IASTA-2010

D–O–1

Variations in Atmospheric Aerosol Concentration of Various Sizes during the Total Solar Eclipse of 22 July 2009

Deepti Saxena1 and Adarsh Kumar*

1Meerut College, Meerut, UP – 251 004

Department of Physics, Apeejay College of Engineering, Sohna, Gurgaon – 122 103

Corresponding Author : E mail : adarsh_phy@yahoo.co.in

ABSTRACT : The effect of total solar eclipse on aerosol size and concentration has been studied. Continuous measurements of aerosol particle concentration having various sizes (0.3- 0.5, 0.5-1.0, 1.0-2.0 and 2.0-5.0 ìm) along with some meteorological parameters in conjunction with the total solar eclipse of 22 July 2009 were made from 21 to 23 July 2009 at Roorkee (29052’ N, 77053’ E, 275 m above sea level). Roorkee observed 90 % maximum obscurity of eclipse since it was close to the path of totality. The event started from 05:28 AM and ended at 07:42 AM with the total eclipse at 06:30 AM. Results of the measurements show an increase in the aerosol number concentration in the sub-micron and micron size ranges during the eclipse with respect to those made on any other day. This may be explained in terms of the effect of condensational growth due to the increased relative humidity.

Key words : Aerosols, Solar eclipse, air temperature, elative humidity.

Results

321

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

References

1.Manohar G K, Kandalgaonkar S S and M K Kulkarni, J Geophys Res (USA), 100 (1995) 20805.

2.Singh A K, Nivas S, Kumar A, Rai J & M J Nigam, Indian J Radio & Space Phys, 28 (1999) 1.

3.Bhattacharya A B, Kar S K, Bhattarcharya R & J Basak, Indian J radio & Space Phys, 25 (1996) 173.

4.Chandra H, Vyas G D & S Sharma, Indian J Radio & Space Phys, 26 (1997) 30.

5.Pahwa D R, Singhal S P & L T Khemani, Study of aerosols at Delhi, Mausam, 45(1), 49-56, 1994.

6.Zhang X Y, Arimoto R, Cao J J, Zhi S A, & D Wang, Atmospheric dust aerosol over Tibetan Plateau,

Journal of Geophysical Research, 106, 18471-476, 2001.

7.Devara P C S & P E Raj, A lidar study of atmospheric aerosols during two contrasting monsoon season, Atmosfera, 11, 199-204, 1998.

8.McCormick M P, Swissler T J, Chu W P & W H Fuller, Post-volcanic stratospheric aerosol decay as measured by Lidar, Journal of Atmospheric Science, 35, 1296-1303, 1978.

9.Parameswarn K & G Vijaykumar, Effect of atmospheric relative humidity on aerosol size distribution,

Indian J Radio & Space Physics, 23, 175-188, 1994.

10.Sharma D K, Rai J, Israil M & Pratap Singh, Summer variations of the atmospheric aerosol number concentration over Roorkee, India, Journal of Atmos Solar-Terres Physics, 65, 1007-1019, 2003.

322

IASTA-2010

D–O–2

The Information of the Background Aerosols from the Measurements of the Coarse Particle and Ion- mobility Spectra at Maitri (Antarctica)

Devendraa Siingh, V. Pant and A.K. Kamra

Indian Institute of Tropical Meteorology, Pune, India

devendraasiingh@tropmet.res.in

ABSTRACT : Simultaneous measurements of aerosol size distribution of coarse particles and ion mobility spectrum made at Maitri, Antarctica have been used to get information on the background aerosols from KL approximation.

Introduction

The ion mobility spectrum is correlated with the size distribution of aerosol particles. Therefore, measurements of the ion mobility spectra have often been used to arrive at aerosol particle size distribution. Several methods have been used for this conversion. For example, Misaki [1964] converted the ion mobility spectra into aerosol size distribution using the method of Bricard [1962] and assuming the aerosols to be neutral, singly or doubly charged. Dhanorkar and Kamra [1993] converted the ion mobility spectra into size spectra of aerosol particles using the procedures of Israel [1970] and Hoppel [1977], and assuming the presence of only neutral and singly charged particles.

If a steady state of charge equilibrium is assumed, KL model of Tammet [1992] for the particle size distribution and least square fitting of the mobility measurements can be used for the conversion of ion mobility sprectra to aerosol size distribution. Horrak et al. [1994, 1995] used this method to estimate the aerosol size distribution from the 6 - month and 14 - month averaged data of ion mobility. In this paper, we test this conversion procedure using the simultaneous measurements of the ion mobility distribution and aerosol size distribution made with and ion – spectrometer and an Aerodynamic Particle Sizer (APS) respectively, at Maitri, (700 45’ 55’’ S, 110 44’ 03’’E, 117 m above sea level ) Antarctica.

