NASA Technical Reports Server (NTRS) 20000068520: South American Summer Monsoon of 1997/1998 and 1998/1999

4 Pages·0.34 MB·English

by  NASA Technical Reports Server (NTRS)

24'_AnnualClimate Diagnostics andPredictionWorkshop, I-5November, 1999,Tucson,AZ
South American Summer Monsoon of 1997/98 and 1998/99
William K.-M. Lau and Jiavu Zhou"
Code 913. Lztborator3' for Atmosphere.v
NASA/God&lrd Space Flight Center
Greenbelt. MD 20771
Introduction
It is well known that during El Nifio years, severe
drought occurs in the area of Amazon and northeastern
Brazil. According to the linear model result (Silva Dias
et al. 1983) the reduced latent heating over the Amazon
may lead to a weaker than normal upper tropospheric
Bolivian high. As a result, some studies have suggested
a weaker South American summer monsoon (SASM)
during El Ni6o years (Bell et aL 1999).
Using reanalysis data, Zhou and Lau (1999) found
a statistically significant positive correlation between
the tropical eastern Pacific sea surface temperature
(SST) and the strength of low-level jet (LLJ) along the -v"
eas, eo,ontoh,i,,h, et,o ica,-.Au n,, oep..,ca,
Douglas (1999) also showed a strong LLJ at Santa Cruz, "
Bolivia during a special pilot ballon observation period _0_,_., _:_.:. :::;_?_:"- _,___7_r_ %------_.'x....._
in 1997/98 El Nifio austral summer. Since this LLJ is g"= >_.9_.'>,"."-.L_:...:::_:::_ ___ ._
a1n99i8n)tegirnal theparstumofmetrhteimme,onsoothnesesyrsetesmults (Zihnoduicateadnd Lthaaut _-::t,S_ )k ..::'_' ......it_i.__--_ A_
SASM could be stronger than normal in E1 Nifio years. EQ__a-d-+_,_ r I,' /_---_
To clarity this issue, we conducted an _,_,>,_.'_-'h"3"_]"ZIQ%_ _ M_ ; k,y'-]
30¢ "! 7_'-I ' z" ' J,// _ ", " _t 05_
investigation on SASM anomaly in the recent ENSO I_ "s_/<_(.CTP__ 'c''''-o i
event of 1997/98 E1Nifio and 1998/99 La Nifia. In the 6os - Q-.______s ---------_
following, we first __ivea brief review on SASM and the -_---77_--:'__----'"_'_''O ---'_ -'_
interannual variability of summer rainfall over South <:-----°._'_
908 1
America. Then, the impact of 1997-99 ENSO on the 60E :20_ ;80 !20_ 60W 0 60E
South American regional thermal structure and its
dynamical consequences to SASM will be discussed. Fig. 1 Regression of the DJF CMAP rainfall principal
component of the ENSO mode with GEOS-1 SLP
Data and 850-hPa wind (upper) and that with 850-hPa
streamfunction (lower). (From Zhou and Lau
The data used in this study are the National Center
1999
for Environmental Prediction (NCEP) reanalysis, the
NCEP SST and the monthly mean CPC (Climate Gran Chaco of South America, as a result of rapid
Prediction Center) Merged Analysis of Precipitation heating over the South American land mass following
(CMAP). the seasonal migration of the sun. The low level flow,
which is an integral component of the South American
SASM and interannual variability of summer summer monsoon migrates from the northwestern
rainfall African continent, flows along the northern equatorial
Atlantic Ocean, crosses the equator over the east of the
From austral spring to summer, sea level pressure
tropical Andes and brings the rainy season to the
increases over northwest Africa and decreases over
Corresp,mding author address: Dr William K.-M. Lau, C_de 91_.N..\SA/GSFC. GreenbeLt. MD"()771
E-mail: lauCa'_.c_l_j_.matg.:gsfc_n.asa.g!.)E
" SM&A Corporation
2,{* Annual Climate Diagnostics and Prediction Workshop. I - 5 November. 1999. Tucson. AZ
subtropical South America. In November, the forefi'ont
<)theavy precipitation advances ,;outhw'ard more than l0
degrees, which signifies the SASM onset. During DJF
(December, January and February) SASM prevails over
most tropical-subtropical land area. The maximum
rainfall axis tilts northwest-southeastward, coinciding
with the South Atlantic convergence zone (SACZ),
which is most active of this time of the ye.ar. The
SASM rainfall regime can be clearly identified
according to the seasonal rainfall concentration, which
shows that the subtropical South American continent
(25°-10°S) receives almost 50% of the local annual
rainfall from SASM precipitation in DJF. This rainfall
regime is distinct from that over northern Northeast
Isl ,+41o.. " _oo - -
Brazil and the equatorial Atlantic, where large _t' ¢ !,_¢_,. \/ " .... -.'_]
concentration of annual rainfall is in austral fall.
