Tropical Precipitation in Relation to the Large-scale Circulation. Courtney Schumacher

Transcription

1 Tropical Precipitation in Relation to the Large-scale Circulation Courtney Schumacher A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy University of Washington 2003 Program Authorized to Offer Degree: Department of Atmospheric Sciences

2 University of Washington Abstract Tropical Precipitation in Relation to the Large-scale Circulation by Courtney Schumacher Chair of the Supervisory Committee: Professor Robert A. Houze, Jr. Department of Atmospheric Sciences This study uses Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) observations from to characterize the convective and stratiform components of the tropical precipitation field. National Centers for Environmental Prediction (NCEP) reanalysis fields of relative humidity, temperature, and zonal wind are jointly analyzed with the TRMM PR observations to highlight the large-scale environmental characteristics of stratiform rain (i.e., rain associated with deeper cloud systems and in which precipitation processes occur above the 0ºC level). This study also models the steady-state atmospheric response to the four-dimensional latent heating field implied by the TRMM PR tropics-wide convective and stratiform precipitation climatology. The TRMM PR provides an illuminating view of the ensemble of precipitating clouds across the tropics; shallow convective rain elements dominate the outer fringes of the tropical rain area but give way to deeper, more organized convective systems and associated stratiform areas toward heavy rain regions. The PR shows geographical and temporal variations in the proportion of rain that is stratiform, most notably a trans-pacific gradient with low proportions of stratiform rain over the maritime continent and high proportions over the eastern-central Pacific. This gradient becomes even more pronounced during El Niño. Variations in stratiform rain production may be attributed to the environment s ability to sustain moderate convection and differences in zonal wind shear and momentum transport. Horizontal variations in the vertical heating profile implied by the PR stratiform rain fraction pattern lead to variations in the height and vertical extent of circulation anomalies not present when the model is forced with latent heating derived from a geographically-uniform stratiform rain fraction. Accurate representation of convective and stratiform precipitation (as detected by the TRMM PR) is essential in modeling a realistic large-scale circulation response to precipitating cloud systems in the tropics.

3 TABLE OF CONTENTS List of Figures ii List of Tables vi Chapter 1: Introduction Chapter 2: The mechanics of convective-stratiform rain classification TRMM Precipitation Radar Rain type classification and rain estimation Reclassification of shallow, isolated echo Retrieval uncertainties in the PR convective-stratiform rain statistics Long-term validation at Kwajalein Chapter 3: Stratiform rain climatology Zonal averages Annual-mean patterns Land versus ocean Seasonal patterns Interannual variations Quasi-steady circulations Chapter 4: Large-scale environmental characteristics of stratiform rain Relative humidity Temperature Zonal wind Chapter 5: Latent heating and the tropical dynamical response Latent heating estimates Model Annual-mean experiments Seasonal cycle El Niño Effects of nonprecipitating convection and cloud radiative forcing Chapter 6: Conclusions References i

4 LIST OF FIGURES Number Page 2.1 PR 2.5º observations from for annually averaged a) rain accumulation, b) stratiform pixel count (rain types 10-14), c) convective pixel count (rain types 20-25), and d) shallow, isolated pixel count (rain types 15, 26-29) PR 2.5º average stratiform rain fraction from based on a) TRMM product 2A23 version 5 stratiform (rain types 10-15) and convective (rain types 20-29) classifications and b) TRMM product 2A23 version 5 convective-stratiform classifications with the stratiform shallow, isolated pixels (rain type 15) considered convective Kwajalein radar (KR) and PR monthly averages within a 150 km radius of the Kwajalein radar from June-December 1999 and 2000 for a) rain accumulation and b) stratiform rain fraction PR 2.5º zonal averages from for a) stratiform rain fraction and b) rain accumulation PR a) total rain, b) convective rain, c) stratiform rain, and d) stratiform rain fraction based on 2.5º grid averages for PR mean a) conditional convective rain rate, b) conditional stratiform rain rate, and c) convective-stratiform (CS) rain rate ratio based on 2.5º grid averages for Two-dimensional histograms of PR monthly stratiform rain fraction versus a) stratiform area fraction, b) CS rain rate ratio, and c) rain accumulation with bin sizes of two, 0.35 and 14, respectively Two-dimensional histograms of PR monthly stratiform rain fraction versus the monthly mean a) convective rain rate over land, b) stratiform rain rate over land, c) convective rain rate over ocean, and d) stratiform rain rate over ocean Seasonal PR stratiform rain fraction based on 2.5º grid averages for Annual averages of PR stratiform rain fraction based on 2.5º grid averages for 1998, 1999, and, ii

5 Number Page 3.8 PR a) stratiform rain fraction and b) rain accumulation longitudinally averaged over 20ºN-20ºS for January-April 1998 (El Niño) and January-April 1999 (La Niña) Quasi-steady circulation regions defined as I (intertropical convergence zone), M (seasonal monsoon regions), and C (semi-permanent equatorial convective zones) a) The average relative humidity profile for the monsoon, semi-permanent equatorial convection (C), and ITCZ circulation regimes assuming at least 50 mm mo -1 of rain accumulation. b) The difference in relative humidity between when the stratiform rain fraction is greater than and less than the 2.5º grid element mean The vertical gradient in relative humidity differences between higher and lower stratiform rain fraction conditions from a) mb, b) mb, c) mb, and d) mb a) The average temperature profile for the monsoon, semi-permanent equatorial convection (C), and ITCZ circulation regimes assuming at least 50 mm mo -1 of rain accumulation. b) The difference in temperature between when the stratiform rain fraction is greater than and less than the 2.5º grid element mean The vertical gradient in temperature differences between higher and lower stratiform rain fraction conditions from a) mb, b) mb, and c) mb a) The difference in Reynolds and Smith SST when the stratiform rain fraction is greater than and less than each 2.5º grid element mean for b) As in a) except that the El Niño months of JFMA 1998 were excluded a) The average zonal wind profile for the monsoon, semi-permanent equatorial convection (C), and ITCZ circulation regimes assuming at least 50 mm mo -1 of rain accumulation. b) The difference in zonal wind between when the stratiform rain fraction is greater than and less than the 2.5º grid element mean The vertical gradient in zonal wind differences between higher and lower stratiform rain fraction conditions from a) mb and b) mb iii

6 Number Page 5.1 PR 2.5º annually averaged rain accumulation from The precipitation has been linearly decreased from values at 20ºN and S to zero at 35ºN and S in order to isolate the tropical heat source a) Idealized stratiform (SF), deep convective (DC), and shallow convective (SC) latent heating profiles. b) Total latent heating profiles assuming 3.6 m yr -1 rain accumulation Horizontal distributions of the annually averaged latent heating at a) 7.4 km (~400 mb) and b) 2.2 km (~800 mb) derived from the TRMM PR rain-type fractions and tapered rain amounts. c) Vertical cross section of annually averaged latent heating at 10ºN The 400 mb latent heating (shaded) and the resulting 250 mb streamfunction anomalies (contours) from a model run using a resting basic state forced with the heating derived from the PR annually averaged precipitation field and geographically uniform stratiform rain fractions of a) 0% and b) 40% Vertical cross sections of ω (shaded) and zonal wind (contours) anomaly fields averaged along the equator from 8.5ºN-8.5ºS from a model run using a resting basic state forced by heating derived from the PR annually averaged precipitation and stratiform rain fractions of a) 0%, b) 40%, c) 70%, and d) PR-observed The 250 mb streamfunction anomalies from a model run using the NCEP seasonal basic state forced with the heating derived from the PR seasonally averaged precipitation and stratiform rain fraction fields for a) JJA and b) DJF. The NCEP eddy streamfunctions for c) JJA and d) DJF The 400 mb latent heating (shaded) and the resulting 250 mb streamfunction anomalies (contours) from a model run using the NCEP basic state for a) La Niña (JFMA 1999) and b) El Niño (JFMA 1998) forced with the latent heating derived from the PR precipitation and stratiform rain fraction fields for the same periods Vertical cross sections of ω (shaded) and zonal wind (contours) anomaly fields along the equator from 8.5ºN-8.5ºS from a model run using the NCEP basic state for El Niño (JFMA 1998) forced by heating derived from the PR-observed precipitation and stratiform rain fractions of a) 40%, b) 70%, and c) PR-observed iv

7 Number Page 5.9 Idealized latent heating profiles of nonprecipitating cumulus (dotted) in a) deep and b) shallow convective rain Idealized cloud radiative forcing profiles for shallow convective, deep convective, and stratiform rain areas assuming 100% cloud cover Profiles of total heating (including nonprecipitating convection and CRF as well as precipitating clouds) for 0, 40, and 70% stratiform rain fraction and 0, 75, 90% stratiform rain area fraction assuming 3.6 m yr -1 rain accumulation, 90% cloud cover, and 10% of the rain and rain area is shallow convective Vertical cross sections of ω (shaded) and zonal wind (contours) anomaly fields along the equator from 8.5ºN-8.5ºS for a) model results based on PR DJF precipitation and stratiform rain fraction, b) NCEP DJF reanalysis, and c) model results based on PR DJF precipitation and stratiform rain fraction with additional heating from nonprecipitating convection and cloud radiative forcing v

8 LIST OF TABLES Number Page 2.1 TRMM product 2A23 version 5 rain type classifications Tropical mean convective-stratiform rain statistic sensitivities to PR reflectivity threshold Tropical mean convective-stratiform rain statistic sensitivities to the exponent, b, in the Z-R relation Land versus ocean tropical mean convective-stratiform rain statistics stratiform rain fractions for quasi-steady circulation regions vi

9 Acknowledgements I d first like to thank my husband, David Kalil, whose love, support, and culinary skills kept me happy and healthy throughout my graduate years. I also very much appreciate the encouragement of Dave s and my family. I thank my advisor, Bob Houze, for being such an excellent friend and mentor. I reserve the role model spot for Marcia Baker. Stacy Brodzik and other members of the Mesoscale Group provided both wonderful friendship and scientific camaraderie. The other graduate students in the Atmospheric Sciences department were indispensable. I can t imagine all of these years without my first year classmates, Rolling Thunder, the Supercool Hydrometeors, the powerful bonds with the other women atmospheric scientists, and the assorted and sundry coffee breaks and lunches that kept us sane. vii

10 1 Chapter 1. Introduction There are two main classifications of tropical precipitation: convective and stratiform. These classifications are based on the microphysical growth processes of precipitation particles and on the vertical distribution of latent heating associated with precipitation processes (see, e.g., Houze 1982, 1989, 1997). The convective classification refers to regions where precipitation is falling from young, active convection (i.e., regions of strong, non-hydrostatic vertical motions). Numerous updrafts on the order of meters per second refresh the cloud liquid water content such that the droplets and ice particles growing in these regions increase in mass through collection processes, namely, coalescence and/or riming. The stratiform classification refers to regions of older, less active convection, where weaker vertical air motions predominate, and precipitation particles increase in mass primarily through vapor deposition. Some uses for convective-stratiform partitioning are to validate cumulus parameterizations (Donner et al. 2001), provide data for assimilation in general circulation models (Hou et al. 2001), and further refine area-time integral (ATI) rainfall estimation (Doneaud et al. 1984; Yuter and Houze 1998). Previous observational studies delineating convective and stratiform precipitation in the tropics have focused on individual events or locations (e.g., Cheng and Houze 1979; Gamache and Houze 1983; Churchill and Houze 1984; Houze and Rappaport 1984; Leary 1984; Chong and Hauser 1989; Goldenberg et al. 1990; Steiner et al. 1995; Short et al. 1997; Yuter and Houze 1998; Schumacher and Houze 2000; Rickenbach et al. 2002). The fraction of rain that was stratiform in these studies varied from 25-85%, suggesting that the relative contribution of convective and stratiform precipitation varies spatially and/or temporally across the tropics. Until now there has been no clear indication of the spatial and temporal variations of convective and stratiform precipitation beyond these regional studies carried out for limited time periods. In November 1997, NASA and the Japanese space agency, NASDA, launched the Tropical Rainfall Measuring Mission (TRMM) satellite. Aboard the satellite was the first quantitative Precipitation Radar (PR) to be placed in space. The PR is perhaps the best available remote-sensing instrument with which to differentiate regions of convective and

11 2 stratiform precipitation because of its ability to see the three-dimensional structure of the precipitation field at high horizontal and vertical resolution. Reasonable agreement has been found between instantaneous PR convective-stratiform observations and other airborne and ground-based radar convective-stratiform observations (Heymsfield et al. 2000; Schumacher and Houze 2000; Liao et al. 2001). The coverage of the TRMM PR (35ºN- 35ºS) allows multi-year regional intercomparisons of convective and stratiform precipitation across the tropics and subtropics with data available since the launch of the satellite. One of the primary goals of TRMM was to determine the four-dimensional distribution of latent heating in the tropics in order to better understand global climate (Simpson et al. 1988). There has always been a lack of high-resolution data of tropical heating estimates over large time and space scales. It is possible to calculate the horizontal and vertical distribution of total tropical diabatic heating as a residual using the equations of motion and observed winds and temperatures based on reanalysis fields (e.g., Valdes and Hoskins 1989; Nigam 1994; Wang and Ting 1999), but sounding data are sparse over the tropics, and the accuracy of this method is somewhat uncertain. It is also difficult to separate the total heating field into the latent, sensible, and radiative heating components using the residual method. Bergman and Hendon (1998, 2000) calculated tropical cloud radiative heating using International Satellite Cloud Climatology Project (ISCCP) observations and a radiative transfer model. They then subtracted the radiative heating from the total heating calculated as a budget residual to obtain the latent heating (assuming the sensible heating to be negligible), but the latent heating remains dependent on the residual calculation. Tao et al. (1990, 1993) and Olson et al. (1999) introduced methods to derive the horizontal and vertical tropical latent heating field from satellite passive microwave observations in conjunction with cloud modeling. However, passive microwave radiometers provide very limited information regarding the vertical distribution of hydrometeor types on which such methods are dependent. Houze (1982, 1989) demonstrated that the relative proportions of convective and stratiform rain amounts amount can be used to infer the vertical structure of latent heating in tropical precipitating systems. The heating associated with convective rain is positive

12 3 throughout the profile with the height of maximum heating depending on the size spectrum of the convective elements (Houze et al. 1980; Johnson et al. 1999). The heating profile of stratiform regions is dominated by heating above the 0ºC level (~5 km in the tropics) and cooling below. These profiles combine to produce strong net heating in the mid-to-upper troposphere and weak net heating in the lower troposphere. Higher fractions of stratiform rainfall are associated with an upward shift in the level of maximum heating and an increase in the vertical heating gradient in the upper troposphere. Using observations of surface rainfall separated into convective and stratiform components, the TRMM PR is able to provide an estimate of the horizontal and vertical structure of latent heating across the tropics on seasonal to annual time scales. The ability to estimate the four-dimensional distribution of latent heating allows the large-scale dynamical response to horizontal variations in the vertical heating profile associated with tropical precipitating systems to be investigated because the generation of potential vorticity in the tropics is directly proportional to the local vertical gradient in the heating profile (Haynes and McIntyre 1987; Mapes and Houze 1995). Hartmann et al. (1984) determined that the large-scale response to the heat released by tropical precipitation required a heating profile based on an ensemble of both convective and stratiform elements. The work of Hartmann et al., along with other studies that have investigated how variations in the vertical structure of tropical heating influence the dynamical response (Geisler 1981; DeMaria 1985; Sui and Lau 1989; Wu et al. 2000; Chiang et al. 2001), assume geographically uniform heating profiles. However, heating profiles of precipitating cloud systems have been observed to vary over space and time across the tropics (e.g., Thompson et al. 1979; Frank and McBride 1989). This dissertation presents a three-year climatology ( ) of tropical convective-stratiform rain statistics observed by the TRMM PR, relates this climatology to National Centers for Environmental Prediction (NCEP) reanalysis fields of relative humidity, temperature and zonal wind, and examines the large-scale circulation response of a climate model to the spatially varying heating profiles implied by the TRMM-observed fields of rainfall and stratiform rain fraction.

