1. Executive Summary

2. Programmatic Linkages

3. Scientific Components

3.1 Introduction

3.2 Evolution and Variability of the SAMS

3.2.1. Annual cycle of precipitation over South America

3.2.2. Diurnal cycle of precipitation during the SA monsoon

3.2.3. Variability of the SA monsoon

3.3 Moisture Budgets and Major Moisture Sources for the Region

3.4 The Atlantic ocean impact on Climate Variability over South America

4.1 Empirical studies

4.2 Numerical Model Requirements

4.2.1 Downscaling Methods.

4.2.2 Assessment of energy budgets in regional models.

4.3 SAMS Data Requirements

4.4 Enhanced Monitoring and Field Experiments.

4.4.1 Special atmospheric sounding networks for monitoring climate variability

4.4.2 Special raingauge neworks for monitoring climate variability

4.4.3 Field process studies

4.5 South American low-level jet east of the Andes (SALLJ) Program

5. Readiness and Priorities.

References
 
 

The first meeting of the VAMOS Working Group on the South American Monsoon System, sponsored by the PACS (OGP/NOAA) program, took place at AOML (Atlantic Oceanographic and Meteorological Laboratory), Miami. The workshop was convoked to promote interaction and coordinate efforts of scientists interested on advancing the understanding and predictability of summer precipitation over South America. The goals of the meeting were 1) to assess the current state of knowledge on the South American Monsoon, 2) to identify gaps in this knowledge and field experiments (existing or already planned) designed to close these gaps, 3) to prioritize problems and 4) define special observing needs and numerical experiments as required for specific topics. The workshop report is intended to further develop the science and implementation planning for VAMOS that was initiated with the CONAM meeting in Mexico City in 1997 and the VAMOS workshop earlier this year in Sao Paulo. A special session on the South American Monsoon Systems will take place during the 6th International Conference on Southern Hemisphere Meteorology and Oceanography (6ISHMO), to be held April 3-7, 2000, at the Diego Portales Convention Center located in Santiago, Chile, to expose the science plans to a broad audience of interested scientists, further develop implementation planning, and entrain greater participation in the program. It is expected that issues raised in this session will enhance the awareness in the scientific community (in particular in South -American countries) of the benefits to be gained by developing a better understanding of the South American monsoon system. Such awareness is likely to translate on local support of scientists embarked on pertinent research and endorsement and sponsorship of necessary field experiments.

The VAMOS/SAMS meeting was attended by E. H. Berbery (University of Maryland), M. Douglas (National Severe Storms Laboratory), D. Enfield (AOML), R. Garreaud (JISAO and University of Chile), A. Grimm (University of Parana, Brazil), C. Jones (UCSB), V. E. Kousky (CPC/NCEP/NOAA), R. Lawford (OGP/NOAA), B. Liebmann (CDC, University of Colorado), R. Mechoso (UCLA, California), G. Miranda (Institute of Ecology, Bolivia), K. Mo (CPC/NCEP/NOAA), M. Nicolini (University of Buenos Aires, Argentina), J. N-Paegle (University of Utah), M. Patterson (OGP/NOAA), G. Podesta (RSMAS), P. Silva-Dias (University of Sao Paulo, Brazil) and C. Vera (University of Buenos Aires, Argentina).

Participants demonstrated a keen interest in advancing knowledge on the SAMS through a program of complementary activities on empirical studies, enhanced observational systems and development of numerical models that address the unique challenges presented by the South American steep orography and variable surface conditions.

The workshop discussed components, evolution and variability of SAMS, the relative importance of the Altiplano and Amazon heat sources, atmospheric teleconnections between the Americas during austral summer, numerical modeling applications for the region and ocean-atmospheric interactions. Current plans for a two-month field-experiment of the low-level jet east of the Andes were extensively discussed as well as a proposal for an expanded observational network for five years to improve monitoring of climate variability over the region. The discussion was lively and indicative of the support and interest that this research topic currently generates in several South-American countries and the USA. The role of international initiatives such as those of CLIVAR through VAMOS, the IAI and national funding institutions was also addressed. There was a general consensus in the readiness of the region to embark and support collaborative research on SAMS.

Such readiness is partly due to the enhanced regional awareness on the impact that summer precipitation and its variations have on water resource management, energy production, agriculture and health. In addition, links between summer precipitation over South America and inter-annual to intraseasonal processes such as El Niño-Southern Oscillation (ENSO), Atlantic sea surface temperatures and the Madden-Julian Oscillation are indicative of the inherent predictability of this precipitation regime. Improving the knowledge on these physical processes is a pre-requisite to substantial gains in the prediction of summer rains.

To better understand the dynamics of these links it is necessary to more completely quantify these relationships in the historical record. This quantification should be multidisciplinary and involve assessments of related variability in water resources, agricultural yields, tropical diseases and availability of energy. The impact of rainfall and temperature variability on human activities needs to be determined to assess vulnerability and to develop prediction techniques for better resource management.

The discussion isolated the following questions that limit our understanding of the SAMS:

• dynamical processes responsible for the onset, demise and character of the monsoon over different regions of South America.
• cause-effect relationships for monsoon modulations in scales from intraseasonal to interdecadal, including teleconnection patterns that span the Pan-American region.
• variability of the atmospheric and terrestrial components of the water and energy cycles at various time scales, including the role of low-level flow in the vicinity of the Andes.
• physical processes that cause the organization of nocturnal meso-scale convective systems in the La Plata river basin, and whether these are linked to low-level jets similar to those found over the North American plains.
• effect of the ITCZ and Atlantic SST's in the modulation of the SAMS
• land-surface effects in various time scales.

 

There are four sections in this report. Section 2 discusses programmatic linkages between the SAMS program and other institutional arrangements, and discusses potential collaboration with other efforts. It is also argued in this section that a relatively small incremental funding properly timed with other ongoing or already planned observational campaigns may add substantial value to the resulting data base. Section 3 presents a review of current knowledge of key aspects of the SAMS, while the emphasis on section 4 is on goals designed to address unresolved issues and data requirements. These requirements are addressed through proposed enhanced monitoring and a strategy for field experiments designed to improve our understanding of regional climate variability and enhanced its predictability. Such data sets will also be useful to assess model performance and evaluate seasonal forecasts. Finally, section 5 discusses the readiness of the program and identifies initial priorities.
 

The report on the VAMOS/PACS workshop (March 30-April 2, 1998, Sao Paulo, Brazil) reviews programmes relevant to VAMOS. The SAMS working group identifies as a main goal advances in the seasonal to interannual climate prediction over South-America, with an initial focus on improving summer precipitation prediction. This goal is closely aligned with those of the PACS (CLIVAR/GOALS 1995 - 2005) programme. Perspectives from LBA and GCIP are also relevant to advance the understanding of hydrological processes over South-America. IAI and IRI objectives to promote multi-disciplinary and international cooperative climate research is essential for South-America, where the main river basins of the region cross international borders. Research on the SAMS will benefit from data gathered from other field programmes and observational efforts listed in Table 1 of the VAMOS/PACS workshop, such as EPIC, PIRATA, CORC, N. Chile, SACC, PACS/SONET, IAS, etc.).

The SAMS working group recognized the unique opportunity of the expanded observational capabilities currently planned for the period 2000 - 2002 to enhance our understanding of the SAMS. NASA missions for the period 1998 - 2003 may be found at http://www.hq.nasa.gov/office/ese/missions/spacecraft.html. Planning efforts are also underway for a Coordinated Enhanced Observing Period (CEOP) in 2001-2002 (http://monsoon.nagaokaut.ac.jp/ceop/) .

The GEWEX Hydrometeorology Panel (GHP), at its third session (Sapporo, Japan; September 1997) established a Working Group to address the basic science objectives and practical implementation and coordination issues associated with the CEOP. Consequently, the CEOP Working Group recommended an international experiment designed to carry out a case study of the influence of continental hydroclimatic processes on the predictability of global atmospheric circulation and changes in water resources on time scales up to seasonal. It was also recommended that the CEOP take place over the two year period starting on 1 October 2000 and ending on 30 September 2002. The first year of the CEOP (CEOP1) is to be used as a build-up and preparation phase leading to a highly coordinated enhanced observing period among all the participants during the second year (CEOP2) to assure that a high quality global data set is collected. The CEOP overall science objective is closely related with the SAMS working group goals. The requirement of additional data to advance our understanding of SAMS processes has been identified as the first priority for the SAMS program as outlined in section 4 below. A strategy consistent with the goals of CEOP is proposed here with enhanced monitoring activities over South America building up observational capacity in a timely basis to focus on the summer of the year 2002 in a field experiment of the low level jet east of the Andes. This time table is also consistent with efforts currently underway with the LBA in Brazil.
 
 

3.1 Introduction
 

Many of the descriptive features of the South American Monsoon System discussed in São Paulo were also discussed in Mexico City and have already been included in the VAMOS chapter of the CLIVAR document. We first review some of the characteristics of the SAMS and we then focus on a regional variations.

The South American monsoon system (SAMS) develops over a land mass characterized by a large area at the equator, very high mountains to the west that effectively block air transport, and surface cover that varies from tropical forests over Amazonia to high altitude deserts over the Bolivian altiplano. Plentiful moisture supply from the Atlantic maintains a precipitation maxima over central Brazil. In addition, the subtropical plains of South America benefit by moisture transport from tropical latitudes which result on abundant precipitation over northern and central Argentina and Uruguay. The combination of this heat source and the orography results in seasonal evolution of convection unique to this region. Furthermore, there is an important influence of mid-latitude systems organizing tropical precipitation. Many important questions remain on the relative roles of orography, the Bolivian altiplano, the Brazilian planaalto, the Andes mountains and tropical heat sources (both over South America and other continents) on regulating circulation features over South America. Modelling and theoretical studies have given partial answers to these questions, but validation of these results require observational confirmation with more complete data sets.
 

