Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 3 27

www.elsevier.com/locate/palaeo

Annual and inter-annual variability of the present climate in northern

South America and southern Mesoamerica
German Poveda a,*, Peter R. Waylen b, Roger S. Pulwarty c
a
Escuela de Geociencias y Medio Ambiente, Universidad Nacional de Colombia, Medelln, Colombia
b
Department of Geography, University of Florida, Gainesville, FL, USA
c
Cooperative Institute for Research in Environmental Sciences/Climate Diagnostics Center, University of Colorado, Boulder, CO, USA
Received 21 September 2004; accepted 20 October 2005

Abstract

Present climate of northwestern South America and the southern Isthmus is detailed in terms of major hydro-climatic controls,
supported by evidence from station records, reanalysis data and satellite information. In this tropical region, precipitation is the
principal hydro-climatological variable to display great variability. The primary objective is to view the controls that operate at
intra-seasonal to inter-decadal time scales. This is a topographical complex region whose climate influences range in provenance
from the South Atlantic to the Canadian Prairies, and from the North Atlantic to the Eastern Pacific. The situation is further
complicated by interactions and feedbacks, in time and space, between these influences, which are interconnected over various
scales. The greatest single control on the annual cycle is the meridional migration of the Inter-tropical Convergence Zone and its
pattern of associated trade winds. Consideration of these alone and their interaction with the Cordilleras of the Andes and Central
America produce a variety of unimodal and bimodal regimes. Regionally, two low level jet streams, the westerly Choco jet (58N)
and the easterly San Andres jet (12148N), and their seasonal variability, have tremendous significance, as do mesoscale
convective storms and mid-latitude cold fronts from both the northern (bnortesQ) and southern (bfriagemsQ) hemispheres. There
are many examples of hydro-climatological feedbacks within the region. Of these the most notable is the interaction between
evaporation over the Amazon, precipitation onto the eastern Andes and streamflow from the headwaters of the Amazon. This is
further compounded by the high percentages of recycled precipitation over large areas of the tropics and the potential impacts of
anthropogenic modification of the land surface. The El Nino-Southern Oscillation phenomenon (ENSO) is the greatest single cause
of interannual variability within the region, yet its effects are not universal in their timing, sign or magnitude. A set of regional
physical connections to ENSO are established and their varying local manifestations are viewed in the context of the dominant
precipitation generating mechanisms and feedbacks at that location. In addition, some potential impacts of longer run variations
within the ocean-atmosphere system of the Atlantic are examined independently and in conjunction with ENSO. This review of the
climatic controls and feedbacks in the region provides a spatial and temporal framework within which the highly complex set of
factors and their interactions may be interpreted from the past.
D 2005 Elsevier B.V. All rights reserved.

Keywords: Climatology; El Nino-Southern Oscillation; Northern South America; Choco jet; San Andres jet

1. Introduction

* Corresponding author. This paper reviews the present climate of northern

E-mail address: gpoveda@unalmed.edu.co (G. Poveda). South America and southern Central America, mainly
0031-0182/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.palaeo.2005.10.031
4 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
focusing on precipitation and significant atmospheric
circulation features and their variability at annual and
interannual time scales. The region covers the northern
tropical Americas, (equator158N, and 608W858W)
encompassing Costa Rica, Panama, Colombia and
Venezuela. It abuts the tropical North Atlantic, Carib-
bean and eastern tropical Pacific. Regional atmospheric
circulation and climate arise from interactions with the
neighboring oceans, the Andes, the Cordillera of Cen-
tral America, the VenezuelanColombian Llanos and
the Amazon basin. The classic work of Snow (1976)
provides an excellent initial understanding of regional
climatology. It is the intent of this paper to detail the
complexity and interaction of the factors governing
climate, at various time and space scales, over the
region through the use of recent data, including at-
station records, climatic reanalysis and satellite data.
Hence, one goal of this paper is to more fully analyze
and describe the regional climates of the meso-Amer-
icas and their varying sensitivities to allow paleoclimate
researchers to better reconcile modeling studies, proxy
datasets and present-day observations.
The principle controlling factors will be shown to be
the Inter-tropical convergence zones (ITCZ) of both the
Atlantic and Eastern Pacific Oceans and the trade
winds, the Hadley cell circulation, and interactions
occurring between land surface and atmosphere at dis-
tinctive time scales. These in turn govern the strength
and location of low level jets in the area and the
propensity for the generation, path and penetration of
such transient features as cold fronts and tropical
storms. The excellent long paleo-sedimentary record
available (14,000 yrs.) from the Cariaco Basin off the
north coast of Venezuela has been the source of much
speculation concerning variations in the position of the
ITCZ, the strength of the trade winds, contrasts in
Atlantic sea surface temperatures and rainfall on the
contributing contiguous continental margin (e.g., Haug
et al., 2001; Lea et al., 2003). El Nino-Southern Oscil-
lation (ENSO) will also be shown to be an important
part of interannual climatic variations over broad areas
of the circum-Pacific region for millennia (e.g., Rodbell
et al., 1999).
Section 2 reviews the annual climatological cycle,
focusing mainly on precipitation patterns associated
with the migration of the ITCZ, and discussing im-
portant physical mechanisms interacting within the
ITCZ regionally. The roles of mesoscale convective
systems (MCSs), important feedbacks between the
Andes and the Amazon, and land surfaceatmosphere
interactions are also discussed. Section 3 deals with
interannual variability, mostly associated with the ex-
treme phases of ENSO, but also with oceanatmo-
sphere activity over the North Atlantic Ocean. The
effects of ENSO are thoroughly discussed, and a sum-
mary of regional physical mechanisms associated with
the occurrence of El Nino is presented. Final remarks are
provided in Section 4.
2. Annual cycle
The annual hydro-climatic cycle is dominated by the
meridional migration of the ITCZ, which in turn con-
trols the various dynamics of the trade winds over
oceans and land masses. However, the important role
of the landatmosphere system in shaping the climate
of the region has been largely overlooked. Diverse
regional and local atmospheric circulation patterns in-
teract with the ITCZ and its migration to modify the
annual cycle of precipitation. Among these, we identify
two low-level jets, one easterly over the Caribbean
around 108128N (Poveda and Mesa, 1999; Magana
et al., 1999; Amador and Magana, 1999), and the other,
the Choco jet, westerly at about 58N over the Pacific
(Poveda and Mesa, 1999, 2000). The former is linked to
two unique features of precipitation on the isthmus; a
boreal winter maximum on the Caribbean flank and a
bmid-summer droughtQ throughout the Pacific flank.
The latter is associated with MCSs (Velasco and Frisch,
1987; Mapes et al., 2003; Zuluaga and Poveda, 2004).
A third easterly jet is confined to the 600700 hPa level
over South America and the eastern equatorial Pacific
(Hastenrath, 1999).
In deep tropical America, the activity of MCSs
becomes a significant source of prolonged, intense pre-
cipitation events. The study of Velasco and Frisch
(1987) pioneers the understanding of the dynamics
and thermodynamics of MCSs over the region. Recent
data from the Tropical Rainfall Measuring Mission
(TRMM, Kummerow et al., 1998) permit a deeper
understanding of the annual precipitation cycle. A con-
siderable portion of the study region is occupied by the
upper Amazon basin. The relevant climatic roles that
Amazonia plays at both the planetary and regional
scales are discussed, drawing attention to land surface
atmosphere interactions in the creation of annual and
inter-annual variability in the study region.
This review purposely ignores other mechanisms
and atmospheric circulation patterns associated with
climatic variability over the region, including the dy-
namics of tropical easterly waves, squall lines over the
Amazon basin, the South American low level jet (Bye-
rle and Paegle, 2002; Campetella and Vera, 2002), and
the intra-seasonal variability associated with the activity
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
of the 3060 day oscillation. All of them need to be
understood in terms of their modulation and coupling to
the annual and inter-annual time scales.
2.1. Meridional migration of the ITCZ in the region
There is no single, universally held definition of the
ITCZ. The classical view defines it as a region close to
the equator of trade wind convergence, ascending air,
low atmospheric pressure, deep convective clouds and
heavy precipitation (Henderson-Sellers and Robinson,
1986). In turn, Hartmann (1994) defines it as the axis of
the broad trade wind current of the tropics where the
northerly and southerly trade winds meet, either in
narrow bands or in the broader convergence zones
over South America, Indonesia and Africa. Gu and
Zhang (2001) describe it as a zonally elongated (i.e.,
parallel to the equator), latitudinally confined, rain or
cloud band, composed mainly of non-propagating, ran-
dom deep convective clouds and non-deep clouds.
Hastenrath (2002) invokes a suite of variables including
low surface pressures, negative values of divergence of
horizontal wind, high rainfall rates and low values of
Outgoing Longwave Radiation (OLR). Regional posi-
tions of the ITCZ and the concomitant location of
maximum rainfall can be traced in the seasonal distri-
bution of OLR. Assuming that disturbances in the
dynamic fields are associated with deep convective
activity (Gu and Zhang, 2001), Fig. 1 shows the
extremes of the seasonal cycle of OLR, using data
derived from the 40 year Reanalysis Project (Kalnay
et al., 1996) of the National Center for Environmental
Prediction-National Center for Atmospheric Research
(NCEP-NCAR). The quantity of OLR is inversely re-
lated to deep convective cloudiness, wave motions in
the tropics and higher rainfall rates (Gruber, 1974; Nitta
et al., 1985; Hendon and Liebmann, 1991; Wheeler and
Kiladis, 1999).
The ITCZ has been explained in terms of synoptic-
scale disturbances in the circulation, many of which
propagate westward (Holton et al., 1971; Chang, 1973;
Lindzen, 1974; Hess et al., 1993; Gu and Zhang, 2001,
2002), although other hypotheses have been put for-
ward (e.g., Pike, 1971; Philander et al., 1996; Tomas
and Webster, 1997; Ferreira and Schubert, 1997). The
meridional oscillation of the ITCZ responds to the
seasonal insolation cycle, lagging the zenithal position
by approximately one month. This oscillation exerts a
strong regional control on the annual hydroclimatolo-
gical cycle. During the austral summer the continental
ITCZ is located over the Amazon basin, southern
Colombia and Ecuador, while those portions over the
5
eastern equatorial Pacific migrate no further south than
38N (Alpert, 1945, Hastenrath, 1966, 2002). Northward
displacement occurs in the boreal summer, covering the
upper Amazon and Orinoco basins, the Guyanas,
Colombia, and the Central America isthmus, as far
north as the Costa Rica/Nicaragua border. Central
Colombia and the western Andean cordillera experi-
ence a bi-modal annual cycle with peaks during April
May and OctoberNovember, and less rain in Decem-
berFebruary and JuneAugust, mainly as the result of
the double passage of the ITCZ (Figs. 2 and 3). A uni-
modal cycle (MayOctober) is witnessed over the Ca-
ribbean coast of Colombia and the Pacific flank of the
southern isthmus, reflecting the northernmost position
of the ITCZ over the continent and eastern equatorial
Pacific, respectively (Portig, 1965; Hastenrath, 1967,
1990, 1991, 2002; Waylen et al., 1996a). A slight
reduction in JulyAugust precipitation, known locally
as the bVeranillos de San JuanQ, bCanculaQ, or mid-
summer drought (Magana et al., 1999) is present
throughout the Pacific flank of Central America. Its
causes are complex, but its effects have been shown
to be modulated by the east Pacific bwarm poolQ bor-
dering Central America, which attains maximum tem-
peratures during this time, and to be associated with the
development of a low-level jet at 158N over the Carib-
bean (Pulwarty et al., 1998). A large-scale movement of
deep convection way from northern South and Central
America has been observed synchronous to the drought
(Magana et al., 1999). A single peak is also evident
over the eastern piedmont of the Eastern Andes, mainly
due to the advection of moisture from the Amazon
which encounters the orographic barrier of the Andes,
thus focusing and enhancing deep convection and rain-
fall on the eastern flank of the cordillera, with maxi-
mum rainfall occurring during JuneAugust.
Unlike the major cordillera of Costa Rica and the
Andes, the mountains of Panama run from west to east.
Their comparatively low relief and the narrowness of
the southern isthmus, allow complex influences of
moisture-laden winds from circulations above both
the Caribbean and Pacific. The Caribbean slope exhi-
bits an almost uniform precipitation regime, with min-
imum values of 100 mm/month. The Pacific slope
experiences a rainy season from late Aprilearly May
to late November, and a dry season in DecemberApril,
during which 2 maxima are witnessed in June and
October, as a response to the ITCZ migration.
Venezuela can be divided into three major climatic
zones: (1) the north and west, comprising the Mara-
caibo basin, the eastwest orientated Cordillera de la
Costa (b 1500 m high), and the prominent southwest
6 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327