Observations

An ion counter which was being used for measurements of ion concentration at Maitri during the 24th Indian Scientific Expedition to Antarctica, was used in the mobility- measuring mode for measurement of mobility spectra at different times on two days (17 January, 2005; and 18 February, 2005). Ion counter consists of three Gerdien condensers. Inner electrode of each condenser consists of two coaxial electrodes of equal diameters but insulated from each other. All the three condensers are fixed on an end-plate of a cylinder. A fan fixed on the other end of cylinder, sucks the air through all the three condensers.

323

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Mobility range and corresponding diameter of the ion- spectrometer is given in Table-1. Details of the instrument are described by Siingh et al. (2005, 2007).

Table 1. The Ion Mobility and Particle Size Ranges of Different Types of Ions

a Estimates of equivalent diameter of particles have been calculated assuming single charged particles

Aerosol number size distribution in the size range 0.5 - 20 μm diameter was made with an Aerodynamic Particle Sizer (APS, TSI model 3321). The instrument was kept inside Kamet observatory and the air samples for the APS systems was drawn at the rates of 3.3/6.6 lpm through conductive silicon tubes of 0.5 cm internal diameter each and 1.0 m. The inlets of tube were cleaned daily with butanol and allowed to dry. Measurements of meteorological parameters were made with a weather-station with its sensors mounted at 10 m height. All instruments were operated side by side on both days.

KL parameter

The mobility spectrum of air ion space charge (Z) = d /dZ and the size distribution of the number concentration of the particle n(r) = dn/dr, are correlated through the relation (Salm, 1988),

(1)

In the above equation, q is number of elementary charge, rqZ- - is the radius of an aerosol particle that corresponds to the pre-arranged values of Z and q, p--q(rqZ) is the probability

differential distribution function of atmospheric air from r-space to z-space at a known radius rqZ.

Several empirical distribution functions have been proposed from time to time for the description of the aerosol spectrum of the number concentration in a size range from 0.01 to 20 m. For the calculation of aerosol particle size spectrum is represented by the simple four parameter model i.e KL model (Tammet,1992) which was chosen to fit the data in the present case. The size distribution is given by

(2)

where a, rx is the co-ordinates of the interaction point of the asymptotes, L is ascent of the left asymptotes, K is descent of the right asymptotes. The model description in detail is given by Tammet (1992). If the n(r) in equation (1) is replaced by the expression of the KL distribution, the equation will contain four unknown scalar quantities.

324

IASTA-2010

The values of the parameters a, rx, K and L are estimated according to the principle of the least square fitting of the mobility spectra.

Results and Discussion

The aerosol particle size distribution measured by the APS and calculated from ion- spectrometer measurements (average of 8 spectra and 4 spectra on January 17 and February 18, 2005 respectively) or plotted in Figure 1. Diameters of aerosol particles have been corrected as suggested by Tammet (1995) for aerosol particles assuming particles density of 1.6 g/cm3 (Park et al., 2004). Best fitting lines are then drawn from each spectra and values of constants K, L, a and rx for KL model are then determined from the slopes and intersection of these lines. Fig.1 also shows the size distribution of aerosol calculated from KL model over the whole size range. The estimated KL parameters K, L, a and rx derived from the data of Jan 17, 2005 are 2.4579, 0.61099, 521 cm-3 and 20.9 nm respectively and the same based on the data of February 18, 2005 are 1.9616, 0.5083, 325 cm-3 10.15 nm, respectively. In comparison to the study of Horrak et al., [1998], although the values lines of shapes K and L are comparable, the values of a and rx are much lower reflecting the values of clean environment in Antarctica. In the whole particle size range, KL model closely represents the derived spectra from mobility measurements. It is known that the KL model is valid for the approximation of smoothed size spectra of aerosol particles, when the fine structure in the spectra is not essential (Kikas et al, 1996). Horrak et al. (1998) have shown good correlation between the aerosol particle size distribution measured directly by aerosol spectrometer and calculated from the intermediate and large ion measurements at Takhuse rural station, Estonia using the appropriate procedures of conversion between mobility distribution and size distribution.

Figure 1. The aerosol particle size distribution measured by the APS and calculated from ion-spectrometer measurements (average of 8 spectra and 4 spectra on January 17 and February 18, 2005 respectively)

Conclusion

Using the aerosol size distributions, directly measured with the APS and indirectly

325

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

derived from the mobility measurements with an ion spectrometer, KL approximation can be used to get information on the background aerosol at a remote site.

Acknowledgment

Authors gratefully acknowledge the National Centre for Antarctic and Ocean Research (NCAOR) Goa, India for their participation in the 24th Indian Scientific Expedition to Antarctica (ISEA) and to the India Meteorological Department for providing the Metrological data. One of us (AKK) acknowledges the support under INSA Senior Scientist Programme.

References

Bricard, J., 1962. La fixation de petits ions atmospheriques sur les aerosols ultra-fins. Geofis. Pura Appl. 51, 237- 242.

Dhanorkar, S. and Kamra, A. K., 1993. J. Geophys. Res. 98, 2639-2650.

