__i_̧iiiiiiiiii_iliiiiiiiiiiiiiii.....__ _!ii_iiii_ii_iiiiiiiiiiiiiiiiii!iiii

_?_!i_i!_!i_iii_iii!iii_iiiiiiii!iiiiiii_i_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii!i___iii_i_ii_i_i_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiill

he numerical simulation of precipitationhelps scientists understand the complex

mechanisms that determine how and why

rainfall is distributed across the globe.

Simulation aids in the development of forecast-

ing efforts that inform policies regarding the

management of water resources. Precipita_on

modeling also provides short-term warnings for

emergencies such as flash floods and mudslides.

Just as precipitation modeling can warn of an

impending abundance of rainfall, it can help

anticipate the absence of rainfall in drought.

What constitutes a drought?

A meteorological drought simply means

that an area is getting a significantlylower amount of rain than usual over

a prolonged period of time.

An agricultural drought is based on thelevel of soil moisture.

As we learn more about what drives precipitation, we can anticipate extreme fluctuations in rainfall, leading to droughts and floods.

57

i!i_i!_i,i!ii_!'i!_ii_iiii_ii _.... •

iiiiiiiiiilii_

Precipitation is a vital link in the hydrological cycle that moves water through land, sea, and ain

A hydrological drought focuses on

lower precipitation levels that reducebodies of water on the surface, such

as lakes and streams.

A socioeconomic drought takes intoaccount human activities, such as theconstruction of dams and terrain

shaping, that alter the distribution ofsurface water.

A numerical model for drought prediction is

shaped by the drought criteria that are selected.

The meteorological definition of drought focuses

primarily on the amount of precipitation. The

agricultural and hydrological definitions require

model components not only for precipitation but

also for soil moisture absorption and evapora-

tion of water. The socioeconomic view of

68

iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiJiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_Ee_p[_t[O..Niiiiiiiiiliiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii_iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii!iiiiiiiliiiiiiiiiiiliiiii!iliiiiiiiii!ii!iiiiiiiiiiiii!iiii@iiiIii_!@i_iiiiiiii i iiiiil!iiiiiiii_iiiii_i_i_i_i_i_i_i_i_i_i_i:_i_i_i__i_!_i_i_i_i_i_i_i___________________________________.............._ _:ii iliiiiiiiii!ili!iiii!ii!iiii!i!iiiiiiiiiiiiiiiiiiiiiii!iiiiiiiiiiiiiii!iiilliiiiiii!!iiiili!iiii!ii!!iiiiiii!iiiiiiiiii!i!ili!i!!ii!iiiiiii_'_'_i'i'i'_'_'ii_'_i_'_'

Monsoons

and August means _e end of the long d_ season... A monsoon that is either too strong or too weak

may lead to flooding or drought. _!b help predict monsoon streng_ and.to understand the interac,

tion between monsoon and other atmospheric phenomena, we must understand why monsoons

exist, what accounts for the circulation field of a monsoon, and why monsoon onset is:abrupt.

The two types of monsoon are summer and winten Despite their names, boO'monsoons take place

during_ ...........the July-August. season. The main difference .is in their location, The su er monsoon

takes place more than 10 degrees north of the Equator, where the season_is summer.. Likewise, the

winter monsoon is located south of the Equator, where July is in the middle of winter.

The summer monsoon is a con_nental_size convective system &aracterized by a l_ge preapita-

tion region. This region is identified as an off-equator_aI inter t con nce zone (ITS)' The

strongest summer monsoons r near southern Asia. Monsoons also r near Australia, North.

and Sou_ America, and Africa.

In a northern summer ;monsoon, the prevailing winds at the low levels are from the sou st. At

that does not occur elsewhere in the tropics.. . , -.