ill\tw _'\ i4VS_,_I _111L_I£_ ,<.: loe a:: .,j:
The ENSO impact on the South American summer
rainfall has been explored extensively by many previous
studies, which showed during El Nifio years severe
drought occurred in the area from Northeast Brazil to Fig. 3 DJF 200-500 hPa mean temperature (left) and
central Amazon and flood over Urugray-southern Brazil 200 hPa geopotential height (right) of climatology
and the west coast of Ecuador. Figure 1(from Zhou and (upper) 1997/98 (middle) and 1998/99 (lower).
Lau 1999) shows the dynamical response of the ENSO
anomaly in the lower troposphere. It reveals that during are the integral parts of the global ENSO response.
El Nifio years the northwest African high is enhanced, Consequently, the flow around the outskirts of the
and the subtropical highs of the South Pacific and the tropical South Atlantic high and the summer monsoon
western North Atlantic are weakened. These anomalies
flow over the equatorial Atlantic and along the eastern
100 - • 200 foothill of the Andes are substantially enhanced. These
200 210 220 210'>'5_-
230 2_0_ two branches of anomalous wind diverge over
300 -240- 240_
400 -- 26205_0__ 226500 northeastern Brazil and converge over southern Brazil.
p 500 270 _270 Over the west coast of Ecuador, anomalous convergence
800
_, 700 280- -280_ is encountered due to above normal westerlies from the
i.., 800 displaced Walker circulation over the eastern equatorial
9OO_Zg0Climatology ,_zg0_ Pacific and the increased easterlies of enhanced summer
I000
Ioo monsoon circulation along equatorial Amazon Basin.
200 These changes imply abnormal low-level moisture
a 300
convergence, which is consistent with the
,._ 400
soo aforementioned anomalous rainfall distribution over
G00 South America.
7oo
c_ BOO
900 '1997/98 Anomaly --_0, _--,5__,-. , -- SASM anomalies in 1997198 and 1998/99
tO00
I00 E:'_;-:<c:?_-'__2-_-:--:_-=,_" . .-__-c_--_ o--O5 J_> --" " The years 199%1999 witnessed the strongest El
200
a 300 _ 0 . Nifio event of the century. During this time peric,d. the
-_ 400 typical ENSO-related rainfall anomaly pattern was
5oo -2 -_..
= 600 --1L5 . J -05 reported by a number of studies (Bell et al. 1999). The
700
w following analysis will focus on the SASM system and
8oo -05 _ -05
the cause of the anomaly.
goo i99B/99 Anoi.,n-al-y_ --tc' 05
lO00_so _6ow _4ow t20w t00w _ow sow 4ow z0w a. Temperature (ltlOtrl_l[V _lttd hvdro.Ytatic re.vpott.ve
Fig. 2 Vertical cross-section of DJF temperature
As a deep convecti,,e system, summer monsoon is
ah)ng 15°S. From top to bottom shows the
thermally driven and ',erticallv extended from the
climatology, the 1997/98 anomaly and the surface to the upper troposphere. %) search for the
1998/99 anomaly, respectively. ENSO impact on the rnons_on -;,,,stem. a _,erti,..'al cross-
24t_Annual Climate Diagnostics and Prediction Workshop, ! -5 November, 1999. Tucson, AZ
temperature gradient increases in the subtropics, which
has great impact on the strength of the subtropical
westerlies in the upper troposphere. Figure 4 show's the
vertical cross-section of zonal mean subtropical
westerlies along a north-south strip over South America•
In DJF 1997/98 a jet core of greater than 30 m/s is
1oooi.1>_to,_'y located near 33_S at 200 hPa, where the westerlies are
_oo. il almost 10 m/s (5 m/s) stronger than that in DJF 1998/99
2OO
(climatology). At the north of the jet, anomalous zonal
'* 300 '
400 - wind share produces significant positive vorticity
500 anomaly, which is dynamically consistent with the
=. 600
enhancement of Bolivian high.