Hoppel, W.A., 1977. Ion aerosol attachment coffecients and the diffusional charging of aerosols In: Dolezalek, H. Reiter, R. (Eds.), Electrical Processes in Atmospheres. Dr. Dietrich Steinkopff Verlag, Darmstadt, 60-69

Horrak, U., Iher, H., Luts, A., Salm, J. and Tammet, H., 1994. J. Geophys. Res. 99, 10697-10700. Horrak, U., Salm, J. and Tammet, H., 1995. J. Aerosol Sci. 26, S429-S430, Suppl 1.

Horrak, U., Mirme, A., Salm, J., Tamm, E. and Tammet, H., 1998. Atmos. Res. 46, 233-242.

Israel, H., 1970. Atmospheric electricity, Vol. 1. Israel Program for Scientific Translations, Jerusalem, 317pp. Kikas, U., mirme, A., Tamm, E. and Raunemaa, T., 1996. J. Geophys. Res. 101, 19319-19327.

Misaki, M, 1964. J. Geophys. Res. 69, 3309-3318.

Park, J., Sakurai, H., Vollmers and K., McMurry, P.H., 2004. Atmos. Envor. 38, 5493-5500. Salm, J., 1988. Res. Lett. Atmos. Electr. 8, 21-24.

Siingh, Devendraa, Pawar, S.D., Gopalakrishnan and Kamra, A.K., 2005. J. Geophys. Res. 110, D18207,doi:10.1029/ 2005JD005765.

Siingh, Devendraa, Pant, V. and Kamra, A.K., 2007. Journal of Geophysical Research, Vol. 112, D13212, doi:10.1029/ 2006JD008101.

Tammet, H., 1992. Acta Comm. Univ. Tartunsis 847, 136-139. Tammet, H., 1995. J. Aerosol Sci. 26, 459-475.

326

IASTA-2010

D–O–3

Effect of Aerosols on Cloud Droplet Effective Radius as Observed in CAIPEEX Aircraft Measurements : Implications to Aerosol Indirect Effect

G. Pandithurai, S. Dipu, R. S. Maheskumar, J. R. Kulkarni and B. N. Goswami

Indian Institute of Tropical Meteorology, Pune

Introduction

The effect of aerosols on cloud microphysical and radiative properties has the greatest uncertainty of all known climate forcing mechanisms. Cloud droplets form on a subset of aerosol particles called cloud condensation nuclei (CCN). In general, an increase in aerosol leads to an increase in CCN and an increase in cloud droplet concentration. Thus, for the same amount of liquid water in a cloud, more available CCN will result in a greater number but smaller size of droplets (Twomey, 1977). A cloud with smaller but more numerous droplets will be brighter and reflect more sunlight to space, thus exerting a cooling effect. This is the first aerosol indirect effect or “cloud albedo effect or Twomey effect”. In addition, aerosol perturbations to cloud microphysics may involve feedbacks; for example, smaller drops are less likely to collide and coalesce; this will inhibit growth, suppressing precipitation, and possibly increasing cloud lifetime (Albrecht et al. 1989). In this case clouds may exert an even stronger cooling effect. Aerosol indirect effects (AIE) are generally referred to any aerosol induced modification of cloud microphysics, cloud life time, precipitation etc. All these effects are based on examining relative changes in cloud droplet size by aerosol. Model-based estimates of the first AIE alone range from - 0.22 to -1.85 Wm-2 in terms of the radiative forcing (Forster et al. 2007), although the principles pertaining to the first AIE have been adequately understood at the cloud scale with in situ (Twomey and Wojciechowski, 1969; Martin et al., 1994; Hudson and Yum, 1997; Snider and Brenguier, 2000) and laboratory (Latham and Reed, 1977) experiments.

A wide variation is noted for the sensitivity of cloud microphysics to aerosol and some were attributed to the different observational / analysis methods and natural variation. All previous studies showed Twomey effect (droplet size decreases for increase in aerosol) and some studies showed Anti-Twomey effect (cloud droplet size increases by adding aerosol) mainly from satellite data analysis but detailed investigations are sparse (Sekiguchi et al., 2003; Yuan et al., 2008). Using MODIS data, Yuan et al (2008) observed dominant Anti-Twomey effect over southeastern US and southeastern China but could not find any dependence of aerosol and cloud droplet effective radius (DER) over Indian region.

In the present study, extensive aircraft measurements carried out over Indian subcontinent were used to document the aerosol and cloud microphysical properties under Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX) during pre-monsoon, active and break monsoon conditions.

327

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Cloud effective radius is defined as the ratio of the third to the second moment of the droplet size distribution. It is one of the key variables used to estimate the radiative properties of liquid water clouds such cloud optical thickness, single scattering albedo and asymmetry parameter. The main objectives of this paper is (i) to find observational support to Twomey / Anti-Twomey effect in Continental and Monsoon clouds, (ii) to study the variations in aerosol loading and their effect on cloud droplet effective radius in different regions over Indian sub-continent with different synoptic conditions and (iii) to estimate aerosol indirect effect (AIE) for different cloud regimes.