3.2 Evolution and Variability of the South American Monsoon System

3.2.1. Annual cycle of precipitation over South America

The climate of tropical and subtropical South America (SA) is characterized by a regular and pronounced annual cycle in rainfall (e.g., Horel et al. 1989). Austral summer (DJF) is the rainy season for a large portion of the central region of South America, whereas austral winter (JJA) is the dry season. Some parts of South America, such as Colombia, southern Venezuela, northwestern Brazil, and some regions of southern South America do not experience a dry season and, therefore, do not have a well defined rainy season. From the northwestern parts of South America to central and southeastern Brazil, the rainfall season changes progressively from JJA to DJF, following the annual migration of the deep tropical convection and the establishment of a heat low in central SA in summer.

A pentad outgoing longwave radiation climatology for the South America sector was documented by Kousky (1988). This has been recently updated with a more extensive climatology (Kousky, personal communication). Deep convection over Central and South America displays considerable seasonal variability. For most of South America between the equator and 20°S there are distinct dry and wet seasons each year. These correspond approximately to the southern winter and summer seasons, respectively.

The onset of the rainy season begins in the western Amazon Basin in late August. Deep convection and rainfall, as represented by outgoing longwave radiation (OLR) values less than 200 W/m*m, shifts southward along the east slopes of the Andes, and southeastward towards the Brazilian highland during September and October (Fig. 1). By late November, deep convection covers most of western South America from the equator south to 20°S, while there is a noticeable absence of deep convection over the eastern Amazon Basin and Northeast Brazil. Throughout this period the deep convection, associated with the ITCZ, is confined to the central Atlantic between 5°N and 8°N.

Figure 1. The 1979-1995 mean evolution of the 220 W/m2 contour of outgoing longwave radiation (OLR) for the period 16 August-29 November. Analysis is based on the pentad climatology, available from the Climate Prediction Center, Washington, DC. From Kousky (1999), personal communication.

From late November through mid-February (Fig. 2), the mature phase of the South American Monsoon System, there is little change in the areal extent of the deep convection, except over the eastern Amazon Basin, which gradually experiences an increase in deep convection throughout the period. As deep convection spreads into the eastern Amazon, deep convection associated with the Atlantic ITCZ weakens.

Figure 2. Same as Figure 1 but for the period 29 November-12 February.

Beginning in March, the South American Monsoon System begins to weaken as the area of deep convection retreats northwestward, especially over central and western sections (Fig. 3). However, deep convection over the north coastal regions of Brazil doesn't begin to weaken until late April. Rao and Hada (1990) show that the rainy season in coastal NE Brazil takes place during April through June. Throughout the demise phase of SAMS deep convection associated with the Atlantic ITCZ remains relatively weak.

Figure 3. Same as Figure 1, but for the period 12 February-28 May.

At the height of the austral summer (DJF), latent heat released from deep cumulus convection over these regions of intense rainfall dominates the total diabatic heat ing over South America. According to estimates by Zhou and Lau (1998), based on GEOS-1 and NCEP/NCAR reanalysis, the diabatic heating is maximum in the middle and upper troposphere (~ 5°K/day) over the Altiplano and the southern Amazon basin, collocated with a thermal ridge around 2°S, leading to an effective production of available potential energy (Fig 4).

Figure 4. Latitudinal average from 26°S to 16°S of (a) temperature deviation from the local zonal mean (100° - 10°W averaged), (b) total diabatic heating, (c) turbulent sensible heating, and (d) latent heating. Units are °K for (a) and °K/day for (b)-(d). From Zhou and Lau (1998)

Sensible heat flux near the surface constitutes an additional heat source of the SAMS. Sensible heating dominates the lower troposphere and it reaches its maximum strength (~ 4°K/day) before the onset of the Monsoon, when the spring-to-summer increment of surface insolation over the subtropical latitudes is redistributed to the troposphere by dry, shallow convection. The interaction between these two heat sources and the tropospheric circulation is essential in shaping the climatological South America monsoon system (SAMS), and perhaps important in the year-to-year variability of the warm season circulation and rainfall over South America (e.g., Lenters and Cook 1997; Zhou and Lau 1998; Fu et al. 1998). Most of the current knowledge on this subject is based on modeling efforts aimed to reproduce the mean conditions of the SAMS during its mature phase, particularly the interaction between the latent heat sources and the upper level circulation (e.g., Silva Dias et al. 1983; Buchmann et al. 1986; Kleeman 1989; Figueroa et al. 1995; Lenters and Cook 1997). Less is known about the interaction of the heat sources and the mean circulation in the lower troposphere, including the formation of a low-level jet along the eastern slope of the subtropical Andes. Considerable effort must also be placed in the understanding of the physical mechanisms responsible for the onset / demise of the SAMS, that tends to occur as a rapid meridional shift of the area of convection around October and April, respectively.

The seasonal migration of the ITCZ in the eastern tropical Pacific and Atlantic does not follow the apparent trajectory of the sun, which crosses the equator twice a year (Li and Philander 1996; Mitchell and Wallace 1992). In October, sea surface temperatures at and to the south of the equator are low, and the ITCZ is farthest north. That is not the case over the continents, where the maximum cloudiness is near the equator. During the subsequent months, the continental convective zones move southward, but over the waters of the eastern Pacific and Atlantic, the ITCZ remains in a northerly position. By April, the waters south of the equator have warmed, and the ITCZ moves toward the equator. In the Pacific it only reaches the equator, and over the Atlantic, it reaches as far as 10° S, near the Brazilian coast. In this last region, the ITCZ remains slightly south of the equator from February through April. There are suggestions that the asymmetries favor the Northern rather than the Southern Hemisphere, with the warmest waters and the ITCZ involving the details of the local coastal geometries and low-level stratus over cold waters (Philander et al. 1996).

3.2.2. Diurnal cycle of precipitation during the SA monsoon
 

Convective cloudiness exhibits a well defined diurnal cycle tied to the diurnal march of the insolation and influenced by regional factors. This fluctuation is particularly relevant over the tropical and subtropical regions, where synoptic variability is weak, and it has a clear signature in the rainfall climatology. This diurnal cycle has climatic relevance in its own right, and its large amplitude also is thought to sup- press low-frequency (intraseasonal) oscillations over the central part of the continent (e.g., Horel et al.1989; Hendon and Salby 1994). The pronounced diurnal cycle of rainfall and convective cloudiness during the SA monsoon has been partially documented on the basis of satellite imagery and ground-based observations (e.g., Kousky 1980; Meisner and Arkin 1987; Negri et al. 1994; Garreaud and Wallace 1997). A detailed, continental-scale picture of the climatological march of convective cloudiness at the height of the austral summer (DJF) is presented in Fig. 5 on the basis of 3-hourly infrared GOES imagery with 0.5° x 0.5° of horizontal resolution (the ISCCP B3 product).

Figure 5. (a) Daily mean frequency of convective cloudiness (TB<235°K) during the austral summer (DJF). (b) Diurnal standard deviation of the 3-hourly frequency of convective cloudiness during DJF. (c) Frequency of convective cloudiness at 2100 UTC (PM conditions) during DJF. (d) Frequency of convective cloudiness at 1200 UTC (AM conditions) during DJF. The statistics are calculated based on 11 years of LSCCPB3 data (1983-1991). Data resolution is 0.5° x 0.5° lat/lon, 3-hourly. See Garreaud and Wallace (1997) for further details.

3.2.3. Variability of the SA monsoon

There are variations in the SA monsoon in several time scales. The day-to-day variability of rainfall over the subtropical SA and western Amazon basis is largely explained by frequent northward incursions of mid-latitude systems to the east of the Andes. The deep northward intrusion of midlatitude systems is the result of their interaction with the Andes topography (Kousky 1979; Garreaud and Wallace 1998). As in wintertime, these incursions occur with periodicity of about 7 days. They produce a band of enhanced convection (forced by the strong low-level convergence at the leading edge of the cool air) that extends from the eastern slope of the Andes southeastward into the Atlantic and moves from the subtropics (35°S) into the tropics in about four days (Garreaud and Wallace 1998, Fig. 6 below).

Figure 6. Composite maps of low-level wind (1000-850 hPa) and LC (Wm2) anomalies for days -1, 0, and +1. The compositing analysis is based on the dates with intense convection over the subtropical plains of the continent (25°S, 60°W). The anomalies are calculated as the composite maps minus the long term mean. Black area indicates terrain elevation in excess of 3000 m. Low-level winds are from NCEP/NCAR reanalysis. See Garreaud and Wallace (1998) for further details.

In intraseasonal time scales (periods less than 90 days), the OLR maximum variance is in the SACZ and in central SA (Paraguay, southern Brazil, and northern Argentina), and the minimum is over the southern Amazon Basin, where mean convection is at a maximum (Liebmann et al. 1998). The strong diurnal cycle over the Amazon Basin is probably responsible for the weak intraseasonal variability in this region (Zhao and Weare 1994). Precipitation and OLR spectra display several significant peaks corresponding to periods less than 90 days, especially in the SACZ region, with the relative proportion of higher frequency power increasing towards the southeast (SACZ) (D'Almeida 1997; Liebmann et al. 1998).

Intraseasonal variations of SA summer circulations and precipitation have been associated with the 30-60 day oscillation by several authors (e.g., Casarin and Kousky 1986; Kiladis and Weickmann 1992a; Kousky and Kayano 1994; Grimm and Silva Dias 1995; Nogues-Paegle and Mo 1997, Mo and Higgins, 1998). However, the intraseasonal variations in the 6-30 day band also play an important role in the SA monsoon variability, especially in the subtropics. The precipitation spectrum for the State of Sao Paulo (Brazil), in the SACZ region, shows a 25-day peak with higher significance level than the 50-day peak (D'Almeida 1997). Nogues-Paegle and Mo (1997) noted an intraseasonal variation with half-period of around 10 days. Kodama (1992) also reported an approximate 10-day half-period of convection in the subtropical convergence zones.