Fig. 1. Extremes of the seasonal cycle of average long-term (19681996) Outgoing Longwave Radiation (OLR in Wm 2) over the study region:
JuneAugust (top), and DecemberFebruary (bottom). Lowest OLR values correspond to higher clouds, deeper convection and higher rainfall rates.
The meridional migration of the ITCZ is reflected accordingly. Data source: NCEP-NCAR Reanalysis 19681988. (Kalnay et al., 1996). Images
from the NOAA-CIRES Climate Diagnostics Center, Boulder, Colorado (http://www.cdc.noaa.gov/).

northeast oriented Cordillera de Merida (up to 5000 m
high); (2) the central area, known as the bLlanosQ, a
vast, low-lying basin; and (3) the Guyana Highlands in
the southeast, with elevations between 400 and 1400 m
(Pulwarty et al., 1992). Variability of precipitation in
northern South America, and Venezuela, in particular, is
made more complex by the geography of the region.
The northsouth trending cordillera and the presence of
the major moisture source (Caribbean) to the north of
the landmass, produce very different conditions from
the classic boreal bmonsoonQ climates of South Asia
and West Africa (Pulwarty et al., 1992). Nevertheless, a
narrow equatorial trough can be identified in lower
tropospheric wind shifts, zonal wind divergence, and
pressure height deviations around 58N. Bimodal rainfall
distributions are widespread at stations west of 708W, in
sharp contrast to the JulyAugust maximum of the
Llanos. Further east, along the coastal zone, rainfall is
more directly governed by the migration of the ITCZ.
Remnants of old cyclonic systems and cold fronts
(friagems) from Brazil and Argentina may yield
heavy precipitation events over the Llanos (58108N)
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Fig. 2. Months of maximum mean monthly precipitation arranged by triad.
and occasionally into the Caribbean (Pulwarty et al.,
1992).
Bimodal maxima (usually AprilMay and Septem-
berOctober) can be sub-divided into four subtypes.
Coastal, low annual totals of about 430 mm and July
August and November December maxima: Northwest
lowlands, annual totals of 1000 mm with a larger peak
in fall and a secondary one in spring; Venezuelan
Andes, higher totals and both fall/spring peaks; and
Eastern Guyana Highlands, 12501300 mm annually,
a primary early boreal summer maximum and a sec-
ondary one in late fall. Orographic controls strongly
amplify the JulyAugust minimum in the Venezuelan
Andes and sharpen the transition between unimodal
and bimodal regimes (7.48.68N, 71.572.58W).
2.2. Regional mechanisms interacting with the ITCZ
2.2.1. The Choco low-level jet
Oceanatmosphereland interactions over the east-
ern Pacific have considerable influence on Colombias
hydro-climatology. It has also been suggested that the
Andes influence rainfall off Colombias Pacific coast
through thermally driven gravity waves (Mapes et al.,
2003). The Pacific lowlands are among the worlds
7
rainiest regions, with mean annual precipitation in
excess of 8000 mm and even beyond 12000 mm
(e.g., Lopez, 1966; Arnett and Steadman, 1970;
Snow, 1976; Nieuwolt, 1977; Eslava, 1993; Poveda
and Mesa, 1997, 2002; Mapes et al., 2003). These
copious totals arise from the interaction of a low level
tropical westerly, Choco jet from the Pacific, with the
dynamics of mesoscale convective systems (Poveda
and Mesa, 2000).
The Choco jet, thermally driven by the SST gradient
between the EcuadorPeru cold tongue and the Colom-
bian Pacific, is further favored by the change of direc-
tion of the cross-equatorial austral trades which become
westerly at this latitude (Poveda and Mesa, 2000; Has-
tenrath, 2002), and by the latent heat release of rainfall
in MCSs over the region which favors surface conver-
gence. The Choco jet is clearly visible in the seasonal
cycle of airflow at 925 hPa (Fig. 4), transporting large
quantities (mean rate of 3.78  106 kg s 1 or 3774 m3
s 1) of moisture inland where it contributes directly to
the combined annual discharges (~5000 m3 s 1) of the
Atrato and San Juan rivers, which exhibit among the
worlds highest runoffs (Poveda and Mesa, 1999,
2000). The Choco jet is forced to ascend the topogra-
phy of the western Andes and to interact with the
8 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327

Fig. 3. Percentages of mean annual precipitation falling within the triads of maximum precipitation. Larger symbols indicate greater seasonality.

easterly trade winds, thereby further enhancing deep
convection. Its seasonal strengthening (SeptemberNo-
vember) and weakening (FebruaryMarch), partially
explains why the OctoberNovember rainy season is
more intense than that of AprilMay over central and
western Colombia. At paleoclimatic time scales, there
is evidence for the presence of La Nina-like conditions
in the eastern equatorial Pacific and a stronger Choco
jet in the northern Andes during the last glaciation
(MartVnez et al., 2003).
Fig. 5 depicts the October mean meridionalverti-
cal cross-section of zonal winds at 808W, between
58S and 208N. The Choco jet, is confined below 800
hPa and focused around 58N at 950900 hPa, with
maximum wind speed of 78 ms 1. Two easterly
jets, the San Andres (148168N, and 900 hPa;
Magana et al., 1999; Poveda and Mesa, 1999), and
the Equatorial Mid-tropospheric Easterly Jet (EMTEJ)
at 600 hPa, centered on the equator (Hastenrath,
1998; 1999), show up as negative values and are
discussed below.
2.2.2. San Andres and the equatorial mid-tropospheric
equatorial jets
The low level jet in the Caribbean trade winds
(Poveda and Mesa, 1999; Amador and Magana, 1999;
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Fig. 4. Seasonal cycle of horizontal wind over the tropical Americas at 925 hPa. Note the southwesterly Choco jet moving from the Pacific inland at around 58N. The easterly San Andres jet (Section
2.2.2) can be found around 128148N. Data source: NCEP-NCAR Reanalysis.

9
10 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327

Magana et al., 1999), acts as a link between the eastern Table 1

Pacific and the Caribbean on interannual and shorter
time scales. Its activity over the Caribbean has potential
effects on tropical convective systems, such as easterly
waves, and their organization into tropical cyclones.
Core velocities (138N158N, below 900 hPa) are high
in JulyAugust (12 m s 1) and DecemberFebruary
(10 m s 1), and least during SeptemberNovember
(6 m s 1) and MarchMay (8 m s 1).
The EMTEJ exhibits an annual cycle (Table 1)
which is almost the inverse of the Choco and San
Andres jets: more intense during MarchMay and
JuneAugust, with core velocities larger than 10 m
s 1, and weaker core velocities during SeptemberNo-
vember (6 m s 1) and DecemberFebruary (8 m s 1).
The EMTEJ experiences almost no meridional oscilla-
tion, being located over the Equator during MarchMay
and over 28S during SeptemberNovember.
After crossing the isthmus, part of the San Andres jet
deviates to the southeast, feeding the Choco jet in
conjunction with downstream portion of the cross-equa-
torial flow from the southern hemisphere. As a result,
the San Andres jet is weaker at 858W than at 808W. The
physics of the San Andres jet is still elusive, but prob-
ably it is associated with the dynamics of the tropical
Western Hemisphere warm pool (WHWP), a region that
extends over parts of the eastern North Pacific, the Gulf
of Mexico, the Caribbean, and the tropical North At-
lantic (Wang and Enfield, 2001, 2003; Wang, 2002,
2003).
1
Seasonal cycle of average wind velocities (m s ) at the core of the
three jets discussed in the text
MAM JJA SON DJF
EMTEJ jet 10 (0) 10 (0) 6 (0) 8 (0)
San Andres jet 8 (1) 12 ( 4) 6 ( 2) 10 (2)
Choco jet 2 (0) 4 ( 1) 6 ( 2) 4 ( 2)
Negative figures indicate easterly wind direction. Figures in paren-
theses represent computed changes in mean core wind velocities
warm phase years minus cold phase years of ENSO.
2.2.3. Mesoscale convective systems (MCSs)
These systems (MCSs) deliver large quantities of
moisture to the region (Fig. 6). Velasco and Frisch
(1987) hypothesize that their formation and dynamics
are related to the development of warm nucleus vortices
of mesoscale size, forced by latent heat. Most are noc-
turnal and continental. The systems are defined in terms
of black body infrared temperatures ( 40 and 62 8C),
and a required continental area in the order of 50,000
km2 (Maddox, 1980), which may drop to 30,000 km2 at
low latitudes. The development and evolution of MCSs
is closely linked to the activity of low-level jet streams
and deep convection (Bonner, 1966; Raymond, 1978;
Maddox, 1980; Lemaitre and Brovelli, 1990; Stensrud,
1996).
The role of the Choco jet in the development of
numerous MCSs in the Panama Bight is paradigmatic
(Velasco and Frisch, 1987; Poveda and Mesa, 2000),
and both exhibit a strong diurnal cycle (MejVa and
Poveda, 2005; Mapes et al., 2003; Poveda et al.,

Fig. 5. October long-term (19681996) mean meridionalvertical cross-section of the zonal winds at 80EW, between 5ES and 20EN. The core of
the westerly Choco jet is depicted in gray. Data source: NCEP-NCAR Reanalysis.
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327 11

Fig. 6. Monthly distribution of mesoscale convective systems (MCSs) over the tropical Americas, during 1998, according to criteria by Nesbitt et al.
(2000). The size of the triangle is correlative with rainfall intensity within the MCSs (Zuluaga and Poveda, 2004). Data source: Satellite sensors of
the Tropical Rainfall Measuring Mission (TRMM).

