1. Introduction
The purpose of this study is to measure the performance of a general circulation model in its simulation of precipitation in the Tropics. The motivations for a model validation of this kind are numerous. Between 30°S and 30°N, vast acreage is devoted to the production of citrus, corn, cotton, rice, wheat, and sugar (Espenshade 1995). According to figures obtained from the Population Reference Bureau, as of mid-2002, over 2.8 billion people were living in the Tropics, 1.5 billion alone in Southeast Asia. Residents of the Tropics constitute approximately 45% of the earth's population of around 6.2 billion (PRB 2003). Thus a large part of the human population benefits from the accurate prediction of long-term changes in precipitation within these latitudes.
Additionally, precipitation in the Tropics is the source of latent energy that comprises three-fourths of the energy used to drive the general circulation (Kummerow et al. 1998). Therefore, variability in the horizontal distribution and intensity of tropical convection has global effects, as evidenced in the teleconnections of the El Niño–Southern Oscillation (ENSO) phenomenon (see Ropelewski and Halpert (1987, 1989), Montroy (1997), Mo and Higgins (1998), Lau and Wu (2001), and Adler et al. (2000), among others).
General circulation models (GCMs) use Newton's equations of motion and the laws of thermodynamics along with parameterizations for subgrid-scale processes to simulate the atmospheric and oceanic circulations and the land surface. Essentially, they are models of the entire climate system and can be used to assess the impacts of variability in the mean state of the atmosphere. Long simulations (of several tens of years) can be used to predict climate change. Validation studies of the precipitation simulated by GCMs are numerous, many concentrating on specific regions of the earth. Examples include the studies of Kirkyla and Hameed (1989) and Chen et al. (1996; United States), Chen and Yen (1994; Indian monsoon), Busuioc et al. (1999; Romania), Trigo and Palutikof (2001; Iberian peninsula), and the 32-model intercomparison of Sperber and Palmer (1996) which examined the simulation of interannual rainfall variability in the Brazilian Nordeste, African Sahel, and the Indian subcontinent.
Since tropical precipitation is so important to the circulation of the atmosphere and thus to the climate system itself, its simulation by these models should be evaluated. This kind of evaluation may be performed by comparing the model with observations. In recent years, an important new dataset has become available, that of the Tropical Rainfall Measuring Mission (TRMM) satellite, which is a joint venture between the National Aeronautics and Space Administration (NASA) and the Japan National Space Development Agency (NASDA). One of the primary goals of this mission, as stated by Simpson et al. (1988), is to help modelers of the general circulation know the location and amount of latent energy released in the atmosphere to improve climate modeling (Kummerow et al. 2000). Launched 27 November 1997, the TRMM satellite orbits at an altitude of approximately 401 km, boosted from an altitude of 350 km in 2001,1 and is equipped with a microwave radiometer and a precipitation radar to infer rainfall rates between roughly 38°N and 38°S latitude. The microwave radiometer (TRMM Microwave Imager, or TMI) measures the electromagnetic energy emitted by the earth–atmosphere system in channels between 10.65 and 85.5 GHz. The size of the instrument's field of view depends on the frequency, ranging from 63 km × 37 km at 10.65 GHz to 7 km × 5 km at 85.5 GHz (Kummerow et al. 1998). Precipitation retrieval algorithms rely primarily on emission of radiation from hydrometeors (Kummerow et al. 2000).
The TRMM Precipitation Radar (PR), operating at a 13.8-GHz frequency, can measure the three-dimensional rainfall distribution. The energy backscattered by precipitation and received by the radar is converted to an equivalent precipitation rate, as explained in Kummerow et al. (2000). At present, the TRMM satellite remains in orbit, and fortuitously, it observed most of the strong 1997/98 El Niño event, allowing modelers to take advantage of this rich dataset for evaluating their simulations of ENSO. Our research uses the TRMM satellite data from both instruments, that is, the merged dataset based on PR/TMI algorithm 2B31 (Kummerow et al. 2000), to evaluate the National Center for Atmospheric Research (NCAR) Community Climate Model, version 3 (CCM3) in the simulation of tropical precipitation.
2. Methods
a. TRMM data
This study uses the 3G68 dataset, which was obtained from the TRMM Science Data and Information System (TSDIS). It consists of essentially instantaneous precipitation rates derived from TMI, from PR, and from the combination of both instruments averaged over 0.5° × 0.5° latitude–longitude boxes between 38°S and 38°N. For validating CCM3, we use the precipitation rates from the combination of TMI and PR. Generally, the TRMM satellite is able to observe a given location in the Tropics about once per day, at a different time each day, with a cycle of 46 days, the period of its orbital precession (Negri and Bell 2002). Therefore, for the 44-month period considered here (January 1998–August 2001), there are about 1320 observations of each of the aforementioned boxes. Precipitation is highly variable in both space and time, and the incomplete nature of the satellite's sampling introduces an error in the retrieval relative to actual ground truth. This error is known as sampling error and has been the subject of several studies (see Shin and North 1988; Bell and Kundu 1996, 2000; Bell et al. 2001; and most recently Bell and Kundu 2003, for example). According to Shin and North (1988), TRMM sampling errors in monthly mean rain rates in wet areas of the Tropics would be less than 10% when averaging the observations on areas of 5° × 5°. And the results of Bowman et al. (2003) show that very long-term averages of satellite-derived rain rates compare remarkably well with those measured by rain gauges on areas as small as 1° × 1°.
b. GCM simulations
CCM3 is a three-dimensional global spectral model. For this study, simulations are carried out at T42 horizontal resolution (approximately 2.8° latitude × 2.8° longitude) with 18 vertical levels. There is a rigid lid at 2.9-hPa pressure. The model uses a hybrid terrain-following vertical coordinate with sigma coordinates at lower levels that transition to pure pressure coordinates at upper levels. Physical tendency parameterizations include those for clouds (Slingo 1987, 1989; Ebert and Curry 1992; Cess 1985; Liou 1992), radiative fluxes (Ramanathan 1976; Ramanathan and Downey 1986), surface fluxes (Holtslag and Boville 1993), boundary layer height (Vogelzang and Holtslag 1996), and gravity wave drag (McFarlane 1987; Lindzen 1981). Adjustment physics consist of a convective parameterization following Zhang et al. (1998), Zhang and McFarlane (1995), and Hack (1994), large-scale stable condensation, and dry convective adjustment. One of the inputs for the model is a monthly mean sea surface temperature boundary condition. For this study, we used sea surface temperatures provided to us by the Program for Climate Model Diagnosis and Intercomparison (PCMDI) at Lawrence Livermore National Laboratory (Taylor et al. 2000). Additionally, the model requires a time-variant ozone mixing ratio boundary dataset, and an initial conditions dataset that includes initial values of prognostic variables (Kiehl et al. 1996).
