Characteristics of precipitation estimates for rate and amount from three global high-resolution precipitation products (HRPPs), four global climate data records (CDRs), and four reanalyses are compared. All datasets considered have at least daily temporal resolution. Estimates of global precipitation differ widely from one product to the next, with some differences likely due to differing goals in producing the estimates. HRPPs are intended to produce the best snapshot of the precipitation estimate locally. CDRs of precipitation emphasize homogeneity over instantaneous accuracy. Precipitation estimates from global reanalyses are dynamically consistent with the large-scale circulation but tend to compare poorly to rain gauge estimates since they are forecast by the reanalysis system and precipitation is not assimilated. Regional differences among the estimates in the means and variances are as large as the means and variances, respectively. Even with similar monthly totals, precipitation rates vary significantly among the estimates. Temporal correlations among datasets are large at annual and daily time scales, suggesting that compensating bias errors at annual and random errors at daily time scales dominate the differences. However, the signal-to-noise ratio at intermediate (monthly) time scales can be large enough to result in high correlations overall. It is shown that differences on annual time scales and continental regions are around 0.8 mm day−1, which corresponds to 23 W m−2. These wide variations in the estimates, even for global averages, highlight the need for better constrained precipitation products in the future.

Intermittency is a core characteristic of precipitation, not well described by data and very poorly modeled. Detailed analyses are made of near-global gridded (about 1°) hourly or 3-hourly precipitation rates from two updated observational datasets [3-hourly TRMM 3B42, version 7, and hourly CMORPH, version 1.0, bias corrected (CRT)] and from special runs of CESM from January 1998 to December 2013 to obtain hourly values. The analyses explore the intermittency of precipitation: the frequency, intensity, duration, and amounts. A comparison is made for all products using several metrics with a focus on the duration of events, and a new metric is proposed based on the ratio of the frequency of precipitation at certain rates (0.2–2 mm h−1) for hourly versus 3-hourly versus daily data. For all seasons and rain rates, TRMM values are similar in pattern to CMORPH, but durations are about 80%–85%. It is mainly over land in the monsoons that CMORPH exceeds TRMM rain durations. Observed duration of precipitation events in CMORPH over oceans are 12–15 h in the tropics and subtropics, much less than the ~20 h for CESM. Hence, the observational results differ somewhat but both are considerably different from the model, which has too much precipitation overall, and it precipitates far too often at low rates and not enough for intense rates, with the divide about 1–2 mm h−1. There is a need to properly represent precipitation phenomena and processes either explicitly or implicitly (parameterized).

Despite decades of research on the role of moist convective processes in large-scale tropical dynamics, tropical forecast skill in operational models is still deficient when compared to the extratropics, even at short lead times. Here we compare tropical and Northern Hemisphere (NH) forecast skill for quantitative precipitation forecasts (QPFs) in the NCEP Global Forecast System (GFS) and ECMWF Integrated Forecast System (IFS) during January 2015–March 2016. Results reveal that, in general, initial conditions are reasonably well estimated in both forecast systems, as indicated by relatively good skill scores for the 6–24-h forecasts. However, overall, tropical QPF forecasts in both systems are not considered useful by typical metrics much beyond 4 days. To quantify the relationship between QPF and dynamical skill, space–time spectra and coherence of rainfall and divergence fields are calculated. It is shown that while tropical variability is too weak in both models, the IFS is more skillful in propagating tropical waves for longer lead times. In agreement with past studies demonstrating that extratropical skill is partially drawn from the tropics, a comparison of daily skill in the tropics versus NH suggests that in both models NH forecast skill at lead times beyond day 3 is enhanced by tropical skill in the first couple of days. As shown in previous work, this study indicates that the differences in physics used in each system, in particular, how moist convective processes are coupled to the large-scale flow through these parameterizations, appear as a major source of tropical forecast errors.