2005). The annual cycle of MCSs and their atmo-
spheric environments during 19982002 are studied
by Zuluaga and Poveda (2004), using satellite infor-
mation from the Tropical Rainfall Measuring Mission
(TRMM, Kummerow et al., 1998). On average, a total
of 44,709 rainfall events occurred, of which 19,875
(44.4%) and 24,834 (55.6%) were over land masses
and oceans, respectively. Their monthly distribution
and intensity for 1998 are shown in Fig. 6. A low
number of systems actually develop into MCSs
(4929), 2761 over oceans and 2168 over land, yet
their contribution to total annual precipitation is dis-
proportionate, with an average of 70% during the
19982002 period. The distribution of events devel-
oping MCSs is strongly controlled by the position of
the ITCZ, yet during 1998 (warm phase ENSO) the
most favorable region for MCSs was located over
western Colombia and the adjoining Pacific. The Mag-
dalena river valley, northern Antioquia, and the Am-
azon and Orinoco regions of Colombia also exhibit
high frequencies of MCSs. Least favorable regions are
northern Colombia, Venezuela, and western Peru.
MCSs appear more frequently off the Pacific coast
of the isthmus than the Caribbean, but there is a
notable dearth of events in July and August (midsum-
mer drought) in these Pacific waters, and a slight
preponderance of the limited number of December
events in Caribbean waters. Events are almost totally
absent from the isthmus during JanuaryApril. The
studies of Mapes et al. (2003) and Warner et al.
(2003) provide diagnostics and a modeling framework
for rainfall over western Colombia and off the Pacific
coast.
2.2.4. Feedbacks between the Andes and the Amazon
and the role of land surfaceatmosphere interactions.
The Amazon River basin provides an excellent ex-
ample of coupling and feed backs in the land surface
12 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
atmosphere system, due to its large area (N6.4106
km2), tropical setting, and complex eco-hydro-climato-
logical dynamics which exert a global influence (e.g.,
Avissar and Nobre, 2002; Roberts et al., 2003). Mod-
eling results suggest changes in global, regional and
local atmospheric circulation patterns associated with
deforestation or perturbations in the land surfaceatmo-
sphere system (Shuttleworth, 1988; Zeng and Neelin,
1999). Zeng et al. (1996) propose that deforestation
would reduce the SST gradient over the tropical Atlan-
tic, thereby affecting global atmospheric circulation
through perturbations of Walker and Hadley cells,
which is confirmed by the studies of Zhang et al.
(1996) and Werth and Avissar (2002). Relevant features
of the atmospheric circulation over the Amazon are
studied by a number of authors (e.g., Kousky, 1985;
Kousky and Kagano, 1981, 1994). Major factors affect-
ing the Amazon basin include the meridional oscillation
of the ITCZ and the South Atlantic Convergence Zone
(SACZ, Kousky, 1988), which is defined as a persistent
cloud band oriented in the northwestsoutheast direc-
tion, and encompassing southern Amazonia through to
the South-Central Atlantic, being most prevalent during
summer. The presence of the Andes affects consider-
ably the climate of Amazonia because they exert a
strong influence on air masses distributions, constitute
an orographic barrier for the easterly winds, and funnel
tropical maritime air east of the Andes towards the
southeast.
Land surfaceatmosphere interactions constitute sig-
nificant drivers of the climate and weather variability of
Amazonia. The important role of precipitation in con-
trolling vegetation activity is reflected by the Normal-
ized Difference Vegetation Index, NDVI (Poveda and
Salazar, 2004). Evapotranspiration (ET in Fig. 7) from
Fig. 7. Cartoon of feedbacks (not to scale) between the Andes and the Amazon River basin during MaySeptember. The variables are
represented as follows; P: Precipitation, Q: River discharges, S: Sediments, N: Nutrients, C: Pollutants and Constituents, Vq: Moisture transport
by the trade winds, and ET: Evapotranspiration. Precipitation in the Amazon, P, consists of both recycled precipitation and moisture transported
from outside the basin.
vegetation is itself a key control on precipitation recy-
cling (Pr in Fig. 7; Salati, 1985; Eltahir and Bras, 1994;
Trenberth, 1999). Over the Amazon, precipitation exhi-
bits a uni-modal annual cycle, with a wet season during
DecemberApril and a dry season in JuneOctober
(Obregon and Nobre, 1990; Zeng, 1999). According
to Horel et al. (1989) the establishment of the rainy
season over the Amazon basin occurs relatively fast,
taking one month or less, due to dynamic and thermo-
dynamic mechanisms. Conversely, the cessation of the
rainy season occurs more slowly.
The seasonal march of OLR depicted in Fig. 1
shows differences in the behavior of the ITCZ and
the SACZ over land and ocean regions, due to season-
ally varying landatmosphere interactions and bound-
ary layer dynamics, which, by means of soil moisture
and evapotranspiration, control precipitation and weath-
er patterns over tropical land masses (Delworth and
Manabe, 1993; Elthair, 1998). Moist deep convection,
the predominant regional mechanism for intense
storms, is partially controlled by the dynamics of the
atmospheric boundary layer and land surfaceatmo-
sphere interactions. Thus the seasonal cycle of rainfall
over the Amazon and adjoining regions reflects the
combined effects of large-scale forcing such as the mi-
gration of the ITCZ, moisture transport from the Atlan-
tic Ocean (Pa in Fig. 7), and land surfaceatmosphere
interactions, including precipitation recycling.
Feedbacks are also present in the AndesAmazon
bio-geophysical system (Fig. 7). The tropical Andes
constitute the Amazon headwaters, simultaneously
importing atmospheric water from the Amazon, and
exporting surface waters (Q), sediments (S), nutrients
(N), and constituents and pollutants (C) to it. This
feedback is most active during MaySeptember, when
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
the moisture laden southeast trades (Vq), and their
orographic lifting over the eastern Andes, contribute
to convective precipitation (P). The interactions and
feedbacks are felt: (1) blocallyQ through precipitation
recycling within the Amazon, (2) bregionallyQ by means
of fluxes between the Amazon and Eastern Andes, and
(3) bhemisphericallyQ through the strengthening/weak-
ening of Hadley cell circulation (Werth and Avissar,
2002). They also indicate the potentially important
anthropogenic control on deforestation in both the Am-
azon and the Andes may exert on the functioning of the
system.
Recent studies over the Amazon basin (Nesbitt et al.,
2000; Petersen et al., 2002) indicate that wet-season
convection exhibits a vertical structure that is interme-
diate to that observed over tropical oceans (less vertical
development) and other continents (more lightning, larg-
er vertical development). The convection also displays
westerly and easterly modes (Cifelli et al., 2002), which
are correlated to changes in 850700 hPa zonal wind
direction, and to the dynamics of the South American
low level jet (Byerle and Paegle, 2002; Campetella and
Vera, 2002). Each mode promotes different vertical
structures and precipitation statistics (Petersen et al.,
2002). The easterly one favors westward propagating
precipitation as Amazonian squall lines (Greco et al.,
1990; Cohen et al., 1995; Warner et al., 2003), exhibiting
a strong diurnal cycle and cross the basin in 23 days.
The rainbands appear to involve both moisture advection
and ducted gravity wave dynamics in the presence of the
EMTEJ (Poveda and Mesa, 1997; Warner et al., 2003).
Also, during the easterly regime, MCSs and rain cells
exhibit closely related propagations, mostly associated to
the midlevel mean flow (Laurent et al., 2002), and
enhancement of MCSs convective cells, and more in-
tense in terms of kinematic and microphysical features,
comparatively to the westerly regime (Carvalho et al.,
2002; Cifelli et al., 2002). Instead, during this latter
regime, the propagation of both MCSs and rain cells
appear to be more disorganized, with apparently no
clear relationship with the mean atmospheric flow at
any level (Laurent et al., 2002).
3. Interannual variability
3.1. ENSO
ENSO is the main forcing mechanism of climatic
variability in tropical South America from seasons to
decades. El Nino refers to the unusual warming of sea-
surface temperatures (SSTs) over the eastern and central
tropical Pacific. Important components of this anomaly
13
are the deepening of the oceanic thermocline in the
eastern Pacific, and the weakening of the dominant
surface easterly trade winds. During an El Nino event
there is a shift in the center of convection from the
western to the central Pacific. The accompanying South-
ern Oscillation, the bseesawQ of the atmospheric mass
that produces a pressure gradient between the western
and the eastern equatorial Pacific, is characterized by the
Southern Oscillation Index (SOI). Negative values of the
SOI are associated with warm events (El Nino), while
positive values accompany cold events (La Nina) and a
stronger Walker cell (westeast) circulation. ENSO is an
aperiodic oscillation with a recurrence interval of be-
tween two and ten years, and an average of four (Tren-
berth, 1991), which appears to have become more
frequent since the mid 1970s (Trenberth and Hoar,
1996). El Nino events begin during the boreal spring
and exhibit strong phase locking with the annual cycle
(Tziperman et al., 1994; Chang et al., 1994; Webster,
1995; Neelin et al., 1998). Those events, which encom-
pass two calendar years, are generally characterized by
increasing SST anomalies during boreal spring and fall
of the first year (year 0), which peak in winter of the
following year (year +1). Anomalies then decline in
spring and summer of year +1. Kiladis and Diaz
(1989) recognize the following years of El Nino event
during the past century: 1902, 1904, 1911, 1913, 1918,
1923, 1925, 1930, 1932, 1939, 1951, 1953, 19571958,
1963, 1969, 19721973, 19761977, 19821983, and
19861987. More recently, El Nino occurred in 1991
1992, 19941995, 199798, 20022003, and 2004. La
Nina events occurred in 1949, 1954, 1964, 1970, 1973,
1975, 1988, and 1998. Rossel (1997) provides a thor-
ough review of the diverse classification schemes for
phases of ENSO.
3.2. ENSO at paleoclimatic time scales
ENSO has been an important part of interannual
climatic variations over broad areas of the circum-Pa-
cific region for millennia (Rodbell et al., 1999). Several
recent studies focus on the reconstruction of the paleo-
climatic ENSO based on instrumental records (Whetton
and Rutherford, 1994), tree-ring proxies (Jacoby and
DArrigo, 1990; Stahle et al., 1998); isotopic analysis
from coral reefs (Dunbar et al., 1994; Urban et al.,
2000; Tudhope et al., 2001), ice cores (Thompson,
1992), and multiproxy records (Mann et al., 2000).
Changes in the frequency, magnitude and duration of
ENSO, as derived from proxy land-based data sets
around the Pacific Basin, have been observed through-
out the last millennium with significant decreases in
14 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
ENSO activity during the Medieval Warm Period 900
1300 AD (Diaz and Pulwarty, 1994). Haug et al. (2003)
and Hodell et al. (1995) even extend these discussions
of the effects of the ITCZ, trades and ENSO to the
northern most limits of this area in discussing the
collapse of Mayan civilization. As the following dis-
cussion will illustrate, there are large uncertainties in
understanding of the interrelationship between ENSO
characteristics and the background mean climate at the
present.
ENSO tends to be phase-locked to the annual cycle so
that the largest SST anomalies occur in the boreal winter.
Model simulations of the annual cycle at various times in
the past imply a somewhat shifted and greatly reduced
annual cycle in equatorial SSTs (warmer conditions in
September and cooler in March) 3000 to 12,000 years
ago relative to today (Trenberth et al., 1998). Interan-
nual SST variability has been shown to have de-
creased during the mid-Holocene and to have
increased during the Late Glacial Maximum (LGM).
There is no scientific consensus on the causes of the
suppression of ENSO activity during the mid-Holo-
cene, or on the causes of the strengthening of ENSO
activity during the LGM. Clement et al. (2000) pro-
pose that ENSO may be highly sensitive to orbitallly
driven changes in the seasonal cycle of solar radiation
in the tropics. Koutavas et al. (2002) reconstructed
SSTs, using Mg/Ca ratios in foraminifera from sea-
floor sediments near the Galapagos Islands, from the
LGM to the present. Cold-tongue SSTs varied coher-
ently with precession-induced changes in seasonality
during the past 30,000 years. SST descend only 1.28C
implying a relaxation of tropical temperature gradi-
ents, weakened Hadley and Walker circulation, a
southward shift of the ITCZ, and a persistent El
Nino-like pattern in the tropical Pacific (Koutavas et
al., 2002). This is contrasted with mid-Holocene cool-
ing suggestive of a La Nina-like pattern with enhanced
SST gradients and strengthened trade winds. As noted
by Rosenthal and Broccoli (2004), these observations
are intriguing but raise questions as to whether they
are representative of the broader tropical Pacific or of
only local conditions. Enfield et al. (2001) speculate
that early to middle Holocene climate may have been
stabilized by smaller pole to equator temperature con-
trasts and reduced ENSO variability. A smaller merid-
ional temperature contrast is assumed to reduce the
need for poleward heat transport as observed during
ENSO.
In a model-based study of the increased ENSO
activity over the past 20 years (relative to the previous
two decades), Sun (in press) shows that the strengthen-
ing of ENSO may be a consequence of an increase in
the equatorial radiative heating or equivalently an in-
crease in the tropical maximum SST. The author further
suggests that the warmer equatorial thermocline water
could be a factor in the suppression of ENSOthe
reduced difference between the tropical maximum
SSTs and the temperature of the equatorial thermocline
water makes the coupled system more stable. Whether
these proposed mechanisms are actually independent of
each other is not known. Subtle variations of the Walker
circulation and associated airsea interactions however
modify the climatic teleconnections over South Amer-
ica somewhat limiting the comprehensiveness of deri-
vations of bcanonicalQ impacts (Diaz and Pulwarty,
1994).
3.3. Present effects of ENSO in tropical America
In general, there is a coherent pattern of climatic and
hydrological anomalies over the region during extreme
phases of ENSO. Negative anomalies in rainfall, soil
moisture and river flows, as well as warmer air tem-
peratures, occur during El Nino. The reverse is true for
the cold phase (La Nina), although there are regional
differences in timing and amplitude. The climatic and
hydrological effects of ENSO in the tropical Americas
have been the subject of intense research (e.g., Hasten-
rath, 1976, 1990; Hastenrath et al., 1987; Waylen and
Caviedes, 1986; Aceituno, 1988, 1989; Rogers, 1988;
Kiladis and Diaz, 1989; Ropelewski and Halpert, 1987,
1996; Chu, 1991; Glantz et al., 1991; Lau and Sheu,
1991; Halpert and Ropelewsky, 1992; Diaz and Kiladis,
1992; Pulwarty et al., 1992; Diaz and Markgraf, 1993,
2000; Pulwarty and Diaz, 1993; Marengo, 1992; Ma-
rengo and Hastenrath, 1993; Waylen et al., 1996a,b;
Enfield and Mayer, 1997; Enfield and Alfaro, 1999;
Poveda and Mesa, 1997; Poveda et al., 2001, 2003;
Waylen and Poveda, 2002).
3.3.1. Summary of regional physical mechanisms
associated with ENSO
ENSO has affected the hydro-climatology of north-
ern South America for centuries. For instance, Shongart
et al. (2004) using dendrochronology reconstruction
identified an increase in El Nino severity on Amazonia
during the last two centuries. The present hydro-clima-
tological anomalies associated with ENSO in north-
western South America can be diagnosed as a combi-
nation of mechanisms as follows (Fig. 8):
(1) The reduction of the SST gradient in the eastern
Pacific weakens the Choco jet, decreases mois-
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Fig. 8. A summary of hydro-climatological mechanisms which com-
bine to explain drought conditions in tropical South America during
El Nino. Numbers and symbols correspond to mechanisms and pro-
cesses listed in text.
ture transport inland (Poveda and Mesa, 2000;
Poveda et al., 2001), particularly through the
associated reduction in the intensity and number
of mesoscale convective complexes (Velasco and
Frisch, 1987). Fig. 9 reflects this reduction during
El Nino (and intensification in La Nina) in the
annual cycle of moisture transport by the Choco
jet at 925 hPa. Such anomalies in moisture trans-
ported by the Choco jet are important contributors
to the rainfall anomalies reported over central and
western Colombia.
Fig. 9. Long-term annual cycle of average moisture transport by the zonal winds of the Choco jet at 925 hPa, in units of ms
Nino years (triangles), and La Nina years (squares).
15
(2) Changes in the flow of moisture onto the conti-
nent establish anomalous Hadley cell circulation.
The subdued ascent of moist air and associated
convective precipitation explain the anomalously
high surface pressure over the region, particularly
during DecemberFebruary, as noted by Rasmus-
son and Mo (1993) in the 198283, 198687, and
199192 El Nino events and diagnosed by Yasu-
nari (1987) and Aceituno (1988). Characteristics
(position and horizontal and vertical structure) of
the convective thermal forcing over the tropical
Pacific during boreal winters appear to be deter-
minants of the phase and amplitude of the ENSO-
related anomalies over the Americas (Ambrizzi
and Magana, 1999).
(3) Changes in pressures over tropical South America
(Fig. 8) contribute to:
(4) he displacement of the centers of convection
within the ITCZ in the eastern Equatorial Pacific
towards the west and south of their normal posi-
tions (Pulwarty and Diaz, 1993).
(5) The reduction in the feedback between precipita-
tion and surface convergence in tropical South
America associated with Hadley cell circulation
(Numaguti, 1993) and Caribbean trade winds.
(6) The disruption of landatmosphere interactions,
and the anomalies eastward travel with time, due
to the regional coupling between anomalies of
precipitation, soil moisture, vegetation and eva-
potranspiration (Nepstad et al., 1994; Jipp et al.,
1 1
g kg , during El
16 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
1998; Poveda and Mesa, 1997; Zeng, 1999;
Poveda and Jaramillo, 2000; Poveda et al., 2001;
Poveda and Salazar, 2004). A reduction in evapo-
transpiration also contributes to decreased precip-
itation from evaporation recycling. Diminished
cloudiness promotes increased solar irradiance
and surface temperatures, thus reinforcing dry
conditions. Even in wet tropical climates, water
limitation (as during El Nino events) imposes
conditions of water stress on forests (Oren et
al., 1996). Land surface hydrological processes
must be playing a major role in determining the
velocity at which the anomaly propagates over
the continent through positive landatmosphere
feedbacks (Fig. 10).
(7) Initially, during the boreal summer of Year 0, the
northeast trade wind fields intensify (weaken)
during El Nino (La Nina), however in concor-
dance with the changes in surface pressures (3) in
the boreal winter, the fields reverse in Year+1,
triggering a change in SSTs over the Caribbean
and the tropical North Atlantic, physically
explained by Curtis and Hastenrath (1995).
(8) SSTs and the strength of the trades are two im-
portant controls on the intensity of tropical east-
erly waves and tropical storms over the North
Atlantic and Caribbean (Frank and Hebert,
1974; Gray and Sheaffer, 1991).
3.3.2. Costa Rica and Panama
Fig. 11 shows the regionally dominant positive as-
sociation between the SOI and annual precipitation
(Estoque et al., 1985; Ropelewski and Halpert, 1987).
Fig. 10. Schematics of the land surfaceatmosphere interactions associated with extended dryness during El Nino in tropical South America. ET and
Rn represent evapotranspiration and net radiation. Adapted from Fig. 14.3 of Sellers (1992). Reprinted with permission from ****.
However, significant associations are geographically
restricted to the Pacific slope. Negative associations
implied along the Caribbean slope of Costa Rica, con-
tinue into the Boca del Toro region of northwestern
Panama. The strongly unimodal regime of the Pacific
slope (Figs. 2 and 3) is closely associated with the
position of the ITCZ and cross-equatorial westerlies
(Hastenrath, 2002). Their southern and western dis-
placement during warm phases of ENSO brings about
a general reduction in rainfall, which is most apparent
(Fig. 12) during the veranillos of July and August
(Waylen et al., 1996b). A compounding factor may be
the synchronous strengthening of the San Andres jet
(Table 1), enhancing the rain-shadow effect on the
Pacific slope, jetting through topographic gaps (Le-
geckis, 1988; McCreary et al., 1989), thereby reducing
SST (Magana et al., 1999) and convective activity over
the eastern Pacific warm pool. This is clear in analyses
of seasonal streamflow (George et al., 1998) and flood-
ing (Waylen and Laporte, 1999; Quesada and Waylen,
2004) in the region, where even the slightly higher risk
of flooding in portions of the rainy season preceding the
veranillos, correspond to the reduction in the San
Andres jet during MAM.
By contrast, the Caribbean regime is uniform (Figs. 2
and 3). Deviations found in association with warm phase
ENSO (Fernandez and RamVrez, 1991; Waylen et al.,
1996b) are shown in Fig. 11 and Table 1, reflecting
changes in the strength of the San Andres jet: a slight
reduction in MAM, marked increases in JuneAugust,
and a potential reduction in DecemberFebruary. The
offsetting seasonal changes in precipitation are respon-
sible for the apparent lack of association of annual
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327 17