Vincent et al. (1998) showed that in the Indian and Pacific Ocean (until the dateline), significant OLR peaks in the 30-60 day oscillation band are confined to a 20-degree latitudinal belt straddling the equator. In contrast, significant signals in the 6-30 day band straddle the 20° parallels in either hemisphere and extend southward near the South Pacific Convergence Zone (SPCZ). Maximum activity occurs during the austral summer when midlatitude wave activity intrudes into the SPCZ region (Schrage and Vincent 1996). A similar behavior tends to occur over SA.

Kiladis and Weickmann (1992a,b), focusing on intraseasonal time scales, reported cases in which tropical convection in the Pacific Ocean (associated with the 30-60 day band) forces circulation anomalies that propagate first poleward and then equatorward over South America (see also Berbery and Nogues-Paegle 1993), as well as cases in which SACZ variations (in the 6-30 day time scales) are forced by westerly perturbations originating in the extratropics. Enhanced convection is activated by upper-level troughs over the region. The convection occurs in the upward motion induced by the advection of cyclonic vorticity ahead of the trough axis, as in a midlatitude baroclinic wave. The troughs are accompanied by the intrusion of cold fronts into the tropics from higher southern latitudes. Regions in which upper-level westerly flow lies near a tropical convergence zone (as the SPCZ and the SACZ) are prone to larger interaction between westerly disturbances and tropical convection.

Variations of the SACZ convection appear to be related to a Rossby wave originating over the Pacific, whose structure is nearly equivalent barotropic west of South America but baroclinic in the SACZ region. This wave-pattern seems to associate an enhanced (suppressed) SACZ with rainfall deficits (excess) over southern Brazil, Uruguay, northeastern Argentina (Paegle and Mo 1997; D'Almeida 1997; Liebmann et al. 1998), and central Andes (Aceituno and Montecinos 1997; Garreaud 1998). The influence functions for target points near the centers of opposite anomalies in this Rossby wave-pattern over southern South America show that upper-level divergence associated with anomalous convection in the subtropical South Pacific is very efficient in inducing these circulation anomalies (see Fig. 7 below from Grimm and Silva Dias 1995).
 

Figure 7. Influence functions with the January basic state for the indicated target points (from Grimm and S. Dias, 1995).

Composites of OLR and circulation anomalies associated with wet and dry phases of the monsoon rainy season in the State of Sao Paulo (in the SACZ region), as well as model results, confirm the role of convection anomalies in the Pacific region, as shown in Fig. 8 for a situation associated with one phase of the 30-60 day Oscillation (Grimm and Silva Dias 1995) . Simulations by D'Almeida (1997) also show this connection.
 

Figure 8. Idealized upper-level tropospheric divergence anomaly corresponding to a SPCZ slightly displaced to the east (above) and the associated anomalous streamfunction response (below) (from Grimm and S. Dias, 1995).

A crucial aspect of the tropical-extratropical interaction is the atmospheric basic state on which the energy propagates. Frequent troughs and associated frontal zones propagate into the tropics from the SH westerlies over Australia, the western South Pacific, South America, and southern Africa, giving rise to the "cloud band" activity over these regions (Kiladis and Weickmann 1992b). The contribution of the deep Amazon convection (and of the other deep tropical convection) to the basic state probably determines the preferred position of the SACZ (and of the other "cloud bands"). The zonal mean wind variability on the intraseasonal timescale also seems to influence wave propagation, as confirmed in the composites of D'Almeida (1997).

In the interannual time scales, the correlation between the Southern Oscillation and subtropical rainfall is maximum in the transition season OND, when the circulation anomalies associated with El Niño events tend to reinforce the dynamical mechanism that leads to rainfall in this season. However, there is little correlation (or even anticorrelation, at least in January) during the peak of the monsoon season (DJF) (Rao and Hada 1990; Grimm et al. 1998a). In the tropical region, there are strong influences of El Niño and La Niña events in some regions of the Amazon Basin during the austral summer and also autumn-winter (Ropelewski and Halpert 1987; Aceituno 1988; Marengo 1992; Grimm et al. 1998b). Recent results show that there are areas in the SACZ region with consistent El Niño and La Niña impacts, but the anomalies change signal during the monsoon season, from spring to summer (Grimm and Ferraz 1998a and b). The summertime precipitation over the Altiplano also exhibits substantial interannual variability, with a tendency to dry conditions during the warm phase of ENSO (Aceituno and Garreaud 1995).

Significant interdecadal variations in rainfall have also been detected in several regions of South America. For example, Castaneda and Barros (1994) disclosed long term variations in rainfall over northern Argentina. Robertson and Mechoso (1998) showed long-term variations and trends in streamflows of rivers in southern South America. Grimm et al. (1998c) showed significant shifts in the precipitation over southern and southeastern Brazil in the 70's. The period and region of maximum variation suggest an ENSO-like impact on longer time scales. Interdecadal variations were also identified in the Amazon Basin, as reported by Marengo (1998).

Besides leading to gains in the prediction of summer rains in South America, improving the understanding of the variability of the South American monsoon is also potentially important for the better understanding and prediction of the atmospheric variability in the Northern Hemisphere. Nogues-Paegle et al (1998), have shown that erroneous simulation of convection over tropical South-America in 8 day forecasts of the NCEP reanalysis model leads to southward shifts of the subtropical jet over North-America, a result that is consistent with those of Grimm and Silva-Dias (1995).
 

Figure 9.. Idealized upper-level tropospheric divergence anomaly corresponding to enhanced convection over Amazon and SACZ (above) and the associated anamalous streamfunction response (below) (from Grimm and S. Dias, 1995).

Figure 9 shows that the atmospheric response to enhanced upper-level divergence (associated with enhanced heat source) over the Amazon and the SACZ region involves teleconnections between the two hemispheres. Furthermore, the anomalous compensating subsidence (and related upper-level convergence) over the Caribbean, associated with a shifted or anomalous convection over Amazonia, is efficient in producing circulation anomalies over North Atlantic and other regions in the Northern Hemisphere, as can be seen in Fig. 10 (Grimm and Silva Dias 1995).
 

Figure 10. Influence function with the January basic state for the indicated target point (from Grimm and S. Dias, 1995).

3.3 Moisture budgets and major moisture sources for the region

The Parana and Paraguay rivers in subtropical South America constitute the basin known as Cuenca del Plata, and they are the water resource for one of the most densely populated regions of South America. Moreover, several hydroelectric power plants regulate the river flow and, in turn, can affect the navigability of these natural waterways. These characteristics, together with the fact that harvests and livestock are an important asset in the region, highlight the need to understand the regional hydrological cycle and its components. Research has been conducted to investigate the region's precipitation and streamflow variability on different time scales (e.g., Mechoso and Perez-Iribarren 1992, Pisciottano et al. 1994, García and Vargas 1998). The atmospheric component is less known, despite a few studies that have investigated the character of the low-level winds and related moisture flux.

Virji (1981) used satellite data to provide observational indications of the existence of a northerly/northwesterly low-level jet east of the Andes. However, observational data are limited, and global analyses have been increasingly used to further the description of the moisture fluxes. Nogués-Paegle and Mo (1997) using NCEP/NCAR reanalyses documented the moisture fluxes from the Tropics into Argentina and southern Brazil. Their study highlights the relationship between the low-level jet and larger scale variability in the Tropics and the South Atlantic Convergence Zone (SACZ). Nogués-Paegle and Mo also point out that some differences exist between the low-level jet over the United States Great Plains and that over South America: while the moisture source of the former is a water mass (the Gulf of Mexico), the latter would have a continental moisture source. In addition, other differences are noted, e.g., the Great Plains low-level jet has a marked annual cycle, with the jet developing during the warm season. According to global analyses estimates, the South American counterpart appears to be present most of the year.

Figure 11. Annual cycle of ERA 850 hPa meridional wind

Figure 11. presents the mean monthly 850 hPa meridional wind computed from 15 years of ECMWF reanalyses (ERA): SACZ has a well-defined annual cycle with a maximum during the warm season, but near the Andes the maximum wind appears to be present during all the year. The annual cycle estimated from other global analyses, while revealing similarities over SACZ, depict very different character close to the Andes (not shown). Part of the problem is due to the difficulty for global analyses to represent a phenomenon (the low-level jet) whose transverse scale is of mesoscale nature.

Use of global analyses to describe the moisture fluxes in the larger scales is frequent and the results can be deemed reliable, but a recent study by Wang and Paegle (1996) has brought a note of caution on their use for moisture budget computations. They found that primarily due to large uncertainties in wind analyses, and consequently in the moisture flux convergence, estimates of moisture budgets from different data sets reveal an unacceptable degree of disagreement. The uncertainty is not reduced by using the more recent reanalysis products, as noted in Higgins et al (1996) for North America and by Min and Schubert (1997) for regions around the world, including one in Argentina east of the Andes.

A somewhat different approach is to compute moisture budgets from GCM simulations. Lenters and Cook (1995) followed this method to evaluate the moisture budgets in different regions of South America. According to their results, precipitation in the central Andes is the result of orography and wind convergence. In contrast, evaporation, wind convergence and moisture advection are positive contributors to the balance over SACZ, while orography has no influence and transients have a negative impact. The simulations of Lenters and Cook do not reproduce the precipitation maximum in southern Brazil/northeastern Argentina that could be related to the exit region of the low-level jet. The hope for enhancing the reliability of moisture budgets particularly at regional scales relies in higher resolution analyses or forecasts (Giorgi et al. 1994, Berbery et al. 1996). Berbery and Rasmusson (1999) show that both the diurnal cycle and the smaller scale features of the circulation need to be resolved at higher resolutions to avoid sampling errors that affect significantly the estimates of the moisture budgets at regional scales.