Climate models may exhibit considerable internal variability or noise, partly due to fluctuations on synoptic time scales. Consequently, as has been shown by Barnett (1995) in his GCM simulations, a single model simulation of an interannual climate event or a climate forecast is woefully inadequate for the accurate evaluation of a model's performance. We will show that this finding is also valid for CCM3. To distinguish the model's response to natural variations in the SST boundary condition (external variability) from its response to its own internal variability, it is helpful to look at statistics from an ensemble of simulations. Therefore, for this project, we carried out eight separate CCM3 simulations, each forced by exactly the same sea surface temperature boundary condition and differing only in their initial conditions. While the sensitivity of extended-range forecast models to initial conditions is quite significant (Lorenz 1963; Tracton and Kalnay 1993), the actual initial conditions are largely irrelevant to climate forecasts (Barnett 1995), since a climate simulation “forgets” its initial conditions after some limit of deterministic predictability. Therefore, to generate our ensemble, we modified initial conditions among the members by adding random perturbations to the temperature field of a 4-month spinup run (1 September 1996–1 January 1997). By the beginning of the TRMM observing period, in late November of 1997, the realizations have decorrelated from each other and can be treated as statistically independent. Intermember correlations are never identically zero since the observed sea surface temperature field, common to all the members, exerts a common forcing.
The precipitation rates of each simulation, both convective and large scale, are saved as hourly averages. Unless otherwise stated, CCM3 monthly, seasonal, or annual means shown here are ensemble monthly, seasonal, or annual means, and all TRMM results are either monthly, seasonal, or annual means averaged onto the CCM3 grid. However, since the sampling errors associated with TRMM are too large on this grid, both datasets will be spatially averaged over much larger regions (greater than 5° × 5°) for the model validation.
3. Results
Before discussing the comparison, we briefly address the variability across the ensemble. For the purposes of analyzing the model's internal variability and for the comparison that follows, we divided the Tropics into 48 separate, nonoverlapping regions with partitions resting halfway between two adjacent grid points. See Fig. 1a which is a map of their locations. Boxes are numbered down the columns and then across the rows, each box labeled with a suffix of “W” or “E” depending on whether it is west or east of the prime meridian. Time series are computed from the monthly means averaged over all grid points within a box. For example, the results to be shown for box 1W are monthly means averaged over all grid points between longitudes 178.6°E and 136.4°W, and between latitudes 19.5°N and 30.7°N. Longitudes and latitudes are rounded off to the first decimal place for brevity. Figure 1b shows area-averaged monthly mean precipitation rates for all eight ensemble members for box 14W. The ensemble mean is shown in heavy black. Figure 1c shows the results for box 15E.
For box 14W, the model exhibits remarkably little variability across the ensemble. Each of the eight curves falls very close to the ensemble mean. This is not the case for box 15E. Here, the model exhibits considerably more spread among the simulations. Notice that a clearly defined seasonal oscillation is evident, which is the model's response to SST variations.
The comparison of the simulated and observed precipitation rates is now discussed. Figure 2 shows a map of the climatological annual-mean precipitation rate in the Tropics, as computed from the CCM3 simulations and from the TRMM observations for the period 1 January 1998–31 August 2001. These results are presented on the model's approximate 2.8° × 2.8° grid.
Features of the large-scale general circulation are evident in the annual-mean precipitation intensity. The model clearly depicts the intertropical convergence zone (ITCZ) regions straddling the equator and the areas of subsidence in the subtropics. It also captures the ascent and descent regions of the Walker circulation in the Pacific Ocean. Along the equator, more intense precipitation occurs over the Indian Ocean and the Maritime Continent, while less precipitation occurs in the eastern Pacific. The geographical distributions of mean precipitation agree fairly well with the TRMM climatology, but there are regions of overestimation by the model (see Fig. 2c, which plots the absolute differences). These include the Caribbean and eastern Pacific, equatorial Africa, and the Indian Ocean. There are also regions where the model underestimates precipitation. These are generally smaller in area and include parts of northwestern and southeastern South America, the central equatorial Atlantic, western equatorial Africa, extreme southeastern Asia, and parts of the Maritime Continent.
Maps of the climatological seasonal means (Fig. 3) show the tendency for CCM3 to simulate larger time-mean precipitation rates than observed by TRMM. At this point, it should be noted that the TRMM seasonal means are the time averages of 12 months of TRMM data (four 3-month seasons). CCM3 results, on the other hand, are computed using 8 times as much data. In addition, the CCM3 rain fields are available much more frequently than the TRMM observations. Therefore, when averaging over these shorter time periods, the CCM3 maps are generally smoother than the corresponding TRMM maps; this is particularly evident in maps of monthly means (not shown). Despite the substantially greater amount of averaging, the CCM3 data have numerous localized regions with higher precipitation rates than seen by TRMM. For example, during December–January–February (DJF) simulated precipitation rates are quite high over portions of eastern South America and the western equatorial Pacific Ocean, compared to the TRMM observations. During March–April– May (MAM), large localized differences are evident in equatorial Africa and in the western equatorial Atlantic. In June–July–August (JJA), the largest differences are concentrated in two regions: Central America (and the adjacent waters) and in the northern Indian Ocean. The differences in these regions are large not only in magnitude but also in spatial scale. In September–October– November (SON), model–satellite differences are more localized. Relative to TRMM observations, CCM3 simulates too much precipitation in portions of the Caribbean, the central equatorial Atlantic, equatorial Africa, the central Indian Ocean, and the western equatorial Pacific.
To provide a more detailed view of the time structure of the differences between the model and the data, we compare the fields month by month in each of the 48 boxes described earlier. Note that the 48 boxes range in size from 45° × 8.3° to 45° × 11.2° and thus are large enough to compare monthly means without worrying about the sampling errors associated with the TRMM observations. However, it is important to observe two points regarding these time series. First, the monthly means associated with these boxes are averages over at least 48 2.8° × 2.8° model grid boxes, while the earlier maps of the annual and seasonal means are produced on the model's approximate 2.8° × 2.8° grid. Therefore, some of the details evident on the model's grid can be blurred or hidden in the averages over these larger boxes. Second, due to its orbital precession, TRMM requires at least 1.5 months to completely sample the diurnal cycle of rainfall over a given grid box. Therefore, in a given month, boxes with a strong diurnal cycle may experience most of their rain in hours unobserved by the satellite, creating a low bias in the satellite monthly mean. It has been noted by Lin et al. (2002) that this phenomenon may introduce a spurious signal into a monthly time series. It is difficult to estimate the size of this effect without a thorough investigation of the diurnal cycle simulated by the model. Such an investigation is currently underway.