13 4 Chapter 2. The mechanics of convective-stratiform rain classification This chapter describes the TRMM PR and its capability to subdivide precipitation into convective and stratiform components. This chapter also explains a modification to the official TRMM PR convective-stratiform separation algorithm necessary to produce more realistic distributions of convective and stratiform precipitation in regions of small rain accumulation. Uncertainties in PR stratiform rain fraction statistics due to the scanning geometry and sensitivity of the PR, the choice of a reflectivity-rain rate (Z-R) relation, and temporal sampling of the satellite are described. In addition, Kwajalein ground radar observations are compared to PR observations over Kwajalein for two rainy seasons. 2.1 TRMM Precipitation Radar The PR scans 17º to either side of nadir at intervals of 0.35º with a vertical resolution of 250 m. Until August 7, 2001 the TRMM satellite had an operating altitude of 350 km giving the PR a swath width of 215 km and a horizontal footprint of 4.3 km at nadir. The TRMM satellite has an orbital domain extending from 35ºN-35ºS and has a precessing orbit so that it samples the full diurnal cycle. The PR operates at K u band (2.17 cm wavelength) and is thus subject to strong attenuation. Owing to power constraints, the PR has a relatively low sensitivity of ~17 dbz (which corresponds to a precipitation rate of approximately 0.4 mm h -1 ). However, the PR can detect some reflectivities down to 14 dbz (particularly over the ocean) but not in proportion to their actual occurrence (Schumacher and Houze 2000). This study uses only data > 17 dbz to avoid the ambiguity of sampling at lower reflectivities. The PR is downward-looking but it cannot sense rain directly above the ground because of contamination from the surface return. Thus, this study uses the reflectivity observed closest to the surface that is free from clutter (normally 1-2 km). Further details on the TRMM satellite and the PR can be found in Kummerow et al. (1998) and Kozu et al. (2001). Uncertainties in stratiform rain statistics due to the scanning geometry and sensitivity of the PR are discussed in Secs. 2.3 a-c.

14 5 2.2 Rain type classification and rain estimation Convective precipitation regions are generally identified with intermittently strong vertical velocities ( w > 1 m s -1 ), high rain rates (> 5 mm h -1 ), and small (~1-10 km horizontal dimension), intense, horizontally inhomogeneous radar echo. Stratiform precipitation areas are characterized by statistically small vertical velocities ( w < 1 m s -1 ), low rain rates (< 5 mm h -1 ), and widespread (~100 km horizontal dimension), horizontally homogeneous radar echo. The microphysical processes dominating the growth of particles to precipitation size in convective regions is collection, while the dominant precipitation growth mechanism in stratiform regions is vapor diffusion (Houze 1993). Extensive stratus and stratocumulus cloud decks that have tops well below the 0ºC level, and occasionally produce drizzle or light rain, are considered stratiform by standard terminology (Glickman 2000). However, the 17 dbz sensitivity threshold of the TRMM PR guarantees that most, if not all, of this stratiform rain is excluded from this study. The stratiform rain to which this study refers is from deeper cloud systems evolving from or attached to deep convective clouds. In these stratiform clouds, the primary precipitation processes occur in the ice layer above the 0ºC level. Fallout and evaporation occur below this level. Radar reflectivity observations cannot always be unambiguously separated into regions of convective and stratiform precipitation, which occasionally leads to a third transition or intermediary classification. Using wind data from Doppler radar and soundings, Mapes and Houze (1995) showed that only two modes account for most of the large-scale atmospheric response to convection and that these modes correspond to the divergence signatures in convective and stratiform precipitation regions. The transition/ intermediary region therefore represents an inability to definitively separate precipitation into convective and stratiform components in the radar reflectivity field, not a physically distinct phenomenon. Houze (1997) argues that the transition/intermediary region is more closely related to the stratiform region in terms of microphysical and dynamical processes such that the two categories in this study represent precipitation from young, vigorous convection (convective) and precipitation from older, weaker convection (transition/inter-

15 6 mediary and stratiform). This study uses a modified version of the PR convective-stratiform rain classifications from version 5 of TRMM product 2A23 1 (Awaka et al. 1997). The PR 2A23 algorithm determines whether the echo is convective or stratiform based on the vertical profile of reflectivity (from which the brightband, echo top height, and maximum reflectivity in the vertical profile are identified) and the horizontal variability of the echo. The horizontal variability criterion within the algorithm follows Steiner et al. (1995). The PR convectivestratiform algorithm also assigns echo to a category called other when there is no brightband detected, the convective reflectivity threshold is not met, and any of the observations below the 0ºC level are noise. Thus, the other category represents either noise or regions of precipitation aloft with no precipitation near the surface. Because of the ambiguity of the other category and its very small contribution to total rain (< 0.2% when using the 17 dbz threshold), this study uses only reflectivity defined as convective or stratiform. It is difficult to validate quantitatively the convective-stratiform echo classification; however, uncertainty is expressed qualitatively in the PR algorithm by assigning to each data point a confidence category based on how strongly the point satisfies the horizontal and vertical criteria. In version 5 of TRMM product 2A23, convective confidence is separated into ten categories, and stratiform confidence into six categories (Table 2.1). In addition, the PR convective-stratiform separation algorithm designates pixels as shallow, isolated when the echo top is lower than the climatological 0ºC level by more than 1.5 km and the pixel is separate from other rain-certain areas. Version 5 of TRMM product 2A23 places the majority of the shallow, isolated pixels in a stratiform subcategory; however, as will be argued in Sec. 2.3, the shallow, isolated echoes most likely represent warm rain processes and as such should be classified as convective. This study reclassifies the stratiform shallow, isolated pixels as convective in order to obtain more reasonable patterns of stratiform rain contribution over the tropical oceans. Version 6 of the algorithm will reflect this refinement (personal communication, J. Awaka 2002). Since the convective-stratiform designation of some of the echoes has changed, the 1. TRMM products are available at daac.gsfc.nasa.gov.

16 7 rain rate at each pixel is recomputed. This study places the near-surface reflectivity from version 5 of TRMM product 2A25 (Iguchi et al. 2000) into 1 db bins at 2.5º resolution for each month in The reflectivity is converted into rain rate using version 5 convective and stratiform initial Z-R relations of Z=148R 1.55 and Z=276R 1.49, respectively. In the PR 2A25 algorithm, these Z-R relations are subject to adjustment, based on an attenuation correction and an index of spatial nonuniformity. For simplicity, the calculation is done using the initial Z-R relations without adjustment. Sec. 2.4d addresses some of the uncertainty resulting from the use of different Z-R relations. The average convective and stratiform rain rates in each monthly 2.5º grid box are then multiplied by the probability of rain for each rain type to obtain the convective and stratiform rain accumulations. For most of this study, regions that have less than an equivalent of 0.6 m yr -1 of average rain accumulation are not included in order to decrease uncertainty from sampling. Stratiform rain and rain area fraction may be artificially inflated when the total rain accumulation is too low. The 0.6 m yr -1 accumulation threshold gives less noisy results in the stratiform rain statistics. This thresholding is further supported by Bell and Kundu (2000) if the relative sampling error for monthly rain amounts is extended to stratiform rain statistics. The stratiform rain statistics are calculated using the histograms of rain information taken in aggregate for the regions and time periods under consideration, i.e., stratiform rain statistics are based on total rain volume when considering multiple 2.5º grids. 2.3 Reclassification of shallow, isolated echo According to the Glossary of Meteorology (Glickman 2000), warm rain is rain formed from a cloud having temperatures at all levels above 0ºC, and resulting from the droplet coalescence process. In order for rain to fall from a cloud with a top below the 0ºC level, the amount of collision/coalescence required would need to occur in a convective cloud (Houze 1993). Stratus or stratocumulus would be insufficient for any precipitation other than drizzle (which, as noted in Sec. 2.2, is not observable by the PR due to sensitivity issues). The Glossary distinguishes warm rain from the warm rain process,

17 8 which is defined as growth by collision-coalescence and limitations to growth by drop breakup... Moreover, the warm rain process occurs in clouds having sufficient liquid water, updraft, and lifetime to sustain collision-coalescence... and is found to be active in both shallow and deep convection... Thus, warm rain is produced by the warm rain process, but the process may occur in deeper convective clouds. From these definitions, it follows that the warm rain process is likely associated with convective clouds. Stratiform precipitation, as described above, occurs in the deep ice-cloud regions of previously more active convective cloud. The warm rain process does not occur in such cloud regions. The last stratiform subcategory (rain type 15) and the last four convective subcategories (rain types 26-29) have an additional designation of shallow, isolated (Table 2.1). Based on the physical arguments above, the stratiform shallow, isolated subcategory (rain type 15) is likely warm rain and as such should be considered to be a convective subcategory. As will be shown later in this section, the spatial pattern of rain type-15 echoes across the tropics is also consistent with these echoes being convective rather than stratiform and addresses the occurrence of anomalously low stratiform echo heights that Short and Nakamura (2000) found in their study of TRMM-observed shallow precipitation. For each 2.5º x 2.5º grid element in the TRMM PR domain, the number of pixels > 17 dbz were tabulated for each convective and stratiform subcategory listed in Table 2.1 for Maps of annually averaged convective and stratiform pixel counts were constructed excluding the shallow, isolated subcategories (Figs. 2.1 b and c). The annually averaged convective and stratiform shallow, isolated pixel counts were combined to obtain Fig. 2.1d. The PR average annual rain for the three years is shown as reference in Fig. 2.1a. These maps illustrate how the population of precipitating clouds varies over the tropics. Figures 2.1 b and c show that the stratiform and convective patterns are qualitatively similar to each other across the tropics. Both patterns resemble the tropical rainfall pattern in Fig. 2.1a; regions of maximum stratiform and convective pixel counts occur in regions of high rain accumulation. These similarities are consistent with the greatest rain accumulations being produced by mesoscale systems containing both deep convection and

18 9 stratiform precipitation. The shallow, isolated pattern (Fig. 2.1d) is quite different from the stratiform and convective patterns, with maximum pixel counts over ocean regions where the rain accumulation is low. This category gives an illuminating indication of where the precipitating cloud population consists mainly of shallow convective clouds (probably cumulus congestus and isolated cumulonimbus). Comparison of Figs. 2.1 b-d shows how the regime of primarily shallow precipitating convective clouds populating the outer edges of the tropical rain region gives way to deeper convection toward the center of the tropical rain region. Comparison of Figs. 2.1b and c shows further that stratiform precipitation is intimately connected to the deeper convection. Figure 2.2a shows the average stratiform rain fraction (i.e., the percent of total rainfall accounted for by stratiform precipitation) for based on version 5 TRMM product 2A23 convective-stratiform classifications. Figure 2.2a indicates high (> 60%) stratiform rain fractions in regions of very low rain accumulation (c.f. Fig. 2.1a) over northern Africa, the southeast tropical Indian Ocean, the equator in the central Pacific, the northeast tropical Pacific off the coast of Baja, the southeast Pacific off the coast of South America, and the subtropical south Atlantic. This pattern is suspect, as high stratiform rain fractions wouldn t be expected outside the main precipitation zones but rather in the centers of the rainy areas, where deep convection and mesoscale convective systems thrive. Very high stratiform rain fractions are also observed in broad regions with low to moderate precipitation poleward of 20ºN and 20ºS; these features are most likely associated with extratropical systems occasionally affecting these latitudes. Figure 2.2b is the same as Fig. 2.2a except that it has been modified by placing the shallow, isolated stratiform subcategory (rain type 15) in the convective classification. The modified pattern of stratiform rain fraction is more physically reasonable. The maxima over the low rain accumulation regions of the tropical oceans have all been reduced substantially. Thus, the gradient of stratiform rain fraction has been reversed in these areas and appears to be more reasonable since large mesoscale convective systems in areas peripheral to the main precipitation zones of the tropics are not expected. The intertropical convergence zone (ITCZ) has more moderate stratiform rain fractions and is more clearly

19 10 differentiated from regions outside of the ITCZ, especially the double ITCZ structure in the central-to-eastern Pacific (which is discussed in more detail in Sec. 3.4). The stratiform rain fractions in regions of low to moderate rain accumulation poleward of 20º N and 20ºS remain high but have become distinct from precipitation regions closer to the equator. This separation by latitude is reasonable since the belts of high stratiform fraction poleward of 20º N and 20ºS appear to be caused by extratropical frontal systems. These systems are probably more stratiform, as indicated, even though the validity of the convective/stratiform separation algorithm (tuned for tropical convection) is questionable at these latitudes. A few regions with very low precipitation rates still exhibit unrealistically high stratiform rain fractions (e.g., northern Africa, 10-40ºE, 20-35ºN, and the northeast Pacific, ºW, 20-30ºN). Since the stratiform rain fraction is the ratio of stratiform rain to total rain, areas of extremely low rain accumulation can produce artifacts of high stratiform rain fraction. Although shallow precipitating clouds do not contribute much to overall tropical rain accumulation, they can locally account for a significant amount of the total rain (Short and Nakamura 2000). As seen in Fig. 2.2, the reclassification of shallow, isolated echo substantially affects the deduced pattern of convective and stratiform rain proportions. For reasons noted in Sec. 1, the proportion of rain that is stratiform is a direct indicator of how the vertical profile of heating varies over the tropics. An error in this stratiform proportion leads to an error in the pattern of the vertical profile of heating over the tropics. An incorrect convective-stratiform classification can also lead to errors in other convective-stratiform rain-based statistics (e.g., storm height distributions and rain rate intensities). The misclassification of rain type 15 echo is partly a result of the PR s low sensitivity. The PR algorithm is based partially on the horizontal-texture method of Steiner et al. (1995), which assumes a pixel of radar echo is convective only if it exceeds a specified high intensity or stands out against the background echo intensity. The Steiner et al. (1995) technique assumes that if the echo at a given pixel location is weak, and there are no surrounding background pixels with detectable echo, the pixel is not convective because the echo fails both the intensity and peakedness criteria. The pixel therefore gets

20 11 classified as stratiform by default. This horizontal method of Steiner et al. (1995) was based on the assumed availability of radar data with a full range of sensitivity to weak rain (as is usually the case with a ground-based radar). However, the low sensitivity of the PR to weak rain most likely causes the PR algorithm to alias weak, isolated convection into the stratiform category because pixels with substantial rain rates are often left without a background with which to compare their intensities. Data from the Kwajalein Atoll (~9ºN, 168ºE) ground radar (see Sec. 2.5 for details on the Kwajalein radar data set) were used to assess the performance of the convectivestratiform algorithm when reflectivity < 17 dbz is excluded. While Kwajalein is not located in the main regions of shallow, isolated echo seen in Fig. 2.1d, the atoll receives moderate amounts of rain type 15. Two versions of the convective-stratiform algorithm were applied to all the ground radar data in December The first version uses all of the echo observed by the Kwajalein radar and the second version uses only echo > 17 dbz. About 3% less pixels were classified as convective when the 17 dbz threshold was applied. This 3% decrease led to a 1% increase in the percent of total rain that was stratiform. This result does not change when data are interpolated to 2 km x 2 km or 4 km x 4 km horizontal resolution. The misclassification of rain type 15 echo can also be attributed to the horizontal resolution of the PR. Convective cells are normally on the order of 1-2 km in horizontal dimension such that they would appear less intense when observed with the PR s 4 km footprint. The convective-stratiform algorithm depends on the intensity of the echo and it is difficult to tune the algorithm to unambiguously classify echo as convective at low reflectivities. When the Kwajalein radar data is interpolated to 4 km x 4 km horizontal resolution instead of 2 km x 2 km, the stratiform rain fraction increases by ~10%. 2.4 Retrieval uncertainties in the PR convective-stratiform rain statistics a. Scanning geometry and wavelength of the PR The quasi-vertically pointing PR has relatively good vertical resolution (250 m at

