Theor. Appl. Climatoi. 48, 179 185 (1994)

Theoretical and Ap21ied Climatology © Springer-Verlag 1994 Printed in Austria 551.555

Department of Geophysics, Banaras Hindu University, Varanasi, India 2 Metallurgical and Engineering Consultants (l) Ltd, Ranchi, India

Atmospheric Dispersal Capacity over North India D. V. Viswanadham ~ and V. V. S. N. Pinaka Pani 2 With 9 Figures Received June 29, 1993

Summary In this paper the dispersal capacity of the atmosphere has been determined over North India by studying spatial variations of inversions, isothermals, lapse conditions, mixing heights and ventilation coefficients. Seasonal variations of these parameters have also been studied. The western part of North India is best for good dispersal of pollutants while the eastern part is poor at all time of year.

there are about eleven stations, although there are wide gaps especially in the central parts of North India. In the present paper the spatial variation of surface-based and elevated isothermals, inversions and lapse conditions for every 50 hPa up to 500 hPa, MH and VC is studied over North India, which has vast potential for industrial development.

1. Introduction

Prevailing meteorological conditions determine the capacity of the atmosphere to disperse and dilute the pollutants emitted into it from various sources. The vertical thermal structure and horizontal wind play a dominant role in the atmospheric dispersal capacity for any given region. The thermal structure can be obtained directly from radiosonde stations. In addition, it is essential to obtain the vertical and horizontal extent of mixing in the form of Mixing height (MH) and Ventilation coefficient (VC). A spatial variation of all these parameters would aid planners in identifying favourable and unfavourable regions for good dispersal of pollutants. Earlier studies were due to Holzworth (1964, 1967, 1972), Szepsesi (1967), Padmanabhamurty and Mandal (1976, 1980), Vittal Murty et al. (1980 a, b) and Viswanadham and Santosh (1989). Although, studies were carried out in India the data were rather sparse and additional stations would give a more realistic picture. In North India itself

2. Materials and Methods

The radiosonde data at 0000 G M T were collected for a five-year period (1981-1985) for eleven stations in North India (Fig. 1) for the months of January, April, July and October, which are representative of the winter, premonsoon, monsoon and postmonsoon seasons, respectively. The maximum and minimum temperatures are also collected for the same stations. The intensities of inversions, lapse conditions and isothermals of dry bulb temperature are computed at every 50 hPa level from the surface to 500 hPa at 0000 GMT. The total percentage frequency ofisothermals, inversions and lapse conditions irrespective of their intensities was obtained for all stations at 0000 G M T and isolines were drawn over North India for the four months. The minimum MHs, maximum MHs morning VCs and afternoon VCs have been computed according to Holzworth (1972) by using the 0000 GMT radiosonde data. MH is obtained by ex-

180

D.V. Viswanadham and V. V. S. N. Pinaka Pani

70° I

~ I.

3_s°

BOOl

)

( )

("

L

-~

90"1

9s°l

//

rj I t

re PATIALA 0

, f~

t.

\~

~-.

DELHI

('l JODHPoUR ~.5°

85° I

GWALQIOR 0

;--" "~..

L.

~

j

(,~GARTALA ~

~A--N~AMADABAD

RANCHIo "\( NAGPUR 0

I

/ ~ j j

I

f

P~

m

"~-1 J

I

I

Fig. 1. Location map of the stations in North India

TEMPERATURE ------

PROFILE

DRY ADIABAT

\\

LJJ Q p<

MAX. MIXING HEIGHT

\ / \ \ \ \ \ \

~

\

\

\

\ \

MIN. M~iXING HE~HT

\

\ \

MINIMUM TEMPERATURE T EMPERATURE

MAXIMUM TEMPERATURE

--i~-

Fig. 2. Computation of mixing heights tending a dry adiabat from the surface temperature to its intersection with the early morning rural temperature profile (Fig. 2). The height of the point of intersection from the ground is known as MH. If the dry adiabat is extended from the minimum (maximum) temperature, then one gets the minimum (maximum) MH. However, since the radiosonde observations are taken in suburban or rural areas, we get the M H pertaining to the suburban areas. To obtain urban MHs, one has to account for the so-called urban heat effect which is rather pronounced during the early morning and it is relatively small (perhaps negligible) during the afternoon. Holzworth suggested a value of 5 °C to be added to the rural minimum