Time series of monthly mean precipitation rates are shown in Fig. 4. In order to quantitatively measure how well one series tracks the other, the Spearman rank correlation coefficient is computed for each box. It is a nonparametric statistic that assumes no a priori knowledge of the distributions of the monthly means (Deshpande et al. 1995), and its value may span the real numbers from −1 to 1. In Fig. 4, the coefficient is printed in the top left-hand corner of each region's plot. A deficiency of the difference map of Fig. 2c is that it is not useful for discerning the model's underestimations or overestimations relative to what TRMM observes. While the absolute differences are more important for evaluating the model's simulation of latent heating, the relative differences are important for appraising the model's success with precipitation; relative differences can be large in regions where TRMM observes little to no rainfall. Therefore, the absolute and relative differences in total precipitation as simulated by the model and as observed by the satellite are computed for each of the 48 regions. Total precipitation refers to the sum over all 44 months of the monthly mean precipitation rates weighted by the number of days in each month. The absolute difference in total precipitation refers to CCM3 total precipitation − TRMM total precipitation, where the sign of the difference is retained. That is, an absolute difference that is less than zero indicates that when rainfall is totaled over the comparison period, the model exhibits a dry bias. The relative difference is the ratio of this absolute difference to the TRMM total precipitation multiplied by 100. In Fig. 4, the absolute and relative differences are printed in the top right-hand corner of each region's plot, the absolute differences expressed in centimeters per year, and the relative differences in %. We divided the Tropics into six geographical regions, labeled A through F and discuss each region in turn. Where absolute and relative differences in total precipitation are quite large, the region is further subdivided into smaller boxes for closer inspection. However, even these smaller boxes are large enough that we expect TRMM sampling errors to be sufficiently small. Within these subregions, large model–satellite differences are diagnosed as either wet season, dry season, or seasonally invariant, based on relative differences between the monthly means.
a. Region A: The central and eastern Pacific
Over the central and eastern Pacific, CCM3 performs reasonably well in it simulation of precipitation. For example, the model simulates well the annual cycle of precipitation, where it is discernible in the observations. The simulations of boxes 3W, 4W, and 10W are good examples. The simulation of box 4W in the central Pacific is particularly noteworthy in that CCM3 tracks TRMM closely during the relatively wet El Niño period of 1998 and then continues to simulate the normal annual cycle thereafter. In boxes where the model's phase of the monthly precipitation is simulated well, the correlation coefficient is high, as would be expected. Even where the annual cycle is somewhat less pronounced, the model results remain close to the observations, such as in box 6W. However, where the model correctly captures the phase of the annual cycle, its precipitation rates do not always agree so well with the observations in magnitude. Absolute and relative differences are particularly high in boxes 5W, 11W, and 12W. For a better understanding of where and when the largest of the differences occur, these regions are further subdivided, as seen in Fig. 5.
Over this region of the southern tropical Pacific, differences in the monthly means are generally independent of season and are highest in boxes 1 and 2, where, compared to the satellite observations, the model simulates at least 83 cm more precipitation per year during the comparison period. Figure 3 suggests that the large model–satellite differences here are caused by the model's generating a South Pacific convergence zone (SPCZ) which is both too large and too intense. The absolute differences in total precipitation taper toward the South American coast, as do the monthly mean precipitation rates in general. However, relative differences are larger over this part of the region. For example, the model simulates 328% too much precipitation in box 4 and in excess of 100% too much precipitation in box 3. In spite of its difficulties with simulating magnitude in these dry areas, the model is reasonably successful with capturing the annual cycle. The success is most evident in boxes 1 and 2 where it correctly simulates maxima in DJF and minima in JJA. However, the model's maxima and minima are large relative to the observations.
b. Region B: Central America, the Caribbean, the Gulf of Mexico, and the far eastern Pacific
In contrast to region A, where the model–satellite differences are relatively small except in the southernmost boxes, differences are quite large in most of region B, encompassing Central America, the Caribbean, the Gulf of Mexico, and the far eastern Pacific. While the simulation in box 7W is satisfactory, the model exhibits problems with magnitude elsewhere. In box 8W, the model simulates two precipitation peaks per year, which are not seen by the satellite. Overestimation by the model appears both in JJA and in DJF. In boxes 13W and 14W, it is largely confined to JJA. Differences in the monthly means are considerable in box 14W, where simulated precipitation rates are some 3 times higher than observed by TRMM during this period, resulting in excessive rates of near 14 mm day−1. In box 9W, where absolute and relative differences in total precipitation are small, the model appears to be having trouble capturing the annual cycle. This region is further subdivided for a more precise evaluation in Fig. 6.
Over this region, model–satellite differences are much more seasonally dependent than they are over the southern tropical Pacific. In box 5, the model produces 3 times as much precipitation in the wet season despite fair agreement with the satellite observations during the dry season. In box 4, the model's overestimation is as large as 100% in both the wet and dry seasons despite fair agreement during the transitional periods of the year. In box 7, the simulation is relatively satisfactory during the wet season, but significantly wet-biased during the dry season, estimating between 2 and 6 times the satellite-measured precipitation during this time. Box 3 is drier than the other boxes examined here. However, in this box, the relative difference in total precipitation is large. Relative to the observations, the annual maximum is simulated late in the year and is about twice as high. In contrast to boxes 3, 4, 5, and 7, where absolute and/ or relative differences in total precipitation are significant, these differences are almost negligible in box 6. Here, the simulation disagrees with the observations in the phase of the annual cycle. Annual maxima are placed reasonably correctly, but annual minima are consistently about 1/3 of a year (4 months) behind their location in the data. The model is comparatively successful with both magnitude and phase in boxes 1 and 2, except for a model wet bias of around 100% in the August–September time frame. In terms of the total precipitation over the comparison period, the absolute and relative differences in these two boxes are quite a bit lower than those of boxes 4 and 5 to the south.
c. Region C: South America and the tropical Atlantic
Compared to the rest of the tropical Western Hemisphere, the simulation over the continent of South America (boxes 15W–18W) is superior. In general, the CCM3 annual cycle and magnitudes of precipitation are in good agreement with TRMM. Over the northern tropical Atlantic (boxes 19W, 20W, and 21W), agreement is also quite good. Boxes 19W and 20W are dry compared to the rest of region C, and the model's low monthly mean precipitation rates closely follow those observed. Box 21W is considerably wetter throughout the year, but the CCM3 monthly means are still quite close to those observed, and the annual cycle evident here is similarly well simulated. As a whole, region C shares a similarity with central and eastern Pacific region A: the simulation and observations are in good agreement except in the southern oceanic boxes, where simulated magnitudes are too large. The most problematic portion of the region resides in southeastern boxes 22W and 23W, where the model tends to be consistently too wet throughout the period. Absolute and relative differences in total precipitation are quite large in these two boxes. See Fig. 7, in which this area is further partitioned.