Fig. 11. Distribution of simple Pearson product moment correlation coefficients of annual precipitation and the Southern Oscillation Index in Costa
Rica. Red Figures indicate positive correlations and blue are negative. Solid symbols are statistically significant at the 0.05 level.

precipitation totals with ENSO (Fig. 8). Rivers on the
Caribbean slope (George et al., 1998; Quesada and
Waylen, 2004) similarly show an increase in JuneAu-
gust and a decline in DecemberFebruary, although this
signal may be limited to elevations of less than 1000m.
Above that altitude rainfall records indicate a response to
ENSO more characteristic of the Pacific slope. It may be
noteworthy that the greatest effect of ENSO on the San
Andres jet appears to be felt below 900 hPa.
The relationship between boreal winter precipitation
(Figs. 3 and 4), which is generally associated with the
passage of fronts, or nortes, originating over North
America (Hastenrath, 1967; Klaus, 1973; Schultz et al.,
1997, 1998), and ENSO, is unclear. However, Fig. 12
and analysis of flood timings (Quesada and Waylen,
2004) suggest that the earlier norte season (Novem-
berDecember) experiences a reduction in rainfall dur-
ing warm phases of ENSO, while the latter period
(JanuaryMarch) is wetter. Schultz et al. (1998) provide
some physical basis for this possibility in terms of the
origins of the anticyclones and the positions of the mid-
latitude and sub-tropical jets. During cold phases of
ENSO these patterns are generally reversed as implied
by observations of year +1 and +2 in Fig. 12.
The regional response of precipitation to ENSO in
Panama is not as clear as in neighboring countries.
The general tendency (with the exception of the ex-
treme northwest) is for drought during El Nino (Esto-
que et al., 1985; Hastenrath, 1991), and excessive
rainfall during La Nina. The interaction of topography
and the regional jets appears to play a major role in
controlling interannual variability in Costa Rica and
Colombia. However, the latitude of Panama (798N)
falls between the cores of the Choco (468N) and San
Andres (11188N) jets and its topography runs parallel
to them.
3.3.3. Colombia
Both ENSO phases are associated with hydro-clima-
tological anomalies in Colombia which amplify the hy-
drologic regime through complex and non-linear
interactions. Seasonal cross-correlation analyses con-
firm that El Nino produces drier and more prolonged
dry seasons, and La Nina wetter rainy seasons. Fig. 13
maps seasonal cross-correlations between the Multivar-
iate ENSO Index (MEI), which synthesizes the ocean
atmosphere conditions of the tropical Pacific in both
phases of ENSO (K. Wolter, NOAA http://www.
cdc.noaa.gov/~kew/MEI/), and discharge records at the
marked stations. Correlations are significantly positive
and high for MEI during SeptemberNovember and
DecemberFebruary with river discharges during De-
18 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Fig. 12. Monthly anomalies of regional precipitation observed along the Southern Pacific and Caribbean slopes of Costa Rica during El Nino years
(Waylen et al., 1996b).
cemberFebruary. Correlations decay during March
May and beyond.
Streamflow and precipitation from western and cen-
tral Colombia indicate significant positive correlations
with MEI (~0.7 to 0.8) during DecemberFebruary
(Poveda et al., 2001), the season of ENSOs greatest
strength, and thus of greatest teleconnections and direct
impact on Colombia. Forecasts of DecemberFebruary
discharges are possible as a result of the high correla-
tions with the MEI of the preceding SeptemberNo-
vember and JuneAugust (Poveda and Penland, 1994;
Carvajal et al., 1998; Poveda et al., 2002, 2003). The
skill of hydrological forecasting based upon the climate
of the tropical Pacific is reduced during the boreal
spring, as witnessed by the very low correlations with
the MEI in MarchMay.
Analyses of diverse atmospheric and terrestrial fields
of the NCEP/NCAR Reanalysis and hydrologic records
lead to the conclusion that the hydrology is most af-
fected by ENSO during DecemberFebuary (year +1),
SeptemberNovember (year 0) and JuneAugust (year
0), while MAM (years 0 and + 1) are the least affected
(Poveda et al., 2001).
Soil moisture is considerably depleted during El
Nino, but the exact extent and severity is dependent
on land cover (Poveda and Jaramillo, 2000). Coupling
between the vegetationsoil system and land cover
modulates hydrological (space-) time variability in the
tropics, and the effects of El Nino-induced dry spells
may potentially be modified. Similar strong regional
patterns are apparent in correlations of vegetation cover
(NDVI) with ENSO and with SOI (Myneni et al., 1996;
Asner et al., 2000; Poveda et al., 2001; Poveda and
Salazar, 2004). Decreased plant activity during El Nino
is entirely consistent with negative anomalies in pre-
cipitation, streamflow, soil moisture, and actual evapo-
transpiration (Poveda et al., 2001).
Streamflow anomalies are more pronounced in De-
cemberFebruary (year + 1) and JuneAugust (year 0)
as a result of the combined anomalies in precipitation,
soil moisture, and evapotranspiration. During Septem-
berNovember (year 0), effects are felt but ameliorated
by smaller soil moisture and precipitation anomalies.
Large-scale forcing and land surface hydrology play a
key role in the dynamics of hydro-climatic effects of
ENSO over the region (Marengo and Hastenrath,
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327 19

Fig. 13. Maps of seasonal cross-correlations between the Multivariate ENSO Index (MEI) and river discharge anomalies (Q) gauged at the marked
points, during SeptemberNovember (SON, top), and DecemberFebruary (DJF, bottom), with following seasons. The diameter of the circle
indicates the value of correlation, scaled as shown in the bottom right, with filled symbols representing significant correlations at the 0.05 level.