temperature to get the urban MH. One of the authors (Viswanadham, 1983) studied the spatial distribution of temperature continuously for four days at a tropical station namely Cochin, according to which a value of 3 °C was found to be more realistic for Indian coastal stations. However, since there is no coastal radiosonde station in North India, the value of 3 °C would undoubtedly be an underestimate. Keeping in view the inland characteristics, a value of 5 °C is used except for the monsoon season wherein a value of 3 °C is chosen. The VCs were computed by multiplying the corresponding MH with the mean wind speed through the mixing layer. Since the mean wind speed is not available, only surface wind speeds are used; hence, VCs could be underestimated. Before the results are discussed, it should be born in mind that this study is subject to the following limitations: 1. The radiosonde observations are available at two times per day only. Thus the computation of M H with the help of the morning temperature profile is likely to create some errors. 2. The data used are from rural areas while the M H are computed for urban areas, based on an assumption about the urban heat island. 3. Assumptions about urban heat island intensities need not be made if actual observations are available. However, only a small amount of experimental evidence is available. . The construction of isolines from a small network of stations is likely to result in some error. 5. The non-availability of data below 500 hPa for computation of inversions is a limitation. 3. Results and Discussion Figure 3 shows the spatial variation of the percent frequency of occurrence of inversions over North India. In all four months, maximum frequencies of inversions are observed in the eastern and southwestern regions, with minimum values in the western parts. In the seasonal comparison, maxim u m values are observed in January and October followed by April and July. The highest values in January can be attributed to the winter conditions, which are always characterised by inversion conditions especially in inland regions. It may be noted that North India can be considered as a

Atmospheric Dispersal Capacity over North India

%

~ "-

L

{

e'

18 l

JANUARY

-

3o°

25"

1

~ o°

'? "-'~1

35 °

N

~ --)

;ULY

-'

L

OCTOBER

1

I

~f

25 °

) 70 °

75=

1, 80 °

85 °

90 °

70 °

95 °

./5°

/,2"~,\

80 °

8 5°

"

I

F..,1% "--~

.¢ ...

30°

/2~

%¢ .,- .F

I

/~

l

I

I

JANUARY

-k ~

I

\~'j

I

95 °

normally does not lead to the development of inversions. Figure 4 depicts the spatial variation of the percent frequency of occurrence of isothermals. Maximum values are observed in January followed by July, April and October. The maximum has not exceeded 4% in any of the cases. The patterns do not reveal any significant feature because the percent frequency is so low. However, most of the northern parts show the lowest values in all cases unlike the situation with inversions, where reversals of such cases are also observed. Figure 5 shows the spatial variation of percent frequency of occurrence of lapse conditions. Consistently higher values are observed in the western parts in all four months. The most intense lapse

typical inland region. Even in October, which is supposed to represent the postmonsoon features, inversions are more frequent. In this part of the country, postmonsoon conditions tend to create this anomalous condition because the earth's surface cools much faster than the atmosphere above, with the result that temperature increases with height especially at night when only cooling takes place, reaching a maximum during the minimum temperature epoch. The low frequencies in July can be explained as due to the overcast conditions prevailing most of the time, which do not allow these anomalous conditions to develop. Although April is typical of premonsoon conditions, summer conditions prevail in this month. It is well known that any typical summer condition

35 °

Fig. 3. Spatiai variation of percent frequency of inversions over North India at 0000 GMT

,! 90 °

t

I--

I

APRIL

J ~"b

,-, a o_x~_

ixr~-i

"

I t 35 ° i

(

/.

t.

I

I

I

JULY

'

L F OCTOBER

"" ~

[ J

.3

)

\,.,>

J

30 °

25 °

:~

1.Z

!

