note: If nesting is available (optional), the area defined above should be the inner most domain. Model domain and resolutions are subject to change depending on computational ability.)

Nudging

• To keep the simulation as realistic as possible, the ECMWF grid-data is suggested to be nudged in the outer domains every 6 hr. Nudging coefficients are set to be the same as those suggested by Fridlind et al. (2009). But nudging should not be on in the innermost research domain.

Tracer

• Mean tracer lifetime= 6 hours (exponential decay)
• Initialisation: tracer=1 in source layer; 0 elsewhere
• Source: tracer kept to 1 in source layer throughout the run

Output

Schedule and Expected Outcomes

Deadline

Publications

Download ECMWF analysis data for LAM participants only