1993) and in feedbacks within the regional land sur-
faceatmosphere system (Poveda and Mesa, 1997).
The ENSO signal propagates to the east in northern
South America, leading hydrological anomalies by one
month over western Colombia (Poveda and Mesa,
1997) and by six-to-ten months in the Amazon River
basin (Richey et al., 1989; Chu, 1991; Eagleson,
1994). Diagnostics of the physical mechanisms asso-
ciated with these hydro-climatological anomalies are
discussed in Section 3.2.1.
3.3.4. Venezuela
Land-sea temperature contrasts in the tropics strong-
ly affect the evolution of monsoons. In most other
monsoonal regions this gradient is driven by land tem-
perature changes and therefore primarily by insolation.
During ENSO events, the contrast in northern South
American is also created by the large SST anomalies in
the eastern Pacific. Pulwarty and Diaz (1993) observed
that differences of up to 2 8C modulate the migration of
MarchMay convection, and that deep convection
moved westward during warm ENSO events, in contrast
to north and eastward shift during cold events. This
MarchMay convection in southern Venezuela and
Colombia may result not only from warm SSTs off
Peru but also from the westward migration of convec-
tion associated with the South American monsoon.
Some portions of central South America may therefore
experience diminished seasonal rainfall through redis-
tribution of convection rather than total regional sup-
pression. A corresponding eastward displacement was
observed during the strong La Nina event of 198889.
Further study of this eastwest landsea temperature
contrast and its modulation during ENSO, especially
by remote heating along the Peruvian coast, are needed
(Yu and Mechoso, 1999).
As in the remainder of the study area, the response to
ENSO depends on the stage of the events life cycle
(Giannini et al., 2000). The strengthened trade winds
inhibit convective activity over the southern Caribbean
during July and August (Year 0) of a warm event. But
precipitation increases the following year (+ 1) is asso-
ciation with weakened trades and positive SST anoma-
lies in the tropical North Atlantic (Pulwarty, 1994;
Enfield and Mayer, 1997; Poveda and Mesa, 1997;
Giannini et al., 2000). Large-scale controls on interan-
nual variability of Venezuelan climate may be summa-
rized as follows. Wetter than average rainy seasons
require; a persistent positive SOI, below normal SSTs
in the eastern Pacific, above normal SSTs in the Carib-
20 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
bean in Year 0, weaker easterlies between 108N and
208N at 200 mb, below normal SSTs in the tropical
South Atlantic, and the northward displacement of the
South Atlantic High reaching into Brazil (Pulwarty et
al., 1992). Regional variations in climatological con-
trols are also manifest in relationships with ENSO,
through both direct atmospheric links and the lagged
influence of eastern Pacific SSTs. Rainfall of the Mar-
acaibo region is significantly modulated by the eastern
Pacific, as it is in the Andes region and the northern
Cordillera de la Costa.
Coastal regions of Venezuela and Colombia are also
subject the influence of bnortesQ. In December 1999,
during a particularly heavy event (512 mm in two days
i.e. more than the annual total for the north coast),
convection was pushed onto the Cordillera de la Costa
by an extratropical cold front moving off North America,
which penetrated the tropics (Lyon, 2003). Over 15,000
people died, a toll greater than any previously associated
with tropical storms during the past century.
3.4. North Atlantic
The North Atlantic is a potentially important region-
al source of moisture and control on the strength of the
trade winds, propagation of easterly waves and tropical
storms. The North Atlantic Oscillation (NAO) is one
macro-climatic indicator of the state of the atmosphere
ocean system over the maritime area. First postulated
by Sir Gilbert Walker, it is analogous to the Southern
Oscillation (SO), reflecting differences between the
subtropical and sub-polar North Atlantic at monthly
and seasonal time scales. The NAO manifests itself
in; westerly mid-latitude winds, SSTs, and the climate
on the adjacent continental regions. Similar to the SO,
the NAO is defined in terms of pressures under the sub-
tropical anticyclone (e.g., Azores), and sub-polar sur-
face cyclone (e.g., Iceland). A positive mode of the
NAO arises when the subtropical high is anomalously
strong and the Icelandic low is very deep. Westerly
winds of the North Atlantic then become stronger.
During negative NAO both centers are anomalously
weak. Temporal variations in the NAO are linked to
weather disturbances over Europe, the eastern United
States, and the Mediterranean basin. Rogers (1988)
suggests that precipitation over the Caribbean and the
tropical Americas vary during extremes of the NAO,
due to its influence over the strength of the North
Atlantic trade winds. Hurrell et al. (2003) provide a
detailed and updated discussion on the NAO.
Seasonal cross correlations of NAO and streamflow
in Colombia (Poveda and Mesa, 1996; Poveda et al.,
1998) indicate significant correlations, declining from
north to south. Most exist between the NAO in
MAM and discharges in DJF, and NAO in DJF
with flows in JJA and SON. Fig. 14 depicts results
from 10 rivers, with the correlation coefficient be-
tween the NAO index during the quarter shown on
the abscissa, and river discharges during the seasons
(quarters) marked by the different lines. These obser-
vations may be explained by the intensity of the trade
winds over the tropical North Atlantic and the Carib-
bean, as influenced by the position and strength of
the North Atlantic Anticyclone. The exact nature of
this relationship and its interaction with the annual
cycle of the trades, more intense in DecemberFeb-
ruary less intense during JuneAugust, warrants fur-
ther research.
As noted previously, the strength of the trade winds
is closely related to the veranillos of the Pacific slope
and increased rainfall along the Caribbean, both of
which show marked association with ENSO. Vargas
and Trejos (1994) propose that changes in pressures
in the North Atlantic are responsible for many low
frequency changes observed in precipitation records
within Costa Rica. An analysis of seasonal precipitation
totals throughout Costa Rica (Waylen and Quesada,
2002) supports earlier conclusions (Enfield and Alfaro,
1999; Giannini et al., 2000), that regional precipitation
may be sensitive to SST of the tropical North Atlantic.
Monthly SSTs since 1950 over the Tropical North At-
lantic and Nino 3.4 (www.nnic.noaa.gov/data/indices/
index.html) are used to place each month into one of
four classes depending on whether oceanic temperatures
are above or below long run median. In Fig. 15, the pair
of maps beneath the heading bCold PacificQ can be
viewed as being roughly analogous to the cold phase
of ENSO and those under bWarm PacificQ are suggestive
of warm phase, although this is far from a one to one
correspondence. The horizontal divisions indicate the
state of the tropical North Atlantic. Each chloropleth
map depicts the class mean veranillos (JA) precipita-
tion, expressed as a standard normal deviate from the
undifferentiated station mean, under each set of com-
bined oceanic temperature conditions.
The seasonal drought on the Pacific slope and excess
rains on the Caribbean, associated with a warmer Pa-
cific, are particularly marked during periods of a cooler
Atlantic, with the regional cordillera clearly delineating
the two effects. If the Atlantic is warmer, the drought is
more ubiquitous but not as severe. A cooler tropical
North Atlantic implies higher atmospheric pressures, a
strengthening of the trades and a more pronounced
windward/leeward effect. By contrast, a cooler Pacific
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327 21

Fig. 14. Cross-correlation coefficients between the North Atlantic Oscillation (NAO) and ten river discharge stations in Colombia. Correlations are
shown for the NAO during the season on the abscissa and river discharges during the season denoted by the symbols.
22 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327

Fig. 15. Mean seasonal (JulyAugust) precipitation totals in Costa Rica under four sets of temperature conditions in the Equatorial Pacific and
tropical North Atlantic. All observations are expressed as standard normal deviates using the long run seasonal mean and variance at each station.

