Atmospheric response to tropical denuding of vegetation

(1)

1352-2310(94)00291-6

A T M O S P H E R I C R E S P O N S E T O T R O P I C A L D E N U D I N G O F V E G E T A T I O N

R. C.

RAGHAVA

Centre for Atmospheric Sciences, Indian Instttute of Technology, Delhi, New Delhi 110016, India

and

K. LAVAL, R.

SADOURNY and J. POLCHER

Laboratoire de M6t6orologie Dynamique du CNRS, Ecole Normale Sup6rieure, 24 Rue Lhomond, 75231 Paris Cedex 05, France

(First received 3 February 1993 and retinal form 15 July 1994)

Abstract--Two simulations of atmospheric circulations dunng June, July and August 1988 have been made with LMD Atmospheric General Circulation Model using a classified vegetatmn global cover with and without the tropical vegetation separately. The initial conditmns prepared from ECMWF analysed data were used, while the Reynolds' monthly blended analysis, i e., the blend of m situ, AVHRR satellite and ice data, were taken to prescribe the sea surface temperatures. The global charts of mean monthly precipitation and assocmted velocRy potentmls at 200 and 850 mb have been compared and analysed for June, July and August 1988. The temporal evolutmns of precipitation averaged over a specific region of Indian summer monsoon de,ring Rs regime from onset to retreat have also been discussed. Consequently, a pronounced impact of tropical vegetation on the precipitation has been observed so as to characterise a forest as one of the local rain inducing agents. Moreover, the tropical vegetation appears to modulate the Indian summer monsoon also for the contributive preopitatlon over India.

Key word index: Precipitation, velocity potential, biosphere-atmosphere interaction, atmospheric general circulation.

1. INTRODUCTION

The past several decades have witnessed the scientific interest to a grea~L deal to understand the role of tropical forcing in the maintenance of general circulation in the atmosphere. Diagnostic investigations of Riehl (1965), Mare,be and Smagorinsky (1967), Mak (1969), Charney (1969) and Manabe et al. (1970) reveal the low latitude condensation process and lateral coupling with the higher latitude energy sources as the most important driving force in the Tropics. A theoretical analysis by Webster (1972) exhibits the dominance of latent heat release over sensible and radiational heating in the tropical atmosphere while assessing the role of orography, the release of latent heat, the effect of ocean - - continental contrast and longitudinal variation towards the production of the standing eddies in the tropical atmosphere. Manton (1985) found that the variations in the surface heat flux can induce horizontal pressure gradients in the convection layer, thereby, leading to the significant geostrophic flows such as cyclonic heat flows in the vicinity of local maxima in the surface heat flux.

The characteristic physiographic features of the Tropics in the form of arid regions of the Sahara,

the Middle East and the Indian subcontinent, the Himalayan Mountains and oceans straddling the equator predominate the precipitation distribution in the Tropics with its large longitudinal variation Ramage (1968) has discussed the importance of the continental relative precipitation maxima. The recent evolution of General Circulation Models (GCMs) has contributed in the visualisation of the scenarios that emerge from various experiments to understand the atmospheric responses to prescribed changes in the land surface boundary conditions. Whereas Charney et al. (1977) showed significant changes in the large- scale atmospheric conditions due to changes in the land surface albedo, Walker and Rowntree (1977), Shukla and Mintz (1982) demonstrated large feedback effects of changes of available soil moisture on the continental climate. The influence of land surface roughness on the convergence of horizontal water vapour transport in the atmospheric boundary layer of a G e M has been found by Sud et al. (1986).

Sellers et al. (1986), Dickinson et al. (1986) and Ducoudr6 et al. (1993) among others brief how the vegetation evolves an intricate phenomenon of interaction with the atmosphere through the viably prominent mechanisms of radiation absorption, bio- 1963

(2)

1964 R. C R A G H A V A et al

physical control of evapotranspiration and momentum transfer, etc., thereby, affecting the above- mentioned factors. It, obviously thus, arouses a curiosity to visualise the emergent scenario of the atmospheric general circulation over the globe with its tropical section stripped of any extant vegetation whatsoever to its extreme to investigate the role of tropical vegetation towards the maintenance of the atmospheric general circulaUon. It is with this motiva- tion that present studies have been conducted.

2. T H E M O D E L

A grid point Atmospheric General Circulation Model (hereinafter referred to as L M D AGCM) de- veloped at Laboratoire de Mrtrorologie Dynamique (LMD), Paris and described in various studies (Laval and Picon, 1986; Sadourny and Laval, 1984) has been invoked for the present study. Its characteristic features, however, are mentioned briefly as follows for a quick glance.

The three-dimensional domain of atmosphere has been so taken to consist of 50 latitudinal circles that the sines of their latitudes form the regular grid from north pole to south pole and the meridians to divide the latitudinal circles into 64 regular horizontal grids.

The equatorial grid size along eastwest direction is, thus, about 625 km and along northsouth, about 225 km. The vertical resolution has been taken to have 11 sigma layers comprising of four finer layers in the atmospheric boundary layer.

Whereas, the nonlinear transfer of enstrophy towards the subgrid scale is modelled with the ap- plication of bi-Laplacian operator on the potential enthalpy as well as on the rotational part of the flow, the suppression of gravity waves is affected by operat- ing upon the divergent part of the flow with a Lap- lacian. The dissipation process of vertical turbulent diffusion is parameterised based on the classical ap- proximations of vertical diffusion of wind, temperature and humidity due to Smagorinsky et al. (1965) and Deardorff (1966). The radiation parameterisation due to Fouquart and Bonnel (1980) has been adopted and the model's built-in cloud generation scheme has been described by Le Treut and Laval (1984).