The characteristics of the vertical coordinate in the Eta model (Mesinger et al. 1988) appear to be well suited to represent the sharp slopes of the Andes. Thus, it is not unexpected that several studies use this coordinate to investigate aspects of the circulation over South America (Figueroa et al. 1995; Tanajura 1996; Collini et al. 1997; Nicolini, section 4.5 of this document). Tanajura (1996) used a version of NCEP's Eta model over South America nested into a GCM. In agreement with Lenters and Cook analysis, Tanajura's mountain/no mountain experiments also assert the importance of the Andes in organizing the low-level convergence and precipitation in the region. He concluded that the Andes and the low-level jet to the east are crucial for precipitation over northern Argentina.

To understand the processes affecting the components of the moisture budgets, Berbery and Collini performed a series of short term forecasts using NCEP's Eta model with NCEP/NCAR reanalyses as initial and boundary conditions. Then, the 12-36 hour forecasts were averaged to produce monthly fields.

Figure 12. Eta model forecast precipitation and two observed estimates.

Figure 12 shows that during November 1997 the Eta model develops precipitation in the northeastern part of Argentina/Paraguay/southern Brazil, although not as intense as the observed precipitation. However, precipitation is a difficult variable to measure reliably, and even different observational techniques differ, as evidenced in the lower panels. The Eta model develops precipitation in other regions of interest, like SACZ and southern Chile (observations for southern Chile were not available).

The importance of the low-level jet in transporting moisture to higher latitudes is evidenced in Fig. 13a, that depicts the vertically integrated moisture flux as calculated in the model's computational grid. The exit region of the low-level jet portrays a large area of moisture flux convergence (Fig. 13b) co-located with the maximum in precipitation. Consistent with water balance concepts, the results suggest that moisture flux convergence related to the low-level jet, is a key component in the processes generating precipitation over northern Argentina/southern Brazil.

Figure 13. Mean Vertically integrated moisture flux and its convergence

3.4 The Atlantic ocean impact on Climate Variability over South America

The wind driven oceanic circulation in the eastern equatorial Atlantic is important in determining the patterns of SST variability (Wallace et al., 1989). The South Atlantic ocean circulation is dominated by the wind-driven Subtropical gyre and the Antarctic Circumpolar Current. The western boundary current associated with the subtropical gyre is the southward Brazil Current. The cold subantartic waters of the Malvinas Current flow north along the continental shelf break until they meet the Brazil Current. Both currents then flow eastward toward the interior of the south Atlantic basin. The Brazil-Malvinas Confluence zone (BMCZ) contains one of the world's most intense oceanic fronts and complex SST fields. Water masses with vastly different properties derived from numerous, distant sources are brought into close proximity there, thus setting the stage for globally-significant heat and freshwater exchange. The configuration of the Confluence Zone is thought to depend upon the relative strengths of the Brazil and Malvinas currents. The collision of these two opposing and intense currents spawns one of the most spectacular eddy fields in the global ocean (Olson et al., 1988) and thus a highly complex regional SST pattern that couples to atmospheric wind and to fluxes of heat and moisture.

Figure 14. SST mosaic for 1-5 November 1988 that exhibits the formation of a warm-core eddy from the Brazil current. Image recorded by the Argentine Meteorological Service as part of a collaboration agreement with the Rosenstiel School, University of Miami.

Variations in the circulation and SST in the Southern Atlantic ocean occur over time scales ranging from subseasonal to the seasonal and interannual. These variations are influenced to a large extent by interactions between the opposing flows of the Brazil Current and the Malvinas Current which are affected by the basin scale wind field and probably by the absorption of the Agulhas rings. The spatial characteristics of this variability include changes in the latitude of separation of these currents from the western boundary and changes in the geometry of their eddies into the interior of the BMCZ. Between, 36°S and 39°S, the Brazil current separates from the continental margin and turns to the interior of the Atlantic ocean. Unlike the Gulf stream, the separation latitude of the Brazil current fluctuates considerably. At any given time, therefore, SSTs in this area may depend whether it is occupied by waters from the warm BC, or the cooler Malvinas current, or a mixture of both.

Significant intraseasonal variability occur in the tropical and extra-tropical Atlantic. The production of transient cold-core eddies from the Malvinas Current and warm-core eddies for the Brazil Current serve to make the BMCZ one of the most energetic regions of the World Ocean. A conspicuous precursor to the production of warm-core eddies is the anomalous poleward migration of the BC which forms a complicated intrusive pattern leading to a set of meanders and rings. SST anomalies in these situations can be 10°C on scales of nearly 1000 km in the north-south direction, occurring over time scales of about two months. The dominant signal in SST variability is the annual cycle, which weakens toward the south. There is also a semi-annual component that is near zero amplitude at 30S° and increases to nearly half the magnitude of the annual signal at 50°S. This is likely to be in response to the semi-annual cycle in zonal winds over the Southern Ocean (Podesta et al., 1991). On longer time scales, Venegas et al. (1997) identified three leading modes of variability of the atmosphere-ocean coupled system over the South Atlantic ocean. While the first two modes are dominated by interdecadal variability, the third mode is associated with ENSO time scales. Therefore, SST variations within the WSA appear to be linked to ENSO by mechanisms that remain to be determined.

The modulation of monsoonal rainfall over South America by SST variations in the tropical Atlantic is suggested by several studies (Namias, 1972; Hastenrath and Heller, 1977). A recent study by Diaz et al. (1998) obtained significant relationships between anomalies in rainfall over southeastern South America and in SST in the Pacific and Atlantic oceans. Robertson and Mechoso (1998) examined the annual streamflow of four rivers in southeastern and south-central South America (the Negro, Paraguay, Parana, and Uruguay Rivers) for the period 1911-93. They found the following features 1) a nonlinear trend, 2) a near-decadal component, and 3) interannual peaks with ENSO timescales. The trend and near-decadal components are most marked in the two more central rivers, the Paraguay and Parana, with ENSO timescales variability most pronounced in the Negro and Uruguay rivers in the southeast. An analysis of the results confirmed the influence of ENSO on the streamflow variability of the Negro and Uruguay Rivers (e. g., Mechoso and Perez-Iribarren, 1992), with El Niño associated with enhanced streamflow. On the decadal timescale, high river runoff is associated with anomalously cool SSTs over the tropical North Atlantic. The near-decadal component of the Paraguay and Parana Rivers is strongest in the austral summer. Diaz et al. (1998) confirmed results of previous studies, such as the links between ENSO events in the equatorial Pacific ocean and rainfall anomalies in Uruguay and southern Brazil during late austral spring-early summer and late austral fall-early winter. They also found other relationships that have not been reported before, such as links between SST anomalies in the southwestern Atlantic ocean and rainfall anomalies in the entire region during October-December and April-July. The fact that the years which show positive peaks in these links do not necessarily coincide with those obtained for the Pacific ocean, indicate that the SST anomalies in the Atlantic, may contribute on their own to rainfall anomalies over Uruguay and southern Brazil.

The midlatitude atmosphere forces large SST anomalies through fluctuations in the surface winds, temperatures and humidities. However, the reverse influence of Atlantic SST anomalies on the atmospheric circulation over the North Atlantic remains poorly understood. For example, several atmospheric GCM studies have investigated the impact of northwest-Atlantic SST anomalies, but have failed to reach a consensus (Palmer and Sun (1985); Peng et al. (1995); Robertson et al. (1997); Lau and Nath, 1990 ). It has been found using an atmospheric GCM (Robertson et al., 1999) that Atlantic SST anomalies, particularly those over the South Atlantic, enhance interannual variability to near-realistic values over the North Atlantic. Robertson et al. (1999) formulated a hypothesis on a mechanism for interhemispheric teleconnections. This mechanism is based on the following processes:

 
(1) Anomalously warm SSTs over the tropical and subtropical South Atlantic enhance the low-level meridional temperature gradients between subtropical and midlatitudes. Sea-level pressures rise over the warm SSTs and decrease over the cooler SSTs over the southeast Atlantic south of 30°S, consistent with thermal wind balance. The associated low-level anomalous anticyclone is most intense over the southwest Atlantic, to the west of the warm SST anomalies, and extends into South America. Composites of the GCM's latent-plus-sensible heat flux anomalies are weak over the South Atlantic (not shown), and there is little evidence of direct thermal forcing. In the equatorial region, the anomalous surface fluxes are directed into the ocean and would thus act to reinforce the warm SST anomalies there.

(2) Low-level convergence into the southern portion of the South American convergence zone is amplified while the SACZ is suppressed, giving rise to a net southward and westward shift of convective activity .

(3) The sectorial Hadley circulation emanating from Amazonia is intensified and displaced southward, giving rise to region of anomalous upper-level convergence dryness over Caribbean.

(4) The impact on the NH extra-tropics seen in the GCM may be a barotropic response to the upper-level convergence anomaly over the Caribbean (see Hoskins and Sardeshmukh (1987))

Further work is required to understand the details of how SST anomalies over the South Atlantic influence the South American summer monsoon, and how changes in the latter influence the North Atlantic. A low-resolution GCM compromises the ITCZ and SACZ, which may play important roles in the atmospheres response to SST anomalies over the South Atlantic. In the model by Robertson et al. (1999), SST anomalies do not appear to influence directly the thermally-direct monsoonal circulation. Rather, they appear to modify midlatitude pressure gradients which then in turn appear to influence the convection zone over South America. Longer simulations with higher resolution models are required to determine whether these are due to SST variations over the North Atlantic, as opposed to sampling variations.

The atmospheric circulation in large scales and low frequencies can be considered as a mean state within which cyclones and other phenomena interact. Changes in the mean state are expected to produce changes in the behavior of the transients. The WSA portion just off the South American coast between 30 S and 45 S is one of the region with highest cyclogenetic activity within the Southern Hemisphere. During winter, some eastward travelling cyclonic systems are enhanced after reaching the coast of Uruguay and southern Brazil and adjacent ocean and their trajectories tend to be parallel to the coast. Saraiva and Silva Dias (1997) performed and they showed that SST horizontal gradients have a significative influence on cyclone trajectory and also on surface sensible and latent heating.