21 12 nadir) with which to define the brightband, a good indicator of stratiform rain. However, brightband detection efficiency is beam-angle dependent and can be as low as 20% at the antenna scan edge (TRMM PR algorithm instruction manual, Version 1). The PR s horizontal resolution is > 4 km which is marginal for defining individual convective cells (normally on the order of 1-2 km). The PR s horizontal resolution became ~5 km after the increase in operating altitude in August 2001, exacerbating this problem. The PR operates at an attenuating wavelength (K u band) that significantly reduces the apparent intensity of convective cells near the surface. The PR s convective-stratiform separation (Awaka et al. 1997) is applied to reflectivity data that are not corrected for attenuation. Therefore, the PR s low horizontal resolution and attenuation may lead the algorithm to alias some convective echoes into the stratiform category. b. Changes of reflectivity with height The convective-stratiform proportions may differ depending the elevation at which the reflectivity is observed because of the differential change of reflectivity with height in convective and stratiform profiles (Zipser and Lutz 1994; DeMott and Rutledge 1998; Steiner and Houze 1998). Over land and ocean, stratiform reflectivity profiles tend to be constant with height below the 0ºC level. Over the tropical oceans, convective reflectivities increase toward the surface while over land in the tropics, the maximum in the convective profile is often elevated to a height between the surface and the 0ºC level. The horizontal method in the PR convective-stratiform algorithm uses the lowest level that is free from surface clutter (ranging from 1 km at nadir to 2 km at the antenna scan edge). Thus, over the ocean, one would expect the stratiform rain fraction to decrease toward the surface while the opposite trend would occur over land. c. PR sensitivity The PR has limited sensitivity to very light precipitation (~17 dbz). While it can capture all but ~3% of surface rain amount, it can miss up to 50% of the rain area greater than 0 dbz observed by a ground radar (Schumacher and Houze 2000). In addition, the

22 13 low sensitivity of the PR affects the average convective and stratiform rain rates (especially compared to what a rain gauge or ground radar might measure). Table 2.2 shows the trend that a decreasing PR sensitivity would have on tropical mean stratiform rain statistics 1. Assuming a 22 dbz reflectivity threshold, stratiform rain fraction decreases to 38%, stratiform rain area fraction decreases to 69%, convective rain rate increases to 8.3 mm h - 1, stratiform rain rate increases to 2.3 mm h -1 and the CS rain rate ratio decreases to 3.6. An opposite trend should occur with increasing sensitivity but cannot be determined quantitatively with PR data. d. Z-R relation In a study of the Darwin, Australia, coastal radar, Steiner and Houze (1997) found that the uncertainty in the ground radar s monthly stratiform rain fraction due only to the choice of Z-R relation is about + 20%. Half of this uncertainty is from the choice of the exponent, b, while the other half is from the choice of whether to use a single or multiple Z-R relations. Work by Ciach et al. (1997) and Yuter and Houze (1997) suggests that for radar echoes over the tropical ocean there is no statistical justification for applying separate convective and stratiform Z-R relations when convective and stratiform rain areas are distinguished solely by typical radar observations. The choice of b is therefore more relevant to this discussion. Table 2.3 shows that the range of the mean tropical stratiform rain fraction is 42-50% when the exponent, b, in a single Z-R relation (300R b ) is varied between 1.3 and 1.6. These b values represent expected climatological values (Doelling et al. 1998; Steiner and Smith 2000). Similarly, the mean convective rain rate ranges from mm h -1, the mean stratiform rain rate ranges from mm h -1 and the CS rain rate ratio ranges from when the b is varied between 1.3 and 1.6. Stratiform rain area fractions are not affected by the choice of the Z-R relation. The PR uses separate convective and stratiform Z-R relations but its estimate of mean stratiform rain fraction is similar to the single Z-R relation means when b = 1.3. This study uses the initial Z-R rela- 1. The mean stratiform rain statistics in Tables 2.2 and 2.3 are defined and discussed in Sec. 3.2.

23 14 tions from the PR algorithm because the resulting stratiform rain statistics are similar to what is available to the broader community; however, there is as yet no definitive evidence supporting the use of a single or double Z-R relation for rain rate estimation. 2.5 Long-term validation at Kwajalein Although Kwajalein represents only one point in the tropics, it is a useful exercise to validate PR observations for monthly time scales with a continuously operating ground radar. The Kwajalein validation site (8.72ºN, ºE) is equipped with an S-band (10 cm), dual-polarization, Doppler weather radar with a beamwidth of 1.12º that collects a three-dimensional volume of data roughly every 10 min (Schumacher and Houze 2000). Operationally, the Kwajalein radar (KR) data is interpolated to 2 km in the horizontal and 1.5 km in the vertical. Monthly rain maps are created using all available radar data (if less than 75% of the monthly data is available, no rain map is created). First, the reflectivity values of the KR are calibrated to be consistent with those of the PR by matching the area covered by echo > 17 dbz at 6 km (a height not likely to be affected by attenuation). Second, climatological convective and stratiform reflectivity profiles based on data within 50 km of the KR from over 1300 radar volumes are applied to correct for the height of the lowest beam above the ground. Third, the radar reflectivity data is converted to rain rate using a single climatological Z-R relation, Z=175R 1.5. Finally, a gap-filling algorithm is employed to account for missing data. These steps are further described online at and Houze et al. (2003). When comparing monthly observations of the PR and KR, some caveats exist. The KR convective-stratiform algorithm utilizes only horizontal information (at 2-km resolution) while the PR convective-stratiform algorithm utilizes both horizontal information (at 4-km resolution) and vertical information (at 250-m resolution). The discrete elevationangle scans of the ground radar do not provide sufficient vertical resolution to identify the brightband across the whole area covered by the radar. However, brightbands observed close to the radar were used in the tuning of the algorithm (Steiner et al. 1995; Yuter and

24 15 Houze 1997). The KR monthly rain and stratiform rain fractions are calculated with nearly continuous sampling while the PR visits are relatively infrequent. The KR reflectivities have been corrected to represent surface values whereas the PR reflectivities represent values at near surface. There is no explicit correction in version 5 of the PR rain profiling algorithm to bring the PR reflectivities to the surface. The KR rain amounts are based on one local Z-R relation applied to both convective and stratiform precipitation, while the PR rain amounts are based on two global relations, one for convective precipitation, the other for stratiform. Figure 2.3 compares the average monthly values of rain and stratiform rain fraction as seen by the PR and the KR from July to December 1999 and 2000 over the Kwajalein validation site (a 150-km radius around the ground radar or approximately a 2.5º grid). The intermittent sampling of the PR causes greater month-to-month fluctuations (see Shin and North 1988 and Bell et al for discussion concerning calculations of satellite sampling error) compared to the practically continuous observations by the ground radar. Over the two six-month periods, the PR observed an average monthly rain accumulation of 161 mm and an average stratiform rain fraction of 46% while the KR showed an average monthly rain accumulation of 203 mm and an average stratiform rain fraction of 34%. Thus, the PR observes less rain and a higher stratiform rain fraction than the KR. A correction to extend PR reflectivities to the surface would increase the PR monthly rain amounts and decrease the stratiform rain fraction, bringing the two radars into better agreement. If the Kwajalein climatological Z-R relation is used for the PR data, the PR average rain amounts and stratiform rain fractions both increase. Therefore, the monthly rain amounts of the two radars become even closer but the stratiform rain fraction bias remains, suggesting that the PR may be overclassifying rain as stratiform.

25 16 a) Rain accumulation 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W 20 N b) Stratiform pixels (10 4 ) EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W 20 N c) Convective pixels (10 3 ) EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W 20 N d) Shallow, isolated pixels (10 3 ) EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 2.1 PR 2.5º observations from for annually averaged a) rain accumulation, b) stratiform pixel count (rain types 10-14), c) convective pixel count (rain types 20-25), and d) shallow, isolated pixel count (rain types 15, 26-29).

26 17 a) TRMM product 2A23 stratiform rain fraction 20 N EQ 20 S 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) Modified stratiform rain fraction 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 2.2 PR 2.5º average stratiform rain fraction from based on a) TRMM product 2A23 version 5 stratiform (rain types 10-15) and convective (rain types 20-29) classifications and b) TRMM product 2A23 version 5 convective-stratiform classifications with the stratiform shallow, isolated pixels (rain type 15) considered convective. The color bar represents stratiform rain fractions between 15-75%, black indicates values > 75%, white indicates values < 15%.

27 18 a) b) Figure 2.3 Kwajalein radar (KR) and PR monthly averages within a 150 km radius of the Kwajalein radar from June-December 1999 and 2000 for a) rain accumulation and b) stratiform rain fraction. See text for processing of each data set.

28 19 Table 2.1: TRMM product 2A23 version 5 rain type classifications. 2A23 Stratiform 2A23 Convective 2A23 Other 10 V=sf (BB exists) H=sf 11 V=sf (BB exists) H=other 12 V=other (BB possible) H=SF 13 V=sf (BB exists) H=conv 14 V=other (no BB) H=sf 15 V=other H=sf shallow, isolated 20 V=conv (no BB) H=conv 21 V=other H=conv 22 V=conv H=other 23 V=conv (BB exists) H=conv 24 V=conv H=sf 25 V=sf (BB possible) H=conv 26 V=conv H=conv shallow, isolated 27 V=conv H=other shallow, isolated 28 V=other H=conv shallow, isolated 29 V=conv H=sf shallow, isolated 30 V=other H=other 31 V=other H=other shallow, isolated

29 20 Table 2.2: tropical mean convective-stratiform rain statistic sensitivities to the PR reflectivity threshold. Reflectivity threshold (dbz) Stratiform rain fraction (%) Stratiform area fraction (%) Conditional convective rain rate (mm h -1 ) Conditional stratiform rain rate (mm h -1 ) CS rain rate ratio Table 2.3: tropical mean convective-stratiform rain statistic sensitivities to the exponent, b, in the Z-R relation. b Stratiform rain fraction (%) Conditional convective rain rate (mm h -1 ) Conditional stratiform rain rate (mm h -1 ) CS rain rate ratio PR initial

30 21 Chapter 3. Stratiform rain climatology This chapter presents a three-year climatology ( ) of tropical convectivestratiform statistics observed by the TRMM PR. The statistics are posed in terms of the percent of rain area classified as stratiform (stratiform rain area fraction), the percent of total rain accounted for by stratiform precipitation (stratiform rain fraction), and the ratio between convective and stratiform rain rates (convective-stratiform [CS] rain rate ratio). The analysis emphasizes stratiform rain fraction because of the direct link between rain and total column latent heating. The discussion focuses primarily on patterns observed in the three-year climatology that lead to an interpretation of the role of stratiform rain contributions in the general circulation of the tropics. 3.1 Zonal averages Figure 3.1 shows the latitude dependence of PR stratiform rain fractions and rain accumulations between 35ºN and 35ºS for Zonally averaged stratiform rain fractions are nearly constant at ~40% from 20ºN-20ºS and rise rapidly (up to values > 50%) in latitudes outside of this equatorial zone (Fig. 3.1a). As mentioned in Chapter 2, regions outside of 20ºN and S can receive much of their total rain from midlatitude frontal systems, especially during the local winter. The convective-stratiform algorithm is tuned for the tropics and may not be applicable to systems subject to extratropical influences. Zonally averaged rain amounts are low (~0.7 m yr -1 ) and relatively constant outside of 20ºN-20ºS while rain amounts are high (up to 1.7 m yr -1 ) within 20ºN-20ºS (Fig. 3.1b). Since rain accumulations outside of the equatorial region are a factor of two to three less than inside of the equatorial region (in agreement with other rain climatologies, Janowiak 1992), the bulk of the net latent heating occurs from 20ºN-20ºS. Because of possible issues of algorithm stability (outside the tropics and in areas of low rain accumulation) and a desire to concentrate on regions that contribute the most to latent heating, this chapter will focus on stratiform rain statistics between 20ºN and 20ºS.

31 Annual-mean patterns a. Stratiform rain fraction As would be expected from the widely varying estimates of tropical stratiform rain fraction previously found in the literature (25-85%), stratiform rain amount and rain fraction have considerable geographical variability in the PR averages between 20ºN and 20ºS for Figure 3.2 shows the annually averaged total rain, convective rain, stratiform rain, and stratiform rain fraction for the three-year period. Total rain accumulations are highest (>2.5 m yr -1 ) over central Africa, the maritime continent, and Central and South America. Well-defined ITCZ are located in the Pacific and Atlantic Oceans between 5-10ºN. Convective rain amounts are also highest (> 1.5 m yr -1 ) over central Africa, the maritime continent, the and Central and South America. Stratiform rain amounts tend to be more evenly distributed over land and ocean with values rarely > 1.5 m yr -1. Figure 3.2d shows the stratiform rain fractions resulting from the above rain distributions. Figure 3.2d is based on the same data as Fig. 2.2b except that areas with annually averaged rain of less than 0.6 m yr -1 were excluded because of sampling concerns (Sec. 2.2) and the color scale represents a smaller range of values than in Fig. 2.2b in order to highlight the stratiform rain fraction variations within the equatorial belt. The mean stratiform rain fraction for the three-year period is 40% for 20ºN-20ºS. In general, land regions tend to have lower stratiform rain fractions while oceans have higher fractions. Stratiform rain fractions are lowest (20-30%) over central Africa, parts of the maritime continent, and the Caribbean with other regions < 35% over the Arabian Sea, the Bay of Bengal, eastern Brazil, and the east Atlantic. Stratiform rain fractions are highest (55-60%) over the ITCZ region of the eastern-central Pacific. The contrast between this maximum of 55-60%, over the eastern-central Pacific, and the stratiform rain fractions of 25-30% over the maritime continent have profound implications for the mean east-west circulation over the tropical Pacific (the Walker cell) since these differences imply a strong difference in the vertical profile of latent heating between the western and eastern equatorial Pacific. Chapter 5 will

32 23 address this topic. Physical interpretations of the variations in stratiform rain fraction across the tropics will be discussed further in Secs In particular Sec. 3.5 will discuss how the strong west-east gradient of stratiform rain fraction across the Pacific becomes more pronounced during El Niño. b. Stratiform rain area fraction Stratiform rain area fraction is defined as the percent of total rain area covered by stratiform rain. While the percent of rain mass that is stratiform indicates the vertical structure of net latent heating by the precipitating clouds, the percent of rain area covered by stratiform rain indicates the vertical structure of the cloud radiative forcing within the precipitating region (Houze 1982, 1989; also, see Sec. 5.6). Geographical variations in stratiform rain area fractions are very similar to variations observed in the stratiform rain fractions and are thus not shown. The relationship between stratiform rain and rain area fractions is discussed in Sec. 3.2d. The average stratiform rain area fraction between 20ºN and 20ºS for is 73%. Therefore, while stratiform precipitation accounts for less than half of the total rain accumulation, it covers the majority of the raining region in the tropics. c. Convective/stratiform (CS) rain rate ratio The convective-stratiform (CS) rain rate ratio is defined as the mean convective rain rate divided by the mean stratiform rain rate over the 2.5º grid. Rain rates are conditional, i.e., only instantaneous observations of raining pixels are used in the rain rate average. Previous radar studies in the tropics (with horizontal resolutions ranging from 2-4 km) have observed that average convective rain rates range from 9-14 mm h -1 and average stratiform rain rates range from 1-4 mm h -1 (Gamache and Houze 1983; Leary 1984; Bell and Suhasini 1994; Steiner et al. 1995; Yuter and Houze 1997). The CS rain rate ratios from the above studies range from As with the stratiform rain and rain area fractions, there is considerable geographical variability in the PR convective and stratiform rain rates and CS rain rate ratio averages

33 24 between 20ºN and 20ºS for Convective rain rates range from 3-20 mm h -1 and are highest (> 10 mm h -1 ) over the continents and more moderate (5-7 mm h -1 ) over the oceans (Fig. 3.3a). Stratiform rain rates range from mm h -1 ; however, there is little difference between land and ocean (Fig. 3.3b). The CS rain rate ratio ranges from 2-14 and is > 5 over most land areas and < 4 over much of the oceans (Fig. 3.3c). The overall average convective rain rate observed by the PR is 7.3 mm h -1 while the average stratiform rain rate is 1.8 mm h -1. The resulting mean CS rain rate ratio is 4.1. A notable anomaly from the land-versus-ocean paradigm is the values of about 4 over the Amazon. This region of lower values extends in over the continent from the Atlantic Ocean and is neatly encircled on the west by high CS rain rate ratios over the Andes. The lower ratios over the Amazon basin are consistent with other studies (Nesbitt et al. 2000; Petersen and Rutledge 2001), which indicate that convection over the Amazon basin is, climatologically, somewhat weaker (or perhaps smaller-scale) than typical continental convection. One reason for this fact may be the extension of oceanic environmental characteristics over the land. Garstang et al. (1994) showed the movement of large oceanic mesoscale systems into the Amazon from the Atlantic. The Amazon canopy may also provide the boundary layer some oceanic characteristics (Garstang et al. 1990). Using data collected from a C-band radar located at approximately 10ºS, 62.5ºW during the TRMM Large-Scale Biosphere Atmosphere (LBA) field campaign, Rickenbach et al. (2002) showed that convective rain rates were lower with the presence of the South Atlantic Convergence Zone (SACZ), while the stratiform rain rates had little regime dependence. Thus, the CS rain rate ratio during LBA decreased dramatically when the stationary fronts that characterize the SACZ were present. This conclusion is consistent with the lower CS rain rate ratios seen by the PR over the southwestern Amazon in Fig. 3.3c. More general physical explanations for the convective and stratiform rain rate variations are addressed in Secs. 3.3 and 3.4.