The scheme called SECHIBA, i.e. Schrmatisation d'Echanges Hydrique a l'Interface Biosphere et Atmo- sphrre, for the treatment of flux exchanges of energy, mass and momentum at the biosphere and atmosphere interface as detailed by Ducoudr6 et al. (1993) forms one of the characteristic constituents of L M D AGCM. Briefly, it computes transpiration and interception loss for each type of canopies which may be present in one mesh. The aerodynamic and architec- tural resistances control interception loss. It accounts for the diffusion of evaporated water from inside the canopy and also for the variations of evapotranspiration within the canopy due to variations of wind, specific humidity and radiation. SECHIBA manages

the soil water content and calculates the bare soil evaporation. The soil moisture is kept in two reser- voirs with the upper one having variable depth to allow a rapid reaction of evaporation to a shower.

3. DATA

The atmospheric imtlal conditions of 1 June 1988 for horizontal wind, pressure, temperature and specific humidity were prepared from the E C M W F (Euro- pean Centre for Medium-range Weather Forecasting) observed analysis, while Reynolds' Monthly blended analysis, i.e., a blend of in situ, AVHRR satellite and ice data of sea surface temperatures as supplied by COLA (Centre for Ocean, Land and Atmosphere in- teractions) of the Umversity of Maryland, U.S.A. for the year 1988 were prescribed for integrating L M D AGCM over the months of June, July and August with surface conditions of albedo, soil moisture, sea ice, snow cover and surface pressure prepared also from E C M W F observed analysis.

The vegetation data characterlsed by the Leaf Area Index (LAI) and spatial distribution were extracted from the Atlas published by Matthews (1983) for eight categories of vegetaUon: bare soil, tundra, herbaceous plants, steppes, savanna and three types of forest, namely, deciduous, Semper Viren and humid tropical.

The global distribution of leaf area density defined as the sum of the products of the leaf area index and cover area normalised over the grid area of the eight types of vegetation available on the grid area has been shown in Fig. la while superficially denuded tropical strip can be perceived in Fig. lb.

4. T H E P E R F O R M A N C E O F L M D A G C M

The degree of simulative dependability on the L M D AGCM is confined a priori to its capability of capturing reasonably well some of the distinct prominent features of immensely contrastive Indian summer monsoons during the years 1988 and 1987. The simulative study of these features with a reasonable degree of success as discussed by Raghava et al. (1992) and mentioned briefly in this section renders it an incentive for the adoption of this model and the year 1988 for the present work

The model was integrated to simulate summer monsoons of 1988 and 1987 that were the years of copious rainfall and severe drought, respectively, with regards to Indian subcontinent. The data set of June 1 of the corresponding years as mentioned above and that for the year 1987 acquired also from the same source, were used as initial and surface conditions.

Analysing the simulations of monsoons for June, July and August during these years, it can be inferred that the model succeeds to simulate the wet summer monsoon in 1988 and dry season in 1987 over India. These features are associated with variations of circulation

(3)

gO

ot

50- . 30- 10-

Leaf Area Density for Summer control run

MAX: 8.00 MIN: 0.00 B =o "O a O ,< a)

• 180 -130 -80 -30 20 70 120 170 LONGITUDE oz ~z ~o 3z ~ 5z ~ 7a sz ~'ig. 1. (a) Global distribution of chmatological leaf area density, (b) same as (a) but with tropical belt from 30°N to 30°S turned bare soil. The contour interval is 1.0 each.

(4)

9O 70- 50- 30- 10- -10. -30. "50--

Leaf Area Density for Summer sensitivity run Ip I

MAX: S.O0 MIN: 0.00 Ib)

• 180 -130 -80 -30 20 70 120 LONGITUDE U 1.0 2.0 3.O ~ $,0

170 Fig. 1. (Continued)

(5)

Atmospheric response to tropical denuding of vegetation 1967 which appear similar to the E C M W F analysis. These

results seem to show that good and poor monsoons over India are predominantly associated with variations of oceanic temperature.

5. SENSITIVITY E X P E R I M E N T

This experimentation consists of two components:

control run and sensitivity run. The former is an integration of L M D AGCM over three months June, July and August initializing it with the data set of

1 June 1988 as described previously and prescribing the global vegetation distribution as shown in Fig. la.

For the latter, we ~tntegrate the model over the same period prescribing the same initial conditions and the vegetation distribution but with tropical vegetation from 30°N to 30'S turned bare soil as shown in Fig. lb.

6. RESULTS A N D D I S C U S S I O N

A set of panels for mean monthly precipitation and velocity potenUal fields at 200 and 850 mb characteristic of the associal.ed convective activities were pro- duced and analysed. For the global distributions of these fields, the figures identified by (a-c) have their causative association with the control run and sensitivity run, and their difference ( b - a), respectively.

While analysing the sensitivity run generated fields for the impact of tropical denuding of vegetation, the comparative terms' drying, wetting, excess, deficient, etc., have been taken m the sense relative to control run. The acronym TDV, hereinafter, refers to the Tropical Denuding of Vegetation.

6.1. Precipitation

During June, July and August, the monthly averaged precipitations are shown separately in Figs 2a-c through 4a-c, respectively. In June, when the Indian summer monsoon is in its onset phase, not so significant change in the monthly mean precipitation appears to occur over whole of the globe during the first month of the tropical denudation except over the region of south Pacific about 15°S at date line (Fig. 2c) where pronouncedly additional precipitation of as much as about 8 mm d-1 amounting to about 100% mcrease occurs due to TDV. It can be at- tributed to the pronounced change in divergent circulation over that region as illustrated in Fig. 6c.