I "f

20° 70 Q

75 °

BO°

85 °

90 °

95 °

70 °

75 °

80 °

85 °

90 ~

95 °

Fig. 4. Spatial variation of percent frequency of isothermals over North India at 0000 GMT

182

D.V. Viswanadham and V. V. S. N. Pinaka Pani I

,1.

I

" ,•

35 °

""

I

I

" [*'

I

I

JANUARY

~"

t

I

I

)

APRIL

i

I

(

~o° 25 °

20 °

I

¢,

3¢

30 °

"~

~ ,,-.I

JULY

.~--sz---H

..

I

")

"~-~/

)

'-,

(

f

r ~ "-]

I

I

I

OCTOBER

(

r

25° ::,

~'1

I'~ [ /

20° 70 °

.~o

BO°

~5o

90"

%0

70°

7~

~0"

t,5"

9¢

conditions are observed in July when the percentage is more than 90% everywhere. The second most intense lapse conditions are observed in April as expected. As already discussed earlier, April can be considered as typical of summer which, due to the extreme heat input from the surface, almost always results in lapse conditions. The case of July, however, is slightly different in the sense that it not only shows occasional heat input during the non rainy days but also shows overcast conditions that represents typical neutral conditions, both of which lead to the high frequency of lapse conditions. The lowest values in the months of January and October can be explained as due to the winter characteristics. Figure 6 shows the spatial variation of minimum MHs over North India. In the seasonal compari-

I

.L t~

35 o

' ~% . . . .

f

JANUARY

,(

.I

I

,,

,

•%

95'

Fig. 5. Spatial variation of percent frequency of lapse conditions over North India at 0000 G M T

son, the highest values are noticed in April followed by July, October and January. The extreme in April and January are mainly because of summer and winter characteristics, respectively. The heat input at the surface is so low that even after adding the urban heat island effect, the values hardly rise to 400 m in January. The high number of inversions also sometimes results in low mixing. Even in the case of April when typical summer characteristics prevail, the values hardly reach 600 m. July also shows values in the same range as those of April. The extreme western parts do not experience the monsoonal effect especially in the desert region where in the absence of cloud/rain, the heat input is large leading to a relatively high mixing height. By any standards, the values are extremely low in all the months, making pollution prob-

i

JULY

,ll

3o° #~/

~I6W

,M " . . e

.."

25 °

Y-.---.A:}E ~s.

/

, K"1~, "..

,

rpJ

(

3¢

"'-Q

322"

3¢

L)

APRIL

".

-~--,

OCTOBER

0"

'-. ....

2¢ 7o~

7¢

8¢

85°

90 °

95°

70°

75 °

80~

85o

90 o

95~

Fig. 6. Spatial variation of minimum mixing heights (× 10m) over North India

Atmospheric Dispersal Capacity over North India

the principal reason for such high values. Thus the western parts and to some extent the southern central parts of the region appear to be favourable for good dispersal of pollutants. On a seasonal basis, April and October appear to be most favourable for dispersal. The combination of minimum MHs and maximum MHs reveals that the extreme western parts are favourable for good dispersal of pollutants. The southern central parts and the extreme eastern parts are also favourable for good dispersal. The extreme northern parts and to some extent the southern parts do not seem to have enough assimilative capacity to disperse the pollutants, if of course, only the vertical mixing is considered. In almost all cases the patterns follow the surface temperature distribution. Vertical mixing alone cannot be used to identify the potential areas for reasonably good dispersal of pollutants. In fact a combination of this vertical mixing with the horizontal wind, serve much better and perhaps more directly for identifying potential areas. Figure 8 depicts the spatial variation of morning VCs over North India. In all months, the common feature is the occurrence of the maximum morning VCs in the extreme western parts of the region. In fact barring the western parts, the values are alarmingly low. All seasons show equally poor conditions in this regard. The worst season is October followed by January, April and July. An interseasonal comparison reveals April and July to be relatively more favourable for the dispersal of pollutants. The relatively higher values