During summer the southwestern tropical Atlantic is characterized by the formation of the South Atlantic Convergence Zone (SACZ), which is in some ways analogous to the South Pacific Convergence Zone in the southwestern Pacific. However, the SACZ is less well known in terms of oceanographic consequences than its Pacific counterpart. Kalnay et al. (1986) diagnosed and modeled a strong SACZ event and they found that low-level cyclonic vorticity values were associated with negative SST anomalies. This correlation therefore would indicate that the atmospheric anomalies were causing the SST anomalies, which in turn provide a negative feedback to the atmosphere. However, there are some indications that warm SST anomalies may have a role in the onset of the convection over tropical South America. Therefore the possible influence on intraseasonal time scales, of WSA SST anomalies on SACZ and convection onset should be further explored.
 
 

This section revisits some of the issues addressed in the above review in the context of identification of unresolved issues and establishing priorities.

4.1 Empirical studies
 

A rough documentation of climate variability of the SAMS region has emerged over the years. The annual cycle is the dominant source of convective variability. Deep convection moves rapidly southward from northwest Amazonia near the beginning of September and is fully established by the beginning of October, reaching its maximum intensity by the end of February. It withdraws more gradually, such that the driest period in the central Amazon is in August.

Diurnal variability in the Amazon is also quite large. The time of maximum convection varies depending on location, but the presence of a diurnal cycle is believed to be an important component in forcing many of the observed features of the South American sector (e.g., Figueroa et al. 1995).

Other time scales of variability, while exhibiting variances than those associated with annual or diurnal cycles in the Amazon, also have large impacts on hydroelectric generation, agriculture and human populations.

Sea surface temperatures are believed to exert the single largest systematic influence on interannual variability. Climate variations associated with Pacific SST anomalies (i.e., El Niño) are known to occur on the equatorial west coast, the Amazon basin, northeast brazil, the Parana basin, the central coast of Chile, the cuenca del Plata, and northeast Brazil. Atlantic anomalies may be of equal or greater importance, especially in the eastern part of the continent, although their influence is still largely undocumented, except in northeast Brazil.

In most locations, intraseasonal variability is larger than interannual. On synoptic time scales, cold air surges, stronger during winter, often propagate to the Amazon. At lower frequencies, wave trains, originating in the midlatitude or tropical Pacific, are associated with a dipole structure, with centers of action in the SACZ and subtropical plains. In addition to the dipole, which occurs preferentially at a period of about 40 days and is related to the Madden-Julian oscillation, the SACZ also varies on several distinct, shorter time scales.

Many issues remain to be resolved. They include:
 

1) Decadal impact: Large decadal-scale fluctuations have been observed, but it is difficult to make conclusive statements without longer data records. Fluctuations in the Atlantic are thought to play a major role on these scales.

2) The impact of ENSO: Although we have a qualitative picture of how ENSO affects mean rainfall amounts, better data are needed to determine qualitative relationships. Furthermore, how ENSO affects the dates of the onset and demise of the monsoon, the frequency of extreme rainfall events, and the influence of the intraseasonal oscillation, have yet to be determined.

3) Remote forcing: Although wave trains originating over the Pacific have a clear influence on climate variability on intraseasonal time scales, the ability to predict these events remains elusive. The interannual variability of these wave trains is not known.

4) Local forcing: The relative importance of the complex Andean topography and that of air-sea interaction needs to be further understood.

5) The role of the low-level jet in transporting moisture into southern South America.
 

4.2.1 Downscaling Methods.
 

Simulation of monsoonal circulations requires addressing a number of key modeling issues pertaining to spatial scale interactions. This can be done by nesting limited area models within GCM or driving mesoscale models with global data sets to regionally enhance gridded analysis at a coarser resolution. This approach is discussed in the CLIVAR/VAMOS report (CLIVAR Principal Research Areas). Also, variable resolution global models may be used which allow for meso-scale processes to influence the large scale. This issue is relevant in prognostic efforts when the resolved subdomain includes strong latent heat release that could modify global scale circulations.

There is little doubt that high local resolution is required to reproduce the several key elements controlling regional monsoonal oscillations. Adequate resolution of phenomena such as low-level-jets flanking both sides of the Andes and Rockies, moisture surges above the narrow sea of Cortez, and the full blocking potential of the central and South American Cordillera which locally extend through the lower half of the troposphere, may all require local grid size on the order of a few tens of km. Such resolutions are currently impractical over the entire globe, particularly for climate simulation.

Methods to provide regionally enhanced depictions are clearly needed over South America, a continent of sharp orographic and surface vegetation contrasts. The Joint PACS-GCIP Modeling Workshop held on 1-3 October 1997, Silver Spring, Maryland, also identified the importance of accurate downscaling to produce regional precipitation forecasts from GCM-predicted planetary scale flow patterns over North-America. Feedbacks from the local to the global scales remain to be documented, partly because of the lack of previous high resolution global climate simulations.

A second approach would be to telescope, or regionally scale down a global model with reasonably conservative numerics to the point that local resolution suffices to resolve key mesoscale elements. This approach is relatively novel, but an increasing number of research and operational centers are now exploring the technique as an alternative to nested mesoscale models in real time prediction. It would be interesting to compare the relative benefits of mutually interacting mesoscale and global models with regionally enhanced variable resolution global models for the purposes of climate simulation. SAMS provides an optimal geographic and scientific setting for this purpose.

4.2.2 Assessment of energy budgets in regional models.
 

In a recent review paper, Betts et al. [1996] emphasized the critical role of surface energy budgets in determining the boundary conditions that control weather and climate at different time scales. Consequently, the subject has been the interest of researchers for many years. From a climate diagnostics perspective, surface energy budgets help understand the physical processes by which the atmosphere gains energy [see, e.g., Betts et al. 1996]. From a modeling standpoint, there is a need to evaluate the way surface energy budget components are treated in the respective parameterizations, so that the models can address reliably issues related to climate assessments, seasonal-to-interannual predictions or climate change scenarios. As a result, many articles evaluate the performance of models in terms of surface energy processes [e.g., Gutowski et al. 1991, Garratt et al. 1993, 1998, Berbery et al. 1999].

A major roadblock in analyzing continental surface energy budgets is the lack of extensive observational information to validate the models' products. However, limited (in space or time) data sets and estimates from satellites may help to add reliability to the models' assessments. Even in the case in which no data are available, comparing similar products of different models is useful since the differences may shed information on the accuracy of their respective parameterizations [Gutowski et al. 1991].

Operational seasonal forecasting for South America is currently available from global circulation models. The International Research Institute (IRI) maintains a web site with the operational forecasts from NCEP, NCAR and the Max Plank Institute. The ECMWF products are also available in the web in a delayed mode. The Brazilian Center for Weather Forecasting and Climate Studies (CPTEC) also performs operational seasonal forecasting with the ensemble technique. The experience in the last El Niño episode was quite satisfactory in large scale aspects but lacked regional geographical definition. However, there is a tendency to nest regional models into the global models with the purpose of capturing the regional effects such as the topography and land-surface processes.

The global model forecast drift produced by the CPTEC model indicates a certain lack of latent heat release in the tropical sector of South America (Silva Dias et al. 1998). As a result, important features of the summer tropospheric circulation are not well represented such as the Bolivian High which shows a westward drift with increased forecast time as well as a weakening of the upper tropospheric low off the NE coast of Brazil. The latent heat release along the South Atlantic Convergence Zone is also reduced with possible important implications for the climate in southern Brazil, northern Argentina, Paraguay, Bolivia and Uruguay. The NCEP forecast drift, although somewhat smaller than the CPTEC also indicates significant problems in the location and intensity of the heat sources over land areas. An important improvement can be achieved by using higher resolution mesoscale models (Giorgi et al 1994). Still some deficiencies persist. It has been shown that over the North American continent, there are biases in the short wave radiation at the surface of about 25-50 W/m² and 10-25 W/m² in the surface heat fluxes, with important consequences in the Bowen Ratio (Berbery et al. 1999). The forecast experience at the Department of Meteorology of the University of Sao Paulo with the Regional Atmospheric Modeling System RAMS in short range (48 hours) forecast as well as in climate regional modeling in the Amazon have indicated excessive moistening at low level leading to extensive cloudiness either at the surface (fog) in the morning or low stratiform cloudiness. As a result of the extensive stratiform cloudiness in the morning, there is a slow surface warming and a significant delay in the mixed layer growth. Thermodynamical instability is also reduced with significant impact on the convective precipitation. There are indications th at the model lacks vertical heat and moisture transport above the mixed layer. Diagnostic studies indicate that shallow moist convective parameterization needs to be improved. Recent experiences with the RAMS model support this statement (Silva Dias et al., 1998).

Models indicate large sensitivity on the soil moisture initialization. Heterogeneous soil moisture patterns in the initial condition may have an impact during several days on the surface temperature and moisture with a detectable impact on the precipitation pattern. These effects are particularly evident in the tropical rain forest and neighboring deforested areas with the Cerrado vegetation. There are indications that the memory is related to the root depth (larger in the tropical forest and in the Cerrado). Whether the soil moisture memory effect is realistic or not remains to be shown in observational studies (one of the LBA targets).

Based on the current experience with the global climate seasonal forecasting and the regional modeling efforts in the South American region, it was perceived the need to perform detailed evaluations by comparisons to conventional and non-conventional information. Higher resolution satellite data provides an important data set for the surface energy balance as well as the cloud morphology and structure of convective systems. Besides the assessment of the models with the conventional observational surface and upper air network, special data sets as those to be provided by planned field experiments such as LBA and the SAMS LLJ field experiment will be crucial in improving the land surface parameterizations that have an important effect on the seasonal and weather forecasting.