An almost 50% drying effect now on the Ama- zonian region evolves in July apparently because of a shift of the precipitation to its adjoining Atlantic oceanic region (Fig. 3c). A similar trend becomes per- ceptible about Bay of Bengal but here precipitated water transport happens to be meridional, i.e., from the Indian ocean to the northeast Indian region. In- terestingly, the precipitation scenario over western India features the split of the region about 15°N with

its northern part witnessing deficient rain, while the southern an excess. It, thus, displays TDV as to inhibit the monsoonal precipitation from proceeding north- ward over northwest India. The most significant drying to the extent of about 50% occurs over Guatemala and over Taiwan. A similar impact of TDV becomes conspicuous by deficit precipitation to the tune of about 50% over maritime continent of southeast Asia also. Characteristically, the occurrence of the quantitative and distributive changes in the precipitation is, however, confined to the entire tropical belt undergoing the physiographic change due to TDV. Interestingly, a spillover of this change is exhibited to occur as widely as to cover the temperate Pacific also. It appears, probably due to the coupling of low latitude condensation process with high latitude energy source as revealed by Charney (1969) and Manabe et al. (1970).

In August, the drying, though as moderate as about 25%, is observed over Amazonian region, Sudanese region and whole Indian subcontinent except Himalayan region where significantly excess precipitation by about 50% is noticed (Fig. 4c). In the south Pacific, the pattern of excess and deficient precipitation reverses over most of regions with the transition of Indian summer monsoon from its modulation to retreat phase. Comparing the Figs. 4a, b it is evident that prominent characteristics of precipitation distribution are so featured as to be suggestwe of the predominance of ocean and land mass contrast over atmospheric general circulation.

Notably, the TDV induced diminutive precipitation figures most prominently over the regions of dense forests, Le., the Amazonia, Congo basin and the maritime continent of southeast Asia during the modulation regime of Indian summer monsoon, i.e., in July. The TDV, however, originates a centre of an excess precipitation apparently over Ethiopian high lands with the retreat of Indian summer monsoon in August. Contrary to the July trend, TDV is found to enhance the precipitation by over 50% over southeast China and the China sea. This pattern of excess precipitation, though moderate to the extent of about 10%, extends across the entire maritime continent of southeast Asia also.

The temporal evolution of precipitation averaged over the region bounded by the parallels of 1.15°N and 38.32°N, and the meridians of 56.06°E and 98.44°E as shown in Fig. lb by the inset rectangle, are illustrated in Figs. 5a-c during June, July and August 1988, respectively. Interestingly, the general trend of precipitation induced by TDV is manifested to be in consonance with the control run. The extrems are, however, exhibited so distributed that they appear to be in opposite phase during the modulation and retreat phases of monsoon in July and August, respectively. The impact of TDV on the precipitation spatially averaged over the monsoon region in ques- tion during onset phase of monsoon in June is witnessed as the furtherance of drying and wetting trends

(6)

Smoothed Precipitation (mm day in Jun, 1988 control run

D -1

F~,,X: 52.26 ~IN: 0.00 • ~.=_-_ ~.v ~__~__z.o~~ 1.oX_Zz.o~_____ --- - o---~. _'-4 -90

'a) -180 -150 -120 -90 -60 -30 0 30 60 90 120 1 50 180 LONGITUDE

Fig. 2. (a) The monthly averaged precipitation field (mm d- 1) for June dunng 1988 generated from control run, (b) same as (a) but generated from sensitivity run, (c) the difference field (b - a) m mind- 1. The contour intervals for (a) and (b) are 1, 2, 4, 8, 12, etc. mm d- 1 and for (c) they are - 12, - 8, - 4, - 2, - 1, 0, 1, 2, 4, 8, 12, etc. mm d- 1.

:I: < >.

(7)

,...I

90 ,

Smoothed Precipitation (mm day -t) in Jun, 1988 sensitivity run E

"' 4.0 &'O~r ' <

,o-_-,of ~,oi- ~-,o~ ~,o--~,o,__3

MAX: 56.80 MIN: 0.00 > o 0 ,-t 0 o 0

(b)

-90 I -180

I I ... I I I I I I I I I

-150 -120 -90 -60 -30 0 30 60 90 120 150 LONGITUDE

Fig. 2. (Continued)

180

(8)

Smoothed Precipitation (mm day -1) in Jun, Difference Field (sensitivity-control) 1988

--4 C,

r,=,l

.< (c)

~9

o.Ov.r -o'~

P

o. o o .o~ "

~'~0.0

0 ~'J

o" "o "--'o _,,..r-0.0 "o

-90 I I I I I I I -180 -150 -120 -90 -60 -30 0

LONGITUDE

0.0- I 30 I I 60 90 I 120

0 I 150

0 0 180

MAX: 9.74 MIN: -6.76

> F]g 2.

(Continued)

(9)

i,ml (a)

NN--

SmooLhed PrecipiLaLion (mm day -~) in Jul, 1988 conLrol run

<> 0_._.._.i

.o~ .-)'--4.o~

%

L.O J "~,0 ~.0- -9o I ~ I i -180 -150 -120 -90

o ~ ~'--',oF, oJ <o~

~4 1.0 "" \~--'¢,"~ . _ ';,u~--T3B----/c-5.- t.u ~ 1 n I I I I I' I I -60 -30 0 30 60 90 120

LONGITUDE

Fig. 3. (a-c) Same as Fig. 2a-c, respectively, but for July.