lems really complicated. The highest value of minimum MHs is around 50 hPa when one finds inversion frequencies also to be a maximum. Moreover, these are all urban MHs and industries are normally located near the outskirts where the mixing is expected to be lower. Remedies have to be found out to overcome this situation during early morning hours. As it is, the extreme western parts and the south eastern parts seem to be favourable for relatively better dispersal compared to the other parts, but still not very good. The spatial distribution of maximum MHs is shown in Fig. 7. The most striking feature of this spatial variation is the consistent occurrence of the highest values in the extreme western region of the parts. The extreme northern parts and the central parts, to a large extent, show slightly lower values. In fact, April and October seem to be most favourable for good dispersal of pollutants because of the consistency of high vertical mixing all over North India. The eastern sectors also do not change appreciably from season to season. Except perhaps in July, the eastern sector has heights of about 1600m. The proximity of the sea to this sector is perhaps the main reason for this consistency. Since, as is well known, the coastal characteristics always try to reduce the extremes, the vertical mixing may not be directly connected to these characteristics, but certainly indirectly. The surface temperature, on which the vertical mixing is dependent, is highly influenced by coastal characteristics. An inter-seasonal comparison reveals that April has the highest values followed by October, January and July. The high heat input in April is r 35°

'

c:~"'...-', 1 / t"

i

I

i

..L,

"~( "--":

JANUARY

N

1412~ , ~1G,,(~,~

. . . . . ..-

~'%\\\

183

•

JULY

".. 'J 12

"'~-d2," . -. . .

/'-'

.

"-Y:>'--,

, ,

;, •~

a4

.,

,

,

,

~; ""--'-:

APRIL

,

,

OCTOBER

'~ ", 14

3o"

.

, _

, ~ ~ %

;'. . . . . . "

25"

_~/_x~ 70°

75°

8 0°

85 °

90"

P'-1' 4'

18

r

95'

70°

75°

-I

80o

:, -,)

85°

90 o

j

95 °

Fig. 7. Spatial variation of maximum heights (× 100m) over North India

mixing

184

D.V. Viswanadham and V. V. S. N. Pinaka Pani (.

• ._.

7

35"

"

)

~,~

R-.4

3o°

"'--'--,1 ; 2

(-.

JANUARY

JULY

",,~'~:"~

L?

-%.

2~"~2.: , ' /

2s°

.

,.J,

.

.

.

"~, . . . . )

35°

,

~o°

toa/

zs"

~ ;,~ ._~~ lj,n.#

, ) ,6

-

7""A-, : Y/ t"7, /

.A .... 70°

75*

i

APRIL

('

cU

80"

85°

/~

',, (A~ I

,

,

I (

i

,

;>

"~--)

i

~ "~ i

,

.a.

,

j


3o°

.

.,.

5 /

((2}

,

,.

,

90 °

950

700

,

,~

75 °

..%

(.

ANU RY

L 't

"--....,,.

....~

l

OCTOBER

80 o

85 °

90 °

of April can be attributed to the higher vertical mixing and those in July can be attributed to the relatively stronger winds. The stagnation of pollutants is extremely likely in the months of January and October. Even in April and July, stagnation is not unlikely except in the western parts. Utmost caution must be exercised in releasing pollutants into the atmosphere during the morning period, although one has a reason to believe that these values are underestimates in view of the lack of consideration of the vertical wind variation. However, the inclusion would, undoubtly not help to raise this value very much. In July, the situation may differ slightly due to the rainfall which may perhaps wash out the pollutants. However, the situation in other months

35 °

I

-

2

"..--

,

~5 °

Fig. 8. Spatial variation of morning ventilation coefficients (x 100 m 2 s - 1 ) over North India

especially January and October, is poor because of the frequent fog formation. Only in the extreme western parts of the region are conditions favourable for the dispersal of pollutants. Figure 9 shows the spatial variation of aftern o o n VCs over N o r t h India. The entire eastern half of N o r t h India shows extremely low values in all four months. The western parts, shows consistently high values. The unusually low values of afternoon VCs is a matter of serious concern as far as the eastern parts, and to some extent the central parts of the region are concerned. Even when one considers April and July, where the highest values are noticed, the situation is equally bad for the central and eastern parts, more so in July.

'JULY',

s / ' ~ ' ,o

;

.,~.