Figure 1. Composite vertical profiles of PCASP aerosol concentration and CDP cloud effective radius as observed over three regions during CAIPEEX

Figure 1 shows the composite vertical profiles of PCASP aerosol concentrations and CDP cloud droplet effective radius (Reff) over these three sites. It can be noted that higher aerosol loading with an elevated aerosol layer around 3 km in Pathankot and the concentration decreases towards south. Over northern India during pre-monsoon, the cloud bases were around 3 km and the Reff values are lower with very little height variation, characteristic of continental/super continental cloud regimes and it needs very high cloud depths to initiate warm rain process. Clouds observed over Hyderabad are of both continental and marine clouds. Cloud bases were around 2 km and the Reff values range from 3 to 12 μm, where as clouds observed in Bangalore are of monsoon clouds with cloud bases around 1 km. Figure 2 shows the frequency distribution of cloud effective radius observed over three regions. Pathankot showed a mono-modal distribution with a modal radius of 6 μm and far below from precipitation threshold Reff of 10 μm. Hyderabad and Bangalore showed a bi-modal distribution indicating the influence of continental and maritime influences in the cloud drop size distribution.

Effect of aerosols on cloud droplet effective radius

To investigate the effect of aerosols on cloud droplet effective radius, simultaneous measurements of aerosol concentration, cloud droplet effective radius and liquid water content (LWC) are analyzed. The datasets are grouped into different liquid water bins and the LWP bin of 0-0.25 gcm-2 which is generally found at the cloud base are further used in the aerosol indirect effect analysis.

Aerosol indirect effect (AIE) is described by

328

IASTA-2010

Where Reff is the cloud effective radius and Na is the PCASP aerosol concentration. The decrease in effective radius with increase in aerosol concentration showing negative slope termed as Twomey effect and AIE values shown as positive values. AIE values estimated as described on individual flight days by using aircraft measurements and will be presented alongwith those estimated from MODIS satellite data will be presented.

Figure 2. Frequency distribution of observed cloud effective radius at the above three sites

329

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–O–4

Investigation of Vertical Profiles of Cloud Microphysics During Southwest Monsoon Using Field Experiments Data of Indian (CAIPEEX-2009)

G.R.Chinthalu. T.Venugopal, Devara .P.C.S and A.R.Dhakate

Indian Institute of Tropical Meteorology

Dr. Homi Bhabha Road Pashan , Pune 411008 INDIA

ABSTRACT : The vertical profiles (height meters) of cloud condensation nuclei (CCN_con cm-3), passive cavity aerosol (PCASP_con cm-3), and cloud droplet effective radius (CDP_Re) of the INDIAN CAIPEEX field experiments, have been critically examined during two intensive observational periods (IOP) of the southwest monsoon 2009, These IOPs are from Phase I: 14- 18 June and Phase II: 27-28 June 2009. The former is attributed to the weak phase of monsoon and later is attributed to active phase of monsoon these dates are selected taking in to consideration the daily area weighted rainfall (mm) over the country as a whole for June 2009, and daily flight reports issued by the caipeex team. These observations have been conducted in the region, en-route Hyderabad and Bangaluru. The preliminary analysis of above profile have revealed that CCN_was 7000 (cm-3), PCASP is 6300 (cm-3), CDP_Re (7.3 μm) for weak phase conditions. Where as for active phase these variables show CCN = 3750 cm-3, PCASP =3600 cm- 3 and cloud droplet effective radius about 14-15 μm (two times higher than the weak phase). A graphical relation ship between the Relative humidity (RH) and the above parameters revealed that the RH-CCN, RH-PCASP and RH-CDP_Re show a linear relationship. A further analysis of the SST wind field and upper tropospheric winds have revealed that the source of the CCN and RH is the Arabian sea and which incurs in to the Indian mainland through monsoon winds which in turn gives rise to cloud formation and rainfall of varying intensities over the Indian region. The winds play an important role in cloud formation and precipitation; the significant results will be presented and discussed.

Introduction

The Indian Summer Monsoon Rainfall (ISMR) during June-September shows large interannual and intraseasonal variations, due to embedded active and break monsoon spells. It is also observed that there are some areas of the Indian subcontinent which do not receive sufficient quantity of rainfall and localized drought like conditions prevail even during excess and normal monsoons the problem is very acute during deficient or weak monsoon years for e.g. monsoon 2002 and 2009. Spatial distribution of coefficient of variation of ISMR shows values of 40 to 100 % over peninsular India. ISMR is also low over this region. This region is one of the most vulnerable regions of ISMR. During the weak monsoon years, acute water shortages prevail over the region. To alleviate the stresses generated by water shortages, cloud seeding experiments for rain enhancements have been carried out by state governments of Maharashtra, Karnataka and Andhra Pradesh in recent years. Weather modification experiments for rain enhancements were carried on in the recent past over different parts of the world for a wide variety of applications such as water resource management, hydroelectric power generation and agriculture for e.g. (Komuscu A.U., 2008). Clouds in the tropical atmosphere occur in a wide

330

IASTA-2010

range of sizes, starting from isolated cumulus to large cloud clusters. The large cloud clusters exhibit mesoscale organization and account for most of the rainfall and vertical transport of energy from the atmospheric boundary layer (ABL) to the upper troposphere (Williams 1987). As per the recent study by Gambheer and Bhat (2000), the active Cb clouds occupy about 10% of the area and the remaining area is occupied by the anvil clouds. In monsoon season, the precipitable water at any day at any place is about 5 cm. Average daily rainfall (84 cm /122)

is about 0.7 cm. This simple calculation shows that only 10% of available moisture is converted into rainfall this suggest that there exist ample scope to harvest more rainfall through artificial cloud seeding.