I 150

MAX: 42.62 MIN: 0.00 180

El o= =. S o == I=, o ==

(10)

Smoothed Precipitation (mm day -t) in Jul, 1988 sensitivity run

MAX" 45.32 VIIN" 0 00

70

-,4

ib) -180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180 LONGITUDE

;> .< ;>

(11)

Cl [,-, E-,

(c)

-90

Smoothed Precipitation (mm day -t) in Jul, Difference Field (sensitivity-control) 1988 -180

~. _ MIN: -12.25 o "0 • ~ , . ~o.o O~o.o..~"0. ° .o

"0 CO.o ~ 0.0 0.0 0.0 I I 90 120

-~o.o----~~o.o--/ \~ \o.o

i I I I I I I I I -150 "120 -90 -60 -30 0 30 60

LONGITUDE

Fig. 3. (Contin~

I

150 180

=o ¢3 0 S 0 F," =1 (Ira o 0

(12)

D

90 60- 30-

SmooLhed Precipitation (mm day-') in Aug, 1988 control run

-2.0 ",,,._~ 1.0 -.---..--- 1.0 .-~-- 1.0 O ¢/ -- "o I r ~'3 _ ~r ~.... ,.~~~..,~

o

"0

.~<~ -~ ~,;-~ ,,.or-""- ",.o,, -

o "4.c ~-.---~ ~- 4.~ 4.o-~ 4.o ~ 4- .0

MAX: MIN: 44.22 0.00 > > < >

(a)

-90 I -180

I I I I I I I I I I

-1 50 -120 -90 -60 -30 0 30 60 90 120 LONGITUDE

Fig. 4. (a-c) Same as Fig. 2a-c, respectwely, but for August.

I 150 180

(13)

Smoothed Precipitation (mm day -~) in Aug, 1988 sensitivity run

E.4 l, il E.~ .,¢

.o- - 1.o ~I. 0

o .0~"

~o

, ~.0"-~ ~ n 2.0"- ,~ "0

• .o/"-1:

-90 J I I I I I -180 -150 -120 -90 -60 -30

"0~4.0~ .n~ 4.0 ~~--'.." ~ 4.0 ~ I I I

0 30 60 LONGITUDE

Fig. 4. (Continued)

2.O

7 4.0 ~.__ 4.0 I I I

90 120 150

180

MAX: 54.11 MIN: 0.00 ;> ~t C~ ,<

(14)

Cl D

(c)

Smoothed Precipitation (mm day- ) in Aug, Difference Field (sensitivity-control)

°~>,o o

1988

0.0 O~

0 "0 J J.O -...~"

0 • °_~'-~ 0.0 7 "9°1 i i i , i i i -180 -1 50 -120 -90 -60 -30 0 30

LONGITUDE

O.o I 6O I 9O

o o" I 120

~0.0 I 150

o o

MAX." MIN: 180

16.31 -9.52 >

(15)

Temporal Evolutions in Jun, 1988

' " =~ ' ' ' " ' ' '"' ~ contro] ... sensitivity

10.0 5 / -... "~ 7.5

"

""-'' 7 5.0 2.5

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 (a)

Days

Fig. 5. (a) The temporal evolution of precipitation in the umt mm d- ~ averaged over the specific monsoon region bounded by a rectangle reset in Fig. la dunng June 1988. Sohd and dotted lines refer to the control and sensitivity runs, respectavely, (b) same as (a) but for July, (e) same as (a) but for August

O Q o O

(16)

t- O .u .m

O_

0 G) IL_

0.

(b)

Temporal Evolutions in Jul, 1988 10.0 7.5 5.0- 2.5-

III III I I II IIIII II I II IIIII IIIIIIIIIIIIIIIII IIIIIIIII IIIII II I 2 3 4 5 6 ? 8 9 !0 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Days

Fig. 5. (Continued)

control ... sensitivity

(17)

t- O

t~ Q.

°~ O 1,1

13_ (c)

10.0-

Temporal Evolutions in Aug, 1988 7.5 5.0

°° °o i O°o • °o° . ! °~°°oo° °%°

2.5-

III Illl III Illlll II II II Ill Illllllllllllllll IIlllllllllllllll 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Days

control ... sensitivity 0 0

(18)

1980 R.C. RAGHAVA et al.

by as much as 20% in the latter half of June. However, the precipitation profile follows coherently the monthly trend of monsoon consistent with the reahsm during its summer hierarchy over Indian subcontinent with the most of precipitation occurring in July and August. It, conclusively, appears to signify a forest as a potential rain inducing local agent. An mference can also be made that the tropical vegetation contrib- utes significantly to the precipitation over India through modulation of Indian summer monsoon.

6.2. Veloczty potential at 200 mb

The mean monthly velocity potentials at 200 mb in June have been shown in Figs. 6a-c. The planetary scale development of Asian summer monsoon can easily be noticed in Fig. 6a. It comprises primarily a divergent centre of outflows established north of Bay of Bengal. This pattern resembles with the obser- vation of Krishnamurti et al. (1990) except its axis is somewhat rotated anttclockwise about its centre over central China. Another prominent pattern of convergent inflows across south America also figures in Fig. 6a. Its comparison with the pattern in the study by Krishnamurti et al. (1990), however, indicates downward tilt by nearly 3 ° of the axis of convergence about ~ts western extremity. It, thus, evolves the well- known summer monsoonal composite system of Walker and Hadley type of overturning with their respective transpacific and cross-equatorial transat- lantic major axes.

The impact of denuded tropical global belt between 30°N and 30°S appears to intensify the descending branch of Hadley type of circulation (Fig. 6b) due probably to the ehmination of predominant Ama- zonian and Congo basin forests. The generation of divergent outflow centre in the subtropical Pacific and a secondary circulation across south of Africa can be noticed in Fig. 6c along the Tropic of Capricorn.