is associated with a greater onshore flow to the Pacific
slope, and reduction of seasonal rainfall along the
Caribbean. A warmer than normal tropical North At-
lantic further encourages this onshore flow, particularly
in northwestern areas of the country, where it may
accompany a reduction in local upwelling of cooler
oceanic waters in the Gulf of Papagayo.
Enfield and Mayer (1997) report that rainfall is
enhanced over the Caribbean and Central America
west of the cordillera, when cool SST anomalies are
found in the eastern Pacific and warm ones in the
tropical North Atlantic oceans. Such conditions weaken
the trade winds, allowing enhanced convection over the
Caribbean Islands and greater penetration of cross-
equatorial westerlies along the Pacific coast. A cold
Pacific and warm tropical Atlantic are robust features
for JuneAugust composite SST anomaly differences
and often occur during rapid transformation from warm
to cold ENSO phases, whereas a positive relationships
between Atlantic SSTs and rainfall in the Llanos is
clearly defined, especially a few months in advance.
4. Final remarks
We have reviewed the main features of the present
climate of northern South America and Southern
Central America, focusing on Costa Rica, Panama,
Colombia and Venezuela, and its variability at annual
and interannual time scales. Due to the specific time
scales of our review, we have explicitly ignored
important physical mechanisms controlling intra-sea-
sonal climate variability in the study region. Among
those we did not mention the activity of the 3060
day or MaddenJulian Oscillation (Madden and
Julian, 1972), which is known as the principal
mode of climatic fluctuation in the intra-seasonal
band over tropical South America (Mo and Kousky,
1993). The diurnal cycle of precipitation in the trop-
ical Americas is another important time scale omitted
from our study, but it certainly is associated with
annual and inter-annual time scales in nontrivial fash-
ion (Poveda et al., 2005). Likewise tropical easterly
waves over the tropical North Atlantic, Caribbean and
northern South America are responsible for the for-
mation of many severe tropical weather systems on
the Caribbean, ranging from severe storms to hurri-
canes. (Riehl, 1945; Riehl and Malkus, 1958; Chang,
1970). The strong altitudinal gradients within tropical
mountain ranges particularly in the Andes of Colom-
bia and Venezuela, and the Cordillera of Costa Rica
are important in controlling climatic variability over
space.
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
This synthesis of the principal mechanisms and phe-
nomena which control climate variability over the study
region at annual and inter-annual timescales provides
important insights into variability at longer (paleocli-
matic) timescales, and permits the evaluation of climat-
ic evidence of climate and global change (IPCC, 2001),
including the accelerated retreat of tropical glaciers
(Kaser and Osmaston, 2002), outbreaks of tropical dis-
eases (Patz, 2002), and the most rapid loss of biodiver-
sity in the world (Myers et al., 2000). Besides, the
region possesses urgent basic and applied research
needs prompted by large scale deforestation, erosion
and land degradation, vulnerability and risk of human
populations and settlements, and water pollution.
Acknowledgements
Valuable discussions with colleagues throughout the
years have contributed to shape these ideas. Among
them we thank H. Diaz, S. Hastenrath, O. Mesa, P.
Aceituno, C. Nobre, C. Penland, C. Caviedes, and H.
Riehl. TRMM data set was kindly provided by NASA
Goddard Space Flight Center, Data Archive and Distri-
bution Center (DAAC), and data set from the NCEP/
NCAR Reanalysis by E. Kalnay. Research referred to in
this paper has been variously supported by; CIRES
University of Colorado, Boulder) (GP), Colciencias
(GP), DIME (Universidad Nacional de Colombia at
Medellin) (GP), Inter-American Institute for Global
Change Research (GP and PW), NOAA (PW), National
Science Foundation (PW). M. D. Zuluaga and P. A.
Arias helped to prepare Figs. 6 and 9, respectively.
References
Aceituno, P., 1988. On the functioning of the Southern Oscillation in
the South American sector: Part I. Surface climate. Mon. Weather
Rev. 116, 505 524.
Aceituno, P., 1989. On the functioning of the Southern Oscillation in
the South American sector: Part II. Upper-air circulation. J. Cli-
mate 2, 341 355.
Alpert, L., 1945. The intertropical convergence zone of the eastern
Pacific region. Bull. Am. Meteor. Soc. 26, 426 432.
Amador, J.A., Magana, V., 1999. Dynamics of the low level jet over
the Caribbean, 23th Conference on Hurricanes and Tropical Me-
teorology. AMS, Dallas, pp. 868 869.
Ambrizzi, T., Magana, V., 1999. Dynamics of the impact of El Nino/
Southern Oscillation on the Americas climate. Proc. 14th Conf.
on Hydrology. AMS, Dallas, pp. 307 308.
Arnett, A.B., Steadman, C.R., 1970. Low-level wind flow over
eastern Panama and northwestern Colombia. ESSA Technical
Memorandum ERLTM-ARL 26, U.S. Depart. Commerce, Envi-
ronmental Science Services Administration Research Laborato-
ries, Air Resources Lab.. Silver Spring, Maryland. 73 pp.
23
Asner, G.P., Townsend, A.R., Braswell, B.H., 2000. Satellite observa-
tions of El Nino effects on Amazon forest productivity. Geophys.
Res. Lett. 27, 981 984.
Avissar, R., Nobre, C.A., 2002. Preface to special issue on the large-
scale biosphere-atmosphere experiment in Amazonia (LBA).
J. Geophys. Res. 107, 8034. doi:10.1029/2002JD002507.
Bonner, W.D., 1966. Case study of thunderstorm activity in relation to
the low-level jet. Mon. Weather Rev. 94, 167 178.
Byerle, L.A., Paegle, J., 2002. Influences on the seasonal cycle of
low-level flows flanking the Andes and their interannual var-
iability. VAMOS/CLIVAR/WCRP Conference on South Ameri-
can Low-Level Jet. Santa Cruz de la Sierra, Bolivia, 57 February.
Campetella, C.M., Vera, C.S., 2002. The influence of Andes moun-
tains on the South American low-level flow. VAMOS/CLIVAR/
WCRP Conference on South American Low-Level Jet. Santa
Cruz de la Sierra, Bolivia, 57 February.
Carvajal, L.F., Salazar, J.E., Mesa, O.J., Poveda, G., 1998. Prediccion
hidrologica en Colombia mediante analisis espectral singular y
maxima entropVa. Ing. Hidraul. Mex. 13 (1), 7 16.
Carvalho, L.M.V., Lavallee, D., Jones, C., 2002. Multifractal proper-
ties of evolving convective systems over tropical South America.
Geophys. Res. Lett. 29 (0). doi:10.1029/2001GL014276.
Chang, C., 1970. Westward propagating cloud patterns in the tropical
Pacific as seen from time-composite satellite photographs.
J. Atmos. Sci. 27, 133 138.
Chang, C.P., 1973. A dynamical model of the Intertropical Conver-
gence Zone. J. Atmos. Sci. 30, 190 212.
Chang, P., Wang, B., Li, T., Ji, L., 1994. Interactions between the seasonal
cycle and ENSO frequency entrainment and chaos in a coupled
atmosphereocean model. Geophys. Res. Lett. 21, 2817 2820.
Chu, P.-S., 1991. Brazils climate anomalies and ENSO. In: Glantz, P.-
S., et al., (Eds.), Teleconnections Linking Worldwide Climate
Anomalies. Cambridge University Press, pp. 42 64.
Cifelli, R., Petersen, W.A., Carey, L.D., Rutledge, S.A., Silva Dias,
M.A.F., 2002. Radar observations of the kinematic, microphys-
ical, and precipitation characteristics of two MCSs in TRMM-
LBA. J. Geophys. Res. 107, 8077. doi:10.1029/2000JD000264.
Clement, A., Seager, R., Cane, M.A., 2000. Suppression of El Nino
during the mid-Holocene by changes in the earths orbit. Paleo-
ceanography 15, 731 737.
Cohen, J., Silva Das, M.A., Nobre, C.A., 1995. Environmental
conditions associated with Amazon squall lines: a case study.
Mon. Weather Rev. 123, 3163 3174.
Curtis, S., Hastenrath, S., 1995. Forcing of anomalous sea surface
temperature evolution in the tropical Atlantic during Pacific warm
events. J. Geophys. Res. 100 (C8), 15,835 15,847.
Delworth, T., Manabe, S., 1993. Climate variability and land-surface
processes. Adv. Water Resour. 16, 3 20.
Diaz, H.F., Kiladis, G.N., 1992. Atmospheric teleconnections associ-
ated with the extreme phases of the Southern Oscillation. In: Diaz,
H.F., Markgraf, V. (Eds.), El Nino. Historical and Paleoclimatic
Aspects of the Southern Oscillation. Cambridge University, pp.
7 28.
Diaz, H.F., Markgraf, V. (Eds.), 1993. El Nino. Historical and Paleo-
climatic Aspects of the Southern Oscillation. Cambridge Univer-
sity Press. 476 pp.
Diaz, H.F., Markgraf, V. (Eds.), 2000. El Nino and the Southern
Oscillation, Multiscale Variability and Global and Regional
Impacts. Cambridge University Press, pp. 183 204.
Diaz, H., Pulwarty, R., 1994. An analysis of the time scales of
variability in centuries-long ENSO-sensitive records. Clim.
Change 26, 317 342.
24 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Dunbar, R.B., Wellington, G.M., Colgan, M.W., et al., 1994. Eastern
Pacific surface temperature since 1600 A.D. The y18O record of
climate variability in the Galapagos corals. Paleoceanography 9,
291 315.
Eagleson, P.E., 1994. The evolution of modern hydrology (from
watershed to continent in 30 years). Adv. Water Resour. 17,
3 18.
Eltahir, E.A.B., Bras, R.L., 1994. Precipitation recycling in the
Amazon basin. Q. J. R. Meteorol. Soc. 120 (518), 861 880
(Part A).
Elthair, E.A., 1998. A soil moisturerainfall feedback mechanism: 1.
Theory and observations. Water Resour. Res. 34, 765 776.
Enfield, D.B., Alfaro, E.J., 1999. The dependence of Caribbean
rainfall on the interaction of the tropical Atlantic and Pacific
oceans. J. Climate 12, 2093 2103.
Enfield, D.B., Mayer, D.A., 1997. Tropical Atlantic sea surface
temperature variability and its relation to ENSO. J. Geophys.
Res. 102, 929 945.
Enfield, D., Mesta-Nunez, A., Markgraf, V. (Eds.), 2001. Inter-
hemispheric Climatic Linkages. Academic Press, San Diego.
454 pp.
Eslava, J., 1993. Algunas particularidades climaticas de la region del
Pacfico Colombiano. Atmosfera 17, 45 63.
Estoque, M.A., Luque, J., Chandek-Monteza, M., Garca, J., 1985.
Effects of El Nino on Panama rainfall. Geofis. Int. 24, 355 381.
Fernandez, W., Ramrez, P., 1991. El Nino, the Southern Oscillation
and their effects on Costa Rica: a revision. (in Spanish). Tecnol.
Marcha 11, 3 11.
Ferreira, R.N., Schubert, W.H., 1997. Barotropic aspects of ITCZ
breakdown. J. Atmos. Sci. 54, 261 285.
Frank, N.L., Hebert, P.J., 1974. Atlantic tropical systems of 1973.
Mon. Weather Rev. 102, 290 295.
George, R.K., Waylen, P.R., Laporte, S., 1998. Interannual variability
of annual streamflow and the Southern Oscillation in Costa Rica.
Hydrol. Sci. J. 43, 409 424.
Giannini, A., Kushnir, Y., Cane, M.A., 2000. Interannual variability of
Caribbean rainfall, ENSO, and the Atlantic Ocean. J. Climate 13,
297 311.
Glantz, M., Katz, R., Nicholls, N. (Eds.), 1991. Teleconnections
Linking Worldwide Climate Anomalies. Cambridge University
Press. 535 pp.
Gray, W.M., Sheaffer, J.D., 1991. El Nino and QBO influences on
tropical cyclone activity. In: Glantz, W.M., et al., (Eds.), Tele-
connections Linking Worldwide Climate Anomalies. Cambridge
University Press, pp. 257 284.
Greco, S., Swap, R., Garstang, M., Ulanski, S., Shipham, M., Harriss,
R.C., Talbot, R., Andreae, M.O., Artaxo, P., 1990. Rainfall and
surface kinematic conditions over central Amazonia during ABLE
2B. J. Geophys. Res. 95, 17001 17014.
Gruber, A., 1974. Wavenumberfrequency spectra of satellite-mea-
sured brightness in the tropics. J. Atmos. Sci. 31, 1675 1680.
Gu, G., Zhang, C., 2001. A spectrum analysis of synoptic-scale
disturbances in the ITCZ. J. Climate 14, 2725 2739.
Gu, G., Zhang, C., 2002. Cloud components of the Intertropical
Convergence Zone. J. Geophys. Res. 107 (D21), 4565. doi:10.
1029/2002JD002089.