331

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Gadgil 1999) and economy of the region. It is also important to note that in the Indian subcontinent more than one billion people live and are dependent on monsoon rainfall.

Aerosols Interactions and Cloud Microphysics

Atmospheric aerosol particles have a major impact on the development of clouds. Under certain environmental conditions aerosols serve as cloud condensation nuclei (CCN) and their number size and chemical composition determine the number, size and composition of nucleated cloud droplets. Those aerosols meeting these criteria are called cloud condensation nuclei (CCN) there are two types of aerosol which can act as CCN: a)

The aerosol is insoluble but wettable (or hydrophilic), and sufficiently large. b) A surface is wettable when water collects on it in a thin film rather than bunching into drops (the opposite of a wettable surface is called hydrophobic). Several studies support the argument that the composition and size distribution of aerosol particles are important to the number and size of cloud droplets activated from cloud condensation nuclei (CCN), which in turn has an impact on the formation of precipitation (For example, an increase in the number of aerosol particles suppresses precipitation (Rosenfeld, 1999, 2000; Givanti and Rosenfeld, 2004).

This paper presents the basic objectives and field campaign, Preliminary data is analyzed we report the initial results of cloud microphysics and aerosol interaction.

Materials and Method

332

IASTA-2010

route during the field observations. The flight data is analyzed by simple graphical plotting to understand the cloud aerosol interactions. Cloud condensation nuclei (CCN), Passive cavity aerosols concentration (PCASP), cloud droplet effective radius (CDP) and Relative humidity (RH %), the mean values of above parameters have been computed to understand the cloud microphysics during weak phase (14-18 June) and active phase (27- 28 June) are presented and discussed in the next section.

Conclusions

The analysis of vertical profiles of air craft flight data collected during CAIPEEX-2009 field campaign during June over Hyderabad and Bangaluru following main conclusions are drawn: a) During 14-18 June the Concentration of the Cloud condensation nuclei (CCN) was very high about 7000 cm-3 , PCASP is 6300 (cm-3), CDP_Re (7.3 μm) for weak phase conditions. Where as for active phase (27-28) these variables show CCN = 3750 cm- 3, PCASP =3600 cm-3 and cloud droplet effective radius about 14-15 μm (two times higher than the weak phase). b) A high concentration of aerosols (CCN) suppresses rainfall these results are in agreement with those reported by (Rosenfeld 1999).

Acknowledgements

The authors are thankful to Prof. B.N. Goswami Director IITM Pune for his keen interest in the subject and constant encouragement. We acknowledge the IMD for satellite imagery and various NWP and satellite derived products, TIFR for radiosonde ascents. Authors also Acknowledges the IITM CAIPEEX-2009 field observing team for collecting valuable data utilized in the above paper.

References

Bhat G. S, A. Chakraborty, R. S. Nanjundiah and J. Srinivasan 2002: Vertical thermal structure of the atmosphere during active and weak phases of convection over the north Bay of Bengal: Observation and model results Current Science, Vol. 83, NO. 3, 10

Gambheer A. V. and Bhat G. S. 2000: Life cycle characteristics of deep cloud systems over the Indian region using INSAT -1B pixel Data, Mon. Wea. Rev., 128, 4071-4083.

Yash P. Abrol and Sulochana Gadgil: 1999 Rice - In a Variable Climate.

Ramage, C.S., 1971: Monsoon meteorology, New York: Academic Press, 269 pp.

Rao, Y.P., 1976. Southwest Monsoon. Meteorological Monograph (Synoptic Meteorology), No.1/1976, India Meteorological Department, New Delhi, 366pp.

Parthasarathy, B., Sontakke, N.A., 1988. El-Nino/SST of Puerto Chicama and Indian summer monsoon rainfall - Statistical relationships. Geofisica Internacionale, 27, 37-59.

Komuscu A. U. 2008: An assessment of impact of cloud seeding on local rainfall- A case study of first part of the ISKI rain enhancement program conducted in Istanbul, Turkey during 1990-1991. Weather modification 17-27 pp.

Torrence, C. and P. J. Webster, 1998: The Annual Cycle of Persistence in the El Niño-Southern Oscillation. Q. J. Roy. Met. Soc., 124, 1985-2004

Williams, M. and Houze Jr, R. A., Satellite-observed characteristics of winter monsoon cloud clusters. Mon. Weather Rev., 1987, 115, 505–519.