The TDV-mduced evolution of positive and negative centres in the upper air being significant of divergence and convergence leads to enhanced precipitation and evaporation, respectively, over the underlying areas.

The reduced precipitation over the regions of dense forests because of TDV is reflected in June's scenario (Fig. 2c).

As the northern summer advances through July, the monsoonal component of the pattern over the Tibetan plateau across China persists with further pronounced intensification. So does the South Ameri- can pattern as well, as shown in Fig. 7a. It, thus, demonstrates further deepening of Walker and Had- ley type of overturning. This trend is in agreement fairly well with the observational analysis of Krish- namurti et al. (1990, Fig. 5d). The TDV-indueed differential scenario for July mean at 200 mb as shown in Fig. 7c emerges with the centres of divergent outflows and convergent inflows over the south Pacific and southeast Indian ocean addressing thereby to the secondary Hadley types of overturning over Australia and the south Pacific. The ascending branches of these

circulations appear to have been estabhshed over oceanic regions west of the tropical dense forests.

Interestingly, an adverse effect of TDV on Asmn summer monsoon is noticed as a subsidence appears to occur extensively m the modulation phase of monsoon in July. It is reflected as negaUvely anomalous precipitation over northern India. The adjoinmg positive anomalies over Himalayan region and over pen- insular India in the south appear to exhibit assocmted orography and oceaniclty to be more dominant in the absence of tropical vegetation.

By August, TDV-induced circulatory change as- sumes the planetary scale, probably, w~th the broadening, merging and intensification of divergent and convergent centres of flows as it has been depicted in Fig. 8c. Interestingly, however, local circulations associated with the dense tropical forests of Ama- zonia, Congo basin and maritime continent of southeast Asia also emerge, thus, significant to have their grip on the local atmospheric circulation and take away their shares of local precipitation with them (Fig. 4c), if removed. The control mean monthly velocity potential (Fig. 8a) compares structurally fairly well with the analysis of Krishnamurti et al. (1990, Fig. 5e) for August. It, however, shows a shift of centres of diverging and converging flows and their general pattern rather intensified. An intensified divergent outflows and longitudinal elongation of major axis of convergent inflows appear to occur consequently upon TDV as illustrated in Fig. 8b.

6.3. Velocity potential at 850 mb

In accordance with the structural description of Walker and Hadley circulations necessary reverse type of overturnings at 850 mb as compared to those at 200 mb over corresponding areas are simulated in June as exhibited in Fig. 9a. It shows a major ascent over summer monsoon region and an agreement with the observations by Krishnamurti et al. (1990, Fig. 6a). With the denudation of tropical belt, there appears an emergence of secondary centres of convergent inflows over summer monsoon region, Indian ocean and central Pacific. The interesting aspect of TDV-indueed subsidence over the Indian subcontinent appears to deepen further, down to the level below 850 mb, thereby, reducing precipitation over India (Fig. 2c). The TDV-induced divergent outflows over the Amazonian region (Fig. 9c) appear to signify the initiation of precipitation deficit due probably to the removal of predominant Amazonian forest

During July, under the influence of TDV, the divergent circulations appear to get more localized over the areas of prominent forests consequently further de- clining the precipitation over corresponding areas. It is reflected in the July precipitation chart (Fig. 3c) as negative anomalies over there. To the contrary, the adjoining oceanic regions witness more pronounced effects leading to enhanced precipitation (Fig. 10c).

During August, a regime slgmficant of retreat of Indian summer monsoon, TDV-induced subsidence

(19)

E~

Velocity Potential (108 m ~s-t) at 200 hpa in Jun, 1988 control run

e,~J% ~ ~ " MIN:

. ., ' C

0

"o ,o

\ .."I"

\

.__6-.- "/ "r /

13.32 -9.54

-9o I

i i i i i i i i

-180

-150 -120 -90 -60 -30

0 30 60

(a) LONGITUDE

\

-'~-2 --S""---2 I I I

90 120 150 180

Fig. 6. (a) The monthly averaged field of velocity potential (106 m 2 s- 1) at 200 mb for June, 1988 generated from control run, (b) same as (a) but generated from sensitivity run, (c) the difference field (b - a) in the unit 10 ~ m 2 s- 1. The contour interval is 2 x 10 e m= s- ~ for (a) and (b), and 0.5 x 106 m 2 s- ~ for (c).

.> @ O t% O

(20)

1982 R. C RAGHAVA et al.

ID

0

,m,-i V

!t

e ~ ID ~

.c 7

.-

/ o_./

u o,, ;

~! I ^{t t, o,.} t'

I::I

Q;

O,,, 0

* l - - I

c,;

0

o

li" #

~, / /

/=...S"/'>'

\ o -" / ~ /~-", \

I I I I I

0 0 0 0 0

a a n ~ v q

0 0

0

C'Q 0

!

0

"7

. . - . . ,

(21)

Velocity Potential (10" m ~s -~) at 200 hpa in Jun, Difference Field (sensitivity-control) 90 I I _ l.,~ ~ "-,'-~- '~

0.0 ~ --- ~

ogp.~o.o

/.. .0

\\ ,";: ~ \ "~ f"".o ~..jo~ ;~-,,~..a~'ht//r / "o s--.__-_ ~'~" /" o,,~°'~"--"o °"~o~

-- • "~.8'----...___0'.6 - / -- 'u~-- "d'

-°°-1 ,~ ~o.o~-r~-~o~ °" ¢ _ • / -180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

(c)

LONGITUDE

1988

MAX: 3.76 MIN: -2.06 _> 'O "¢1 O ¢1 I= P,i 0¢1 ¢1 Fig. 6. (Contmued)

(22)

Velocity Potential (100 m ~ s-') at 200 hpa in Jul, 1988 control run

F,,

(a) -90 "1

-180

/

~f2 - :.=.7

2 2~ J'~h

/

I -150

I -10 ~0 ~ /~

_811./

I -120

J I I I I -90 -60 -30 0 30

LONGITUDE

" J

i

. 4 4---. ll; I I I I 60 90 120 150

MAX: 19.61 MIN: -14.86 180

> > > Fig. 7. (a-c) Same as 6a-c, respectively, but for July.