Just as low-level Winds run southwesterly in the nor,hem summer monsoon, they run sou t-

erlv in the southern winter monsoon. Ancient mariners depended on this seasonal &ange of pre-

vailing wind to.sail be_'een India and. Africa.

in the monsoon onset process, the shi_ from near the Equatorto more than 10 degrees away

in days. Compared With the movement of the Ea:rth,s flit toward the Sun, this change is rapid.

Like any Iarge-scaI.e drcutation, the monsoon is highly influenced by the Cofiolis force. TNslforce,

wN.ch is created by the Earth,s rotation, es rd the right of wind dir_on in the Northern

Hemisphere and toward the left of wind air.on in the Southern Hemisphere. Atthe low levels,the air mass flows toward an I in the Northern Hemisphere from the other side of the Equator,

"iNe wind direction, which is influenced by the Coriotis force, enters the northern hemisphere from

the southeast. After the air mass crosses the Equator, the direction changes to a southwesterly flow:

Wind flow also comes .from the north of the .GCZ, but its strength, is much weaken

At upper levels, the: airflow must return in the opposite direction. The circulation .field and the con-

vective heating in the I'tCZ precipitation region::interact; yet, neither causes the other.

59

0ooo00i oi0ooi

drought, with its dependence on human action,

may be the most difficult version to quantify in a

simulation.

Accurate precipitation modeling can also

improve understanding of other environmental

issues. For example, rain droplets absorb certain

airborne pollutants and remove them from the

atmosphere when they fall to the ground.

Therefore, models of atmospheric composition

can benefit by incorporating a precipitation

component.

As scientists studied global precipitation, they

noted a correlation between the amount of rain-

fall in particular areas of the world and other

weather conditions. For example, in certain

regions, droughts are strongly associated with

the warm or cold phase of the E1 Ni_o-Southern

Oscillation cycle. Indeed, evidence links global

patterns of sea surface temperature (SST) to

regional precipitation patterns across the

decades. However, as evidenced in the second

research profiles that follows, SST explains only

a portion of precipitation variability.

Precipitation simulations are usually compared

to actual rainfall measurements to test their accu-

racy. These measurements can be derived from

rain gauges at specific points on the ground.

Remote sensors based on radar and satellite

instruments are a more practical means of col-

lecting data to estimate rainfall over larger..

regions.

References

Browning, K. A., and Gurney, R. J. (Eds,), Global

Energy and Water Cycles, Cambridge University

Press, 1999

Collier, C. G., World Meteorological

Organization Operational Hydrology Report

No. 46: Precipitation Estimation and Forecasting,

2000

Linacre, E., and Geerts, B., Climates and Weather

Explained, Routledge, 1997

Sorooshian, S., "Hydrologic forecasting,"

Encyclopedia of Earth System Sciences, Vol. 2,

Academic Press, 1992

Trenberth, K. E. (Ed.), Climate System Modeling,

Cambridge University Press, 1992

Wilhite, D. A. (Ed.), Drought Volume I"A Global

Assessment, Routledge, 2000

6O

ili_ii!i!¸ • ............................i__i!i!iiiiiiilLiiiiiiiiiiiiiiiiiiiiiiii!ililiiiiiiiii

Research Profile-The Role of Landmass in MonsoonDevelopment ...............

Investigator: _ " _Winston Chao, NASA Goddard Space Flight Center, Climate and Radiation Branch _

Since the 17th century; a fundamental belief has

prevailed that the basic cause of a monsoon is

the contrast in surface temperatures between the

continents and the oceans. Research using

general circulation model (GCM) experiments

demonstrated that this belief should be changed.

In an experiment in which the researchers

replaced the landmasses of Asia and Australia

with ocean, the simulated Asian and Australian

monsoons remained largely intact. The figures

on page 62 show the monsoon precipitation

averaged for the month of August in a 4-year

model integration. The upper panel shows the

control experiment with the continents intact.

For the lower panel results, ocean replaced the

continents.