The regional assessment of the climate and forecasting models will also serve for planning of future continental scale experiments in South America, such as the study of the La Plata River Basin in a combined effort with hydrological and environmental communities.

4.3 SAMS Data Requirements

The three major objectives of the VAMOS program are:

• to describe, model, and understand the mean and seasonal aspects of the American monsoon systems,
• to investigate predictability and to make predictions to a feasible extent, and
• to prepare products in view of meeting societal needs.

The SAMS working group encourages the free and open access to meteorological data. Data access will be achieved through existing data centers, research institutions and universities, rather than through the establishment of a single centralized data center. Data management links will be established through a World Wide Web Home Page for VAMOS. While VAMOS/SAMS will be able to take advantage of existing data sets, a number of specific data sets will be required to meet the scientific goals of the program. The development of an archive of historical precipitation data will be a key activity. This development will build on current activities in IAI, PACS, LBA and other regional and national activities. The SAMS working group encourages the national weather services in South America to maintain and support valuable observations systems, and to participate with the scientific community in developing improved climate prediction techniques.
 

Historical Surface Observations

Precipitation

Riverflow

Real-time Surface Observations

Global Telecommunications System

Automated Surface Observations

Real-time Upper-Air Observations

Radiosonde and Profiler Data

Non-Conventional Observations

a) Commercial Aircraft Data

b) Lightning Data

c) Satellite Soundings/Winds

Field Experiment Data Sets

Additional Data Requirements

Enhanced monitoring of precipitation at high elevations

4.4 Extended Monitoring

The objectives of extended observations are to:

1) provide observations to describe the slowly varying atmospheric conditions over South America, particularly at the intraseasonal to interannual time scales.

2) provide comparisons with the various reanalyses products and other analyses produced by data assimilation systems

3) to provide the countries of the region with observations to validate and improve meso-scale models used both to downscale climate signals (see 4.2.1 above) and to verify operational forecasts on various time scales.

The latter objective, though not central to those of VAMOS objectives, is recognized as important to the overall long-term sustainability of any network in the region.
 
 

4.4.1 Special atmospheric sounding networks for monitoring climate variability

Since the atmospheric sounding network over South America is not sufficiently dense to describe many important aspects of the variability over the region, even on monthly time scales, it will be an objective of VAMOS to establish additional stations that will be able to provide regular wind, temperature and moisture profiles. The exact nature of the observations will depend upon the available resources and technologies, but some aspects can probably be described. As a minimum, additional wind profiles will be needed to describe the large gyre that includes the inflow into the Amazon basin of tropical Atlantic air, and the subsequent exit of this airstream much farther south as the low-level jet over eastern Bolivia and Paraguay. Currently there are no observations that routinely describe the strong east Andean jet; the feature that is clearly shown in most reanalyses.

Justification for additional sounding observations stems from the need to describe the intra to interannual variations in the tropospheric circulation associated with variations in the rainfall over these same time scales. The current sounding network, even when fully functioning, does not have the spatial resolution needed to describe major anomalies in the atmospheric circulation over large parts of tropical South America . By operating stations in key locations over the continent (see Fig. 15) it should be possible to determine whether much of the variability suggested by the routinely produced global reanalyses is indeed real. This will help establish confidence in these products, or alternatively, will help to force changes in them so that they are more realistic.
 

Figure 15. The figure shows a possible configuration of an enhanced monitoring network that should provide a basic description of the flow in the vicinity of the LLJ and also over the neighboring altiplano. The hypothesized network would be part of a larger network, currently in place (PACS_SONET).

Although the need for special observations over South America is important for climate analysis, it is likely that extended monitoring will not be affordable without the support of the meteorological services in the region. Thus, it should be a priority of VAMOS to see that such observations not only are transmitted in near real-time, but that the meteorological services of the region have access to these observations and use them for shorter-range forecasts. If the value of the observations for daily weather forecasting is made apparent, then obtaining help to maintain operation of the network will be easier.
 

4.4.2 Special raingauge neworks for monitoring climate variability

An important additional source of data for the operational monitoring of climate and its variability will be rainfall observations. Currently the routinely available network of rain gauges is too sparse to provide accurate estimates over small regions, and given the irregular nature of convective rainfall, even monthly estimates of rainfall are not possible with high accuracy over regions as large as, for example, Bolivia. To improve the estimation of precipitation, to help calibrate space-based estimates of rainfall, and to help quantify precipitation anomalies it will be a VAMOS objective to establish selected raingauge networks in high scientific priority regions within the VAMOS domain. These observations will serve to calibrate space-based estimates of rainfall. Initial sites would likely be in the Andean altiplano, where high-based convective clouds produce less rainfall than might be estimated from IR-based procedures, and along the eastern slopes of the Andes, where relatively shallow, but persistent cloudiness may produce large rainfall accumulations.

In addition to providing baseline values for satellite precipitation estimates, the rainfall networks should provide accurate estimates of rainfall variability over the regions they represent. Because the networks might be composed of dozens of stations over regions of 1000 to 10,000 square km, they will be more reliable estimates than averages of a sparse network spread over different geographical regions. Such networks would also be insensitive to small changes in the distribution or number of stations within the network.
 

4.4.3 Field process studies

In addition to an extended monitoring program, it will be necessary to make special higher resolution (both in space and time) observations to answer key questions related to the evolution and maintenance of key features of the atmospheric circulation over the VAMOS domain. One feature that has been identified as not only very important in the overall circulation over South America, but a feature that is virtually undescribed by the current atmospheric sounding network is the low-level jet that lies east of the Andes, generally between 10 and 20° S. This feature, which is in many ways comparable to the low-level jet over North America, is present in all months. Although some aspects of this circulation feature can be described through enhanced monitoring, other aspects cannot be addressed. For example, the diurnal variation of the low-level flow has not been described, although mesoscale simulations indicate that it is considerable. Is there a particular time of day when an observation will more closely represent the daily averaged wind? An answer to this question might help improve a long-term monitoring strategy.

The relationship between mesoscale convective system development and synoptic variability of the jet, although seemingly a weather prediction problem, also has important implications for climate variability. A substantial part of the rainfall associated with wet and dry anomalies during the warm season can be due to variations in the number of mesoscale convective systems that form in a region. These mesoscale systems, at least in relatively flat regions, are associated with favorable synoptic-scale flow patterns. Determining these conditions in a global model may allow for statistical prediction of the frequency of mesoscale convective systems, and thus rainfall variability in regions of central South America.

A field campaign would also provide better information on cross-jet details of the moisture and wind field than from a less-dense monitoring effort. Both of these fields need to be well-described to calculate the moisture transport and moisture convergence over regions. While it might appear that some of these questions do not directly relate to climate variability over the region, an improved knowledge of these shorter-time scale features should improve our understanding of the large-scale and mesoscale models to reproduce these features. The design of a more adequate network for monitoring climate variability over the region should be an outcome of these studies.

The host of surface and upper air stations to be set in the frame of the LLJ experiment will be also improve our understanding of the sub-monthly variability of convective rainfall over the Altiplano and the role of South American cold surges on the convection over the subtropical low-lands. Given the recurrence and intensity of cold air incursions along the east side of the subtropical Andes, field experiments in this region will likely capture passages of this phenomenon providing valuable high-resolution observations for its mesoscale analysis and validation of modeling studies (e.g., Garreaud 1999b). On the other hand, hypotheses concerning the important role of the easterly, upslope flow on the variability of the Altiplano rainfall (Garreaud 1999a) can be tested by an east-west transect with pibal and radio-sonde stations.
 
 

4.5 South American low-level jet east of the Andes Field Program

The VAMOS Panel decided during the Sao Paulo meeting to 1) endorse a field program on the South American LLJ, and 2) appoint a South American Monsoon Working Group charged with developing cooperative, international research to investigate the South American Monsoon System (SAMS), including the LLJ. The report from that workshop includes a discussion of the scientific background. During the Miami meeting of the SAMS working group reported here, an initiative on a South American low-level jet program was discussed.

The South American low-level jet program is an international effort to understand the nature of this LLJ as a moisture conveyor from the tropics to the extratropics and its role in regional climate. It includes research and field components. The Planning Committee on this LLJ is presently coordinated by Argentina and is integrated by the following countries and institutions that have confirmed their participation (most of them have appointed representatives and have explicited their in-kind contribution to the Program): Argentina: Dept. of Atmospheric Sciences (Univ. de Buenos Aires)- Argentina National Weather Service. Brazil: Centro de Previsao de Tempo e Estudos Climaticos (CPTEC) -Dept. of Atmospheric Sciences (Universidad de Sao Paulo). Chile: Dept. Geofisica (Universidad de Chile). Paraguay: Direccion de Meteorologia e Hidrologia (DINAC)-LIAPA- Universidad de Asuncion. This Committee strongly encourage a broader international cooperation from other countries involved in the study of this South American feature.

Overarching goals of the LLJ program are: characterization of its dimensions, description of its temporal and spatial variability, and to answer basic questions about its transport properties, diurnally oscillating mechanisms, interaction with large scale circulation and land surface forcing factors including orography, vegetation cover and soil moisture. Additionally, it is expected that an improved description of organized convection and severe weather will lead to advances in parameterizations used in meso-scale models that downscale climate predictions. Special observations from this program will help fill South-American data gaps and therefore improve gridded analysis used both for climate studies and as initial conditions to weather prediction models.