,.-

25°

i

'-~ "----~

35°

,

Z

i

. )

OCTOBER

2,' 4

12

,

-.,

APRIL

-'"~

2S° 70°

75°

80

85

90

,o 95

Y//o ,R.4 , 70 °

75 °

~# " ~ , _2."V' , I ,#°7k, , 8 0°

850

90 °

95 °

Fig. 9. Spatial variation of afternoon ventilation coefficients ( x 1000m 2 s - 1 ) over North India

Atmospheric Dispersai Capacity over North India

4. Conclusion One important outcome of this study is the identification of the western part of North India as, the potentially best region for industrial development because of the consistency of high MHs as well as VCs. The dispersal capacity is so low in the eastern parts that utmost caution should be exercised in dealing with pollutant dispersal. Even the afternoon VCs, which are more or less taken as maximum VCs, do not exceed 2000-3000 m2/s in the eastern parts while the limit for reasonably good dispersal is about 6000m2/s. In fact, this requirement is not met for the entire area under study in any of the months, although most parts satisfy this in the month of April. When the months of April and July, which are supposed to possess very high VCs because of strong convection is the former case and strong monsoonal winds in the latter, do not meet the requirement, the question of fulfilling this requirement in other seasons does not arise. In fact, in January even in the extreme western parts, this requirement is not met° The case of October is also not good. The solution to this problem does not lie in suggesting closure of industries. Probably stricter regulation of emissions could be effected such that a relatively lower emission with a lower ventilation, will have the same effect as a higher emission with a high VC. If one wishes to choose the appropriate sites for industrial development so as to have the least adverse effects, the obvious choice is the extreme western parts. However, the analyses of optimum locations is not so simple for industrial development since most of the area is industrialized already to some extent, even in meteorologically unfavourable places such as in the eastern parts. Hence, the problem to be addressed is the abatement of pollution, since no control can be exercised over meteorological conditions. In broad terms a reduction of emissions of any type of pollutant in the eastern sector

185

and to some extent in the central parts, is the only apparent solution. Individual stations should be chosen for very detailed studies of not only the meteorological parameters that affect the pollutants but also the typical pollutant distribution within a given area. References

Holzworth, G. C., 1964: Estimates of mean maximum mixing depths in the contiguous United States. Mon. Wea. Rev., 92, 235-242. Holzworth, G. C., 1967: Mixing depth, wind speed and air pollution potential for selected stations. J. Appl. Meteor., 6, 1039-1044. Holzworth, G. C., 1972: Mixing heights, wind speeds and potential for urban air pollution throughout contiguous United States. EPA Office of the Air programmes, AP 101. 118 pp. Padmanabhamurty, B., Mandal, B. B., 1976: A note on pollution potential at Delhi during October 1975-March 1976. Vayu Mandal, 6 (2-3), 53-60. Padmanabhamurty, B., Mandal, B. B., 1980: On pollution potential at Visakhapatnam. Mausam, 31,295 302. Szepesi, D., 1967: Meteorological conditions of the turbulant diffusion of atmospheric pollutants in Hungary. Vol. 32, Official publication of the National Meteorological Institute Budapest, Hungary. Viswanadham, D. V., 1983: On the diurnal and spatial variation of heat and moisture over Cochin. Proc. of Water Balance and National Development, pp. 179-182. Viswanadharn, D. V., Santosh, K. R., 1989: Air pollution potential over South India. Bound.-Layer Meteor., 48, 299-313. Vittal Murty, K. P. R., Viswanadham, D. V., Sadhuram, Y., 1980a: Mixing heights and ventilation coefficients for urban centres in India. Bound.-Layer Meteor., 19, 441-451. Vittal Murty, K. P. R., Viswanadham, D. V., Sadhuram, Y., 1980b: Estimates of mean mixing heights over India. Indian J. Air Poll. Control, 3, 18-22.

Authors' addresses: D. V. Viswanadham, Department of Geophysics, Banaras Hindu University, Varanasi, India; V. V. S. N. Pinaka Pani, Metallurgical and Engineering Consultants (I) Ltd, Ranchi - 834 002, India.