700 '\
During 1997-98, the SASM flow pattern is in
I000 agreement with that based on previous ENSOs. Figure
5 demonstrates a sharp contrast of LLJ along the eastern
side of subtropical Andes and that of the South Atlantic
subtropical high between 1997/98 and 1998/99 austral
summer. A systematic strengthening (weakening) of the
monsoon system introduced by El Nifio (La Nifia) is
,xxxt,..J
_. 800 : _ -"'---mJ / t evident.
c. Changes in annual cycle and characteristics of
45S 42S 39S 36S 33S 30S gTS 24S 2IS 18S 15S 12S
SASM development
Fig.4 DJF zonal wind zonally averaged from 80°W to
50°W. From top to bottom shows the climatology, Since the monsoon evolution is part of the regional
the 1997/98 anomaly and the difference between annual cycle, it is important to know how the annual
1997/98 and 1998/1999, respectively. cycle is affected by ENSO. Figure 6 shows the time
evolution of each key component and compares that
with the annual cycle. From the figure we can see the
section of the temperature along 15°S is plotted (Fig. 2).
It shows that in DJF 1997/98 the El Nifio induced annual cycles of all components are amplified during
1997/98 El Nifio and slight weaker during 1998/99 La
tropospheric warming starts from I°K at the surface of
Nifia. The warming of SST over Nifio 3 region started
the Nifio 3 region and enhances upward to a maximum
of 3°K between 500 and 150 bPa. In the upper 500 b.Pi C,eopote_tAal :-Imllh% 880 bPa llrLD.d
(-5000 Ipm) fro/,)
troposphere the warming area expands eastward,
reaching South America at about 1.5°K. In DJF
: ' ( "--> .0 \1
1998/99, the anomaly sign is completely reversed and
the intensity weaker than that in DJF 1997/98, :: ..oJ---i
especially over South America.
i ,t ,,;) .... E., ..... ?.".: ',kl
Figure 3 shows the geographical distribution of
500-200 hPa mean temperature and 200 hPa
::, / _. _1 ...I..:-: --s:-._q?2S.,.._ _....
geopotential height. It is evident that during DJF
1997/98 (1998/99)of El Nifio (La Nifio) Bolivian high '"_ ' ,"'" .... +.... Z£,.2, .')_
is much stronger (weaker') than that shown in the ...... "__ _-, "(_,_ "_- "-'"''' /'_ ...... ;"
climatology as a result of hydrostatic response to the
tropospheric warming (cooling). This result indicates ,-..° - I
the primary importance of the thermal impact on the
intensity of Bolivian high during 1997-99 ENSO
episode. The dynamical forcing (Rossby wave
propagation induced by Amazonian transient convective
heating) suggested by previous studies is secondary in
this case.
h. Dvmtmical impact and a#zomalous circulation Fig.-K DJF 500 hPa geopotential height (left_ and
850 hPa wind (right) of climatology (upper),
In strong El Nifio years, because of the I997/98 !middle) and 1998/99 (lower).
tropospheric warming in the tropics, the meridional
24_nAnnual Climate Diagnostics and Prediction Workshop, I - 5 November. 1999, Tucson, AZ
from June 1997 and ended in August 1998. A two Conclusions
months delay can be found in the progressions of
In this paper v,e have conducted an investigation
tropospheric mean temperature and upper tropospheric
to compare and contrast the variations of the SASM
geopotential high anomalies over Altiplano plateau, and
system in recent two SST extreme years of 1997/98 (El
even longer time delay in the evolution of the South
Nifio) and 1998/99 (La Nifia). Using the National
Atlantic subtropical high anomaly. The upper
Center for Environmental Prediction (NCEP) reanalysis,
troposphere subtropical westerly anomaly seems to be
our result shows a substantial increase of the mid-
more sensitive to the change of SST and, evolves
tropospheric temperature above the Altiplano Plateau in
approximately in phase with the Nifio 3 SST anomaly.