Rosenfeld, D.: TRMM observed first direct evidence of smoke from forest fires inhibiting rainfall, Geophys. Res. Lett., 26, 3105–3108, 1999.

Rosenfeld, D.: Suppression of rain and snow by urban and industrial air pollution, Science, 287, 1973–1976, 2000.

Givati, A. and Rosenfeld, D.: Quantifying precipitation suppression due to air pollution, J. Appl.Meteor., 43, 1038–1056, 2004.

333

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–O–5

Physical Mechanism Behind an Aerosol Cluster Event over the Bay of Bengal during ICARB

Marina Aloysius, Prijith S.S and Mannil Mohan

Space Physics Laboratory,Vikram Sarabhai Space Centre,Thiruvananthapuram

Email : marinaaloysius@gmail.com, prijithsudhakar@gmail.com,mannil_mohan@vssc.gov.in

This study explores the physical mechanism responsible for the formation of an aerosol cluster over the Bay of Bengal (BoB) during March 2006 when ICARB (Integrated Campaign for Aerosols, gases and Radiation Budget) organized by ISRO-GBP, was on. Aerosol distribution over the BoB was regularly monitored using the MODIS sensor in addition to the in-situ observations on board ORV Sagar Kanya with various instruments. The event under study occurred during 21-25 March 2006, in which the mean aerosol optical depth (AOD) over the northern BoB steadily shot up to high values in a matter of 3-4 days and dropped suddenly in one day.

Figure 1. Spatial distribution of MODIS-AOD over the BoB during March 21 – 25, 2006 showing the movement of an intense aerosol cluster from north east coast of India to the BoB. Star denotes the position of the ship on each day at 5:30 UTC

Fig.1 shows the spatial distribution of AOD from MODIS on five days from 21-25 March 2006 depicting the genesis of the cluster in which the mean AOD over the north western BoB increased from 0.35 on 21 March to a peak value of 0.7 on 24 March. On 25 March, the AOD suddenly reduced to low values. The contributions of fine and coarse particles to the total AOD determined using the fine mode fraction from MODIS are presented in Figs.2 (a) and (b) which reveal the dominance of fine mode particles in the cluster.

As the cluster was taking shape, the AODs from MODIS and on board the ship showed a steady build up from March 21 attaining a peak on March 24. In total contrast to this, the surface level aerosol mass concentration monotonically decreased from its maximum on March 21 to low values on March 24 when the ship was in fact engulfed in the cluster (Fig.3). This unambiguously points to a significant enhancement of aerosol

334

IASTA-2010

Figure 2. Spatial distribution of fine mode contribution (a) and coarse mode contribution (b) to AOD for the days March 21 – 25, 2006

concentration in the higher altitudes where aerosols were brought in through long range transport. The surface measurements at Vishakapatnam [Niranjan et al., 2008] and Kharagpur [Beegum et al., 2009] had also shown a substantial increase in BC mass concentration just prior to the appearance of the cluster in the BoB. To identify the mechanism behind the cluster evolution, the day to day variability in the wind field and the associated dynamical parameters like wind vorticity and wind convergence computed from NCEP reanalysis were closely examined for the entire period of the episode.

Figure 3. AOD from MODIS at the ship’s position and QCM measured surface mass concentration onboard the ship from March 18 – 27, 2006

Ocean surface winds from QuikSCAT during the event indicated the winds not being strong enough for any substantial in situ production of marine aerosols over the BoB. On

335

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

the other hand, the upper level (700 hPa) winds showed a remarkable short term deviation from the mean flow favourable for the enhancement of aerosol transport from the main land towards the BoB (Figs.4 and 5). At the beginning of the episode, the anticyclonic circulation over western Arabian Sea shifted eastward, merged with another anticyclone over the land mass, intensified and occupied almost the whole central Indian land mass by ~ March 24 (Fig.4).

Figure 4. Mean spatial distribution of wind vorticity at 700 hPa computed from NCEP reanalysis winds from March 20 -25, 2006

Figure 4. Mean spatial distribution of wind convergence at 700 hPa computed from NCEP reanalysis winds from March 20 – 25, 2006

336

IASTA-2010

The strong upper level winds associated with this anticyclone routed the flow of dust laden air from West Asia to the north western BoB sweeping through central India where it got enriched with pollution aerosols as indicated by the increase in BC concentration at Vishakapatnam and Kharagpur. Meanwhile, after an initial weakening, the convergence in the BoB (Fig.5) started building up from March 22 favoring accumulation of aerosols to form the cluster. On March 24, when wind the was strongest, and convergence at its maximum, the cluster was most intense and had migrated eastward into the BoB. But on March 25, as winds weakened, the convergence declined resulting in the dispersal of the cluster as seen in Fig.1.