34 25 d. Relationships among stratiform rain statistics Figure 3.4 highlights the range of monthly stratiform rain fractions and explores the relationship of stratiform rain fractions to monthly stratiform rain area fraction, CS rain rate ratio, and rain accumulation. Only 2.5º grids that are between 20ºN to 20ºS and have rain accumulations greater than 50 mm mo -1 during are considered in the two-dimensional histograms. Monthly stratiform rain fractions range up to 80% while monthly stratiform rain area fractions are rarely less than 40% (Fig. 3.4a). The correlation between monthly stratiform rain and stratiform rain area fractions is 0.81 which implies that stratiform rain fractions relate closely to the amount of area covered by stratiform rain. Monthly CS rain rate ratios range from 2-18 (Fig. 3.4b), similar to the range of CS rain rate ratios in the literature (2.6-14). The correlation between monthly stratiform rain fraction and CS rain rate ratio is which implies that stratiform rain fractions also relate to the relative rain rate intensity between convective and stratiform rain. A relatively lower convective rain rate and/or a higher stratiform rain rate are associated with higher stratiform rain fractions. Histograms that show the convective and stratiform rain rates separately versus stratiform rain fraction over land and ocean are shown in Fig. 3.5 (Sec. 3.3). Figure 3.4c shows that although the correlation between stratiform rain fraction and monthly rain is low (0.13), there is a tendency for the average monthly stratiform rain fraction to increase with increasing rain amounts; the modal value increases from 35% at 50 mm mo -1 to 50% at 500 mm mo -1. Steiner and Houze (1998) found a similar trend in Darwin, Australia, for daily area rainfall. This relationship moreover suggests a tendency for the latent heating in the tropics to become more concentrated at upper levels as rain amounts increase. 3.3 Land versus ocean Studies involving lightning observations clearly indicate a difference in convective intensities between land and ocean with the majority of lightning occurring over land (Jor-

35 26 gensen and LeMone 1989; Lucas et al. 1994; Zipser and Lutz 1994). These studies are corroborated by works that have classified other indicators of convective strength over land and ocean, e.g., height of the 30 dbz contour, maximum reflectivity at 6 km, and precipitation ice water contents (Nesbitt et al. 2000; Petersen and Rutledge 2001). Also, Steiner et al. (1995) found that near the northern Australian coastline radar-observed land and ocean stratiform rain fractions differed by 13% during the monsoon season. Table 3.1 contains bulk percentages of selected quantities for the latitude belt of 20ºN-20ºS for , split between land and ocean. Stratiform rain accounts for less of the total rain over land (35%) compared to the ocean (43%) but covers slightly more of the total rain area (75% vs. 72%). Average convective rain rates are much higher over land (10.2 mm h -1 ) than ocean (5.8 mm h -1 ). However, stratiform rain rates are of the same intensity leading to much different land and ocean CS rain rate ratios (5.5 and 3.3, respectively). The differences between land and ocean stratiform statistics are most likely related to variations in environmental factors, in particular: land vs. ocean surface properties, boundary-layer thermodynamics, lapse rate in the free atmosphere, and wind shear and relative humidity through the depth of the troposphere. A quantitative investigation of some of these factors is addressed in Chapter 4. However, some insight may still be gained into the stratiform statistics by qualitative awareness of the general differences in these environmental factors between land and ocean. When monthly conditional convective rain rates are strong (> 10 mm h -1 ) over land, stratiform rain fractions are negatively correlated with the convective rain intensity (Fig. 3.5a). One might expect the stronger convection implied by higher convective rain rates to result in larger stratiform rain areas and higher stratiform rain fractions. However, the relationship between the convective rain rate and stratiform rain fraction indicates that the occurrence of significant stratiform precipitation regions is not necessarily indicated by the rain rates of their associated convective cells. Over ocean, average monthly convective rain rates are rarely > 10 mm h -1 so the stratiform rain fraction appears relatively insensitive to convective rain rates (Fig. 3.5c). As long as deep convection is present, ocean regions seem proficient at producing stratiform rain, even if the convective rain rates

36 27 are less than those found over land. Over both land and ocean, stratiform rain rates tend to increase with increasing stratiform rain fraction (Fig. 3.5b and d). The effective production of stratiform precipitation over the oceans probably results from the near moist adiabatic stratification of the free atmosphere in convective regions (Xu and Emanuel 1989) and/or the sustainability of the convection by a warm, moist boundary layer with weak diurnal variation (Yuter and Houze 1998; Kingsmill and Houze 1999). The nearly moist adiabatic lapse rates over tropical oceans imply that buoyancy and vertical velocity at lower levels are reduced over ocean compared to over land, where daytime heating creates lower tropospheric stratification that is closer to dry adiabatic. Nonetheless moderate buoyancy exists at all levels, and convective cells over the ocean often extend to great heights. The generally lower convective rain rates in deep cells over oceans may imply greater generation and transfer of ice hydrometeors aloft into the stratiform regions of oceanic mesoscale convective systems (MCSs). Moreover, the potential for deep convection over the ocean exists through the night, so that it is possible to sustain an MCS for a long time. Because land is subject to a diurnal cycle that shuts down the ability of convective cells to form and stratiform regions to grow, precipitation systems over land do not obtain the stratiform rain fractions observed over the ocean unless some process overcomes the diurnal cycle. In addition, variations in the profiles of relative humidity and wind of the large-scale environment between land and ocean may affect stratiform rain production. In order to illustrate the above properties, three regions of different stratiform rain fractions are highlighted: Africa, the west Pacific warm pool, and the east Pacific ITCZ. Over central Africa (30ºE), strong (Fig. 3.3a) but probably short-lived convection evidently produces large rain amounts (Fig. 3.2a) but relatively little stratiform rain (Fig. 3.2c). Two scenarios concerning the microphysics and dynamics of the cloud systems are as follows: 1) the strong (possibly more erect) updrafts related to the large convective rain rates over Africa produce large quantities of water at low levels that rain out and large ice particles at upper levels that fall out, leaving less ice aloft with which to form a robust stratiform region, or 2) the strong (possibly more sheared) updrafts transport more water/

37 28 ice to upper levels, providing sufficient material to form an extensively raining stratiform region; however, larger-scale environmental factors (such as lower buoyancy aloft or dry air entering into the developing stratiform region) keep the stratiform precipitation region from attaining its full potential. Over northern Africa, the subsiding, dry mid-level Sahara air layer may particularly inhibit the development of moist stratiform regions. The former scenario suggests high convective precipitation efficiency at the expense of the stratiform region; the latter scenario suggests low stratiform precipitation efficiency despite sufficient condensate provided by the convective region. Moreover, the boundary layer stabilizes at night, limiting the time available to build a stratiform region. These factors disfavor the development of large, long-lived mesoscale systems with robust stratiform rain regions. Over the west Pacific warm pool (135ºE), weaker (Fig. 3.3a) but probably more frequent convection produces large rain amounts (Fig. 3.2a) with higher stratiform rain contributions (Fig. 3.2c) than over Africa. Adapting the above two scenarios to the west Pacific: 1) over the equatorial ocean, the vertical velocities associated with the weaker convective rain rates are more moderate (LeMone and Zipser 1980; Zipser and LeMone 1980) yet may extend to high levels to produce smaller ice particles that accumulate in the upper levels and form large stratiform regions, or 2) the more moderate updrafts will transport less water/ice to upper levels but environmental factors, such as a more uniform buoyancy through the depth of the troposphere, less evaporation, and cooperative shear, will allow a large precipitating stratiform region to form. The former scenario suggests a low convective precipitation efficiency that promotes stratiform rain production and the latter scenario suggests a high stratiform precipitation efficiency that overcomes the lessened influx of hydrometeors from the convective region. These processes can continue into and through the night since the boundary layer over the ocean undergoes only slight diurnal variation. The western tropical Pacific is known for large, long-lasting MCSs, termed superclusters by Nakazawa (1988) and super-convective systems by Chen et al. (1996). Over the eastern-central Pacific (135ºW), the total rain is less than over Africa and the west Pacific (Fig. 3.2a) but the stratiform rain fraction is greatest (Fig. 3.2d). The convection is weaker than over Africa (Fig. 3.3a) and possibly not as sustained as the convection over the west

38 29 Pacific warm pool. However, the convective cells in the eastern-central Pacific are able to generate sufficient ice to create and maintain stratiform regions that might not seem large compared to the MCSs observed in the west Pacific, but still occur often enough to accumulate fairly large rain amounts. 3.4 Seasonal patterns a. Continents Stratiform rain fractions vary temporally as well as geographically. Over land, the seasonal stratiform rain fractions range from 20-45% over Africa (Fig. 3.6, 0-40ºE) and 20-50% over Central and South America (Fig. 3.6, 90-40ºW); the highest fractions occur with the local seasonal solar maximum. Laing and Fritsch (1993a) and Velasco and Fritsch (1987) observed the highest frequency of mesoscale convective complexes (MCCs) 1 at times of seasonal maximum insolation over Africa and Central and South America, respectively. MCCs are often associated with a nocturnal low-level jet constrained by topography. This jet feeds high moist static energy air into the boundary layer, and mesoscale convective systems can grow into the night. This process overcomes the limitation of stratiform rain formation by the diurnal cycle. This result and reasoning suggest that larger stratiform rain fractions over land during the season of maximum solar insolation reflect the frequency of very large, organized precipitation systems (such as MCCs). Regions that are subject to monsoon circulations also have a distinct seasonal signal in stratiform rain fraction. The Indian subcontinent stratiform rain fractions range from 20-45% with sharply lower values during MAM, the period before the Asian monsoon (Fig. 3.6b, 70-85ºE). Northern Australia s stratiform fractions range from 25-50% with the lowest values during SON, the period before the Australian monsoon (Fig. 3.6d, ºE). The increase in stratiform rain fraction during the fully developed monsoon is 1. Maddox (1980) defined a mesoscale convective complex (MCC) as a contiguous area of IR temperature < 221 K with an areal extent > 50,000 km 2 embedded in a contiguous area of IR temperature < 241 K with an areal extent > 100,000 km that lasts six hours or more and has an eccentricity of > 0.7 at the time of maximum extent.

39 30 consistent with the monsoon precipitation forming in maritime air, giving the free atmosphere a more oceanic character. In addition, Laing and Fritsch (1993b) and Miller and Fritsch (1991) found that MCCs occur most frequently when the monsoon extent is greatest in India and Australia, respectively. The fact that increased MCC occurrence coincides with higher stratiform rain fractions during the monsoons again suggests that higher stratiform rain fractions over land could be associated with the occurrence of very large, organized precipitation systems such as MCCs. The increase in both rain and stratiform rain fraction during the fully developed monsoon implies an increase in both the total latent heating and the height of maximum heating, which leads to an enhanced upper-level circulation response. The maritime continent (Indonesia/Malaysia) has a relatively weak seasonal cycle (Fig. 3.6, ºE). The lack of seasonality may be explained by the maritime character of the region. In addition, most of the maritime continent does not develop many MCCs (Miller and Fritsch 1991) although large MCSs are frequent (Houze et al. 1981; Williams and Houze 1987). b. Oceans Over most of the tropical oceans, seasonal variations in stratiform rain fraction are smaller than those over land. The west Pacific has the smallest seasonal variability (Fig. 3.6, ºE) which may be a result of relatively constant environmental factors, notably sea surface temperature (SST). The east Pacific and east Atlantic have the largest seasonal variability (Fig. 3.6, ºW and 25-0ºW). This variability is most likely due to ocean warm-tongue/cold-tongue dynamics (Mitchell and Wallace 1992). The east Pacific and east Atlantic have the highest contribution from stratiform rain in JJA and SON, seasons in which the cold tongue in each ocean basin is most pronounced. The increased cold-tongue/warm-tongue thermal contrast implies an enhanced thermally direct meridional circulation on the scale of the ITCZ. It remains to be shown why this surface-forced circulation should enhance the production of stratiform precipitation by mesoscale cloud systems in the ITCZ region.

40 31 Some of the variability in the east Pacific can be linked to the occurrence of the double ITCZ. Often during the boreal spring of non-el Niño years, two bands of convection form and straddle the equator in the east Pacific (Lietzke et al. 2001). The formation of the two bands is favored by a narrow sea surface equatorial cold tongue in a region of low-level wind convergence and otherwise warm SST. The PR observes that there is an asymmetry in stratiform rain fractions between the east Pacific north and south of the equator during March and April of 1999 and 2000 with average stratiform rain fractions of 35% in the northeast and of 50% in the southeast. Rain accumulations were similar for both regions. There is little difference in stratiform rain fractions between the two regions in 1998, an El Niño year. 3.5 Interannual variations In order to highlight interannual variations in stratiform rain fraction, Fig. 3.7 shows the annually averaged stratiform rain fraction maps for 1998, 1999, and The largest differences occur in the central and east Pacific (180-90ºW) during 1998, a strong El Niño year. There are no dramatic differences between 1999, a strong La Niña year, and 2000, a normal year. Stratiform rain fractions averaged over each longitudinal belt for JFMA 1998 and 1999 highlight the differences between the El Niño and La Niña events (Fig. 3.8a). During El Niño, there are lower stratiform rain fractions over the maritime continent (20-30% at ºE) and higher stratiform rain fractions over the central and east Pacific (50-60% at ºW) than during La Niña (~40% for both regions). These signals mirror the well-known rain differences over each region during El Niño, with lower rain amounts over the maritime continent and higher rain amounts over the central and east Pacific (Fig. 3.8b). Similar but smaller oscillations occur over the Indian Ocean (45-90ºE) where stratiform rain fraction and rain amounts increase during El Niño. In this instance, there is a stronger relation between stratiform rain fraction and total rain amount than suggested by Fig. 3.4c. The fact that an oscillation of stratiform rain fraction accompanies the El Niño-Southern Oscillation of rainfall indicates that the vertical distribution of heating varies along with the horizontal distribution of heating during these events. The