Like those of the southern tropical Pacific, the model– satellite differences in the subregions of the southern tropical Atlantic are largely independent of season. Also like those of the southern tropical Pacific, the differences are exclusively ones of magnitude. Where an annual cycle is evident, such as in boxes 1, 2, and 4, CCM3 simulates it well. In this region, the largest absolute difference in total precipitation is in box 1, encompassing extreme eastern South America and the adjacent ocean. Here, the model's monthly means are nearly consistently between 100% and 150% larger than the satellite's monthly means throughout the year, excepting reasonable agreement in November. The same behavior can be seen in box 4, yet, as in box 1, there are single months where the model output and satellite data agree. Relative differences are huge in arid boxes 5 and 6, where there is no annual cycle present. For example, in box 5, over the comparison period, the model simulates 262% too much rain, and in box 6, it simulates over 700% too much, the largest relative difference in total precipitation found in the Tropics of the Western Hemisphere.
d. Region D: Northern Africa and southern Asia
In northern Africa and southern Asia, the comparison does not fare well. In box 3E, the model output and satellite data is not well correlated. CCM3 simulates a semiannual maximum in precipitation which appears unfounded in the TRMM data. Though the model's second precipitation peak (in October) lies relatively close to the observations, the first one greatly exaggerates precipitation in January through April, and the relative minimum between the two peaks actually underestimates precipitation from May to October. A partitioning of this region into smaller boxes, as seen in Fig. 8, shows that the model's behavior with respect to the satellite observations varies considerably on smaller spatial scales.
In equatorial Africa, model–satellite differences are highly spatially dependent. In box 1, while the model agrees well with the satellite during the dry season, it is too dry during the wet season. In box 4, the model is too dry during the wet season and too wet during the dry season, indicating that the monthly means are out of phase. CCM3 overestimates total precipitation in boxes 2, 3, 5, and 6. In box 2, the model's monthly means are in good agreement with those of TRMM during the dry months of DJF, but when there is observed precipitation, they are consistently higher. Exceptions occur during SON of 1998 and 1999. In box 3, the model–satellite differences are seasonally invariant: the model's wet bias is present throughout the year, even in the rainless months of DJF. The model–satellite difference in total precipitation is largest in box 5, where the contrast between the monthly means is dramatic. Here, CCM3 most prominently simulates the semiannual cycle evident in the larger-scale average, but this feature is not well supported by the TRMM data. The model's maxima are consistently higher than those of the observations, and its minima in DJF are far too high.
Returning to Fig. 4b, in south-central Asia and the northern Indian Ocean, the disparity between CCM3 and TRMM is also quite large. In this region, absolute differences in total precipitation are especially large in boxes 8E, 9E, and 10E, and with few exceptions throughout the 44-month period, the model consistently overestimates precipitation in these boxes. This region is subdivided in Fig. 9 for a more detailed consideration.
One of the outstanding features of Fig. 9 is the model's excessive precipitation rates in box 1 (the southern Arabian peninsula) during JJA. Here, the model's overestimation is on the order of 300%. According to TRMM observations, monthly mean precipitation rates are less than 1 mm day−1 every month of the year, in stark contrast to the model's simulation of an average 6 mm day−1 during the month of August. This model– satellite difference can hardly be classified as a wet-season model wet bias, since there is no seasonal variation evident in the TRMM observations. It appears that the model is creating a wet season that does not exist. According to the seasonal-means map (Fig. 3) the source of the artificial wet season in JJA appears to be an extension of a rainfall belt stretching across equatorial Africa, from Guinea on the Atlantic coast, through the Congo, and then northeastward into Saudi Arabia. However, in contrast to CCM3, TRMM data suggest that this belt does not extend as far east and that there is a clearly defined “dry” spot over Saudi Arabia, consistent with its small monthly means.
The precipitation of box 2 is simulated much more satisfactorily. This box is dominated by a monsoon-type climate, wherein strong southwesterlies during the summer months transport moisture from the Indian Ocean into the Indian subcontinent (Pant and Kumar 1997; Kendrew 1953). Here, the model's timing of the onset of the monsoon is about right, according to the TRMM observations. In this box, the model's total wet bias is less than 10 cm yr−1. In boxes 3 to 6, where there is little annual cycle evident, the comparison fares better in boxes 4 and 6 than in boxes 3 and 5. Precipitation in box 6 is simulated remarkably well during all of 1998. By contrast, in boxes 3 and 5, the absolute difference in total precipitation is quite large (over 1 m yr−1), yet it is difficult to discern any systematic pattern to the model's wet bias.
e. Region E: The Maritime Continent and the western tropical Pacific
There appears to be little seasonal variation in the precipitation over the Maritime Continent, both in the observations and in the simulations. In boxes 15E and 16E, monthly means are high throughout the year (>6 mm day−1). There are intermittent periods of model overestimation and underestimation, but the total precipitation received over a course of a year is in good agreement with the observations, relative to the rest of region E. Over the western tropical Pacific, CCM3 does not behave as well. Consider the partitioning of boxes 19E and 20E in Fig. 10 for example.
While the model performs comparatively well in box 1, its wet bias in boxes 2 to 6 lies between 50 and 80 cm yr−1. CCM3 appears to be simulating a strong seasonality to the precipitation in boxes 2 and 3, though such a seasonal variation is not apparent in the observations. Thus, it is difficult to classify the model–satellite differences here, except to conclude that the model appears to behave as it does over the Arabian peninsula, simulating unrealistic wet seasons, during which precipitation rates are too high compared to the observations. In box 6, where an annual cycle appears slightly better supported, the model's wet bias is over 100% during the wet season months.
The most remarkable feature of the model's performance in boxes 21E and 22E of Fig. 4b (partitioned in Fig. 11) is its close agreement with TRMM observations over New Guinea and Papua New Guinea (box 4).