337

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–O–6

Vertical Velocity Characteristics in Clouds and their Effects Observed over Different Regions in India During CAIPEEX-2009

R.R. Joshi*, J.R. Kulkarni, Pournima Ghule, Pranjali Mokashi and P. Ernest Raj

Indian Institute of Tropical Meteorology, Pune 411008

Introduction

Vertical motions of the atmosphere in clear-air and cloudy-air are very important to understand various atmospheric phenomena as they are both product and cause of these processes. Upward motions, for example, modify atmospheric stability and promote the formation of clouds and precipitation. Vertical wind is an important meteorological variable as it is closely connected with upward/downward transport of energy, heat, moisture, momentum and atmospheric constituents like aerosols, ozone etc. (Jorgensen and LeMone, 1989). Also knowledge of the vertical velocities is crucial for modeling the microphysics of cloud formation, their dynamics and the resulting precipitation type.

With the primary objective of understanding cloud microphysics, cloud thermodynamics and cloud-aerosol interactions, IITM has launched the National program ‘Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX)’ in the middle of May 2009. Extensive aircraft observations have been made over different geographical regions of India from May 15 to September 20, 2009. One of the important parameters being measured on board the aircraft is the vertical velocity. It would be interesting to investigate the in-cloud updrafts and downdrafts over different regions during pre-monsoon and SW monsoon and also understand their effects on vertical distributions of aerosols, cloud condensation nuclei (CCN), liquid water content (LWC) etc. which would lead to a better understanding of the dynamical and microphysical processes inside the tropical monsoon clouds and ultimately the precipitation processes in different environments in India. Results of characteristics of vertical velocity inside monsoon clouds over four different regions are discussed in the paper.

Data and analysis

Data used for this study are collected from aircraft N 361 JC equipped with all instruments to measure speed and altitudes required to give vertical velocity directly. Data measured in clouds i.e., whenever LWC (hot-wire liquid water content) is greater than and equal to zero, is only considered for the analysis to examine vertical velocity characteristics. Convective events are defined from time series of vertical air velocity and data are categorized into that in draft and core. Draft is defined when vertical velocity is continuously positive (or negative) for 500 m (3-4 sec. at typical aircraft speed of 140-150 m/s) and exceed an absolute value of 0.5 m/s. Cores are stronger portion of drafts and

338

IASTA-2010

defined as vertical values exceeding 1 m/s. All cores are contained in draft. Although the number of drafts is normally greater than number of cores, reverse may be true since single draft may contain several cores. Cores are responsible for considerable transport of mass, moisture and momentum. Diameter of an event (draft or core) is calculated as product of the duration of the event in sec and true air speed. The mass flux in each draft (, where air density, – mean vertical velocity in draft and D – diameter) is also computed. Cumulative frequency distributions of diameter, average vertical velocity, and mass flux are plotted for drafts and cores. Also percentage frequencies are calculated for updraft and downdraft for every kilometer altitude range from 0-8 Km. Further, spatio- temporal variations in vertical velocity and their effect on cloud liquid water content is investigated.

Results and Discussions

A typical case of aircraft measurements made over Bengaluru region is presented here. On 03 July 2009 as aircraft moved from Bengaluru in westward direction through convective cumulus clouds, strong updrafts were encountered. In the beginning vertical velocities were of the order of 1 - 3 m/s, and as the aircraft entered cloud over western Ghats, values of updraft increased up to 15 - 19 m/s. In a very big cloud, aircraft experienced strong updraft for 10 sec and suddenly experienced downdraft for 3 - 4 sec. In almost all the clouds, values of updrafts were in the range of 2 - 8 m/s. As mentioned above, in- cloud vertical velocity measurements made on 03 July have been taken separately for updrafts and downdrafts and their frequency distributions are shown in Figures 1a and

Figure 1a & 1b

339

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Figure 2a & 2b

1b. It is observed that below 3 km altitude, updrafts are in the range 0 - 3 m/s; whereas above 3 km the spread of updrafts is 0 - 10 m/s. It is seen that on this particular day over peninsular India, mean updrafts are stronger than downdrafts.

A similar result has been reported for cumulonimbus clouds off the west coast of Africa during GATE (LeMone and Zipser, 1980). About 10 % of the updrafts have mean vertical velocity in the range 8 – 18 m/s. Whereas, all the mean downdrafts showed vertical velocity below 6 m/s. However, the mean diameters show that at least 30 – 40 % of the updrafts are broader than downdrafts on 03 July 2009. Mass flux was greater for updrafts and about 10 % of the updrafts had mass flux grater by at least one order of magnitude compared to that of downdrafts. Mean vertical velocities in the height range 0 – 8 km measured during 03 July 2009 are plotted separately for updrafts and downdrafts. It is evident that above 5 km altitude updrafts showed higher vertical velocities almost up to 20 m/s. Below 5 km altitude the downdrafts seem to have slightly higher values of vertical velocity compared to those observed in updrafts.