41 32 strong gradient of stratiform rain fraction between the west and east Pacific implies that the maximum of latent heating rises and becomes more concentrated at upper levels in the east and lowers and becomes more vertically elongated in the west during El Niño. This change in heating profile further implies a stronger upper-level atmospheric response to the heating over the eastern-central Pacific during El Niño. 3.6 Quasi-steady circulations The land-ocean differences discussed in the previous sections help relate the stratiform rain statistics to specific circulation systems in the tropics. Webster (1983) highlighted three tropical large-scale structures: the ITCZ, summer monsoons, and semipermanent equatorial convective zones. Figure 3.9 illustrates regions classified by these large-scale structures. Table 3.2 ranks the regions from lowest to highest stratiform rain fraction and includes the seasonal stratiform rain fractions for each region. Overall, the pre-monsoon regions (west Africa and south Asia during MAM and southern Africa and northern Australia during SON; Table 3.2 lines 3, 4, 7, 12) tend to have the lowest stratiform rain fractions (19-33%) followed by the equatorial convective zones (Fig. 3.9, Table 3.2 lines 1, 2, 6, 9, 10, 15) with values of 26-45% and the monsoon regions (west Africa and south Asia during JJA and southern Africa and northern Australia during DJF; Table 3.2 lines 3, 4, 7, 12) with values of 31-43%. The ITCZ regions (Fig. 3.9, Table 3.2 lines 5, 8, 11, 13, 14, 16-19) have the highest stratiform rain fractions in the tropics (36-53%). This breakdown is consistent with Figs. 3.2d and 3.6 and the discussion in Sec. 3.3 in that land environments tend to support less stratiform rain formation than ocean environments. Most likely the pre-monsoon regimes and equatorial convective zones, the majority of which are continental, are subject to intense surface heating and strong diurnal modulation of the boundary layer and therefore have buoyancy concentrated at low levels. The result is intense, shorter-lived convection that is not able to produce significant stratiform regions. The monsoon regimes and ITCZ regions are subject to maritime environments with only weak diurnal modulation and more uniform buoyancy throughout the troposphere that yield weaker but longer-lived convection that can form

42 33 and maintain significant stratiform regions. Comparisons of the seasonal stratiform rain fractions in Table 3.2 indicate that the monsoon regimes have the highest seasonal variability in stratiform rain fractions (>10%), followed by the ITCZ regions (5-10%), and then the equatorial convective zones (< 5%). These seasonal variations are in agreement with Webster s (1983) definitions and descriptions of the large-scale structures. Monsoons are, by definition, itinerant in time and the stratiform rain fractions vary with the seasonal influx of maritime air. Equatorial convective zones, on the other hand, are considered semi-permanent and accordingly have more stable stratiform rain fractions. However, some seasonality may be introduced over continental equatorial convective zones by conditions (such as topographically constrained nocturnal low-level jets) which lead to the occurrence of MCCs during the season of maximum solar insolation. ITCZ regions are subject to the meandering of the southeast and northeast trades which is not great enough to dramatically affect stratiform rain fraction. Exceptions in seasonal variations occur in the ITCZ regions of the east Pacific and east Atlantic (Fig. 3.9, Table 3.2 lines 5 and 19), where SST gradients are most pronounced during JJA and SON, and the equatorial convective zone over South America (Fig. 3.9, Table 3.2 line 9). These regions have seasonal variations on the order of the monsoons (> 10%). The link between large-scale tropical structures and stratiform rain fraction evident from Table 3.2 further highlights the role of stratiform rain in dictating the vertical structure of latent heating and thus the response of the circulation to precipitating systems. The fact that the large-scale structures are coherent with stratiform rain fraction suggests that the environments related to these structures play an important role in the evolution of convective precipitation systems which can contribute to the horizontal and vertical gradients of heating that may feedback on the large-scale circulation.

43 34 a) b) Figure 3.1 PR 2.5º zonal averages from for a) stratiform rain fraction and b) rain accumulation.

44 35 a) Total rain 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) Convective rain 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W c) Stratiform rain 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W d) Stratiform rain fraction 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 3.2 PR a) total rain, b) convective rain, c) stratiform rain, and d) stratiform rain fraction based on 2.5º grid averages for Areas with annually averaged rain of less than 0.6 m yr -1 were not included in d).

45 36 a) Convective rain rate 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) Stratiform rain rate 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W c) Convective-stratiform (CS) rain rate ratio 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 3.3 PR mean a) conditional convective rain rate, b) conditional stratiform rain rate, and c) convective-stratiform (CS) rain rate ratio based on 2.5º grid averages for Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

46 37 a) b) c) Figure 3.4 Two-dimensional histograms of PR monthly stratiform rain fraction versus a) stratiform area fraction, b) CS rain rate ratio, and c) rain accumulation with bin sizes of two, 0.35 and 14, respectively. The monthly values are from and represent 2.5º grids within 20ºN-20ºS. Areas with rain accumulations less than 50 mm mo -1 were not included. The lines indicate the tropics-wide averages. The contours are counts of 1, 10, 25, 100, and 250.

47 38 a) Land b) Land c) Ocean d) Ocean Figure 3.5 Two-dimensional histograms of PR monthly stratiform rain fraction versus the monthly mean a) convective rain rate over land, b) stratiform rain rate over land, c) convective rain rate over ocean, and d) stratiform rain rate over ocean. Note that the monthly mean rain rates are conditional (i.e., calculated only when rain is present). Bin sizes are 0.6 mm h -1 for convective rain rates and 0.08 mm h -1 for stratiform rain rates. The monthly values are from and represent 2.5º grids within 20ºN-20ºS. Areas with rain accumulations less than 50 mm mo -1 were not included. The lines indicate the tropic-wide averages. The contours are counts of 1, 5, 10, 25, 50, and 100.

48 39 DJF 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W MAM 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W JJA 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W SON 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 3.6 Seasonal PR stratiform rain fraction based on 2.5º grid averages for Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

49 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 3.7 Annual averages of PR stratiform rain fraction based on 2.5º grid averages for 1998, 1999, and, Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

50 41 a) b) Figure 3.8 PR a) stratiform rain fraction and b) rain accumulation longitudinally averaged over 20ºN- 20ºS for January-April 1998 (El Niño) and January-April 1999 (La Niña).

51 42 20 N EQ 20 S M C C M I I M C M C 45 E 90 E 135 E 180 E 135 W 90 W 45 W I I I I C I C I I Figure 3.9 Quasi-steady circulation regions defined as I (intertropical convergence zone), M (seasonal monsoon regions), and C (semi-permanent equatorial convective zones).

52 43 Table 3.1: Land versus ocean tropical mean convective-stratiform rain statistics. Stratiform rain fraction (%) Stratiform area fraction (%) Conditional convective rain rate (mm h -1 ) Conditional stratiform rain rate (mm h -1 ) CS rain rate ratio All Land Ocean

53 44 Table 3.2: stratiform rain fractions for quasi-steady circulation regions. Circulation regions are as defined in Fig. 10. Seasons before the monsoon are indicated by * and seasons of the full monsoon are indicated by **. Seasons with rain accumulations less than 50 mm mo -1 were not included. Circulation region Land/Ocean Location ALL DJF MAM JJA SON 1 C L south-central Africa C L north-central Africa M L west Africa * 31 ** 29 4 M L southern Africa ** * 5 I O/L east Atlantic C O/L maritime continent M L south and south-east Asia * 41 ** 37 8 I L northern South America C L South America C L/O Central America I O north Indian ocean M L/O northern Australian ** * 13 I O south Pacific I O west Atlantic C O west Pacific warm pool I O south Indian ocean I O central Pacific I O north-east Pacific I O south-east Pacific

54 45 Chapter 4. Large-scale environmental characteristics of stratiform rain Shear, moisture, and convective available potential energy (CAPE) control the characteristics of convective cloud systems (Rotunno et al. 1988; Nicholls et al. 1988; Emanuel et al. 1994). A large proportion of tropical precipitation falls from mesoscale cloud systems that develop stratiform precipitation as a result of sheared wind in the environment and/or the natural aging process of convective cells populating the mesoscale cloud systems (Chapter 6 of Houze 1993; Houze 1997). As discussed in Sec. 3.3, the production of stratiform precipitation is especially effective over the oceans, most likely a result of the more uniform buoyancy through the free atmosphere and/or the sustainability of the convection by a warm, moist boundary layer with weak diurnal variation. The horizontal homogeneity of the ocean surface is likely also important. This chapter utilizes the monthly 2.5º gridded data from the NCEP reanalysis 1 and performs a joint analysis with the stratiform rain fraction distributions described in Chapter 3 in order to address more quantitatively how the relative contributions of stratiform rain relate to the large-scale environment: specifically, to relative humidity, temperature, and zonal wind. When relating stratiform rain fraction fields to the NCEP variables, it is difficult to separate cause and effect. Thus, relationships will be identified without necessarily attributing causality (although I may speculate on the matter). The previous chapter (Sec. 3.6) related the ITCZ, summer monsoon, and semi-permanent equatorial convective zone (C) circulation regimes to distinct patterns of stratiform rain fraction. These large-scale structures provide the framework for the discussion of each environmental variable. 4.1 Relative humidity Figure 4.1a shows the average relative humidity profiles for each of the three circulation regimes for Monthly relative humidity values were included in the average if the 2.5º grid box had at least 50 mm of PR-observed rain accumulation that month. The monsoon relative humidity profile has a surface value of 79% and decreases in a fairly 1. NCEP reanalysis data is available at

55 46 linear manner to 42% at 400 mb. A sharp increase in relative humidity occurs above 400 mb, likely because of cirrus clouds associated with deep convection (Gutzler 1993). The C profile is very similar to the monsoon profile except that it has slightly higher relative humidity values. The ITCZ has higher relative humidity than the monsoon and C regimes below 800 mb and lower relative humidity above 800 mb. The moist, fairly constant profile from 1000 to 925 mb is representative of the mixed layer (Lin and Johnson 1996). The mid-level dryness is consistent with other climatologies of relative humidity profiles over tropical oceans (Liu et al. 1991) and may be associated with a minimum in tropical cloudiness between mb (Zuidema 1998). The average relative humidity for each 2.5º grid box in each circulation regime was then calculated when the monthly stratiform rain fraction was greater than and less than the grid box average stratiform rain fraction. The two relative humidity values for each grid box were then subtracted and averaged over the regime domain. The resulting difference profiles are shown in Fig. 4.1b. For each regime, relative humidity is always greater when stratiform rain fraction is higher. This is consistent with previous studies that have composited relative humidity in relation to waves in the west Pacific (Reed and Recker 1971) and east Atlantic (Thompson et al. 1979) and found higher relative humidity near the trough which coincided with the largest rain accumulation and mid and high cloud cover (likely the time of greatest stratiform rain coverage). Serra and Houze (2002) found positive relative humidity differences during periods of maximum convective organization (also likely a time of more stratiform rain) in the east Pacific during the Tropical Eastern Pacific Process Study (TEPPS) cruise, and Halverson et al. (2002) found higher relative humidities at low levels when systems with larger stratiform components occurred in the southwest Amazon during TRMM-LBA. The monsoon regime has a primary relative humidity difference maximum at 500 mb, a secondary maximum at 925 mb, and a minimum at 700 mb. The C regime difference profile is similar but of smaller magnitude. The monsoon regime undergoes the largest seasonal variation in rainfall and stratiform rain fraction such that observed larger differences in relative humidity are to be expected. The maxima and minima in the ITCZ regime

56 47 difference profile are distributed differently than the monsoon and C regimes; the ITCZ profile has minima at 925 and 500 mb and a maximum at 700 (along with a secondary maximum at 400 mb). The solid green line in Fig. 4.1b indicates the ITCZ relative humidity difference profile excluding the El Niño months of JFMA The 700 mb maximum has weakened and the 400 mb maximum has disappeared, leaving a more smooth difference profile with a broad maximum between mb. Upon further investigation, the sharp maxima at 700 and 400 mb were related to variations in the east Pacific during El Niño. The difference profiles for the monsoon and C regimes changed very little when the El Niño months were excluded so the corresponding profiles are not shown. Precipitation systems with a high stratiform rain fraction may increase the relative humidity of the low and mid layers of the atmosphere through evaporation and melting (Lin and Johnson 1996). Alternatively, higher low-level relative humidity may enhance stratiform rain production by promoting the sustainability of convective cells (Yuter and Houze 1998, Kingsmill and Houze 1999b) such that large stratiform regions have sufficient time to form. Johnson and Lin (1997) and Johnson et al. (1999) showed that shallow cumulus and cumulus congestus can grow in a subsiding environment over the west Pacific and that the detrainment from these clouds moistens the lower troposphere, possibly preconditioning the atmosphere for deep convection. Sobel et al. (2003) showed that in the central-western Pacific during KWAJEX, higher values of low-level ( mb) relative humidity preceded active convection, suggesting low level moisture supports convection rather than occurring as a result of convection. The low-level relative humidity difference maximum in the monsoon regime, and to a lesser extent the C regime, is consistent with the notion that near-surface moisture availability promotes the sustainability of convection and thus stratiform rain production. A low-level relative humidity difference is not seen over the ITCZ where there is a much smaller diurnal signal and more constant moisture availability. Reed and Recker and Thompson et al. s wave composites contain a mid-level relative humidity maximum coincident with the trough. Nicholls et al. (1988) performed simulations of tropical squall lines with different relative humidity profiles and showed that

57 48 moist mid-level air tends to be favorable for squall line development, while Sobel et al. s lag analysis showed that higher values of relative humidity at upper levels ( mb) coincided with or followed active convection, thus implying that stratiform regions moisten the mid-to-upper troposphere rather than a previously moist upper layer assisting stratiform growth. In the monsoon and C regimes, higher stratiform rain proportions preferentially moisten the 500 mb layer, while over the ITCZ, higher stratiform rain proportions preferentially moisten the 650 mb layer (Fig. 4.1b). The difference in heights may be related to intrusions of dry air from the subtropics (Mapes and Zuidema 1996) and/or the preferred levels of convective detrainment (Johnson et al. 1999). Mapes and Zuidema (1996) examined 2400 soundings from the Tropical Ocean Global Atmosphere Coupled Ocean Atmosphere Response Experiment (TOGA COARE) and found dry layers advected from the subtropics with frequency peaks near 550, 800, and 950 mb. They hypothesized the dry layers could hinder deep convection by negating essential buoyancy at low levels and detraining mass at higher levels. Johnson et al. (1999) found evidence of a trimodal cloud population in TOGA COARE and attributed it to three prominent stable layers that exist over the tropical oceans: the trade stable layer ~2 km, the 0ºC level ~5 km, and the tropopause ~15 km. The 0ºC level is related to the effects of melting and is the weakest of the three; however, its existence can still serve to inhibit cloud growth and promote detrainment. As will be shown in the next section, the ITCZ regime is more stable at lower levels compared to the monsoon and C regimes. A stronger low-level stable layer and/or weaker convection may promote detrainment at lower levels over the ITCZ. Over the monsoon and C regions, the more intense convection may somewhat overcome the low and mid-level stable layers but will still be weakened. Zuidema (1998) showed that a weak stable layer near the 0ºC level can enhance detrainment just above it (i.e., ~400 mb) when congestus clouds do not glaciate, which may be occurring over the monsoon and C regions in Fig. 4.1b. In order to show geographic variations from the mean relative humidity difference values and to highlight the vertical gradients in the difference profiles, Fig. 4.2 shows maps of the tropical relative humidity difference between levels that exhibit the strongest

58 49 gradients in Fig. 4.1b ( mb, mb, mb, and mb). Blue indicates an increase in relative humidity difference with height and red indicates a decrease in relative humidity difference with height when stratiform rain fractions are higher. Figure 4.2a ( mb) illustrates the regions that contribute to the low-level variations in the mean profiles shown in Fig. 4.1b. This near-surface layer may represent possible changes in the mixed layer depth. The south Pacific causes the small decrease in height in the ITCZ profile while south Asia and the maritime continent cause the small increase in height in the monsoon and C profiles. Parts of the Amazon and the east Atlantic also show an increase in relative humidity difference with height. Between mb (Fig. 4.2b), land regions show the strongest decrease of relative humidity difference with height, while ocean regions show strong increases with height at higher stratiform rain fractions. The relative humidity difference over the Amazon is more like that seen over the ocean basins (very small near-surface differences that increase with height) suggesting that the Amazonian canopy sustains convection more continually than typical land surfaces. The relative humidity differences also increase with height over parts of the maritime continent, most likely due to the more maritime nature of its boundary layer. Figure 4.2c represents the relative humidity difference map between mb. The gradients are opposite of those depicted in Fig. 4.2b. Over land, the relative humidity differences increase with height, including the maritime continent. Over ocean, the differences generally decrease with height. The shift in gradient highlights the regions that have the sharpest minimum (the monsoon areas and continental C regions) and the sharpest maximum (the south and east Pacific and to a lesser extent the southern Indian Ocean and the west Atlantic) at 700 mb. Areas that shift from red to blue in Figs. 4.2 c and d indicate regions that exhibit a secondary maximum at 400 mb (e.g., parts of the south and east Pacific and the Amazon basin). 4.2 Temperature Figure 4.3a depicts the average temperature profiles for the three circulation regimes. The ITCZ profile has a temperature of 26ºC at 1000 mb and decreases to 0ºC