Halpert, M.S., Ropelewsky, C.F., 1992. Surface temperature patterns
associated with the Southern Oscillation. J. Climate 5, 577 593.
Hartmann, D.L., 1994. Global Physical Climatology. Academic Press,
San Diego. 411 pp.
Hastenrath, S., 1966. On general circulation and energy budget in the
area of the Central American seas. J. Atmos. Sci. 23, 694 712.
Hastenrath, S., 1967. Rainfall distribution and regime in Central
America. Arch. Meteorol., Geophys. Bioklimatol., Ser. B Allg.
Biol. Klimatol. 15, 201 241.
Hastenrath, S., 1976. Variations in low-latitude circulations and ex-
treme climatic events in the tropical Americas. J. Atmos. Sci. 33,
202 215.
Hastenrath, S., 1990. Diagnostic and prediction of anomalous river
discharges in northern South America. J. Climate 3, 1080 1096.
Hastenrath, S., 1991. Climate Dynamics of the Tropics. Kluwer,
Dordrecht. 488 pp.
Hastenrath, S., 1998. Contribution to the circulation climatology of
the eastern equatorial Pacific: Lower atmospheric jets. J. Geophys.
Res. 103 (D16), 19443 19451.
Hastenrath, S., 1999. Equatorial mid-tropospheric easterly jet over the
eastern Pacific. J. Meteorol. Soc. Jpn. 77, 701 709.
Hastenrath, S., 2002. The Intertropical Convergence Zone of the
eastern Pacific revisited. Int. J. Climatol. 22, 347 356.
Hastenrath, S., de Castro, L.C., Aceituno, P., 1987. The Southern
Oscillation in the tropical Atlantic sector. Contrib. Atmos. Phys.
60, 447 464.
Haug, G.H., Hughen, K.A., Sigman, D.M., Peterson, L.C., Rohl, U.,
2001. Southward migration of the Intertropical Convergence Zone
through the Holocene. Science 293, 1304 1308.
Haug, G.H., Gunther, D., Peterson, L.C., Sigman, D.M., Hughen,
K.A., Aeschlimann, B., 2003. Climate and the collapse of the
Maya civilization. Science 299, 1731 1735.
Henderson-Sellers, A., Robinson, P.J., 1986. Contemporary Climatol-
ogy. Longman, Essex (UK). 439 pp.
Hendon, H.H., Liebmann, B., 1991. The structure and annual var-
iation of antisymmetric fluctuation of tropical convection and
their association with Rossby-gravity waves. J. Atmos. Sci. 48,
2127 2140.
Hess, P.G., Battisti, D.S., Rasch, P.J., 1993. The maintenance of the
Intertropical Convergence Zone and the large-scale circulation on
a water-covered earth. J. Atmos. Sci. 50, 691 713.
Hodell, D.A., Curtis, J.H., Brenner, M., 1995. Possible role of climate
in the collapse of classic Maya civilization. Nature 375, 391 394.
Holton, J.R., Wallace, J.M., Young, J.A., 1971. On boundary layer
dynamics and the ITCZ. J. Atmos. Sci. 28, 275 280.
Horel, J.D., Hahmann, A.N., Geisler, J.E., 1989. An investigation of
the annual cycle of convective activity over the tropical Americas.
J. Climate 2, 1388 1403.
Hurrell, J.W., Kushnir, Y., Ottersen, G., Visbeck, M. (Eds.), 2003. The
North Atlantic Oscillation: Climate Significance and Environmen-
tal Impact, Geophysical Monograph Series, Am. Geophys. Union,
vol. 134. 279 pp.
IPCC, 2001. Intergovernmental panel on climate change, third as-
sessment report, Latin America. Chapter 14. In: McCarthy, J.
(Ed.), Working Group II: Impacts of Climate Change. Cam-
bridge University Press, Cambridge (UK), pp. 693 734.
Jacoby, G.C., DArrigo, R.D., 1990. Teak (Tectona grandis L. F.), a
tropical species of large-scale dendroclimatic potential. Dendro-
chronologia 8, 83 98.
Jipp, P.H., Nepstad, D.C., Cassel, D.K., de Carvalho, C.R., 1998.
Deep soil moisture storage and transpiration in forests and pas-
tures of seasonally-dry Amazonia. Clim. Change 39, 395 412.
Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D.,
Gandin, L., Iredell, M., Saha, S., White, G., Woollen, J., Zhu,
Y., Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo,
K.C., Ropelewski, C.F., Wang, J., Leetma, A., Reynolds, R.,
Jenne, R., Joseph, D., 1996. The NCEP/NCAR 40-year Reanal-
ysis Project. Bull. Am. Meteorol. Soc. 77, 437 471.
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Kaser, G., Osmaston, H. (Eds.), 2002. Tropical Glaciers. UNESCO,
International Hydrology Series. Cambridge University Press,
Cambridge (UK). 207 pp.
Kiladis, G., Diaz, H.F., 1989. Global climatic anomalies associated
with extremes in the Southern Oscillation. J. Climate 2,
1069 1090.
Klaus, D., 1973. Invasions of cold air into the tropics on the windward
side of the Rocallasas Mountains (in Spanish). Geofis. Int. 13,
99 143.
Kousky, V.E., 1985. Atmospheric circulation changes associated with
rainfall anomalies in tropical Brazil. Mon. Weather Rev. 113,
1951 1957.
Kousky, V.E., 1988. Pentad outgoing longwave radiation clima-
tology for the South American sector. Rev. Braz. Meteor. 3,
217 231.
Kousky, V.E., Kagano, M.T., 1981. A climatological study of the
tropospheric circulation over the Amazon region. Acta Obstetr-
cia. 11, 743 758.
Kousky, V.E., Kagano, M.T., 1994. Principal modes of OLR and
250-mb circulation for the South American sector. J. Climate 7,
1131 1143.
Koutavas, A., Lynch-Stieglitz, J., Marchitto Jr., T., Sachs, J., 2002. El
Nino-like pattern in ice age tropical Pacific sea surface tempera-
ture. Science 297, 226 230.
Kummerow, C.W., Barnes, W., Kosu, T., Shiue, J., Simpson, J., 1998.
The Tropical Rainfall Measuring Mission (TRMM) sensor pack-
age. Notes and correspondence. J. Atmos. Ocean. Technol. 15,
809 817.
Lau, K.M., Sheu, P.J., 1991. Teleconnections in global rainfall anoma-
lies: seasonal to inter-decadal time scales. In: Glantz, K.M., et al.,
(Eds.), Teleconnections Linking Worldwide Climate Anomalies.
Cambridge University Press, pp. 229 256.
Laurent, H., Machado, L.A.T., Morales, C.A., Durieux, L., 2002.
Characteristics of the Amazonian mesoscale convective systems
observed from satellite and radar during the WETAMC/LBA
experiment. J. Geophys. Res. 107 (D20), 8054. doi:10.1029/
2001JD000337.
Lea, D.W., Pak, D.K., Peterson, L.C., Hughen, K.A., 2003. Synchro-
neity of tropical and high latitude Atlantic temperatures over the
last glacial termination. Science 301, 1361 1364.
Legeckis, R., 1988. Upwellings off the Gulfs of Panama and Papa-
gayo in the tropical Pacific, during March 1985. J. Geophys. Res.
93, 15485 15489.
Lemaitre, Y., Brovelli, P., 1990. Role of a low-level jet in triggering
and organizing moist convection in a baroclinic atmosphere. A
case study: 18 May 1984. J. Atmos. Sci. 47, 82 100.
Lindzen, R.S., 1974. Wave-CISK in the tropics. J. Atmos. Sci. 31,
156 179.
Lopez, M.E., 1966. Cloud seeding trials in the rainy belt of western
Colombia. Water Resour. Res. 2, 811 823.
Lyon, B., 2003. Enhanced seasonal rainfall in Northern Venezuela and
the extreme events of December 1999. J. Climate 16, 2302 2306.
Madden, R.A., Julian, P.R., 1972. Description of global scale circu-
lation cells in the tropics with a 4050 day period. J. Atmos. Sci.
29, 1109 1123.
Maddox, R.A., 1980. Mesoscale convective complexes. Bull. Am.
Met. Soc. 61, 1374 1387.
Magana, V., Amador, J.A., Medina, S., 1999. The midsummer drought
over Mexico and Central America. J. Climate 12, 1577 1588.
Mann, M.E., Bradley, R.S., Hughes, M.K., 2000. Long-term variabil-
ity in the El Nino/Southern Oscillation and associated teleconnec-
tions. In: Diaz, H.F., Markgraf, V. (Eds.), El Nino and the
25
Southern Oscillation: Multiscale Variability and its Impacts on
Natural Ecosystems and Society. Cambridge University Press,
Cambridge, pp. 357 412.
Mapes, B.E., Warner, T.T., Xu, M., Negri, A., 2003. Diurnal patterns
of rainfall in northwestern South America: Part I. observations and
context. Mon. Weather Rev. 131, 799 812.
Marengo, J., 1992. Interannual variability of surface climate in the
Amazon basin. Int. J. Climatol. 12, 853 863.
Marengo, J.A., Hastenrath, S., 1993. Case studies of extreme climatic
events in the Amazon basin. J. Climate 6, 617 627.
Martnez, I., Keigwin, L., Barrows, T.T., Yokoyama, Y., Southon, J.,
2003. La Nina-like conditions in the eastern equatorial Pacific and a
stronger Choco jet in the northern Andes during the last glaciation.
Paleoceanography 18 (2), 1033. doi:10.1029/2002PA000877.
McCreary, J.P., Lee, H.S., Enfield, D.B., 1989. The response of the
coastal ocean to strong offshore winds: with application to circu-
lations in the Gulfs of Tehuantepec and Papagayo. J. Mar. Res. 47,
81 109.
Meja, J.F., Poveda, G., 2005. Atmospheric Features pf Mesoscale
Convective systems over Colombia during 1998 using TRMM and
the 40-year NCEP-NCAR Reanalysis data sets (in Spanish). Rev.
Academia Colombiana de Ciencias Exactas, Fsicas y Naturales.
29 (113), 495514.
Mo, K.C., Kousky, V.E., 1993. Further analysis of the relationship
between circulation anomaly patterns and tropical convection.
J. Geophys. Res. 98 (D3), 5103 5113.
Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B.,
Kent, J., 2000. Biodiversity hotspots for conservation priorities.
Nature 403, 853 858.
Myneni, R.B., Los, S.O., Tucker, C.J., 1996. Satellite-based identifi-
cation of linked vegetation index and sea surface temperature
anomaly areas from 19821990 for Africa, Australia and South
America. Geophys. Res. Lett. 23, 729 732.
Neelin, J.D., Battisti, D.S., Hirst, A.C., Jin, F.-F., Wakata, Y., Yamagata,
T., Zebiak, S.E., 1998. ENSO theory. J. Geophys. Res. 103,
14261 14290.
Nepstad, D.C., de Carvalho, C.R., Davidson, E.A., Jipp, P.H.,
Lefebvre, P.A., Negreiros, G.H., Da Silva, E.D., Stone, T.H.,
Trumbore, S.E., Vieira, S., 1994. The role of deep roots in the
hydrological and carbon cycles of Amazonian forests and pas-
tures. Nature 372, 666 669.
Nesbitt, S.W., Zipser, E.J., Cecil, D.J., 2000. A census of precipitation
features in the tropics using TRMM: radar, ice scattering and
lightning observations. J. Climate 13, 4087 4106.
Nieuwolt, S., 1977. Tropical Climatology. John Wiley. 207 pp.
Nitta, T., Nakagomi, Y., Suzuki, Y., Hasegawa, N., Kadokura, A.,
1985. Global analysis of the lower tropospheric disturbances in
the tropics during the northern summer of the FGGE year:
Part I. global features of the disturbances. J. Meteorol. Soc.
Jpn. 63, 1 19.
Numaguti, A., 1993. Dynamics and energy balance of the Hadley
circulation and the tropical precipitation zones: significance of
the distribution of evaporation. J. Atmos. Sci. 50, 1874 1887.
Obregon, G.O., Nobre, C.A., 1990. Principal component analysis of
precipitation fields over the Amazon river basin. Climanalise 5,
35 46.
Oren, R., Zimmermann, R., Terborgh, J., 1996. Transpiration in upper
Amazonia floodplain and upland forests in response to drought-
breaking rains. Ecology 77, 968 973.
Patz, J.A., 2002. A human disease indicator for the effects of
recent global climate change. Proc. Natl. Acad. Sci. 99 (20),
12,506 12,508.
26 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Petersen, W.A., Nesbitt, S.W., Blakeslee, R., Cifelli, R., Hein, P.,
Rutledge, S.A., 2002. TRMM observations of intraseasonal vari-
ability in convective regimes over the Amazon. J. Climate 15,
1278 1294.
Philander, S.G.H., Gu, D., Halpern, D., Lambert, G., Lau, N.-C., Li,
T., Pacanowsky, R.C., 1996. Why the ITCZ is mostly north of the
Equator. J. Climate 9, 2958 2972.
Pike, A.C., 1971. Intertropical convergence zone studied with an
interacting atmosphere and ocean model. Mon. Weather Rev. 99,
469 477.
Portig, W.H., 1965. Central American rainfall. Geogr. Rev. 55,
68 90.
Poveda, G., Jaramillo, A., 2000. ENSO-related variability of river
discharges and soil moisture in Colombia. Biospheric Aspects of
the Hydrologic Cycle, vol. 8, pp. 3 6 (IGBP, December).
Poveda, G., Mesa, O.J., 1996. The North Atlantic oscillation and its
influence on the hydro-climatology of Colombia (in Spanish).
Proc. XVII Latin-American Congress on Hydraulics and Hydrol-
ogy, Guayaqui, Ecuador, IAHR, vol. 2, pp. 343 354.
Poveda, G., Mesa, O.J., 1997. Feedbacks between hydrological pro-