Further experiments showed that in the Asian

monsoon, the change resulting from the replace-

ment of continents with ocean was caused more

by the removal of topography than by the

removal of land-sea contrast. Whereas land-sea

contrast played a minor modifying role in the

Asian summer monsoon, it played an important

role in the African and South American summer

monsoons. However, that role was not irreplace-

able. If an ocean with a high enough SST

replaced Africa or South America, the monsoon

continent.

To understand why monsoons exist is to under-

stand why the ITCZ exists in particular

off-Equator regions during the summer. In an

aqua-planet model with zonally uniform SST, the

ITCZ is zonally uniform and located at the tropi-

cal latitudes where SST is high. The solar energy

absorbed by the surface at these areas heats the

air above and controls the location of convective

precipitation.

The SST is only one factor that determines the

location of the ITCZ. Another influencing factor

is the Earth's rotation, which pulls the ITCZ

toward _o different latitudes at around

13 degrees north and south of the Equator.

Therefore, ITCZ is often observed at both

locations,

When researchers varied the distribution of SST

in the simulation, the ITCZ was no longer

zonally uniform but concentrated somewhat to

the west of the longitudes where SST was

highest. The model then gave a precipitation and

circulation field that was very close to real-world

observations. Therefore, researchers concluded

that landmass is not a necessary condition for

monsoons.

61

August Mean Preci pitation, ................" ................................................................._:_-_:_-_-_-_............................................,_...............;,;_,,,,w,_,,,_...................:_._._......._=,:._`:_:_::_:_:_:_;_:_:_:_:_:_:_:_:_:_:;i_:;_:_:__ii`

•:_ :_,.:...'":::_.... "" :::::ii.... ..,_,_i_i_'.<..=:,!,i:_i:__>...#ii:_" ._:_:::!:_ii:::__:_ii'_ ...... "i!!_i!!!::_' :t='""_>..""_i_i__,_...... _!::_i:!_.::.!" _i_i_" _:.::!:,_._

iii_ii:_;:_`_.............::_iii_ii_::

20: _,_ii_i:_::::_.........',_ii_:_:::_........ '::::::::::::::::::::::...............ii!iii_....• .....................

iiii:!:ii!ii::::ii::i_::ii:ii:::_i:i:!_! _!:!

:8...............................20E::...............................40E:::::::........:::::::_E:::_:::::::::....=====================:::_E•...........................:i_E::_::..................................::_:E:::::.....................:-:::::_'E_..............I:_E::"::::....................:_':!:!:_

AUgUSt":' " Mea_ _:e[ecip|ta|._o'"": nl

:i _:_ii:iiil_ i

.... i iiiii._::...i::::i;_: ::::::::::::::::::::::::::::::::::::.............. >::_s:iN::::1

........ ==========================I__..: ::::::::::::::::::::::::::_! :_iii_iliii_iii:i!iiiii_ii_i_iiiilii%J!i

: :::::::::::::::::::::::::::::::::::::::::...............::::i..... ::_:_::::_.'>":::*::._:_i

0 20E: 40E

:_:::_:_._iiilii_iiiiii::::::i:iiiliiiiilil::_:_:_':.................iiiii!:._:_:_ii:iiiiiiiiiii!&._:_i:!_iiiiiiiiii:iii:iii_i_:ii_i:iiiii:::::_i:ii::_i'_..........====================================.......,..":_ii:iii_iiiii:i:::iiiiiiilii|:::_:_::::_i_:::_:_:_:_:_._:_:_::::_:::_.4."_:::..... =======================================:::::::::::::::::::::::::::::::::::::::::::::::::::....... ::@_i:_" ._f" =======================

_:_:_>..'_::_{_:_#.;_i:_:_:_i:_:_'_J_::_:_:_:_i::l:::::_:_I:_I:1::: ::::::::::::::::::::::::::=====================================:_:::i:>..:_i:_:_..::_:_..::_: .::_?:: .::&: :i:::::::::_::ii:...,:._::li_