5. Readiness and Priorities.

The monsoon system in the Southern Hemisphere has been identified as one of CLIVAR's principal research areas. Workshops in Sao Paolo (March 30 - April 1, 1998) and in Miami (October 22 - 24, 1998, summarized in this report) have highlighted key aspects of the SAMS and identified questions that need to be addressed before substantive advances can be made on this topic. Such advances will have practical benefits for the region as reviewed in previous sections. A point not previously emphasized in this report is the impact that monsoonal circulations over South America have in North America and the North Atlantic winter flow. There is mounting evidence that the wave train excited by convection over tropical South-America exhibit a meridional structure (Grimm and Silva-Dias, 1995, Nogues-Paegle et al, 1998) that extend into the Northern Hemisphere. Errors in the simulation of this convection (Nogues-Paegle et al, 1998) cause a systematic equatorward shift of the North-American subtropical jet in time scales of about 5 days. Maintenance of tropical convection and resulting wave patterns is a pre-requisite for adequate simulation of tropical-extratropical interactions in long-term climate simulations. Furthermore, gains in understanding of low-level circulations over South-America may be transferable to North America, since both region exhibit complex underlying terrain and highly variable surface conditions.

Escalation of SAMS research is expected due to the readiness of the region to embark and support collaborative research on SAMS. There are now a number of meso-scale models that are routinely integrated over different regions of South America (the MM5, the Eta model, the RSM, etc) Also expanded observational capabilities (summarized in section 2.) promise a level of detailed information on South American atmospheric motions never before attained. The availability of retropective analysis of conventional and remotely sensed data by several operational centers (over 40 year of reanalysis/CDAS data from NCEP) makes possible statistically significant studies of processes associated with long term variability.

Workshop discussion recognized data issues as the main priorities for SAMS to focus on. Several recommendations on this topic are listed below:

• use the potential development of a SA gridded precipitation archive as leverage to obtain historical data sets.
• encourage the development of a GHCN monthly data over South America.
• identify opportunities for augmentation of upper level network (radiosoundings from ships, glidersondes, remotely controlled aircraft, wind profilers,etc.)
• encourage agreement between national weather services and airline companies to outfit commercial aircraft to measure and transmit weather data,
• complement proposed activities on programs to measure the LLJ east of the Andes with an extended monitoring program of upper level soundings, and surface fluxes
• trace gas measurements at surface stations in the upper Andes to assess airmass origin,
• instrument energy and water budget sites not covered by LBA (Altiplano, Cuenca del Plata)
• increment the density of precipitation observations in the east slope of Andes,
• assess how the Andes complex orography affect weather radar signals,
• identify a site for holdings of comprehensive data sets for the region, including information and pointers to high-resolution data sets, both from remote platforms, and pseudo-data from regional models.

References

Aceituno, P. F., 1988: On the functioning of the Southern Oscillation in the South America sector. Part I: Surface climate. Mon. Wea. Rev.,116, 505-524.

Aceituno, P. F. and R. Garreaud, 1995: Impacto de los fenomenos El Niño y La Niña en regimenes fluviometricos andinos. Rev. Soc. Chilena Ing. Hidraulica.. 10, 33-43.

Aceituno, P. and A. Montecinos, 1997: Patterns of convective cloudiness in South America during the austral summer from OLR pentads. Preprints of 5th Int. Conf. on S. Hemisphere Meteor. and Oceanography. Amer. Meteor. Soc., pp 328-329.

Berbery, E. H. and J. Nogues-Paegle, 1993: Intraseasonal oscilations between the Tropics and Extratropics in the Southern Hemisphere. J. Atmos. Sci. , 50, 1950-1965.

Berbery, E. H. and E. M. Rasmusson 1999: Mississippi moisture budgets on regional scales. Monthly Weather Review. In press.

Berbery, E. H., E. M. Rasmusson and K. E. Mitchell, 1996: Studies of North American continental-scale hydrology using Eta model forecast products. J. Geophys. Res., 101, D3, 7305-7319.

Berbery, E. H., K. E. Mitchell, S. Benjamin, T. Smirnova, H. Ritchie, R. Hogue and E. Radeva, 1999: Assessment of land surface energy budgets from regional and global models. Accepted in J. Geophys. Res. .

Betts, A. K., J. H. Ball, A. C. J. Beljaars, M. J. Miller and P. Viterbo, 1996: The land surface-atmosphere interaction: A review based on observational and global modeling perspectives. J. Geophys. Res., 101, D3, 7209-7225.

Buchmann, J., P. L. Silva Dias, and A. D. Moura, 1986: Transient convection over the Amazon Bolivia region and the dynamic of the droughts over north-east Brazil. Arch. Meteor. Geophys. Bioklim., A34, 367-384.

Casarin, D. P., and V. Kousky, 1986: Anomalias de precipitacao no Sul do Brasil e variacoes na circulacao atmosferica. Revista Brasileira de Meteorologia, 1, 83-90.

Castaneda, M.E., and V.R. Barros, 1994: Las tendencias de la precipitacion en el Cono Sur de America al este de los Andes. Meteorologica, 19, 23-32.

Collini, E. A., E. H. Berbery and E. Rogers, 1997: Applications of the Eta model to the South American region. 5th International Conference on Southern Hemisphere Meteorology and Oceanography, 7-11 April 1997, Pretoria, South Africa.

D'Almeida, C., 1997: Oscilacoes intrasazonais de precipitacao na estacao chuvosa em Sao Paulo e condicoes atmosfericas associadas. M.Sc. Dissertation, University of Sao Paulo.

Diaz, A. F., C. D. Studzinski and C. R. Mechoso, 1998: Relationships between precipitation anomalies in uruguay and southern Brazil and sea-surface temperature in the Pacific and Atlantic oceans. J. of Climate , 11, 251-271.

Figueroa, S. N., P. Satyamurty and P. L. Silva Dias, 1995: Simulation of the summer circulation over the South American region with an Eta coordinate model. J. Atmos. Sci., 52, 1573-1584.

Fortune, M. A., and V. E. Kousky, 1983: Two severe freezes in Brazil: Precursors and synoptic evolution. Mon. Wea. Rev., 111, 181-196.

Fu, R., B. Zhu and R. E. Dickinson, 1998: How do atmosphere and land surface influence seasonal changes of convection in the tropical Amazon? Submitted to J. Climate.

García, N. O. and W. M. Vargas, 1998: The temporal climatic variability in the Río de la Plata basin displayed by the river discharges. Climatic Change, 38, 359-379

Garratt, J. R., P. B. Krummel, and E. A. Kowalczyk, 1993: The surface energy balance at local and regional scales-A comparison of general circulation model results with observations. J. Climate, 6, 1090-1109.

Garratt, J. R., A. J. Prata, L. D. Rotstayn, B. J. McAvaney and S. Cusack, 1998: The surface radiation budget over oceans and continents. J. Climate, 11, 1951-1968.

Garreaud, R. D., 1999a: A multi-scale analisys of the summertime precipitation over the central Andes. Accepted for publication. Mon. Wea. Rev.

Garreaud, R. D., 1999b: Cold air incursions over subtropical and tropical South America. A numerical case study. Conditionally accepted in Mon. Wea. Rev.

Garreaud, R. D., and J. M Wallace, 1997a: The diurnal march of the convective cloudiness over the Americas. Mon. Wea. Rev., 125, 3157-3171.

Garreaud, R. D., and J. M Wallace, 1997b: Mid-latitude air incursions over tropical South America. Proceedings of the 22th Annual Climate Diagnostics and Prediction Workshop. Amer. Meteor. Soc., pp. 258-261.

Garreaud, R. D., and J. M Wallace, 1998: Summertime incursions of mid-latitude air into tropical and subtropical South America. Mon. Wea. Rev., 126, 2713-2733.

Giorgi, F., S.W. Hosteler and C. Shields Brodeur, 1994: Analysis of the surface hydrology in a regional climate model. Quart. J. Roy. Meteor. Soc., 120,161-183.

Grimm, A. M. and P.L. Silva Dias, 1995: Analysis of tropical-extratropical interactions with influence functions of a barotropic model, J. Atmos. Sci., 52, 3538-3555.

Grimm, A.M. and S.E.T. Ferraz, 1998a: Sudeste do Brasil: uma regiao de transicao no impacto de eventos extremos da Oscilacao Sul. Parte 1: El Niño. Preprints of the 10th Brazilian Meteorological Congress, Brasilia, Brazilian Meteorological Society.

Grimm, A.M. and S.E.T. Ferraz, 1998b: Sudeste do Brasil: uma regiao de transicao no impacto de eventos extremos da Oscilacao Sul. Parte 2: La Niña. Preprints of the 10th Brazilian Meteorological Congress, Brasilia, Brazilian Meteorological Society.

Grimm, A M., S. E. T. Ferraz and J. Gomes, 1998a: Precipitation anomalies in Southern Brazil associated with El Niño and La Niña events. J. Climate, 11, 2863-2880..

Grimm, A. M.; P. Zaratini, and J. Marengo, 1998b: Impacts of the extremes of the Southern Oscillation on rainfall over Amazonia. Preprints of the 11th Conference on Applied Climatology, Dallas, Texas, American Meteorological Society, 103-106.

Grimm, A.M.; I. Muller; C. Kruger and E. Kaviski, 1998c: Variacoes pluviometricas nos Estados de Sao Paulo e Parana entre os periodos pre e pos-1970 e suas possiveis causas. Preprints of the 10th Brazilian Meteorological Congress, Brasilia, Brazilian Meteorological Society.

Gutowski, W. J., Jr., D. S. Gutzler and W.-C. Wang, 1991: Surface energy balances of three General Circulation Models: Implications for simulating regional climate change. J. Climate, 4, 121-134.

Hastenrath, S. and L. Heller, 1977: Dynamics of climatic hazards in northeast Brazil. Quart. J. Royal Met. Soc. ,103, 77-92.

Hendon, H. H. and M. L. Salby, 1994: The life-cycle of the Madden-Julian oscillation. J. Atmos. Sci. , 51, 2225-2237.

Higgins, R. W., K. C. Mo and S. D. Schubert, 1996: The moisture budget of the Central United States in spring as evaluated in the NCEP/NCAR and the NASA/DAO reanalyses. Mon. Wea. Rev., 124, 939-963.