cesses in tropical South America and large-scale ocean-atmo-
spheric phenomena. J. Climate 10, 2,690 2,702.
Poveda, G., Mesa, O.J., 1999. The low level westerly jet (Choco jet)
and two other jets in Colombia: climatology and variability during
ENSO phases (in Spanish). Revista Academia Colombiana de
Ciencias, vol. 23 89, pp. 517 528.
Poveda, G., Mesa, O.J., 2000. On the existence of Lloro (the rainiest
locality on Earth): enhanced oceanatmosphereland interaction
by a low-level jet. Geophys. Res. Lett. 27, 1675 1678.
Poveda, G., Penland, C., 1994. Prediccion de caudales medios en
Colombia usando Modelacion Lineal Inversa. Memorias XVI
Congreso Latino-Americano de Hidraulica e Hidrologa, IAHR,
Santiago de Chile, vol. 4, pp. 119 129.
Poveda, G., Salazar, L.F., 2004. Annual and interannual
(ENSO) variability of spatial scaling properties of a vege-
tation index (NDVI) in Amazonia. Remote Sens. Environ.
93 (3), 391 401.
Poveda, G., Gil, M.M., Quiceno, N., 1998. El ciclo anual de la
hidrologVa de Colombia en relacion con el ENSO y la NAO.
Bull. Inst. Fr. Etud. And. 27 (3), 721 731.
Poveda, G., Jaramillo, A., Gil, M.M., Quiceno, N., Mantilla, R., 2001.
Seasonality in ENSO related precipitation, river discharges, soil
moisture, and vegetation index (NDVI) in Colombia. Water
Resour. Res. 37, 2169 2178.
Poveda, G., Mesa, O.J., Carvajal, L.F., Hoyos, C.D., MejVa, J.F.,
Cuartas, A., PulgarVn, A., 2002. Prediccion de caudales medios
mensuales en rVos Colombianos usando metodos no lineales.
Meteorol. Colomb. 6, 101 110.
Poveda, G., Mesa, O.J., Waylen, P.R., 2003. Non-linear forecasting
of river flows in Colombia based upon ENSO and its associ-
ated economic value for hydropower generation. In: Diaz, H.,
Morehouse, B. (Eds.), Climate and Water: Transboundary
Challenges in the Americas. Kluwer Acad. Publish., Dordrecht,
pp. 351 371.
Poveda, G., Mesa, O.J., Salazar, L.F., Arias, P.A., Moreno, H.A.,
Vieira, S.C., Agudelo, P.A., Toro, V.G., Alvarez, J.F., 2005. The
diurnal cycle of precipitation in the tropical Andes of Colombia.
Month. Weather Rev. 133 (1), 228 240.
Pulwarty, R.S., 1994. Annual and intrannual variability of convection
over tropical South America, Unpublished PhD Dissertation,
Dept. of Geography, University of Colorado, Boulder, Colorado,
USA.
Pulwarty, R., Diaz, H., 1993. A study of the seasonal cycle and its
perturbation by ENSO events over the Tropical Americas. Fourth
International Conference on Southern Hemisphere Meteorology
and Oceanography, March 29April 2. Hobart, Australia.
Pulwarty, R.S., Barry, R.G., Riehl, H., 1992. Annual and seasonal
patterns of rainfall variability over Venezuela. Erdkunde 46,
273 289.
Pulwarty, R., Barry, R.G., Hurst, C., Sellinger, K., Mogollon, L.,
1998. Precipitation on the Venezuelan Andes in the context of
regional climate. Meteorol. Atmos. Phys. 16, 217 237.
Quesada, M.E., Waylen, P.R., 2004. Flood occurrences in basins on
opposing flanks of Costa Rica (in Spanish). Ing. Hidraul. Mex. 19
(3(2)), 129 138.
Raymond, D.J., 1978. Instability of the low-level jet and severe storm
formation. J. Atmos. Sci. 35, 2274 2280.
Rasmusson, E.M., Mo, K., 1993. Linkages between 200 mb tropical
and extratropical circulation anomalies during the 19861989
ENSO cycle. J. Climate 6, 595 616.
Richey, J.E., Nobre, C., Deser, C., 1989. Amazon River dis-
charge and climate variability: 1903 to 1985. Science 246,
101 103.
Riehl, H., 1945. Waves in the Easterlies and the Polar Front in the
Tropics, Dept. Meteor. Univ. Chicago, Misc. Report 17. Depart-
ment of Meteorology, University of Chicago. 79 pp.
Riehl, H., Malkus, J.S., 1958. On the heat balance in the equatorial
trough zone. Geophysica 6, 505 538.
Roberts, D.A., Keller, M., Soares, J.V., 2003. Studies of land-cover,
land-use, and biophysical properties of vegetation in the large
scale biosphere atmosphere experiment in Amazonia. Remote
Sens. Environ. 87, 377 388.
Rodbell, D.T., Seltzer, G., Anderson, D., Abbott, M., Enfield, D.,
Newman, J., 1999. An ~15,000-year record of El Nino-driven
alluviation in southwestern Ecuador. Science 283, 516 520.
Rogers, J.C., 1988. Precipitation variability over the Caribbean
and tropical Americas associated with the Southern Oscillation.
J. Climate 1, 172 182.
Ropelewski, C.F., Halpert, M.S., 1987. Global and regional scale
precipitation associated with El Nino/Southern Oscillation. Mon.
Weather Rev. 115, 1606 1626.
Ropelewski, C.F., Halpert, M.S., 1996. Quantifying Southern Oscil-
lationprecipitation relationships. J. Climate 9, 1043 1059.
Rossel, F., 1997. Influence du Nino sur les regimes pluviometriques
de lEquateur. Unpublished PhD Thesis, University of Montpel-
lier, Montpellier, France. 289 pp.
Rosenthal, Y., Broccoli, A., 2004. In search of paleo-ENSO. Science
304, 219 220.
Salati, E., 1985. The climatology and hydrology of Amazonia. In:
Prance, G.T., Lovejoy, T.E. (Eds.), Key Environment Amazonia.
Pergamon Press, Oxford, pp. 18 48.
Schultz, D., Bracken, W.E., Bosart, L.F., Hakim, G., Bedrick, M.,
Dickson, M., Tyle, K., 1997. The 1993 superstorm cold surge:
frontal structure, gap flow and tropical impact. Mon. Weather Rev.
125, 5 39.
Schultz, D., Bracken, W.E., Bosart, L.F., 1998. Planetary and synoptic
scale signatures associated with Central American cold surges.
Mon. Weather Rev. 126, 5 27.
Shongart, J., Junk, W.J., Piedade, M.T.F., Ayres, J.M., Huttermann,
A., Worbes, W., 2004. Teleconnection between tree growth in the
Amazonian flood plains and the El Nino-Southern Oscillation
effect. Global Chang. Biol. 10, 683 692.
Shuttleworth, W.J., 1988. Evaporation from Amazonian rain forest.
Proc. R. Soc. Lond., B 233, 321 346.
 G. Poveda et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 327
Snow, J.W., 1976. The climate of northern South America. In:
Schwerdtfeger, W. (Ed.), Climates of Central and South America.
Elsevier, Amsterdam, pp. 295 403.
Stahle, D.W., DArrigo, R.D., Krusic, M.K., et al., 1998. Experimen-
tal dendroclimatic reconstruction of the Southern Oscillation.
Bull. Am. Meteor. Soc. 79, 2137 2152.
Stensrud, D.J., 1996. Importance of low-level jets to climate: a review.
J. Climate 9, 1698 1711.
Sun, D.-Z., in press. The control of meridional differential heating
over the level of ENSO activity: A heat-pump hypothesis. In:
Wang, C., et al. (Eds), OceanAtmosphere Interaction and Cli-
mate Variability, Amer. Geophys. Mon. 147.
Thompson, L.G., 1992. Ice core evidence from Peru and China. In:
Bradley, R.S., Jones, P.D. (Eds.), Climate Science A.D. 1500.
Routledge, London, pp. 517 548.
Tomas, R.A., Webster, P.J., 1997. The role of inertial instability in
determining the location and strength of near-equatorial convec-
tion. Q. J. R. Meteorol. Soc. 123, 1445 1482.
Trenberth, K.E., 1991. General characteristics of El Nino-Southern
Oscillation. In: Glantz, R.M., Katz, R., Nicholls, N. (Eds.), Tele-
connections Linking Worldwide Climate Anomalies. Cambridge
University, pp. 13 42.
Trenberth, K.E., 1999. Atmospheric moisture recycling: role of ad-
vection and local evaporation. J. Climate 12, 1368 1381.
Trenberth, K., Hoar, T., 1996. The 19901995 El Nino-Southern
Oscillation event: longest on record. Geophys. Res. Lett. 23,
57 60.
Trenberth, K.E., Branstator, G.W., Karoly, D., Kumar, A., Lau, N.-
C., Ropelewski, C., 1998. Progress during TOGA in understand-
ing and modeling global teleconnections associated with tropical
sea surface temperatures. J. Geophys. Res. 103, 14,291 14,324.
Tudhope, A.W., Chilcott, C.P., McCulloch, M.T., et al., 2001. Vari-
ability in the El Nino-Southern Oscillation through a glacial
interglacial cycle. Science 291, 1511 1517.
Tziperman, E., Stone, L., Cane, M., Jarosh, H., 1994. El Nino chaos:
overlapping of resonances between the seasonal cycle and the
Pacific ocean atmosphere oscillator. Science 264, 72 74.
Urban, F.E., Cole, J.E., Overpeck, J.T., 2000. Influence of mean
climate change on climate variability from a 155 yr tropical
Pacific coral record. Nature 407, 989 993.
Vargas, A.B., Trejos, V.F.S., 1994. Changes in the general circulation
and its influences on precipitation trends in Central America:
Costa Rica. Ambio 23, 87 90.
Velasco, I., Frisch, M., 1987. Mesoscale convective complexes in the
Americas. J. Geophys. Res. 92 (D8), 9591 9613.
Wang, C., 2002. Atlantic climate variability and its associated atmo-
spheric circulation cells. J. Climate 15, 1516 1536.
Wang, C., 2003. ENSO, Atlantic climate variability, and the Walker
and Hadley circulations. In: Diaz, H.F., Bradley, R.S. (Eds.), The
Hadley Circulation: Present, Past, and Future. Kluwer Acad. Publ,
Dordrecht, The Netherlands.
Wang, C., Enfield, D.B., 2001. The tropical western hemisphere warm
pool. Geophys. Res. Lett. 28, 1635 1638.
Wang, C., Enfield, D.B., 2003. A further study of the tropical western
hemisphere warm pool. J. Climate 16, 1476 1493.
27
Warner, T.T., Mapes, B.E., Xu, M., 2003. Diurnal patterns of rainfall
in northwestern South America: Part II. Model simulations. Mon.
Weather Rev. 131, 813 829.
Waylen, P.R., Caviedes, C.N., 1986. El Nino and annual floods on the
north Peruvian littoral. J. Hydrol. 89, 141 156.
Waylen, P.R., Laporte, S., 1999. Flooding and the El Nino-Southern
Oscillation phenomenon along the Pacific coast of Costa Rica.
Hydrol. Process. 13, 2623 2638.
Waylen, P.R., Poveda, G., 2002. El Nino-Southern Oscillation and
aspects of western South America hydro-climatology. Hydrol.
Process. 16, 1247 1260.
Waylen, P.R., Quesada, M.E., 2002. The effect of Atlantic and Pacific
sea surface temperatures on the mid-summer drought of Costa
Rica. In: Garca-Ruz, J.M., Jones, J.A.A., Arnaez, J. (Eds.),
Environmental Change and Water Sustainability. Instituto Pire-
naico de EcologVa, Consejo Superior de Investigaciones
CientVficas, Zaragoza (Spain), pp. 197 209.
Waylen, P.R., Quesada, M.E., Caviedes, C.N., 1996a. Temporal and
spatial variability of annual precipitation in Costa Rica and the
Southern Oscillation. Int. J. Climatol. 16, 173 193.
Waylen, P.R., Caviedes, C.N., Quesada, M.E., 1996b. Interannual
variability of monthly precipitation in Costa Rica. J. Climate 9,
2606 2613.
Webster, P.J., 1995. The annual cycle and the predictability of the
tropical coupled ocean atmosphere system. Meteor. Atmos. Phys.
56, 33 55.
Werth, D., Avissar, R., 2002. The local and global effects of Amazon
deforestation. J. Geophys. Res. 107 (D20), 8087. doi:10.1029/
2001JD000717.
Wheeler, M., Kiladis, G., 1999. Convectively coupled equatorial
waves: analysis of clouds and temperature in the wavenumber
frequency domain. J. Atmos. Sci. 56, 374 399.
Whetton, P., Rutherford, I., 1994. Historical ENSO teleconnection in
the eastern hemisphere. Clim. Change 28, 221 253.
Yasunari, T., 1987. Global structure of the El Nino/Southern Oscilla-
tion: Part I. El Nino composites. J. Meteorol. Soc. Jpn. 65, 67 79.
Yu, J., Mechoso, C., 1999. Links between annual variations of
Peruvian stratocumulus clouds and SST in the eastern Equatorial
Pacific. J. Climate 12, 3,305 3,318.
Zeng, N., 1999. Seasonal cycle and interannual variability in the
Amazon hydrologic cycle. J. Geophys. Res. 104 (D8),
9097 9106.
Zeng, N., Neelin, J.D., 1999. A landatmosphere interaction theory
for the tropical deforestation problem. J. Climate 12, 857 872.
Zeng, N., Dickinson, R.E., Zeng, X., 1996. Climatic impact of
Amazon deforestationa mechanistic model study. J. Climate
9, 859 883.
Zhang, H., McGuffie, K., Henderson-Sellers, A., 1996. Impacts of
tropical deforestation: Part II. The role of large-scale dynamics.
J. Climate 9, 2498 2521.
Zuluaga, M.D., Poveda, G., 2004. Diagnostico de sistemas convecti-
vos de mesoescala sobre Colombia y el oceano PacVfico oriental
durante 19982002. Av. Recur. Hidraul. 11, 145 160.