59 50 ~560 mb. The monsoon and C regimes have 1000 mb temperatures of 28 and 27ºC, respectively, and also decrease to 0ºC ~560 mb. In most of the lower troposphere, the ITCZ regime has colder temperatures and a more stable lapse rate (except from mb) than the monsoon and C regimes. The temperature difference profiles are shown in Fig. 4.3b. For each regime, a less steep lapse rate occurs with higher stratiform rain fraction, especially at low-levels. In the monsoon and C regimes, temperatures below 700 mb are colder and temperatures above 700 mb are warmer when stratiform rain fraction is higher. Also, a shift toward colder temperatures occurs above 200 mb. This is consistent with the cold low-level, warm midto-upper level, and cold near-tropospheric temperature differences found in the trough regions of Reed and Recker (1971) and Thompson et al. s (1979) composites. This pattern also exists in the temperature difference profile in the ITCZ regime when El Niño is excluded (the solid green line in Fig. 4.3b). The solid red line in Fig. 4.3 b represents the difference in temperature when there is at least 50 mm of monthly rain versus no rain threshold, averaged over the three regimes. The upper-level warming is similar to the high stratiform cases, however, the warm low-level difference is in the opposite sense and serves to increase the low-level lapse rate. Cooling very close to the surface may be a result of unsaturated downdrafts (Zipser 1969) and deeper cooling at low levels may be a result of evaporative cooling below the stratiform cloud base (Reed and Recker 1971). Reduced heating of the surface because of cloud cover may also impact the surface energy fluxes. In these ways, the stratiform region may stabilize the larger-scale environment. Less evaporative cooling may occur over the more moist ocean surfaces and sea surface temperature is less likely to change as much as land surface temperatures, explaining why the ITCZ has a smaller low-level temperature difference. Mapes (1993) argues that a tropical MCS (which has an elevated peak in the low-level convergence) causes compensating upward motion (and thus cooling) at low levels in the nearby atmosphere. The upward displacements favor the development of convection nearby, helping sustain convection and the stratiform rain region. A colder, less steep low-level lapse rate may also decrease CAPE and the intensity of convective cells. Halver-

60 51 son et al. (2002) showed that convective systems with large stratiform components over the southwest Amazon occurred with lower CAPE values and, as discussed in Sec. 3.3 (Fig. 3.5), intense convective rain rates are negatively correlated with stratiform rain fraction, especially over land. The solid red curve in Fig. 4.3b suggests that a warmer, more steep low-level lapse rate is beneficial to overall rain production through the initiation of convection but that higher stratiform rain proportions do not require strong low-level buoyancy. The warmer mid-to-upper level troposphere may be a result of the latent heat release or cloud radiative forcing in the upper portions of the large mesoscale convective systems (Xu and Emanuel 1989; Lin and Johnson 1996) or the adiabatic warming in the compensating downward motion to the MCS heating (Mapes and Houze 1995). The less warm difference near the tropopause is most likely related to longwave cooling at the upper limit of the stratiform cloud deck (Webster and Stephens 1980). Figure 4.4 maps the low-level ( mb, mb) and mid-to-upper level ( mb) lapse rate differences across the tropics. Blue indicates the temperature difference decreases with height (creating a more steep lapse rate) and red indicates the temperature difference increases with height (creating a less steep lapse rate) when stratiform rain fractions are higher. The mb layer (Fig. 4.4a) may represent some sensitivity to cooling in unsaturated downdrafts. There is increased cooling at 925 mb over the west coast of Africa, the Asian monsoon region, the maritime continent, the east Pacific, the Amazon basin, and the east Atlantic. There is warming with height in the south Indian ocean and the west and south Pacific. Between mb (Fig. 4.4b), land regions (especially monsoon areas and non- Amazonian South America) have the sharpest decrease in lapse rate. The south and east Pacific show an increase in low-level lapse rate (similar to the solid red line in Fig. 4.3b representing the difference between rain and no rain conditions), which probably results from the changes that occur during El Niño. While very strong convection seems to be inimical to higher stratiform rain fraction, stratiform precipitation production relies on sufficient convection to form large stratiform regions. During El Niño, higher SST shift to the central and east Pacific, sustaining more convection than typically occurs in those regions.

61 52 An increase in the low-level lapse rate would, in this instance, be beneficial to stratiform rain production. The upper-level map ( mb, Fig. 4.4c) shows a weak decrease in lapse rate across the tropics with a stronger decrease in lapse rate over India and in the south and central Pacific. Stratiform rain fraction becomes especially high in the central Pacific during El Niño which may account for the strong signal at upper levels. Lindzen et al. (2001) speculated that an increase in SST over the tropical oceans would enhance convective precipitation efficiency and lead to smaller anvil regions. While stratiform rain fraction is not strictly a measure of anvil coverage, stratiform rain fraction might be expected to decrease if convective precipitation efficiency were to increase. Figure 4.5 shows the SST difference using Reynolds and Smith (1994) 2.5º monthly SST observations from Mean SST for each grid box is found when the monthly stratiform rain fraction is greater than and less than the grid box stratiform rain fraction average. The two values of SST are then subtracted to find the difference. Blue indicates that SST is lower when stratiform rain fraction is higher (in support of Lindzen et al. s hypothesis) while red indicates the opposite. As seen in Fig. 4.5a, much of the Indian Ocean and parts of the Pacific and Atlantic show a lower SST at higher stratiform rain fractions. However, much of the Pacific Ocean shows a higher SST when stratiform rain fraction is greater, especially in regions previously shown to be sensitive to El Niño. When the El Niño months of JFMA 1998 are excluded (Fig. 4.5b), the relationship between SST differences and stratiform rain fraction in the Pacific basin weakens but continues to show higher SST occur in conjunction with higher proportions of stratiform rain. 4.3 Zonal wind Figure 4.6a shows the average monthly zonal wind profiles for each of the three circulation regimes when rain accumulation is at least 50 mm mo -1. The monsoon regime profile has low-level westerlies shifting to easterlies ~750 mb. Near the surface, winds become more westerly with height (negative shear); while above 925 mb, winds become more easterly with height (positive shear). The C regime has easterlies at all levels. The lowest wind speeds occur at the surface and become more easterly up to 500 mb after

62 53 which they remain constant. The ITCZ regime also has easterlies at all levels. Near the surface, winds become more easterly with height up to 925 mb. Easterlies remain constant from 925 to 500 mb above which they become more westerly with height. Figure 4.6b shows the zonal wind difference profiles. In each regime, higher stratiform rain fraction occurs with weaker easterlies/stronger westerlies at low-to-mid levels and stronger easterlies/weaker westerlies at upper levels. Figure 4.6b also indicates that negative shear between mb and positive shear above 700 mb is associated with higher stratiform rain fraction. Lin and Johnson (1996) noted a rapid increase in the midtropospheric vertical wind shear during the time of more mature convective systems during TOGA COARE, viz., strengthening westerlies at low-to-mid levels (with a peak near ~800 mb) and strong easterlies aloft. Keenan and Carbone (1992) showed that the zonal wind shear during the active monsoon over Darwin, Australia had stronger westerlies at low-to-mid levels (with a peak ~900 mb) and stronger easterlies at upper levels compared to the break period. Johnson and Ciesielski (2002) found similar results during the monsoon onset in the northern South China Sea. Also, Serra and Houze (2002) found stronger westerlies at low levels (up to ~700 mb) and stronger easterlies at mid-to-upper levels during the times of the greatest convective organization in the east Pacific during TEPPS. All of these studies are consistent with Fig. 4b. Houze et al. (2000) observed westerly mid-level inflow in the dense populations of mesoscale convective systems displaying large contiguous cold cloud tops (i.e., the type of convection expected to exhibit greater stratiform rain fraction) during TOGA COARE. These super convective systems had a strong mid-level inflow that was largely controlled by the environmental wind speed and direction (Kingsmill and Houze 1999a) such that one would expect a signal between mid-level zonal flow and stratiform rain fraction. Smull and Houze (1987) showed more generally that rear inflow in squall lines with trailing stratiform precipitation occurred at mid levels in both midlatitude and tropical squall lines and that the rear inflow was enhanced when the ambient flow entered the storm. In each squall line in their study, flow relative to the system shifted to front-to-rear at both upper and lower levels. The shift from rear-to-front flow at mid levels and front-to-rear

63 54 flow at upper levels when large stratiform regions are present may be indicated by the profiles in Fig. 4.6b. Houze et al. (2000) highlighted the importance of low-to-mid tropospheric zonal shear in momentum transport within the stratiform regions of large mesoscale cloud systems in TOGA COARE. The large-scale flow in this region was characterized as a Kelvin/ Rossby wave structure. The westerlies lay between the cyclonic gyres to the north and south of the equatorial zone. They found that momentum feedback associated with the mesoscale downdraft in stratiform regions of large cloud systems in the region of lowlevel westerly onset was negative (viz., mid-level easterly inflow transported downward, acting against the development of low-level westerlies). Conversely, they found a positive momentum feedback in the region of strong low-level westerlies (the location of the most widespread occurrence of large mesoscale convective systems according to Chen et al. 1996). In the strong low-level westerlies region, the westerly wind component in the environment increased with height in the lower troposphere, and the mesoscale downdrafts in the stratiform regions of the large cloud systems transported strong westerlies downward to increase the already strong low-level westerlies. It is noteworthy that this positive momentum feedback appears to occur in many regions of the tropics despite different large-scale flow seen in Fig. 4.6a. In addition, the downward transport of anomalous westerlies acts to slow down the climatological easterly low-level flow in the tropics. Moncrieff and Klinker (1997) used a high resolution (T213) spectral model to explore the largescale effects of organized convection and showed that the momentum flux will cause enhanced westerlies at low-to-mid levels and enhanced easterlies at upper levels (at least for eastward-moving cloud systems). Figure 4.7 maps the low-level ( mb) and mid-to-upper level ( ) zonal wind shear differences across the tropics. Blue indicates negative shear differences (winds more westerly with height) while red indicates positive shear differences (winds more easterly with height) when stratiform rain fraction is higher. From mb (Fig. 4.7a), essentially all of the shear differences are negative. A notable exception is over the South China Sea which indicates an increase of easterlies with height when stratiform

64 55 rain fraction is greater. This difference is in contradiction to Johnson and Ciesielski s (2002) results that showed the monsoon over the northern South China Sea had enhanced low-level westerlies. During the South China Sea Monsoon Experiment (SCSMEX), Johnson and Ciesielski observed that localized blocking of the low-level easterly flow occurs previous to the onset of the monsoon due to topography. The NCEP reanalysis optimal interpolation may be adversely affected by this topographic blocking. At upper levels (Fig. 4.7b), most of the shear differences are positive although parts of the Indian Ocean, west Pacific, and west Atlantic exhibit negative shear. Thompson et al. (1979) illustrated that the west Pacific and east Atlantic zonal winds are out of phase in their wave composites. In the trough, the west Pacific shows almost no zonal wind difference at low levels and a westerly difference aloft while the east Atlantic has a strong westerly difference at low levels and an easterly difference aloft. These observations are consistent with the maps in Fig. 4.7.

65 56 a) b) Figure 4.1 a) The average relative humidity profile for the monsoon, semi-permanent equatorial convection (C), and ITCZ circulation regimes assuming at least 50 mm mo -1 of rain accumulation. b) The difference in relative humidity between when the stratiform rain fraction is greater than and less than the 2.5º grid element mean. The solid green line indicates the average relative humidity difference profile for the ITCZ regime excluding the El Niño months of JFMA 1998.

66 57 a) RH N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) RH N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W c) RH N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W d) RH N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 4.2 The vertical gradient in relative humidity differences between higher and lower stratiform rain fraction conditions from a) mb, b) mb, c) mb, and d) mb. Blue indicates an increase in relative humidity difference with height and red indicates a decrease in relative humidity difference with height when stratiform rain fractions are higher. Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

67 58 a) b) Figure 4.3 a) The average temperature profile for the monsoon, semi-permanent equatorial convection (C), and ITCZ circulation regimes assuming at least 50 mm mo -1 of rain accumulation. b) The difference in temperature between when the stratiform rain fraction is greater than and less than the 2.5º grid element mean. The solid green line indicates the average temperature difference profile for the ITCZ regime excluding the El Niño months of JFMA The solid red line indicates the difference in temperature when there is at least 50 mm mo -1 of rain accumulation versus no rain threshold.

68 59 a) T N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) T N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W c) T N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 4.4 The vertical gradient in temperature differences between higher and lower stratiform rain fraction conditions from a) mb, b) mb, and c) mb. Blue indicates the temperature difference decreases with height (creating a more steep lapse rate) and red indicates the temperature difference increases with height (creating a less steep lapse rate) when stratiform rain fractions are higher. Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

69 60 a) SST 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) SST (no JFMA 1998) 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 4.5 a) The difference in Reynolds and Smith SST when the stratiform rain fraction is greater than and less than each 2.5º grid element mean for b) As in a) except that the El Niño months of JFMA 1998 were excluded. Blue indicates lower SST and red indicates higher SST when stratiform rain fraction is higher. Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

70 61 a) b) Figure 4.6 a) The average zonal wind profile for the monsoon, semi-permanent equatorial convection (C), and ITCZ circulation regimes assuming at least 50 mm mo -1 of rain accumulation. b) The difference in zonal wind between when the stratiform rain fraction is greater than and less than the 2.5º grid element mean.

71 62 a) U N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) U N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 4.7 The vertical gradient in zonal wind differences between higher and lower stratiform rain fraction conditions from a) mb and b) mb. Blue indicates winds are more westerly with height (negative shear) and red indicates winds are more easterly with height (positive shear) when stratiform rain fractions are higher. Areas with annually averaged rain of less than 0.6 m yr -1 were not included.

72 63 Chapter 5. Latent heating and the tropical dynamical response 1 The objective of this chapter is to determine whether the latent heating derived from the TRMM PR leads to new insights into the large-scale circulation of the tropics, especially with respect to horizontal and temporal variability in the percent of rain that is stratiform. An idealized global climate model is forced with the PR-derived latent heating to obtain the steady-state, zonally asymmetric dynamical response. Hartmann et al. (1984) used a linear, steady-state model forced with an isolated heat source to demonstrate that a vertical heating profile corresponding to a mature convective system with a high fraction of stratiform precipitation produces stronger flow anomalies aloft, more elevated circulation centers, and larger vertical tilts than a heating profile of a purely convective system. They also found that the heating profile dominated by stratiform rain produces a more realistic Walker Circulation than a convective-only heating profile when a quasi-realistic horizontal distribution of tropical heating derived from climatological outgoing longwave radiation observations along the equator is used. This study extends the work of Hartmann et al. (1984) by using more complete and direct measurements of tropical precipitation, by investigating how geographical and temporal variations in the proportion of rain that is stratiform influence the response of the tropical circulation to latent heating, and by using a nonlinear, time-dependent model with higher horizontal and vertical resolution, improved physical parameterizations, and realistic zonally averaged climates Latent heating estimates To obtain the vertical latent heating profile at each 2.5º grid element, assumed latent heating profiles associated with stratiform and convective precipitation are linearly combined based on the PR-observed rain fraction for each precipitation type, normalized such that the area under the curve equals one, and then multiplied by the PR-observed precipitation at each location. Thus, the three-dimensional latent heating field can be esti- 1. This chapter was done in collaboration with Ian Kraucunas. Ian modified and ran the general circulation model and provided much appreciated feedback.