(23)

Velocity Potential (i0 e m zs-t) at 200 hpa in Jul, 1988 sensitivity run

~-o-'t< ~- L'--,-~(~.,.~-~ - -~tt.,l.~-j

/ - , i / ..J,-~'~.t, ~. -~--a\/" ,,,' / / ~,L..~ ~" ~.. .%J / ,/VJ,: \.'.,Td, o----:L...'./ /, It =. ' .~ I" /" /'% I¢---~----~.- ~-~ / ~ % "- "-

-60

-90 -180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180 (b) LONGITUDE

MAX: 18.87 MIN: -14.06 Fig. 7. (Continued)

=o =t 1¢

E

I=1 I. 2, ^{0 =1}

(24)

Velocity Potential (10 6 m2s-l) at 200 hpa in Jul, 1988 Difference Field (sensitivity-control)

-- I -180 -I;0 -l&I 40 -80 30 SO 90 120 150 180 (c) Fig. 7. (Continued) 3.46 -2.51

(25)

Velocity Potential (I08 m ~s- ,) at 200 hpa in Aug, 1988 control run 90 I_ ~i~ ~__~, '-I

4.,. X ~~.~]~Iz,~w__,~..~__ --~" -2---...~1111. ~ .. .4L.,---~-'4 " 4 MAX: 23.95 ~~. ~ MIN:-13.18 I::I [-, I,,=4 .<

(a)

..2, -.--¢~ -6

i"~-'~f.~/~ 8 "~" " 7 ,j \'.... ,~, 1 r

-180 -150 -120

0

%\ ~-i0---7 r //

, • "4 ~ -4~---.---_4 -4"---'--- - -4 T ~1 I I -90 -60 -30 0 30 60

LONGITUDE

I I 90 120

V -2:-~( I 150 180

o O = O o~ < ~a 0 Fig. 8. (a-c) Same as 6a--c, respectively, but for August.

oo -4

(26)

1988 R C R A G H A V A et al.

0

v

°1~..,4

0 0

0w,,~

0 0 C~

O o O o o

f f a r l ~ i ~ v ~

O

O ¢O

T -

O o 4

O

8

O

o o 4

"7

O t O

"7

O t ~

"7 O

o6

. Q

(27)

Velocity Potential (108 m 2s- ~) at 200 hpa in Aug, Difference Field (sensitivity-control)

90 , ~...-- 0.o "'. ... ... u'v ~ u'u-'-"- O.

l

-'---0.5 ~ 0.5 ~.~ ~~.rL--- --- ~ --z,~ _ ~ ~qmmt,5-- -,if'J5 .... ,,1111' "0

D

E-, .<

(c)

-31

0 •

2.6

#

-~ o.o T 1

~ 0.5 ___.~,O.--- --

T__~---°.°~

-180 -150 -120 -90 -60 -30

w* "0 "0.5 ~ ~I.0 -0.5 ---_0.5 1 I [ T 1 I 0

LONGITUDE

30 60 90 120 150 180

1988

MAX: 5.22 MIN: -5.70 > o "0 2. 0 :3 .=. 0¢1 o .< 0 ~o ^{oo ~D}

(28)

Velocity Potential (108 rn z s- t) at 850 hpa in Jun, 1988 control run 90

/ -2,- "-u -=; -2 - -~ ... ;¢ ----"--2 ,,-. ' MAX: 7.55

-30 w ~ z'~ -60-1 ~ ~-~ I -90 -180 - 150 - 120 -90 -60 -30 0 30 60 90 120 150 180 (a) LONGITUDE

Fig. 9. (a-c) Same as 6a-c, respectively, but at 850 rob.

> C~ >. .< >

(29)

Velocity Potential (10 8 m:s:') at BSO hpa in Jun, 19BB sensitivity run 901 "~-" ..._.--~-'=~

i'-'= --~; -'~ -'= ~---2"~ "" '~ I MAX: 7.61 ...,--- ---J,~ ~ MIN:-12.75 | -.-. 4 ¢1~

?. ,6./ ~ "--'J ,~ "~-~

/

|

I, (b)

-9o I

-180

~-2 I -150

~-Z

-~J "'~'--~-Z --2- ~J I I I I I I I I I - 120 -90 -60 -30 0 30 60 90 120 150

LONGITUDE

180 Fig. 9. (Continued)

;> =o "O oo "0 0 B, 0

E

OQ 0

(30)

Velocity Potential (10 e m Zs- ~) at 850 hpa in Jun, 1988 Difference Field (sensitivity-con_.~_ol) ____ _ ~°.°~~--;0 -0.0"---~-1 ~.