Absolute and relative differences in total precipitation in box 4 are particularly small. However, in the other five subdivisions, the comparison is quite poor. For example, in boxes 1 and 2, CCM3 overestimates DJF precipitation by about 100%. Here, as over the Arabian Peninsula and in subregions of the northwestern tropical Pacific, the model appears to be simulating an unrealistic wet season. The wet bias is of a similar magnitude in box 3, though it resides in JJA, where a wet season actually appears supported by TRMM observations. In boxes 5 and 6, the model appears to be tracking the overall trend in the precipitation observed by the satellite, but there are periods of excessive precipitation, particularly in the latter halves of 1999 and 2000.
f. Region F: The southern Eastern Hemisphere
In contrast to the previous regions discussed, there is very little to be criticized of the model's performance in the southern Eastern Hemisphere. Annual cycles and magnitudes of precipitation are simulated quite well from southern Africa to Australia. Absolute differences in total precipitation are generally less than 0.5 m yr−1 and relative differences are generally less than 60%. Where its measure is useful, the rank correlation between CCM3 and TRMM is high, indicating the model's success in capturing the annual cycle. The largest differences in total precipitation are found in boxes 11E and 12E of the southern Indian Ocean. These are caused by a small, consistent month-to-month model wet bias. Overall, the model's simulation over region F is a testament to the fact that, in certain areas, CCM3 can simulate both wet and dry climates well, since they are both present here; contrast box 23E with box 18E, for example.
4. Summary and conclusions
The results of this study illustrate that there are large regions of the Tropics in which monthly mean precipitation rates, computed from an ensemble of CCM3 simulations, disagree with those computed from observations made by the TRMM satellite. Similarly, there are large regions where CCM3 and TRMM monthly means agree well. We summarize the differences on fairly coarse regional and subregional scales in Fig. 12.
The top two panels (Figs. 12a,b) plot categorical absolute and relative differences in total (44-month) precipitation for the previously examined subregions. The bottom plot (Fig. 12c) labels the differences as occurring in the wet season (W) or in the dry season (D), as phase differences (P), or as differences which are consistent throughout the period or have no discernible seasonal variation to them (C), or which are associated with artificial wet seasons created by the model (A). The model's bias is indicated by a plus sign or a minus sign for wet or dry, respectively.
Where there are differences, they are almost universally positive, indicating the model's wet bias throughout the Tropics. Negative model–satellite differences evident in the annual-difference plot (Fig. 2c) are not nearly as large on the regional and subregional scales considered here; the positive differences are far more prominent. The only exception is in the interior Guinea lands of western Africa (near the intersection of the prime meridian and 15°N). Here, the model consistently underestimates wet-season precipitation. Regions where absolute differences are large and positive correspond to regions where CCM3 simulates too much condensational heating, and unrealistically large amounts of latent energy can have effects on the simulation of the general circulation. In the central South Pacific, the large positive absolute differences between the model and satellite monthly means are most likely caused by the model's generation of a SPCZ which is too large and too intense. Outside this region, the largest positive absolute differences tend to be north of 10°S and along the western edges of the Atlantic, the Indian, and the Pacific Oceans' basins. North of the equator, about half of the largest model–satellite differences are wet biases in the wet season or wet biases in model-generated artificial wet seasons. South of the equator, they are all wet biases with no seasonal dependence.
In locations where absolute differences are large, CCM3 does a poor job of simulating both precipitation and latent heating. However, there are locations where absolute differences are comparatively small but relative differences are extreme. Prime examples include portions of the Arabian peninsula and the southern tropical Atlantic, where absolute differences in total precipitation are less than 75 cm yr−1, but relative differences are greater than 300% and 700%, respectively. In such regions, the model's overrelease of latent energy is not as important as its failure to simulate what is climatologically reasonable “weather.”
It should be noted that the previous analysis has not distinguished between the convective and large-scale stable parts of the model's precipitation. However, the precipitation as simulated by CCM3 in the Tropics is overwhelmingly convective in nature. See Fig. 13, which shows the magnitudes of model-generated precipitation, partitioned into its convective and large-scale stable parts, and TRMM-observed precipitation, totaled over the comparison period. In all regions, the model's large-scale stable portion is smaller than not only its convective portion but also the portion observed by TRMM. Thus the model's wet biases are wet biases in convective precipitation.
In light of the problem regions described earlier, the model's simulation of the annual cycle of precipitation agrees quite well with the TRMM observations over South America and most of the southern Eastern Hemisphere (generally south of 10°S), including Australia and southern Africa. Over these regions, absolute and relative differences between the model- and satellite-derived monthly means are small, and the rank correlations are large. In general, the model's wet bias throughout the Tropics tends to be smaller over the Southern Hemisphere than over the Northern Hemisphere, and it tends to be smaller over land than over ocean. This general conclusion is supported by scatterplots of the regional-mean monthly mean CCM3 and TRMM precipitation rates, specific to hemisphere and land/ocean (see Fig. 14).
In the interest of improving the model, it would be useful to explore how the model's biases in precipitation are related to its biases in features of the general circulation. Unfortunately, describing the “true” general circulation in the Tropics is subject to its own errors since such a description relies on tools like National Centers for Environmental Predictions (NCEP) and European Centre for Medium-Range Weather Forecasts (ECMWF) reanalyses, which are in part based on their own models. Several studies have shown large differences between the reanalyses and observational data and/or large discrepancies between the NCEP and ECMWF reanalyses themselves (see Newman et al. 2000; Rouault et al. 2003; Wu and Xie 2003; and Roads 2003 for examples). Therefore, such an analysis is a complex undertaking and is reserved for a separate study.
Acknowledgments
This material is based upon work supported by a NASA Earth System Science fellowship and by NASA Grant NAG5-4753 to Texas A&M University. The authors thank the Goddard Space Flight Center for making the TRMM satellite data available online via the TRMM Science Data and Information System; the Community Climate Model group at NCAR for developing and assistance with running CCM3; and the Program for Climate Model Diagnosis and Intercomparison at Lawrence Livermore National Laboratory for providing the sea surface temperature dataset in CCM3 format. In addition, the authors thank the anonymous reviewers, whose insightful comments and suggestions helped prepare this manuscript for publication.
REFERENCES
Adler, R. F., G. J. Huffman, D. T. Bolvin, S. Curtis, and E. J. Nelkin, 2000: Tropical rainfall distributions determined using TRMM combined with other satellite and rain gauge information. J. Appl. Meteor, 39 , 2007–2023.