73 64 mated for a chosen subset of the TRMM observational period. This study uses seasonalto-annual average subsets to represent the four-dimensional tropical latent heating field. An example of the total rainfall field used in the calculations is depicted in Fig In order to focus on the latent heating associated with tropical convection and eliminate edge effects at the boundaries of the PR observational domain, the precipitation field outside of 20ºN and 20ºS is multiplied by a scaling factor that decreases linearly from unity at 20º latitude to zero at 35ºN and 35ºS. This modification was found to have little effect on the model results near the equator, and increased the stability of the model by retarding the growth of baroclinic disturbances along the edge of the PR domain. Note that the scaling procedure used to deemphasize the extratropical rain pattern (cf. Fig. 2.1a) minimizes the influence of these midlatitude storms on the heating estimates derived in this section. The assumed heating profiles for stratiform, deep convective, and shallow convective precipitation are illustrated in Fig. 5.2a. The heating profile associated with deep convective rain is positive throughout the troposphere, while the stratiform profile is dominated by heating above the climatological 0ºC level and cooling below. These profiles are based on budget studies of mesoscale precipitation systems observed during tropical field programs (see Figs in Cifelli and Rutledge 1998). In order to use some height information from the PR, the shallow convective classification from TRMM product 2A23 was assigned a separate latent heating profile. Shallow convection does not represent a fundamentally different category from deep convection. Therefore, the profile for shallow convective rain is similar to the profile for deep convective rain, except that it has been vertically compressed to represent the latent heating from low and mid-level precipitating convective clouds. Beyond the separation of shallow and deep convection, fixed vertical heating profiles are used despite the fact that cloud heights vary somewhat in nature. Tao et al. (2001) calculated the tropical latent heating from TRMM PR observations using a database of stratiform and convective latent heating profiles that were based on cloud-resolving model runs in different regions in the tropics. The assumed profiles in this study are generally consistent with those calculated by Tao et al., so the use of a larger database of latent heating profiles would not likely alter the overall results. However, Tao

74 65 et al. s results indicate some of the uncertainty associated with the profiles in Fig. 5.2a. Figure 5.2b illustrates the vertical profiles of latent heating that would be used for a grid element with a total rainfall of 3.6 m yr -1, a shallow convective rain fraction of 10%, and stratiform rain fractions of 0, 40, and 70%. A stratiform rain fraction of 0% (purely convective) has a heating maximum at 4.5 km, a stratiform rain fraction of 40% (which represents the tropics-wide average) leads to a slightly stronger heating maximum at 6.5 km, and a stratiform rain fraction of 70% (the upper end of observed climatological values) produces an even stronger heating maximum at 7.5 km, along with cooling at lower levels. In addition to raising the amplitude and height of maximum latent heat release, increasing the stratiform rain fraction yields a stronger vertical gradient of heating at upper levels. The annually averaged latent heating for the entire TRMM study period is illustrated in Fig Figures 5.3 a and b show the horizontal pattern of latent heating at 7.4 and 2.2 km, respectively, while Fig. 5.3c contains a vertical cross section of the annually averaged heating at 10ºN. Note that the latent heating is stronger in Fig. 5.3a than Fig. 5.3b because the stratiform rain fraction is non-zero, and that the lower-level heating is concentrated over rainy regions with low stratiform rain fractions. The vertical cross section of latent heating at 10ºN (Fig. 5.3c) further illustrates that regions of low stratiform rain fraction (e.g., 40ºE and 85ºW) have latent heating more evenly distributed throughout the troposphere, while regions with high stratiform rain fractions (e.g., 135ºW) have latent heating concentrated at upper levels. 5.2 Model The atmospheric model used for this study is an idealized version of the Community Climate Model, version 3 (CCM3), that has been modified to calculate the timedependent, zonally asymmetric response to a fixed distribution of diabatic heating. The dynamical core of the model uses the spectral transform method in the horizontal domain with T42 resolution, which corresponds to a grid spacing of roughly 2.9º in latitude and longitude. Vertical and temporal aspects are treated using finite differences, with 18

75 66 unevenly spaced vertical levels (12 of which are located in the troposphere) and a timestep of 20 minutes. Further discussion of the model dynamics can be found in Kiehl et al. (1998). In order to isolate the zonally asymmetric response to steady tropical heating, the boundary conditions and physical parameterizations normally employed by CCM3 were replaced with linear damping of temperature and wind perturbations towards a prescribed, zonally uniform basic state. This was achieved by removing topography and water vapor mixing ratio from the model, and locking the zonal-mean values of divergence, vorticity, temperature, and surface pressure to their initial (basic state) values throughout each model run. Experiments were performed using both a resting, horizontally uniform basic state with a static stability in the troposphere equal to the tropics-wide average, and a basic state composed of the zonally averaged winds, temperatures, and surface pressures from the monthly mean NCEP reanalysis fields corresponding to the same time periods as the TRMM PR observations used to derive the latent heating distributions described in the previous section. The frictional damping coefficient applied to zonal and meridional wind perturbations decreases linearly from (2.5 days) -1 at the surface to a constant value of (15 days) -1 at 800 mb and above. The Newtonian cooling uses the same damping profile as the Rayleigh friction below 200 mb, but increases linearly above this level back up to a value of (2.5 days) -1 at the top of the model. The larger damping coefficients near the surface are intended to roughly account for the effects of surface drag and vertical mixing in the boundary layer, while the stronger thermal damping aloft helps limit the growth of wave energy in the stratosphere.the CCM3 dynamical core also includes harmonic ( ) diffusion in the top three model levels to absorb vertically propagating planetary wave energy and weak biharmonic ( 4 ) horizontal diffusion at all levels to control small-scale noise. A limited number of experiments performed with different damping parameters, horizontal diffusion coefficients, and vertical resolutions suggest that our results are not very sensitive to these aspects of the model. 2

76 67 The model is forced by adding the three-dimensional TRMM-derived heating distributions described in the previous section directly to the temperature tendency equation. Since the model is not allowed to have a zonal-mean response, the zonally averaged heating does not need to be removed from the input field. It is assumed that the specified zonal-mean climate includes the dynamical influence of the zonal-mean component of tropical heating, so the model captures only the time-dependent response to the zonally asymmetric component of the input heating field in the absence of nonlinear interactions between forced waves and the zonal mean flow. The model typically reaches an equilibrium solution in about two weeks for experiments performed using a resting basic state. In the experiments performed using realistic basic states, baroclinic instabilities begin to distort the numerical solution in the tropics after days. Therefore, day 14 averages are taken to represent the steady-state, zonally asymmetric dynamical response to the imposed heating in all of our experiments. 5.3 Annual-mean experiments In order to verify that the atmospheric model produces a reasonable dynamical response to horizontal variations in tropical heating, and also provide a simple context for demonstrating how variations in stratiform rain fraction influence the response, several experiments were performed using a resting basic state and the latent heating derived from the PR annually averaged precipitation from and geographically uniform stratiform rain fractions of 0%, 40%, and 70%. The 0% case assumes that all tropical rainfall is purely convective, the 40% case uses the tropic-wide average stratiform rain fraction at all gridpoints, and the 70% case unilaterally imposes the highest climatological stratiform rain fraction observed by the TRMM PR. The column-integrated heating is the same for all three runs, so the differences between the model responses must arise from the differences in the vertical heating structures depicted in Fig. 5.2b. Figure 5.4a shows the 400 mb latent heating distribution and the resulting 250 mb streamfunction anomalies obtained using 0% stratiform rain fraction everywhere. Anticyclonic circulation anomalies are present over south Asia and the Indian Ocean, while

77 68 cyclonic gyres dominate the Pacific basin. This quadrapole pattern is similar to the pattern obtained for an isolated divergence anomaly near the equator (Gill 1980), and is consistent with the interpretation that intense heating over the west Pacific warm pool drives largescale divergence anomalies and vorticity generation aloft that dominate the mean horizontal circulation response in the upper tropical troposphere. Smaller streamfunction anomalies are also present in the Atlantic region. The tropics-wide pattern shows good qualitative agreement with the observed annually averaged streamfunction anomaly field, and also bears a strong resemblance to the horizontal circulation patterns obtained by previous authors (e.g., Hartmann et al. 1984; Nigam 1994; Wang and Ting 1999), which suggests that this approach yields a reasonable large-scale dynamical response. Figure 5.4b illustrates the 400 mb latent heating and 250 mb streamfunction response obtained when the uniform stratiform rain fraction is increased to 40%. The streamfunction and latent heating patterns in Fig. 5.4b are very similar to those obtained in Fig. 5.4a, although both the mid-tropospheric latent heating field and the upper-level anticyclonic and cyclonic circulation features are substantially stronger in the uniform 40% case. This intensification is consistent with the increase in the vertical gradient of latent heating in the upper troposphere noted in Fig. 5.2b and agrees with the results of Hartmann et al. (1984). Figures 5.5 a-c show vertical cross sections of the zonal wind and vertical pressure velocity, ω, anomaly fields near the equator (8.5ºN-S) for the integrations performed using the latent heating derived from geographically uniform stratiform rain fractions of 0, 40, and 70%. In Fig. 5.5a (the 0% case, see also Fig. 5.4a), negative ω anomalies occur over regions of large rain accumulation (20ºE, equatorial Africa; 90ºE-150ºE, the maritime continent and west Pacific warm pool; and 80ºW, Central America and the Amazon basin) and positive ω anomalies occur over regions of lower rain accumulation (45ºE, west Indian Ocean; 180ºE-90ºW, the central and east Pacific; and 5ºW, the east Atlantic). The ω anomalies have maximum amplitude between mb. The zonal wind anomaly pattern has four cells in a square wave pattern over the Indian Ocean (80ºE) and the central Pacific (160ºW) and another couplet over South America and the west Atlantic

78 69 (45ºW). The lower cells are centered at 800 mb and the upper cells are centered around 300 mb. In Fig. 5.5b, the geographically uniform stratiform rain fraction is increased to 40% (as in Fig. 5.4b). The strength of the ω anomalies increases and the centers rise in altitude to mb. The centers of the zonal wind anomalies move upward to 700 mb and 250 mb while intensifying. In Fig. 5.5c, the latent heating based on a uniform 70% stratiform rain fraction produces the strongest ω response with the highest maxima and minima centered at 350 mb and a weaker layer of low-level ω centered at 850 mb. The zonal wind anomalies are also stronger and higher than the previous two scenarios with the lower-level centers above 600 mb and the upper-level centers around 200 mb. In addition, a third layer of zonal wind anomalies begins to emerge near the surface. The assumption of a constant stratiform rain fraction in the above model simulations implies that there is no horizontal variation in the vertical structure of latent heating across the tropics. This lack of variation restricts the vertical structure of the large-scale circulation and leads to idealized circulation centers at constant heights. However, as seen in Fig. 3.2d, stratiform rain fraction varies across the tropics. Figure 5.5d shows the vertical cross section of the equatorial response to the latent heating based on PR-observed climatological stratiform rain fractions from The model response is most similar to the uniform 40% case, which is reasonable since the annually averaged stratiform rain fraction is ~40%. However, the circulation centers have a larger variation in height and vertical extent than the uniform 40% case. For example, the ω anomaly maxima and minima range in height between mb while the low-level zonal wind anomaly centers range between mb and the upper-level zonal wind anomaly centers range between mb. Variation in the heights and vertical extents of the ω anomalies leads to a tilt in the zonal wind field. The zonal gradient in stratiform rain fraction from ~30% over Indonesia to ~60% over the eastern tropical Pacific appears to produce vertical variations in latent heating that create a tilted structure in the equatorial circulation. Hartmann et al. (1984) assumed a heating profile for an idealized mature cloud cluster which closely resembles the profile when the stratiform rain fraction is equal to

79 70 70% (Fig. 5.2b). They found that the mature cloud cluster heating profile forced a more realistic Walker cell response than a purely convective profile, in part, because it produced a westward tilt with height of the simulated zonal mass flux. While the latent heating based on a uniform stratiform rain fraction of 70% produces some variation in the heights of the mid-level circulation centers (Fig. 5.5b), comparison with Fig. 5.5d suggests that a tilted zonal circulation is more realistically simulated with a latent heating field based on the observed horizontal variation in stratiform rain fraction rather than a uniformly applied profile with a very large stratiform component. 5.4 Seasonal cycle In order to see if the model response to the TRMM-derived latent heating could produce a reasonable seasonal cycle, a more realistic set of experiments was performed using the NCEP basic states and the PR-observed precipitation and stratiform rain fractions for June-July-August (JJA) and December-January-February (DJF) of Figures 5.6 a and b show the resulting 250 mb streamfunction anomalies for JJA and DJF, respectively. In JJA, the main region of upper-level heating is located over the maritime continent (120ºE). Cyclonic streamfunction anomalies over south Asia and the Indian Ocean are centered at 70ºE, the anticyclonic Pacific gyre north of the equator is centered just west of the dateline, the Atlantic gyre north of the equator is centered at 60ºW, and the Pacific gyre south of the equator has an elongated center that spans from the dateline to 90ºW. There is also some evidence of a weak wave propagation into the extratropical southern hemisphere in the Pacific sector. In DJF, the main region of upper-level heating has shifted east to over the west Pacific warm pool (150ºE). The centers of the streamfunction anomalies north of the equator have also shifted eastward, while the anticyclonic gyre over the Indian Ocean has weakened and split into multiple centers over southern Africa and Australia, and the elongated cyclonic gyre south of the equator has separated into alternating cyclonic-anticyclonic-cyclonic streamfunction anomalies over the Pacific, South America, and the Atlantic. There is also some indication of enhanced wave propagation into the northern (winter) hemisphere, with reduced extratropical eddy activity in

80 71 the southern (summer) hemisphere. Figures 5.6 c and d illustrate the NCEP reanalysis 250 mb streamfunction field with the zonal mean removed for JJA and DJF, respectively. Between 30ºN-S, the JJA and DJF upper-level streamfunction anomaly patterns from the TRMM-derived latent heating show good qualitative agreement with the NCEP reanalysis for the same time periods. The main difference between the NCEP reanalysis fields and the model response to the TRMM-derived latent heating is that the model response is two to three times smaller than NCEP. This difference in magnitude probably arises because the model is forced with heating derived solely from precipitation. Other components of precipitating systems can redistribute and increase the total heating; some of these components will be addressed in Sec There are small qualitative differences between the locations of the streamfunction anomalies for NCEP and the model response which may reflect nonlinearities associated with the different amplitudes. Outside of 30ºN-S, there is much less agreement between the streamfunction anomaly pattern from the TRMM-derived latent heating and NCEP. The model was forced only with a tropical heat source so it wasn t expected to capture the extratropical wave train which is especially evident in the northern hemisphere of the DJF NCEP streamfunction anomaly field. In addition, the model does not allow transients and wave-mean interactions. 5.5 El Niño The tropics-wide pattern of stratiform rain fraction during the 1999 La Niña event was not very different from the climatology, whereas there were dramatic changes in stratiform rain fraction and precipitation during the 1998 El Niño event (Sec. 3.5). In order to compare the model response to TRMM observations during La Niña and El Niño, another set of experiments was performed using the NCEP basic states and the PR-observed precipitation and stratiform rain fractions for January-April 1999 (La Niña) and January-April 1998 (El Niño). Figure 5.7 shows the 400 mb latent heating distribution and the resulting 250 mb streamfunction anomalies for La Niña and El Niño. The upper-level streamfunction