",m,m _0.~_ 4 ~ _,-~ "O.d MIN: "" Yo ,.," "- 00~

I ~-o:~ o. -o.o-J - "' ~-°.=c-- I -90 I~""'"'"i

"" I "" InnJ I I I I I I I

1.97 -1.82

-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180

,c) LONGITUDE

r < >- Fig. 9. (Continued)

(31)

Atmospheric response to tropical denuding of vegetation 1993

e~l , q -

"7

~ _ __

.~ ~ o - "/ ~, ~ _

.~= ' ~ ~ 1 ~ ~ / ° ~ _

; / ~ ~\I k; 1 7' o

0 -

!~.%~.x~ 1' I

g. , /

a a n £ i £ v q

"i

l -

I=

~2 d~

i

-g

v

(32)

Velocity Potential (10 o m zs-t) at 850 hpa in Jul, 1988 sensitivity run 90 ,,~ -" -" -" -" -" -" --"e-- ~ ~ "qPE") "--"~ ~'~'~'~'~'~'~'~'~~-" MIN:MAX: 8.26

-14.78

, ~ ~o-~~Ro ~ o ~_____6 -e.~L~--- ~

_,,.. =. ~ , .. ~-,,~..~ ___

~-,,~

./ -- ,,.~,'~ ¢~-~4\. "~-->'¢.',,v~ I 7" ¢"~<---.~ "Zn

>o~ I; o- ~

e.,....__ ~ I

! -L ><<", -90 -180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180 (b) LONGITUDE

pc >.

(33)

Velocity Potential (10 8 m zs -t) at 850 hpa in Jul, Difference Field (sensitivity-control)

90 . ~0.0"~0.0---0.0--- 9 0

0 -,q.

y

o I "o o • o ~ 7" " 0 L o I

&

~0.0- 0.0 / 0.-'0

1988

MAX: MIN: o C~ o"

2.38 -1.94 B =o =- '8 5" 0

=_.

I= == 0 (c) -180

I I I I I - 150 - 120 -90 -60 -30 I I I I 0 30 60 90

LONGITUDE

I 120

I 150 180

(34)

Velocity Potential (108 m 2s- t) at 850 hpa in Aug, 1988 control run ~o i__ ~ ..,_~ ~----~ -~ -" --~l'x ~.o~

_ ~-~==,,,~ MIN: -13.49

o~

I -90 I

-180

(a)

Q I -150

(0 ; ,L

j

k.~4. ~0

/

2~- 2. I I I I I I I I - 120 -90 -60 -30 0 30 60 90

LONGITUDE

O~ I I 120 150 180

> > < >

Fig. ll (a-c) Same as 9a-c, respectively, but for August.

(35)

D

,-1 (b)

Velocity Potential (108 m 2 s- ~) at 850 hpa in Aug, 1988 sensitivity run

o -90 I I I I I I I I I I I -180 -150 -120 -90 -60 -30 0 30 60 90 120 150

LONGITUDE

MAX: 9.29 MIN: -13.95 180 Fig. 11. (Continued)

;> O ==r" .% ^{O ==} l= I= o O

(36)

Velocity Potential (108 m zs-') at 850 hpa in Aug, Difference Field (sensitivity-control) 1988

ca

D

,=I

(c)

o\

1.0 - 0.0 -- 0.0 - 0.0 ~ 0 0 ,m~ * - ~ O q

~'I' g"

<> 0

0 I0

~'0

9 0 "d' ~.6~ ~6~

~O.o I r T I I" T -180 -150 -120 -90 -60 -30

D Q~ 0.5. "" J'.O~ 1.1 ~0.6-

~ P--r~ 0 LONGITUDE

30 60 90 120 150 180

MAX: 2.63 MIN: -2.48 ;> :r > ,< >.

(37)

Atmospheric response to tropical denuding of vegetation 1999 over Afro-Asian region appears to be shallow with an

immense and extensive convergence at 200 mb and divergence at 850 rob, thereby, a deficit precipitation appears to persist. It is so also over Amazonian region as illustrated in Fig. 1 lc. A panoramic view of TDV- induced evolutions (Figs 9-11c) appears to suggest that subsidence persists throughout the monsoon season over Afro-Asian region from June to August, thus, leading to dry conditions. The effect is modulated more pronouncedly over the Congo basin in July and moves over to the Indian region in August with the retreat of Indian summer monsoon. A comparative look at the control mean monthly divergence fields at 850mb, (Figs 9-11a), reveals that the Indian monsoon is so modulated as to intensify ascending branch of Walker type of overturning in July over India, a well-known realistic climatic feature of the atmospheric general circulation. In addition, retreat of In- dian summer mon,;oon in August is very well repre- sented by an eastward shift of major axis of ascent to lie over the China sea. The overall spatial distribution of major centres of divergence and convergence in Figs 9-11b has been found to remain indifferent to TDV. However, it brings about configurative and intensity changes of these centres, thus, still showing the dominance of ocean-continent contrast.

Interestingly, the precipitation and convective responses of tropical dense forests as displayed in this study, appear to be in consonance with a theoretical analysis by Webster (1972). Webster (1972) found the dominance of latent heat release for the production of standing eddies in the tropical atmosphere. The tropical dense forests are evidently the strong source of latent heat release through evapotranspiration over the continents.

7. CONCLUSIONS

This paper addresses the role of vegetation on a very prominent section of the globe, i.e., Tropics from 30°N to 30°S which is, per se, a well-recognised primary energy source driving the atmospheric circulation. Given the climatic physiographic conditions on the earth surface and the hypothetical case of mere tropical denudation of vegetation in its extreme, the resultant differential features of the monthly averaged fields of precipitation point to the sweeping away of precipitation to the extent of about 50% from the areas of prominent forest reserves with the vanishing of its vegetation throughout the summer from June to August. Conversely, the excess precipttation of about 100% occurs mostly over oceanic and orographically strong Himalayan region. Moreover, the magnitudes of these changes in precipitation appear to be the most prominent during the period of modulation of the Indian summer monsoon. It, thus, establishes and endorses the legendary folklore of the forest behaving as an attractor of rain. In addition, the tropical veg-

etation appears to play a predominant role towards the modulation of Indian summer monsoon.