Barnett, T. P., 1995: Monte Carlo climate forecasting. J. Climate, 8 , 1005–1022.
Bell, T. L., and P. K. Kundu, 1996: A study of the sampling error in satellite rainfall estimates using optimal averaging of data and a stochastic model. J. Climate, 9 , 1251–1268.
Bell, T. L., and P. K. Kundu, 2000: Dependence of satellite sampling error on monthly averaged rain rates: Comparison of simple models and recent studies. J. Climate, 13 , 449–462.
Bell, T. L., and P. K. Kundu, 2003: Comparing satellite rainfall estimates with rain gauge data: Optimal strategies suggested by a spectral model. J. Geophys. Res.,108, 4121, doi:10.1029/2002JD002641.
Bell, T. L., P. K. Kundu, and C. D. Kummerow, 2001: Sampling errors of SSM/ I and TRMM rainfall averages: Comparison with error estimates from surface data and a simple model. J. Appl. Meteor, 40 , 938–954.
Bowman, K. P., A. B. Phillips, and G. R. North, 2003: Comparison of TRMM rainfall retrievals with rain gauge data from the TAO/ TRMM buoy array. Geophys. Res. Lett.,30, 1757, doi:10.1029/ 2003GL017552.
Busuioc, A., Hvon Storch, and R. Schnur, 1999: Verification of GCM-generated regional seasonal precipitation for current climate and of statistical downscaling estimates under changing climate conditions. J. Climate, 12 , 258–272.
Cess, R., 1985: Nuclear war: Illustrative effects of atmospheric smoke and dust upon solar radiation. Climatic Change, 7 , 237–251.
Chen, M., R. E. Dickinson, X. Zeng, and A. N. Hahmann, 1996: Comparison of precipitation observed over the continental United States to that simulated by a climate model. J. Climate, 9 , 2233–2249.
Chen, T-C., and M-C. Yen, 1994: Interannual variation of the Indian monsoon simulated by the NCAR Community Climate Model: Effect of the tropical Pacific SST. J. Climate, 7 , 1403–1415.
Deshpande, J. V., A. P. Gore, and E. A. Shanubhogue, 1995: Statistical Analysis of Nonnormal Data. John Wiley & Sons, 240 pp.
Ebert, E., and J. Curry, 1992: A parameterization of ice cloud optical properties for climate models. J. Geophys. Res, 97 , 3831–3836.
Espenshade Jr., E. B., Ed.,. 1995: Goode's World Atlas. 19th ed. Rand McNally, 372 pp.
Hack, J., 1994: Parameterization of moist convection in the National Center for Atmospheric Research Community Climate Model (CCM2). J. Geophys. Res, 99 , 5551–5568.
Holtslag, A. A. M., and B. A. Boville, 1993: Local versus nonlocal boundary-layer diffusion in a global climate model. J. Climate, 6 , 1825–1842.
Kendrew, W. G., 1953: The Climates of the Continents. 4th ed. Clarendon Press, 607 pp.
Kiehl, J., J. Hack, G. Bonan, B. Boville, B. Briegleb, D. Williamson, and P. Rasch, 1996: Description of the NCAR Community Climate Model (CCM3). NCAR Tech. Rep. NCAR/TN-420+STR, Boulder, CO, 152 pp.
Kirkyla, K. I., and S. Hameed, 1989: Harmonic analysis of the seasonal cycle in precipitation over the United States: A comparison between observations and a general circulation model. J. Climate, 2 , 1463–1475.
Kummerow, C., W. Barnes, T. Kozo, J. Shiue, and J. Simpson, 1998: The Tropical Rainfall Measuring Mission (TRMM) sensor package. J. Atmos. Oceanic Technol, 15 , 809–817.
Kummerow, C., and Coauthors, 2000: The status of the Tropical Rainfall Measuring Mission (TRMM) after two years in orbit. J. Appl. Meteor, 39 , 1965–1982.
Lau, K-M., and H. T. Wu, 2001: Principal modes of rainfall–SST variability of the Asian summer monsoon: A reassessment of the monsoon–ENSO relationship. J. Climate, 14 , 2880–2895.
Lin, X., L. D. Fowler, and D. A. Randall, 2002: Flying the TRMM Satellite in a general circulation model. J. Geophys. Res.,107, 4281, doi:10.1029/2001JD000619.
Lindzen, R. S., 1981: Turbulence and stress owing to gravity wave and tidal breakdown. J. Geophys. Res, 86 , 9707–9714.
Liou, K., 1992: Radiation and Cloud Processes in the Atmosphere. Oxford University Press, 487 pp.
Lorenz, E. N., 1963: Deterministic nonperiodic flow. J. Atmos. Sci, 20 , 130–141.
McFarlane, N., 1987: The effect of orographically excited wave drag on the general circulation of the lower stratosphere and troposphere. J. Atmos. Sci, 44 , 1775–1800.
Mo, K. C., and R. W. Higgins, 1998: Tropical influences on California precipitation. J. Climate, 11 , 412–430.
Montroy, D. L., 1997: Linear relation of central and eastern North American precipitation to tropical Pacific sea surface temperature anomalies. J. Climate, 10 , 541–558.
Negri, A. J., and T. L. Bell, 2002: Sampling of the diurnal cycle of precipitation using TRMM. J. Atmos. Oceanic Technol, 19 , 1333–1344.
Newman, M., P. D. Sardeshmukh, and J. W. Bergman, 2000: An assessment of the NCEP, NASA, and ECMWF reanalyses over the tropical west Pacific warm pool. Bull. Amer. Meteor. Soc, 81 , 41–48.
Pant, G. B., and K. R. Kumar, 1997: Climates of South Asia. John Wiley & Sons, 320 pp.
PRB, cited 2003: Population Reference Bureau: World population data sheet. [Available online at http://www.prb.org.].
Ramanathan, V., 1976: Radiative transfer within the earth's troposphere and stratosphere: A simplified radiative-convective model. J. Atmos. Sci, 33 , 1330–1346.
Ramanathan, V., and P. Downey, 1986: A nonisothermal emissivity and absorptivity formulation for water vapor. J. Geophys. Res, 91 , 8649–8666.
Roads, J., 2003: The NCEP–NCAR, NCEP–DOE, and TRMM tropical atmosphere hydrologic cycles. J. Hydrometeor, 4 , 826–840.
Ropelewski, C., and M. Halpert, 1987: Global and regional scale precipitation patterns associated with the EI Niño/Southern Oscillation. Mon. Wea. Rev, 115 , 1606–1626.