Horel, J., A. Hahmann and J. Geisler, 1989: An investigation of the annual cycle of the convective activity over the tropical Americas. J. Climate, 2, 1388-1403.

Hoskins, B. J. and P. D. Sardeshmukh, 1987: A diagnostic study of dynamics of the northern Hemisphere winter of 1985-86. Quart. J. Roy. Meteor. Soc., 113, 759-778.

Kalnay, E., K.C. Mo and J. Paegle, 1986: Large amplitude, short-scale stationary Rossby waves in the Southern Hemisphere: Observations and mechanistic experiments to determine their origin. J. Atmos. Sci., 43, 252-275.

Kiladis, G. and K. M. Weickmann, 1992a: Circulation anomalies associated with tropical convection during northern winter. Mon. Wea. Rev., 120, 1900-1923.

Kiladis, G. and K. M. Weickmann, 1992b: Extratropical forcing of tropical Pacific convection during northern winter. Mon. Wea. Rev., 120, 1924-1938.

Kleeman, R., 1989: A modeling study of the effect of the Andes in the summertime circulations of the tropical Americas. J. Atmos. Sci., 46, 3344-3362.

Kodama, Y.-M., 1992: Large-scale common features of subtropical precipitation zones (the Baiu frontal zone, the SPCZ, and the SACZ) Part I: Characteristics of subtropical frontal zones. J. Meteor. Soc. Japan, 70, 813-835.

Kousky, V. E., 1979: Frontal influences on Northeast Brazil. Mon. Wea. Rev., 107, 1140-1153.

Kousky,V.E., 1980: Diurnal rainfall variation in northeast Brazil. Mon. Wea. Rev., 108, 488-498.

Kousky, V. E., 1988: Pentad outgoing longwave radiation climatology for the South American sector. Revista Brasileira de Meteorologia, 3, 217-231.

Kousky, V. E. and M. T Kayano, 1994: Principal modes of outgoing longwave radiation and 250-mb circulation for the South American sector. J. Climate ,7, 1131-1143

Li, T. and G. H. Philander, 1996: On the annual cycle of the eastern equatorial Pacific. J. Climate, 9, 2986-2998.

Lau, N.-C and M. J. Nath, 1990: A general circulation model study of the atmospheric response to extratropical SST anomalies observed in 1950-79. J. Climate, 3, 965-989.

Lenters, J. D., and K. H. Cook, 1997: On the origin of the Bolivian high and related circulation features of the South American climate. J. Atmos. Sci, 54, 656-677.

Lenters, J. D. and K. H. Cook 1995: Simulation and diagnosis of the regional summertime precipitation climatology of South America. J. Climate, 8, 2988-3005.

Liebmann, B.; G. N. Kiladis; J. A. Marengo, T. Ambrizzi and J. D. Glick, 1998: Submonthly convective variability over South America and the South Atlantic Convergence Zone. Submitted to J. Climate

Marengo, J.A., 1992: Interannual variability of surface climate in the Amazon Basin. Int. J. Climatology, 12, 490-504.

Marengo, J., 1998: Observed historical long-term trends in precipitation in the Amazon Basin since the late 1920's. Preprints of the 10th Symposium on Global Change Studies, Dallas, Texas, American Meteorological Society, 135-138.

Mechoso, C. R. And G. Pérez Iribarren, 1992: Streamflow in Southeastern South America and the Southern Oscillation. J. Climate, 5, 1535-1539.

Meisner, B. and P. Arkin, 1987: Spatial and annual variations in the diurnal cycle of large-scale tropical convective cloudiness and precipitation. Mon. Wea. Rev.,115, 2009-2032

Mesinger, F., Z. I. Janjic, S. Nickovic, D. Gavrilov D. G. Deaven, 1988: The step-mountain coordinate: Model description and performance for cases of Alpine lee cyclogenesis and for a case of Appalachian redevelopment. Mon. Wea. Rev., 116, 1493-1518.

Min, W., and S. Schubert 1997: The climate signal in regional moisture fluxes: A comparison of three global data assimilation products. J. Climate, 10, 2623-2642.

Mitchell, T and J. M. Wallace, 1992: The annual cycle in equatorial convection and sea-surface temperature. J. Climate, 5, 1140-1156

Mo., K. C., and R. W. Higgins, 1998: The Pacific South American modes and the tropical intraseasonal oscilla tion. Mon. Wea. Rev.,126, 2009-2032

Namias, J., 1972: Influence of Northern hemisphere general circulation on drought on Northeast Brazil.Tellus,24,336-342.

Negri, A., R. Adler, E. Nelkin and G.Huffman, 1994: Regional rainfall climatologies derived from Special Sensor Microwave Imager (SSM/I) data. Bull. Amer. Meteor. Soc. , 75 , 1165-1182.

Nogués-Paegle. J. and K.-C. Mo, 1997: Alternating wet and dry conditions over South America during summer. Mon. Wea. Rev., 125, 279-291.

Nogués-Paegle. J., K.-C. Mo and J. Paegle, 1998: Predictability of the NCEP-NCAR reanalysis model during austral summer. Mon. Wea. Rev., 126, 3135-3152.

Olson, D. B.; G. P. Podesta, R. H. Evans and O. B. Brown, 1988: Temporal variations in the separation of the Brazil and Malvinas Currents. Deep-Sea Res., 15, 1971-1990.

Pacanowski, 1996: Why is the ITCZ mostly north of the equator. . J. Climate, 9, 2958-2972.

Palmer, T. N. and Z. Sun, 1985: A modelling and observational study of the relationship between sea surface temperature in the north-west Atlantic and the atmospheric general circulation. Quart. J. Roy. Meteor. Soc., 111, 947-975.

Peng S., L. A. Mysak, H. Ritchie, J. Derome, and B. Dugas, 1995: The differences between early and midwinter atmospheric responses to sea surface temperature anomalies in the northwest Atlantic. J. Climate, 8, 137-157.

Podesta, G. P.; O. B. Brown and R. H. Evans, 1991: The annual cycle of satellite-derived sea surface temperature in the Southwestern Atlantic ocean. J. Climate, 4, 457-467.

Philander, S. G. H.; D. Gu; D. Halpern; G. Lambert; N. C. Lau, T. Li, and R.C. Pacanowski, 1996: Why the ITCZ is mostly north of the Equator, J. Climate, 12, 2958-2972.

Pisciottano, G., A. Díaz, G. Cazes and C. R. Mechoso 1994: El Niño-Southern Oscillation impact on rainfall in Uruguay. J. Climate , 7, 1286-1302.

Rao, V. B. and K. Hada, 1990: Characteristics of rainfall over Brazil: annual variations and connections with the Southern Oscillation. Theor. Appl. Climatol., 42, 81-90.

Roberston, A. W., M. Ghil, and M. Latif, 1997: Interdecadal changes in atmospheric low-- frequency with and without boundary forcing. J. Atmos. Sci. submitted.

Robertson, A. W., and C. R. Mechoso, 1998: Interannual and decadal cycles in river flows of southeastern South America. J. Climate, 11, 2570-2581.

Robertson, A. W., C. R. Mechoso, and Y.-J. Kim, 1999: The influence of Atlantic sea surf ace temperature anomalies on the North Atlantic Oscillation. J. Climate, submitted.

Ropelewski, C. F. and M. S. Halpert, 1987: Global and regional scale precipitation patterns associated with ENSO. Mon. Wea. Rev.,115, 1606-1626.

Saraiva, J. M. B. and P. L. Silva Dias, 1997: A case study of intense cyclogenesis off t he southern coast of Brazil: Impact of SST, Stratiform and deep convection. 5th AMS Conference on Southern Hemisphere Meteorology and Oceanography, Pretoria, South Afr ica, 7-11 April 1997, Pre-prints, 368-369.

Schrage, J. M. and D. G. Vincent, 1996: Tropical convection on the 7-21 day timescales over the western Pacific. J. Climate, 9, 587-607.

Silva Dias, P. L., W. H. Schubert and M. DeMaria, 1983: Large scale response of the tropical atmosphere to tran- sient convection. J. Atmos. Sci, 40, 2689-2707.

Silva Dias, P. L., J. P. Bonatti and V. E. Kousky, 1987: Diurnally forced tropical tropospheric circulation over South America. Mon. Wea. Rev.,115, 1465-1478.

Silva Dias, P., J. Aravequia and M. A. F. Silva Dias, 1998: SAMS model requirements.

Tanajura, C. A. S. 1996: Modeling and analysis of the South American Summer Climate. Ph.D. Thesis Dissertation, University of Maryland.

Venegas, S. A., L. A. Mysak, and D. N. Straub, 1997: Atmosphere-Ocean coupled variability in the South Atlantic. J. Climate, 10, 2904-2920.

Vincent, D. G.; A. Fink; J. M. Schrage and P. Speth, 1998: High- and low-frequency intraseasonal variance of OLR on annual and ENSO timescales. J. Climate, 11, 968-986.

Virji, H. 1981: A preliminary study of summertime tropospheric circulation patterns over South America estimated from cloud winds. Mon. Wea. Rev., 109, 599-610.

Wallace, J. M., T. Mitchell, and C. Deser, 1989: The influence of sea-surface temperature on surface wind in the eastern equatorial Pacific. Seasonal and interannual variability. J. Climate, 2, 1492-1499.

Wang, M. and J. Paegle, 1996: Impact of analysis uncertainty upon regional atmospheric moisture flux. J. Geophys. Res., 101, D3, 7291-7303.

Zhao, Y. and B. Weare, 1994: The effect of diurnal variation of cumulus convection on large-scale low-frequency oscillations in the tropics. J. Atmos. Sci, 51, 2653-2663.

Zhou, J., and K. M. Lau, 1997: Does a monsoon climate exist over South America? J. Climate, 11, 1020-1040.