• ...............................>::iNiii:: ::'_iiii!!_i!ii:!'_".::iI_:_iN.. "_:_:" .,_N::_....::%: '_::::::?::i:__:. :_::i::iI_.....:_iilii:Z_I::::_i_I.. ._ii:il_i:::

:_.. ._,:.:.. . :.._..::>..,........... .....

iiii:_i!:_....

i

Monsoon precipitation levelswere calculated for two drasti-

cally different environments.In one simulation (top), theland remained in place. In theother (bottom), the Iand-massess of Asia, Australia, andthe Pacific Islands were

replaced by ocean. When theland was removed from the

model, the Asian andAustralian monsoon regionsstill existed. However, mon-soons did not extend into the

southern China region.

62

..... _i.......1 i _i i ii ¸ i!i ii_iiii!i!i!i! ii i!i i̧

_i!!i!!!!!!!!i!!i!!i!i!ii!!!!!!!!!i!!!!!i!!!!!!!!!!!i!i!!!!i!!!!!!!!!!!!!!!!i!!!!!!!!i!!!!!i!!!i!i!!!i!i_!!_ _ !i_ !_!!iiii_iliiiiii!!ili!i_! _! _!i !!_! i!_ !!!!!!!!i!!!i!!!!!!!i!!!i!!!i!!!i!!!!!!!!!!!i!!!i!!!i!!!i!!!!!!!!!!i!i!!!!!!!!!!!!i!!!!!i!_i!ii_ii_ ii!!ii_!_i__i!!!_!_i!il_ii!ii!i_!!i!!!!!!!!!i!!!!!!!!!!!i!!!!!!!!!!ii!!!!!!!i!!!!!!!!!i!!!i!!!i!!!i!!!!!!!ii!!!!!!i!!i_!!!_!i!!!!!!!!!i!!!_!i!!!!i_!!!!!!!!!!i!i!!!!ii!!!i!!!_i!i!!!!!!!!i!!!i!!!i!i!!!!!!i!!!!!!!!!!i!!!i!!!!!!!!!!!!!!!!!!!i!i!!!ilii!!!!!ii!!i!!!!!!!!!!!!iiii!!i!!!!!!!!!i_ii_i!!!!i!!!!iii!!i!!!i!!!i!!!i!!!!illi!iii_!_i_i!i_!_!iIi!i!_i__!_ii_ii_ii_i_i!i!!i!if!!_ii!_!!!i!iii!i!!i!!_ii!_!!!!ii!!!!!!i!!!i!!!!!!!i!!!i!!!i!!i!!!!!!!!i!!!!!!!!!!!i!!!!!!!ii!!!!!!!iii!i!!!i!iii!_!i_!ii!i!!!!!!!i!!!ii!!!i!!i!!!!ili!iiiii!ii!:_:_:_:::_:::::_:::_:::_:_::_i:::_:_:_::_::::::::::_:_:_:::::::::::__:::_:::::_::::_:::_::_::::::::'_:

ii i i !iliii!iiiiiiiiiiiiiiiiiiiiiiiii!iiiii!iiii!iiii!iiiiiiiiii!iiiiiiiiiii!iiiiiiiiiliililiiiiiiiiiiiiii!ii!iiiiiiiiiiiiiiiiiiiiiiiiiiiii!i ____!ii__!i!iiiiiiiii!_!__i!i_i_!!i!ii_ii_i_i__iiiiii__ii__iiiiiii_!ii!iiiiiiii_iii_i__iiiiiiiiiiiiiiiiii_iiiii_iiiiiii_iiiiiiiiiii_iii_ii__iiiiiiiiiiiii_!iiii!i!_i!!iii

The research project also showed that the

monsoon process is similar to the flipping of a

light switch. When a switch is pushed gently

and persistently, it flips from one stable state to

another very quickly. Likewise, the ITCZ is

pushed toward the poles by the seasonal merid-

ional movement of the peak of the SST. This

movement is countered by another force that is

produced by the Earth's rotation. When the latter

force finally gives way, the ITCZ suddenly jumps

away from near the Equator, and monsoon onset

Occurs.