Ropelewski, C., and M. Halpert, 1989: Precipitation patterns associated with the high index phase of the Southern Oscillation. J. Climate, 2 , 268–284.
Rouault, M., C. J. C. Reason, J. Lutjeharms, and A. Beljaars, 2003: Underestimation of latent and sensible heat fluxes above the Agulhas Current in NCEP and ECMWF analyses. J. Climate, 16 , 776–782.
Shin, K-S., and G. R. North, 1988: Sampling error study for rainfall estimate by satellite using a stochastic model. J. Appl. Meteor, 27 , 1218–1231.
Simpson, J., R. F. Adler, and G. R. North, 1988: A proposed Tropical Rainfall Measuring Mission (TRMM) satellite. Bull. Amer. Meteor. Soc, 69 , 278–294.
Slingo, J., 1987: The development and verification of a cloud prediction scheme for the ECMWF model. Quart. J. Roy. Meteor. Soc, 113 , 899–927.
Slingo, J., 1989: A GCM parameterization for the shortwave radiative properties of water clouds. J. Atmos. Sci, 46 , 1419–1427.
Sperber, K., and T. Palmer, 1996: Interannual tropical rainfall variability in general circulation model simulations associated with the Atmospheric Model Intercomparison Project. J. Climate, 9 , 2727–2750.
Taylor, K. E., D. Williamson, and F. Zwiers, 2000: The sea surface temperature and sea-ice concentration boundary conditions for AMIP II simulations. Tech. Rep. 60, Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Laboratory, 25 pp.
Tracton, M. S., and E. Kalnay, 1993: Operational ensemble prediction at the National Meteorological Center: Practical aspects. Wea. Forecasting, 8 , 379–398.
Trigo, R. M., and J. P. Palutikof, 2001: Precipitation scenarios over Iberia: A comparison between direct GCM output and different downscaling techniques. J. Climate, 14 , 4422–4446.
Vogelzang, D., and A. Holtslag, 1996: Evaluation and model impacts of alternative boundary-layer height formulations. Bound.-Layer Meteor, 81 , 245–269.
Wu, R., and S-P. Xie, 2003: On equatorial Pacific surface wind changes around 1977: NCEP–NCAR reanalysis versus COADS observations. J. Climate, 16 , 167–173.
Zhang, G. J., and N. A. McFarlane, 1995: Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Centre general circulation model. Atmos.– Ocean, 33 , 407–446.
Zhang, G. J., J. T. Kiehl, and P. J. Rasch, 1998: Response of climate simulation to a new convective parameterization in the National Center for Atmospheric Research Community Climate Model (CCM3). J. Climate, 11 , 2097–2115.
(a) Map of the 48 regions of comparison and monthly mean precipitation rates in mm day−1 for Jan 1998–Aug 2001 as simulated by the eight members of the CCM3 ensemble for (b) 14W and (c) 15E. Gray curves represent the individual member means, and the heavy black curve represents the ensemble mean
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
(a) Map of the 48 regions of comparison and monthly mean precipitation rates in mm day−1 for Jan 1998–Aug 2001 as simulated by the eight members of the CCM3 ensemble for (b) 14W and (c) 15E. Gray curves represent the individual member means, and the heavy black curve represents the ensemble mean
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
(a) Map of the 48 regions of comparison and monthly mean precipitation rates in mm day−1 for Jan 1998–Aug 2001 as simulated by the eight members of the CCM3 ensemble for (b) 14W and (c) 15E. Gray curves represent the individual member means, and the heavy black curve represents the ensemble mean
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Climatological annual mean precipitation rate in mm day−1 (a) as simulated by CCM3 and (b) as observed by TRMM (TMI;th+;thPR) for the period 1 Jan 1998–31 Aug 2001. (c) The climatological mean difference (CCM3 − TRMM) in mm day−1. Note that all results are on the model's approximate 2.8° × 2.8° grid
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Climatological annual mean precipitation rate in mm day−1 (a) as simulated by CCM3 and (b) as observed by TRMM (TMI;th+;thPR) for the period 1 Jan 1998–31 Aug 2001. (c) The climatological mean difference (CCM3 − TRMM) in mm day−1. Note that all results are on the model's approximate 2.8° × 2.8° grid
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Climatological annual mean precipitation rate in mm day−1 (a) as simulated by CCM3 and (b) as observed by TRMM (TMI;th+;thPR) for the period 1 Jan 1998–31 Aug 2001. (c) The climatological mean difference (CCM3 − TRMM) in mm day−1. Note that all results are on the model's approximate 2.8° × 2.8° grid
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Precipitation rates (mm day−1) simulated by (top) CCM3 and those estimated by (bottom) TRMM (TMI + PR) data averaged over the four meteorological seasons, from 1998– 2001. Note that all results are on the model's approximate 2.8° × 2.8° grid.> >
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Precipitation rates (mm day−1) simulated by (top) CCM3 and those estimated by (bottom) TRMM (TMI + PR) data averaged over the four meteorological seasons, from 1998– 2001. Note that all results are on the model's approximate 2.8° × 2.8° grid.> >
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Precipitation rates (mm day−1) simulated by (top) CCM3 and those estimated by (bottom) TRMM (TMI + PR) data averaged over the four meteorological seasons, from 1998– 2001. Note that all results are on the model's approximate 2.8° × 2.8° grid.> >
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Time series of monthly mean precipitation rates in mm day−1 as simulated by CCM3 and as observed by TRMM from Jan 1998 through Aug 2001 for 48 separate regions in the Tropics. The solid black curve is the CCM3 ensemble-average mean, and the dashed black curve is the TRMM mean. The rank correlation coefficient is given in the upper left-hand corner of each box while the absolute and relative differences between CCM3 and TRMM total precipitation are given in the upper right-hand corner of each box. Absolute differences are given in cm (per year), and relative differences are given in %
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Time series of monthly mean precipitation rates in mm day−1 as simulated by CCM3 and as observed by TRMM from Jan 1998 through Aug 2001 for 48 separate regions in the Tropics. The solid black curve is the CCM3 ensemble-average mean, and the dashed black curve is the TRMM mean. The rank correlation coefficient is given in the upper left-hand corner of each box while the absolute and relative differences between CCM3 and TRMM total precipitation are given in the upper right-hand corner of each box. Absolute differences are given in cm (per year), and relative differences are given in %
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Time series of monthly mean precipitation rates in mm day−1 as simulated by CCM3 and as observed by TRMM from Jan 1998 through Aug 2001 for 48 separate regions in the Tropics. The solid black curve is the CCM3 ensemble-average mean, and the dashed black curve is the TRMM mean. The rank correlation coefficient is given in the upper left-hand corner of each box while the absolute and relative differences between CCM3 and TRMM total precipitation are given in the upper right-hand corner of each box. Absolute differences are given in cm (per year), and relative differences are given in %
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