austral summer of 1997/98. Accompanied with that. the
The initiation of the positive anomaly of the low-level
Bolivian high is much stronger than normal during DJF
northwesterlies along the eastern foothills of subtropical
1997/98. The South Atlantic subtropical high is
Andes accompanies the enhancement of the upper
enhanced and expands westward into eastern Brazil. As
tropospheric subtropical westerly anomaly, while the
a result, a strong single upper-tropospheric jet over the
termination seems to be more related to the reduction of
subtropical continent and an intense NW LLJ along the
the local geopotential height anomaly, about one month
eastern foothill of the tropical-subtropical Andes are
later (earlier) than that of Bolivian high (South Atlantic
generated. These features characterize a strong SASM.
subtropical high) anomaly.
The situation is completely reversed in 1998/99,
/ '=_,,,osasr I[I ,,04 s_,,_*u,=uo II suggesting a weakening of the SASIVl. The annual cycle
=°1 I._ [ I [ [ ,,,_/_,o-,m=h i I was amplified in 1997/98 and reduced in 1998/99. In
-{ / I 1.1'-"1 1:%
addition, this study demonstrates the primary role of the
thermal forcing in strengthening of Bolivian high during
1997/98 El Nifio episode, which more than compensates
the possible dynamical impact due to reduced heating
over Amazon.
Acknowledgments. We thank Drs. Pingping Xie and
Phillip A. Arkin for providing us the monthly mean
CMAP data used in this study. This research is
supported by the Earth Observing System /
24? ' 28 ; :
24e 2_, Q _ Interdisciplinary Science investigation on hydrological
20
245 processes and climate, and the Global Modeling and
Iz Analysis Program of NASA, Mission to Planet Earth
5_nI BelivtLn High
AI_ Level Set ! Office.
i '1 '" ' ;=
° / .... \ .... ": References
= /' l" I ,'....J:V" "
',001\'.j' I " "r I V ! "" !_/ Bell, G.D., M.S. Halpert, C.F. Ropelewski, V.E.
l -'q ! ...... I
Kousky, A.V. Douglas, R.C. Schnell, and M.E.
060 bJPk _ l_d (@6lr,161) I l
a6oL- t=oo(6sv,15s) i I . t . h i Gelman, 1999: Climate Assessment for 1998. Bull.
Amer. Meteor. Soc., 80, SI-$48.
Fig. 6 Time evolution of key components, named at Douglas, M., 1999: The low-level jet at santa Cruz,
the top and specified at the bottom of each figure. Bolivia during January-March 1998 pilot balloon
The lines marked by the cross sign indicate the observations and model comparisons. I0" Symposium
annual cycle. on Global Change Studies, 10-15 January 1999.
Dallas, Texas. 223-226.
Further examination of the SASIVl development Silva Dias, P.L.. W.H. Schubert, and M. DeMaria,
shows that the convective heating over subtropical 1983: Large-scale response of the tropical atmosphere
Andes is more intense and congregated in the period to transient convection. J. Atmos. Sci., 40, 2689-2707.
from mid-summer to early spring in 1997/98 El Nifio
year. In contrast, during 1998/99 La Nifia episode the Zhou, J. and K.-NI. Lau, 1998: Does a mons(×)n climate
summertime convective activities are weak and spread exist over South America? J. Climate, 11. 1020- 1040.
over a longer time period, in which major breaks can be , and 1999: lnterannual and decadal
identified. In relation to the heating anomaly, a single variability of summer rainfall over South America.
jet in 1997/98 and a double jet in 1998/99 characterize lO"' S3mpo.rium ml Global Cfum_e Stroller, 10-15
the upper-level subtropical westerly. January 1999. Dallas, Texas. 5It5-519