Chao's numerical experiments with an atmos-

pheric GCM have demonstrated the monsoon i

onset process. The circulation field associated

with the ITCZ determines the equatorial surface

winds that, in turn, determine SST near the

equator. Therefore, study of monsoons or the

ITCZ is highly relevant to the study of E1 Nifio.

The experiments used the God dard Earth

Observing System (GEOS) atmospheric GCM,

running on the NCCS Cray SV1 supercomputer.

63

• _i_ili_i!_!__iii__i_!_i/_¸ ii̧ ii_• • _: i:> _¸¸=_

!iiii!iiiiiiii

Research Profile- The Relationship Between Precipitation andSea Surface Temperature on Decadal Time Scales

Investigators:Siegfried Schubert, Max Suarez, and Philip Pegion, NASA Goddard Space Flight Center, DataAssimiilation Office

This research project examined the possibility of

a connection between SST and precipitation on

multiyear time scales. In particular, the research

focused on rainfall over the Great Plains region

of the United States during the summer months.

The researchers ran nine simulations of precipi-

tation, spanning the years 1930 through 1999,

using the NASA Seasonal-to-Interannual

Prediction Project (NSIPP) atmos _heric GCM.

They used 64 of the NCCS Cray T3E computer's

1,360 processors to conduct these 70-year runs.

The associated data was stored on the NCCS's

Sun El0000 UniTree storage system. "

A Cray support member from the NCCS helped

speed up the progress of this effort. He located a

default setting that prevented the model code

from running at full speed. He also showed the

researchers how to buffer the data output and

speed up run times even further. These efforts

reduced run times by 13 percent. Overall, the

Cray T3E calculated more than 225,000 days of

model simulation in only 700 days of computing

time, spread across the 64 processors devoted to

this project.

Each simulation run was forced with the same

set of observed SST measurements. However,

initial atmospheric conditions were different. If

the same SST data produced similar patterns of

precipitation over the Great Plains, even with

different atmospheric conditions, the results

would demonstrate a link of predictability

between SST and precipitation.

The nine runs showed different levels of rainfall;

therefore, overall this experiment did not indi-

cate a strong link between SST and precipitation.

However, the simulated results did share simi-

larities with actual rainfall records. Nearly all the

This graph lays out nine simulations of Great Plains rainfall. Theblack lines represent the individual simulation runs, and the greenline is the ensemble mean. The red and blue lines represent variousobservational estimates.

64

These color maps indicate the negative correlation between certain

meteorological measurements and calculations for precipitation

over the Great Plains region. The bottom map shows SST correla,

tion, and the top map shows the correlation for the altitude at

which air pressure reaches a measurement of 200 millibars (mb).

Because this project is concerned with the absence of precipitation,

negative correlations are mapped.

runs indicated dry conditions during the 1930s, .

followed bywet conditions in the next decade.

This result matches the infamous "Dust Bowl"

drought of the Great Depression era.

In contrast, only one of the nine simulations

replicated conditions similar to another major

Great Plains drought that took place in the 1950s.

In fact, some simulation runs produced multi,

year droughts even when no anomalies in SST

levels were present.

After they completed the simulation runs, the

researchers mapped the correlation between

Great Plains precipitation and SST. The correla-

tion revealed a decades-spanning SST pattern

across the Pacific Ocean that is linked to rainfall

variations in the Great Plains region. When the

Pacific pattern is in its warm phase, the Great

Plains region gets more precipitation than usual.

C onversel)¢ drought conditions tend to occur in

this area when the Pacific SST pattern is in a cold

phase. Several additional atmospheric GCM runs

confirmed this correlation.

Overall, the project results suggested that the

ability to predict Great Plains droughts depends

in part on the ability to predict the long-term

behavior of the Pacific SST pattern.

65