It can also be further inferred from the present study that realistic treatment of biosphere, especially its flora component, is essential for the realisation of climatic studies in particular and enhancement of long-range weather and climate prognostic skill of a general circulation model in general.

Acknowledgements--The computational time was granted by CCVR (Centre de Calcul Vectoriel pour la Recherche), France for this study The co-author (RCR) gratefully acknowledges his scientific visits to LMD financially sponsored by CNRS (Centre National de la Recherche Scientlfique) of France and Indo-French Centre for the Pro- motion of Advanced Research, New Delhi, India under the Project No. 711-1 for this study. The very useful comments from unknown reviewers for this paper are gratefully ac- knowledged.

REFERENCES

Charney J. G. (1969) The Intertroplcal Convergence Zone and the Hadley Circulation of the Atmosphere. Proc

WMO/IUGG Syrup on Numerical Weather Predictwn, Tokyo, Japan, 26 November-4 December 1968, pp. III- 73-III-79. Japan Meteorological Agency, Tokyo.

Charney J. G, Quirk W. J., Chow S. H. and Kornfield J.

(1977) A comparative study of the effects of albedo change on drought in semi-arid regions J. atmos SeL 34, 1366-1385.

Deardorff J. W. (1966) The counter-gradient heat flux in the lower atmosphere and in the laboratory. J. atmos. Sci. 23, 503-506.

Dickinson R. E., Henderson-Sellers A., Kennedy P. J. and Wilson M. F. (1986) Biosphere-Atmosphere Transfer Scheme (BATS) for the NCAR Commumty Chmate Model, NCAR Tech. Note, NCAR/TN-275 + STR, 69 pp.

Ducoudr6 N. I., Laval K. and Perner A. (1993) SECHIBA, a new set of parametenzatlons of the hydrologic exchanges at the land-atmosphere interface w~thln the LMD Atmospheric General Circulation Model. J. Chin. 6, 248-273.

Fouquart Y. and Bonnel B (1980) Computations of solar heating of the earth's atmosphere. A new parameter- ization. Beztr. Phys. Atmos. 53, 35-62.

Knshnamurti T. N., Bedi H. S. and Subramaniam M. (1990) The summer monsoon of 1988. Meteor. Atmos. Phys. 42, 19-37.

Laval K. and Picon L. (1986) Effect of a change of the surface albedo of the sahel on chmate. J atmos. Sc~ 43, 2418-2429.

Le Treut H. and Laval K. (1984) The importance of cloud radiation interaction for the simulation of chmate. In New Perspect,ves in Climate Modelling (edited by Berger A.), pp. 199-222, Elsevier, Amsterdam.

Mak Man-Kin (1969) Laterally driven stochastic motions in the Tropics. d. atmos. Sci. 26, 41-64.

Manabe S. and Smagormsky J. (1967) Simulated chmatology of a general circulation model with a hydrologic cycle II.

Analysis of the tropical atmosphere. Mon. Weath. Rev. 95, 155-169.

Manabe S., Holloway Jr. J L. and Stone H. M. (1970) Tropical circulation in a Ume-integration of a global model of the atmosphere. J. atmos Set. 27, 580-613.

Manton J. (1985) Some effects of convection on geostrophic flow. Q. J. R. Meteor. Soc 111, 173-182.

(38)

2000 R.C. RAGHAVA et al.

Matthews E. (1983) Global vegetation and land-use, new high resolution databases for climate stuches, d. Clim.

AppL Meteor 22, 474-487

Raghava R. C., Laval K. and Sadourny R (1992) Compan- son of the 1988 and 1987 summer monsoons as simulated by LMD AGCM. WCRP-68, WMO/TD-No. 470, pp. 2.

137-2.144.

Ramage C. (1968) Role of a tropical maritime continent m the atmospheric circulation Mon. Weath. Rev. 96, 365-370.

Riehl H. (1965) Varying structure of waves m the easterhes.

Proc. Int. Symp. of Large Scale Atmospheric Processes, Moscow, U.S.S.R., pp. 411-416. Izdatvo Nauka, Moscow.

Sadourny R. and Laval K. (1984) January and July per- formance of LMD general ¢arculation model. In New Perspective m Chmate Modellino (edited by Berger A.), pp. 173-198. Elsevier, Amsterdam.

Sellers P. J., Mintz Y., Sud Y. C. and Dalcher A. (1986) A sunple biosphere model (SiB) for use within general circulation models. J. atmos. Scz 43, 505-531.

Shukla J and Mmtz Y. (1982) Influence of land surface evaporation on the earth's chmate. Science 215, 1498-1501.

Smagormsky J., Manabe S and Holloway J L. (1965) Re- suits from a nine-level general circulation model of the atmosphere. Mon. Weath. Rev. 93, 727-768.

Sud Y. C., Shukla J. and Mintz Y. (1986) Influence of land surface roughness on atmosphenc ctrculation and rainfall.

GCM sensmvity expenment. Extended abstracts, Third Conf. on Chmate and Symposmm on Contemporary Cli- mate 1800-2100, 8-11 January 1985, Los Angeles, Cahfor- ma, 5.2, pp. 93-94.

Walker J. M. and Rowntree P. R. (1977) The effect of sod mmsture on circulation and rainfall m a tropical model Q. J. R. Meteor. Soc 103, 29-46.

Webster P. J (1972) Response of the tropical atmosphere to local steady forcing. Mon. Weath. Rev. 100, 518-541

Atmospheric response to tropical denuding of vegetation