[1] Hourly averaged precipitation rates from an ensemble of the Community Climate Model version 3 (CCM3) simulations for the 44-month period from January 1998 through August 2001 are compared to observations from the Tropical Rainfall Measuring Mission (TRMM) satellite. In order to have adequate sampling by the satellite, comparisons are made for 15°longitude  10°latitude boxes and for larger geographical areas within the tropics. The temporally and spatially averaged hourly precipitation rates from CCM3 and from TRMM are fit to the diurnal harmonic by the method of linear least squares regression, and the phases and the amplitudes of the diurnal cycles are compared. The model's diurnal cycle is too strong over major land masses, particularly over South America (200% too large), and is too weak over many oceans, particularly the northwestern tropical Pacific (57% too small). The model-satellite phase differences tend to be more homogeneous. The peak in the model's diurnal harmonic consistently precedes that of the observations nearly everywhere. Phase differences are large over Australia, Papua New Guinea, and Saharan Africa, where CCM3 leads TRMM by 4 hours, 5 to 6 hours, and 9 to 11 hours, respectively. The model's phase and amplitude biases likely have effects on its hydrologic cycle and its surface and atmospheric energy budgets. Thus the causes for the model's biases need to be investigated.

[1] The climatological diurnal cycle of precipitation in the tropics is analyzed using data from rain gauges on ocean buoys and satellite measurements by the Tropical Rainfall Measuring Mission (TRMM) satellite. The ocean buoy data are from the NOAA/Pacific Marine Environmental Laboratory Tropical Atmosphere-Ocean/Triangle Trans-Ocean buoy Network in the tropical Pacific Ocean. TRMM data are from the precipitation radar (PR) and the TRMM microwave imager (TMI). Climatological hourly mean precipitation rates are analyzed in terms of the diurnal and semidiurnal harmonics. Both data sets confirm an early morning peak in precipitation over ocean regions. The amplitude of the diurnal harmonic over the oceans is typically less than 25% of the mean precipitation rate. Over tropical land masses the rainfall peaks in the afternoon and evening hours. The relative amplitude of the diurnal harmonic over land is larger than over the ocean, often exceeding 50% of the mean rain rate. Previously noted differences between the TMI and PR rainfall retrievals persist in the diurnal cycle. On average the TMI measures more rainfall than the PR and has a larger diurnal variation. Phase differences between the two instruments do not show a consistent bias.

No abstract

The second phase of the North American Monsoon Experiment (NAME) Model Assessment Project (NAMAP2) was carried out to provide a coordinated set of simulations from global and regional models of the 2004 warm season across the North American monsoon domain. This project follows an earlier assessment, called NAMAP, that preceded the 2004 field season of the North American Monsoon Experiment. Six global and four regional models are all forced with prescribed, time-varying ocean surface temperatures. Metrics for model simulation of warm season precipitation processes developed in NAMAP are examined that pertain to the seasonal progression and diurnal cycle of precipitation, monsoon onset, surface turbulent fluxes, and simulation of the low-level jet circulation over the Gulf of California. Assessment of the metrics is shown to be limited by continuing uncertainties in spatially averaged observations, demonstrating that modeling and observational analysis capabilities need to be developed concurrently. Simulations of the core subregion (CORE) of monsoonal precipitation in global models have improved since NAMAP, despite the lack of a proper low-level jet circulation in these simulations. Some regional models run at higher resolution still exhibit the tendency observed in NAMAP to overestimate precipitation in the CORE subregion; this is shown to involve both convective and resolved components of the total precipitation. The variability of precipitation in the Arizona/New Mexico (AZNM) subregion is simulated much better by the regional models compared with the global models, illustrating the importance of transient circulation anomalies (prescribed as lateral boundary conditions) for simulating precipitation in the northern part of the monsoon domain. This suggests that seasonal predictability derivable from lower boundary conditions may be limited in the AZNM subregion.

Simulation of the North American monsoon system by the National Center for Atmospheric Research (NCAR) Community Atmosphere Model (CAM3) is evaluated in its sensitivity to increasing horizontal resolution. For two resolutions, T42 and T85, rainfall is compared to Tropical Rainfall Measuring Mission (TRMM) satellite-derived and surface gauge-based rainfall rates over the United States and northern Mexico as well as rainfall accumulations in gauges of the North American Monsoon Experiment (NAME) Enhanced Rain Gauge Network (NERN) in the Sierra Madre Occidental. Simulated upper-tropospheric mass and wind fields are compared to those from NCEP-NCAR reanalyses. The comparison presented herein demonstrates that tropospheric motions associated with the North American monsoon system are sensitive to increasing the horizontal resolution of the model. An increase in resolution from T42 to T85 results in changes to a region of large-scale midtropospheric descent found north and east of the monsoon anticyclone. Relative to its simulation at T42, this region extends farther south and west at T85. Additionally, at T85, the subsidence is stronger. Consistent with the differences in large-scale descent, the T85 simulation of CAM3 is anomalously dry over Texas and northeastern Mexico during the peak monsoon months. Meanwhile, the geographic distribution of rainfall over the Sierra Madre Occidental region of Mexico is more satisfactorily simulated at T85 than at T42 for July and August. Moisture import into this region is greater at T85 than at T42 during these months. A focused study of the Sierra Madre Occidental region in particular shows that, in the regional-average sense, the timing of the peak of the monsoon is relatively insensitive to the horizontal resolution of the model, while a phase bias in the diurnal cycle of monsoon season precipitation is somewhat reduced in the higher-resolution run. At both resolutions, CAM3 poorly simulates the month-to-month evolution of monsoon rainfall over extreme northwestern Mexico and Arizona, though biases are considerably improved at T85.