CCM3 (solid) and TRMM (dashed) monthly mean precipitation rates in mm day−1 for the southern tropical Pacific Ocean (boxes 5W, 11W, and 12W of Fig. 4a). Rank correlation coefficients are given in the upper left-hand corner of each box, while the absolute and relative differences between CCM3 and TRMM total precipitation are given in the upper right-hand corner of each box. Absolute differences are given in cm (per year), and relative differences are given in %
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
CCM3 (solid) and TRMM (dashed) monthly mean precipitation rates in mm day−1 for the southern tropical Pacific Ocean (boxes 5W, 11W, and 12W of Fig. 4a). Rank correlation coefficients are given in the upper left-hand corner of each box, while the absolute and relative differences between CCM3 and TRMM total precipitation are given in the upper right-hand corner of each box. Absolute differences are given in cm (per year), and relative differences are given in %
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
CCM3 (solid) and TRMM (dashed) monthly mean precipitation rates in mm day−1 for the southern tropical Pacific Ocean (boxes 5W, 11W, and 12W of Fig. 4a). Rank correlation coefficients are given in the upper left-hand corner of each box, while the absolute and relative differences between CCM3 and TRMM total precipitation are given in the upper right-hand corner of each box. Absolute differences are given in cm (per year), and relative differences are given in %
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the Central American–Caribbean–far eastern Pacific region (boxes 7W, 8W, 9W, 13W, and 14W of Fig. 4a)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the Central American–Caribbean–far eastern Pacific region (boxes 7W, 8W, 9W, 13W, and 14W of Fig. 4a)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the Central American–Caribbean–far eastern Pacific region (boxes 7W, 8W, 9W, 13W, and 14W of Fig. 4a)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the southern tropical Atlantic region (boxes 22W and 23W of Fig. 4a)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the southern tropical Atlantic region (boxes 22W and 23W of Fig. 4a)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the southern tropical Atlantic region (boxes 22W and 23W of Fig. 4a)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the equatorial Africa region [boxes 2E (southern half) and 3E of Fig. 4b]
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the equatorial Africa region [boxes 2E (southern half) and 3E of Fig. 4b]
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the equatorial Africa region [boxes 2E (southern half) and 3E of Fig. 4b]
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the south-central Asia–northern Indian Ocean region (boxes 7E–10E and western portions of boxes 13E and 14E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the south-central Asia–northern Indian Ocean region (boxes 7E–10E and western portions of boxes 13E and 14E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the south-central Asia–northern Indian Ocean region (boxes 7E–10E and western portions of boxes 13E and 14E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the northwestern tropical Pacific Ocean region (boxes 19E and 20E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the northwestern tropical Pacific Ocean region (boxes 19E and 20E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the northwestern tropical Pacific Ocean region (boxes 19E and 20E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the western equatorial Pacific Ocean region (boxes 21E and 22E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the western equatorial Pacific Ocean region (boxes 21E and 22E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Same as in Fig. 5 except for the western equatorial Pacific Ocean region (boxes 21E and 22E of Fig. 4b)
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Categorical (a) absolute and (b) relative differences between CCM3 and TRMM total precipitation for the 44-month comparison period. (c) Types of model–satellite differences are plotted, where a plus or a minus sign refers to the model's wet or dry bias, respectively. Each location corresponds to one of the subregions examined above
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Categorical (a) absolute and (b) relative differences between CCM3 and TRMM total precipitation for the 44-month comparison period. (c) Types of model–satellite differences are plotted, where a plus or a minus sign refers to the model's wet or dry bias, respectively. Each location corresponds to one of the subregions examined above
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Categorical (a) absolute and (b) relative differences between CCM3 and TRMM total precipitation for the 44-month comparison period. (c) Types of model–satellite differences are plotted, where a plus or a minus sign refers to the model's wet or dry bias, respectively. Each location corresponds to one of the subregions examined above
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Bar charts representing the magnitudes of 44-month total precipitation in cm yr−1 as simulated by CCM3 and as observed by TRMM. The CCM3 precipitation bar (on the left) is partitioned into a convective part (black) and large-scale stable part (white). The percentage of the CCM3 precipitation that is convective is written above the CCM3 bar. The TRMM (TMI+PR) precipitation bar is represented in dark gray to the right
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Bar charts representing the magnitudes of 44-month total precipitation in cm yr−1 as simulated by CCM3 and as observed by TRMM. The CCM3 precipitation bar (on the left) is partitioned into a convective part (black) and large-scale stable part (white). The percentage of the CCM3 precipitation that is convective is written above the CCM3 bar. The TRMM (TMI+PR) precipitation bar is represented in dark gray to the right
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Bar charts representing the magnitudes of 44-month total precipitation in cm yr−1 as simulated by CCM3 and as observed by TRMM. The CCM3 precipitation bar (on the left) is partitioned into a convective part (black) and large-scale stable part (white). The percentage of the CCM3 precipitation that is convective is written above the CCM3 bar. The TRMM (TMI+PR) precipitation bar is represented in dark gray to the right
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Scatterplots of CCM3 and TRMM regional-mean monthly mean precipitation rates (mm day−1) for (a) Northern Hemisphere, (b) Southern Hemisphere, (c) ocean, and (d) land regions. Rank correlation coefficients are printed in the lower right-hand corner of each plot. The solid black line represents the line of unit correlation
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Scatterplots of CCM3 and TRMM regional-mean monthly mean precipitation rates (mm day−1) for (a) Northern Hemisphere, (b) Southern Hemisphere, (c) ocean, and (d) land regions. Rank correlation coefficients are printed in the lower right-hand corner of each plot. The solid black line represents the line of unit correlation
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
Scatterplots of CCM3 and TRMM regional-mean monthly mean precipitation rates (mm day−1) for (a) Northern Hemisphere, (b) Southern Hemisphere, (c) ocean, and (d) land regions. Rank correlation coefficients are printed in the lower right-hand corner of each plot. The solid black line represents the line of unit correlation
Citation: Journal of Climate 17, 17; 10.1175/1520-0442(2004)017<3319:ACOTPS>2.0.CO;2
The data for this study are derived from observations at its preboost altitude.