81 72 response to the TRMM-derived latent heating for the four-month La Niña season of 1999 (Fig. 5.7a) is very similar to the DJF streamfunction anomaly pattern (Fig. 5.6b). During La Niña, the divergent gyres centered over the west Pacific and the couplet over the Atlantic are in the same location as in DJF but the Pacific gyres are slightly stronger than during an average year, while the Atlantic gyres are slightly weaker. Wave propagation is also enhanced into the northern hemisphere extratropics during La Niña. The TRMM-derived heating and upper-level streamfunction response of the fourmonth El Niño season of 1998 (Fig. 5.7b) is quite different than La Niña. During El Niño, stratiform rain fractions get as low as 20% over Indonesia and reach almost 70% in the east Pacific and the precipitation field (indicated by the latent heating at 400 mb) shifts with warmer sea surface temperature (SST) into the central and east Pacific. The longitudinally averaged precipitation is almost constant across the entire tropical band centered just south of the equator; the variations in upper-level heating are due to the dramatic west-east gradient in stratiform rain fraction. The major changes in the precipitation and stratiform rain fraction during El Niño significantly change the model response in the Pacific. The anticyclonic gyre over the Indian Ocean also elongates across the Pacific and weakens, although this change is not as pronounced as in the northern hemisphere. The circulation anomalies in the Atlantic sector remain relatively unchanged, although the small anticyclones normally present over the western half of South America are absent. To highlight the importance of both the overall magnitude and the horizontal variation of the stratiform rain fraction during El Niño, Fig. 5.8 depicts vertical cross sections of the zonal wind and ω responses averaged along the equator from 8.5ºN-S based on latent heating derived from the PR-observed precipitation during El Niño and geographically uniform stratiform rain fractions of 40 and 70%, along with the PR-observed stratiform rain fraction. The model response to the latent heating based on a uniform stratiform rain fraction of 40% (Fig. 5.8a) is very weak. Comparison with the more idealized uniform 40% model simulation in Sec. 5.3 (Fig. 5.5b) shows that circulation anomalies are centered at similar altitudes in both model runs but are weaker during El Niño. In addition, the main region of upward motion has shifted to the central and east Pacific and the four

82 73 square wave pattern in the zonal wind field over the Indian Ocean and central Pacific has disappeared. When the stratiform rain fraction is increased to 70% (Fig. 5.8b), the circulation centers move upward and strengthen while another layer of ω and zonal wind anomalies appears at lower levels, similar to what occurred in the uniform 70% case in Sec While the tropics-wide average stratiform rain fraction increases only slightly during El Niño (to ~45%), the model response to the latent heating based on the PR-observed stratiform rain fraction shown Fig. 5.8c does not resemble the response to the uniform 40% case shown in Fig. 5.8a. Rather, the strength of the ω and zonal wind response and the presence of an additional circulation layer at lower levels is much more similar to the uniform 70% case; however, negative and positive ω anomalies in Figs. 5.8 b and c are not always collocated (e.g., over 90ºE, the Indian Ocean) and much of the height variation in circulation centers is still lacking in the uniform 70% model run. The structure and tilt in the equatorial cross section appears to come from the trans-pacific gradient in stratiform rain fraction, which is especially pronounced during El Niño. 5.6 Effects of nonprecipitating convection and cloud radiative forcing (CRF) While the above experiments qualitatively simulate seasonal and interannual tropical circulation patterns, the strength of the circulations remains weaker than observed. The TRMM-derived latent heating profiles are based solely on precipitation; however, other components related to the precipitating systems may affect the overall heating profile, including the redistribution of latent heating by nonprecipitating cumulus and cloud radiative forcing (CRF). Model integrations were performed using heating fields that take into account these additional heating components in order to bring the magnitude of the circulation response closer to observations. While the above experiments qualitatively simulate seasonal and interannual tropical circulation patterns, the strength of the circulations remains weaker than observed. The TRMM-derived latent heating profiles are based solely on precipitation; however, other components related to the precipitating systems may affect the overall heating profile, including the redistribution of latent heating by nonprecipitating cumulus and cloud radia-

83 74 tive forcing (CRF). Model integrations were performed using heating fields that take into account these additional heating components in order to bring the magnitude of the circulation response closer to observations. Nonprecipitating clouds do not add latent heat to the overall system, although they do affect the distribution of heating in the vertical (e.g., Nitta and Esbensen 1974; Johnson and Lin 1997). The latent heating profile of a nonprecipitating cloud is positive in the lower part of the cloud, due to net positive condensation, and negative in the upper part of the cloud where detrainment occurs and evaporation exceeds condensation. Johnson et al. (1999) point out that low-level cumulus associated with the trade inversion and mid-level cumulus congestus in the vicinity of the 0ºC level are prominent cloud types in the tropics. Populations of these nonprecipitating clouds are assumed to occur in the same time and space domains as the precipitating clouds seen by the PR. The dotted profiles in Fig. 5.9 a and b are the assumed latent heating profiles for deep and shallow nonprecipitating convection, respectively, and were constructed to be consistent with the budget studies of Nitta and Esbensen (1974) and Johnson and Lin (1997). The vertical integral of each curve is zero, so no net latent heating is added to the system; all each profile does is redistribute latent heating in the vertical. When the deep nonprecipitating convective profile is added to the original deep convective profile (dashed curve, see also Fig. 5.2a), the resulting lowering of the peak heating to 3.5 km (solid curve) had to be within the range of maximum heating heights seen in budget studies of deep convection (e.g., Johnson 1984). It is difficult to make an equivalent comparison in the shallow convective case because the literature lacks budget studies that single out a shallow precipitating regime. The new convective latent heating profiles are substituted for the original convective profiles in order to calculate the four-dimensional latent heating field taking into account nonprecipitating cumulus. Clouds associated with mature precipitating convective systems are largely opaque to infrared (IR) radiation and act to counter the IR energy loss to space that occurs under clear-sky conditions (Gray and Jacobsen 1977). IR cloud radiative forcing (CRF) in mature precipitating convective systems generally warms the troposphere, except for large

84 75 IR losses (or cooling) near cloud top. Deep cloud systems also absorb solar radiation in their upper levels. Over a day s time, IR cooling and shortwave warming at the cloud top largely cancel one another such that CRF only weakly cools at the uppermost levels of deeper cloud systems. Houze (1982) included the radiative heating as part of the total heating profile of a tropical mesoscale system and showed that the CRF in stratiform regions acts to reinforce the upper-level latent heating maximum. Bergman and Hendon (2000) also found that CRF reinforces the circulation from latent heat release in deep convective systems. In addition, Bergman and Hendon (2000) show that the radiative component of shallow clouds tends to be dominated by reflection of solar radiation such that the CRF is negative. Unlike latent heating, radiative heating is associated with the area covered by cloud rather than with the amount of rain produced. Therefore, to estimate a CRF in each rainy area, the TRMM PR measurement of the area covered by precipitation is used. For the relevant time period, each 2.5º grid element in the tropics has a fractional area covered by shallow convective, deep convective, and stratiform radar echo as seen by the PR. However, since the cloudy area generally greatly exceeds the rainy area a factor must be applied to the TRMM PR observed rain area to estimate the cloud cover. The fractional area covered by PR-observed precipitation ranges from 5-10% in regions of large rain accumulation and 1-5% in regions of low rain accumulation, whereas the ISCCP climatology (Rossow and Schiffer 1991; isccp.giss.nasa.gov) indicates cloud fractions 1 of 60-90% in heavy rain regions and 20-50% in light rain regions. So as a rough estimate of fractional cloud cover associated with the PR echoes, the fractional areas covered by shallow convective, deep convective, and stratiform precipitation seen by the PR are multiplied by 10 to estimate the cloud cover in each category. However, the total fractional cloud cover is never allowed to exceed 90%. Figure 5.10 shows the idealized CRF profiles that are applied to the area covered by each type of precipitating cloud category. The dotted curve is the CRF that applies if 1. ISCCP cloud detection is based on variations of visible and infrared radiation from clear sky conditions as observed by a suite of operational satellites. The cloud detection algorithm is described in Rossow and Garder (1993).

85 76 100% of the area of the grid element is covered by deep convective cloud. This curve is essentially a direct cancellation of the clear sky cooling based on Gray and Jacobson (1977). Similarly, the dashed curve shows the CRF that applies if 100% of the area of the grid element is covered by stratiform cloud. Stratiform cloud is assumed to have a base at or above the 0ºC level so more warming occurs at upper levels, and less at lower levels (Webster and Stephens 1980). The shallow convective CRF (dash dot) has a negative CRF in the lower troposphere (Bergman and Hendon 1998). Note that stratus decks are excluded from the PR estimates. The net radiative forcing for the grid element is obtained by multiplying each curve by the actual percentage coverage by the respective cloud type and summing to obtain the net forcing. The net CRF computed in this manner is then added to the net latent heating profile field. Figure 5.11 presents three profiles of net latent plus cloud radiative heating computed from combinations of stratiform rain fraction (for the latent heating) and stratiform rain area fraction (for the CRF) for 3.6 m yr -1 of rain and 90% cloud cover. Comparison to Fig. 5.2b shows that while the additional heating profiles act to increase the magnitude of heating, they also act to alter the vertical gradient of the heating. With the additional heating, the purely convective case has a lower, stronger heating maximum at 2.5 km. In the case of 40% stratiform rain fraction and 75% stratiform rain area fraction, the upper-level maximum strengthens and is elevated to 8 km, and there is a weaker lower-level maximum at 2.5 km. In the case of 70% stratiform rain fraction and 90% stratiform rain area fraction, the maximum around 8 km has sharpened and there is no longer any low-level cooling. The model was run using the NCEP basic state and the heating derived from precipitation only and from precipitation with the additional heating components for DJF The vertical cross sections of zonal wind and ω anomalies averaged along the equator from 8.5ºN-S are shown in Fig The response to the precipitation-only latent heating (Fig. 5.12a) has structure similar to the model response to the annual heating and a resting basic state (Fig. 5.5d) The maxima and minima of ω are centered between mb with magnitudes mb h -1 and a quadrapole structure of zonal wind anomalies

86 77 is over the Indian Ocean and the central Pacific along with a smaller couplet over the Atlantic. The low-level zonal wind maxima and minima range from mb with magnitudes +3 m s -1, while the upper-level zonal wind maxima and minima are around 200 mb with magnitudes +7 m s -1. NCEP (Fig. 5.12b) shows a similar overall circulation pattern, but with significantly stronger upper-level ω anomalies (+ 1.2 mb h -1 ) with centers as high as 250 mb. In addition, there are multiple lower-level ω anomalies centered around 850 mb. Consistent with the ω field, the NCEP upper-level zonal wind anomaly maxima and minima are stronger (+ 15 m s -1 ) and higher in altitude (~150 mb). The centers of the lower-level zonal wind anomalies are also stronger (+ 5 m s -1 ) and closer to the surface ( mb). The addition of the heating from nonprecipitating convection and CRF brings the model response closer to NCEP (Fig. 5.12c). The ω anomalies strengthen (+ 0.9 mb h -1 ) and extend to lower levels. The upper-level zonal wind anomalies strengthen to + 10 m s -1 and the centers move up to ~150 mb. The lower-level zonal wind cells strengthen slightly and extend to lower levels. Despite the lack of land-sea contrasts and topography, the lower-level circulation response looks somewhat reasonable. The additional heating strengthens the circulation and creates a more realistic low-level circulation while not changing the overall pattern originally forced by the spatial variability of the percent of rain that is stratiform.

87 78 20 N EQ 20 S Rain accumulation TAPERED 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 5.1 PR 2.5º annually averaged rain accumulation from The precipitation has been linearly decreased from values at 20ºN and S to zero at 35ºN and S in order to isolate the tropical heat source.

88 79 a) b) Figure 5.2 a) Idealized stratiform (SF), deep convective (DC), and shallow convective (SC) latent heating profiles. The x-axis is meant to be non-dimensional until a precipitation amount is assumed. b) Total latent heating profiles assuming 3.6 m yr -1 rain accumulation.

89 80 a) Latent heating 7.4 km 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W b) Latent heating 2.2 km 20 N EQ 20 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W c) Latent heating 10ºN Pressure (mb) Figure 5.3 Horizontal distributions of the annually averaged latent heating at a) 7.4 km (~400 mb) and b) 2.2 km (~800 mb) derived from the TRMM PR rain-type fractions and tapered rain amounts. c) Vertical cross section of annually averaged latent heating at 10ºN.

90 81 60 S a) SF 0% 30 N EQ 30 S 60 S 60 S b) SF 40% 45 E 90 E 135 E 180 E 135 W 90 W 45 W 30 N EQ 30 S 60 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 5.4 The 400 mb latent heating (shaded) and the resulting 250 mb streamfunction anomalies (contours) from a model run using a resting basic state forced with the heating derived from the PR annually averaged precipitation field and geographically uniform stratiform rain fractions of a) 0% and b) 40%. The streamfunction contour interval is 10 6 m 2 s -1 ; negative contours are dashed.

91 82 a) SF 0% b) SF 40% c) SF 70% d) SF TRMM Figure 5.5 Vertical cross sections of ω (shaded) and zonal wind (contours) anomaly fields averaged along the equator from 8.5ºN-8.5ºS from a model run using a resting basic state forced by heating derived from the PR annually averaged precipitation and stratiform rain fractions of a) 0%, b) 40%, c) 70%, and d) PR-observed. The zonal wind contours are in m s -1 ; easterlies are dashed.

92 83 a) JJA TRMM 60 S 30 N EQ 30 S 60 S b) DJF TRMM 45 E 90 E 135 E 180 E 135 W 90 W 45 W 60 S 30 N EQ 30 S 60 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 5.6 The 250 mb streamfunction anomalies from a model run using the NCEP seasonal basic state forced with the heating derived from the PR seasonally averaged precipitation and stratiform rain fraction fields for a) JJA and b) DJF. The streamfunction contour interval is 10 6 m 2 s -1 ; negative contours are dashed.

93 84 c) JJA NCEP 60 S 30 N EQ 30 S 60 S d) DJF NCEP 45 E 90 E 135 E 180 E 135 W 90 W 45 W 60 S 30 N EQ 30 S 60 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 5.6 (cont.) The NCEP eddy streamfunctions for c) JJA and d) DJF. Note that the contour interval is three times larger than in Figs. 5.6 a and b.

94 85 60 S a) La Niña 30 N EQ 30 S 60 S 60 S b) El Niño 45 E 90 E 135 E 180 E 135 W 90 W 45 W 30 N EQ 30 S 60 S 45 E 90 E 135 E 180 E 135 W 90 W 45 W Figure 5.7 The 400 mb latent heating (shaded) and the resulting 250 mb streamfunction anomalies (contours) from a model run using the NCEP basic state for a) La Niña (JFMA 1999) and b) El Niño (JFMA 1998) forced with the latent heating derived from the PR precipitation and stratiform rain fraction fields for the same periods. The streamfunction contour interval is 10 6 m 2 s -1 ; negative contours are dashed.

95 86 a) SF 40% b) SF 70% c) SF TRMM Figure 5.8 Vertical cross sections of ω (shaded) and zonal wind (contours) anomaly fields along the equator from 8.5ºN-8.5ºS from a model run using the NCEP basic state for El Niño (JFMA 1998) forced by heating derived from the PR-observed precipitation and stratiform rain fractions of a) 40%, b) 70%, and c) PR-observed. The zonal wind contours are in m s -1 ; easterlies are dashed.

96 87 a) b) Figure 5.9 Idealized latent heating profiles of nonprecipitating cumulus (dotted) in a) deep and b) shallow convective rain. The original convective latent heating profiles from Fig. 3a are shown (dashed) along with the resulting convective latent heating profiles after the redistribution of latent heating from nonprecipitating cumulus is taken into account (solid). The x-axis is meant to be non-dimensional, however, it could be thought of as K day -1 for a specified surface rain amount.

97 88 Figure 5.10 Idealized cloud radiative forcing profiles for shallow convective, deep convective, and stratiform rain areas assuming 100% cloud cover.

98 89 Figure 5.11 Profiles of total heating (including nonprecipitating convection and CRF as well as precipitating clouds) for 0, 40, and 70% stratiform rain fraction and 0, 75, 90% stratiform rain area fraction assuming 3.6 m yr -1 rain accumulation, 90% cloud cover, and 10% of the rain and rain area is shallow convective.

99 90 a) DJF TRMM b) DJF NCEP c) DJF TRMM w/ added heating Figure 5.12 Vertical cross sections of ω (shaded) and zonal wind (contours) anomaly fields along the equator from 8.5ºN-8.5ºS for a) model results based on PR DJF precipitation and stratiform rain fraction, b) NCEP DJF reanalysis, and c) model results based on PR DJF precipitation and stratiform rain fraction with additional heating from nonprecipitating convection and cloud radiative forcing.