To examine the effect of vertical velocity on cloud liquid water content (LWC), four typical cases of in-cloud measurements made at 3.3, 3.7, 5.1 and 7.0 km are taken and the temporal variations are shown in Figure 2a. It is seen that whenever strong downdrafts occur, LWC values increase abruptly and significantly. The corresponding scatter plots between in-cloud vertical velocity and LWC shown in Figure 2b reveal that the relationship is significant. It is possible that strong downdrafts at the base of the cloud or in the mid-region of a cumulus cloud can cause vertical transport of cloud water content and cloud drops that are aloft. However, there was one case on 03 July when strong updrafts also increased LWC in 5.0 -7.0 km altitude range. Thus vertical motions

340

IASTA-2010

inside clouds redistribute cloud water and also CCN. This has a strong bearing on the cloud microphysics and ultimately on the precipitation processes. Results of vertical motions in cumulus clouds over other regions of India will be presented.

References

M.A. LeMone and E.J. Zipser: Cumulonimbus vertical velocity events in Gate, Part I: Diameter, intensity and mass flux, J. of Atmos. Sci., 37, 1980, 2444-2457.

D.P. Jorgensen and M.A. LeMone: Vertical velocity characteristics of oceanic convection, J. of Atmos. Sci., 46, 1989, 621-640.

341

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–O–7

Aerosol Transport During the ICARB Campaign Using the ECHAM-HAM General Circulation Model

Ribu Cherian1, Chandra Venkataraman 1, S. Ramachandran2

and Johannes Quaas3

1 Department of Chemical Engineering, Indian Institute of Technology Bombay, Powai, Mumbai, 400 076, India

2 Physical Research Laboratory, Ahmedabad,380 009, India

3 Max Planck Institute for Meteorology, Hamburg, Germany

Corresponding author: Email: chandra@iitb.ac.in

Background and Motivation

Atmospheric aerosols play a significant role in regional climate change by interacting (scattering and absorbing) with radiation and modifying cloud properties. Recent studies found distinctive aerosol chemical, microphysical and optical characteristics in the south Asian region (e.g. Chung et al., 2005; Adhikary et al., 2007). Aerosols exhibit large seasonal variations, which alter the spatial gradient in the atmospheric heating patterns in south Asia (e.g. Menon et al., 2002; Adhikary et al., 2007). Therefore, an understanding of causes of aerosol spatio-temporal patterns over south Asia is needed.

A recent observational campaign, the Integrated Campaign of Aerosols, trace gases and Radiation Budget (ICARB) included extensive observational studies over Bay of Bengal and Arabian Sea during premonsoon season from 18th March to 8th May 2006. Aerosol optical depth and mass concentrations exhibited significant spatio-temporal variations (Moorthy et al., 2008; Kedia and Ramachandran, 2008), which would influence atmospheric radiation and heating. This study investigates the spatial-temporal and vertical structure of aerosols over India during ICARB using a general circulation model.

Methodology and data sets

Atmospheric simulations were made in the general circulation model ECHAM5 (Roeckner et al., 2003) with a horizontal resolution of T63 (about 1.8o x 1.8o) and a vertical resolution of 31 levels (extended from surface to 10 hPa) combined with aerosol model HAM (Stier et al., 2005). The HAM aerosol component comprises of sulfate, black carbon, organic carbon, sea salt and mineral dust (Stier et al., 2005). Global emission inventories provided by AEROCOM (Dentener et al., 2006) combined with regional emission inventories available over India (Reddy and Venkataraman 2002; Venkataraman et al., 2006) were used for biofuel, fossil fuel and wild fire emission categories. Dust and sea salt emissions were calculated online.

Aerosol mass concentrations and optical properties were simulated for the 2006 cruise campaign period. For each simulation we constrain the meteorology by nudging the model to ECMWF reanalysis data sets during 2006 period. The model outputs were sampled

342

IASTA-2010

along the cruise track to obtain daily-mean surface mass concentrations (μg/m3) and aerosol optical depth at 550 nm and 825 nm matching the location and time periods of the cruise observations. The validation of model outputs were carried out using cruise measurement data and using Moderate Imaging Resolution Spectroradiometer (MODIS) derived Terra aerosol optical depth values.

Figure 1. Modeled and measured daily mean surface aerosol mass concentrations for BC, OC, and sulfate (SO4) as a function of longitude for the ICARB cruise period

Results and Discussions

For evaluating modelled mass distributions, we compared the BC, OC, and SO4 daily surface (lowest model level) mass concentrations with surface measurements as a function of longitude for the ICARB cruise period (Figure 1). The model BC and OC surface concentrations agree well (slope: BC=0.97±0.14; OC: 0.79±0.05) with observations at most observational points over the Bay of Bengal and Arabian sea regions, while model sulfate concentration was under predicted by a factor of 2~4 at most of the locations. Possible reasons for the underestimation of sulfate mass concentrations including details of SO4 chemistry scheme and anomalous wet deposition are under investigation. In addition, use of prescribed rather than prognostic oxidant fields, may result in lower SO2 conversion to sulfate.

The model AOD were compared with sun photometer derived daily mean AOD at 550 nm and 825 nm as a function of longitude for the entire cruise period (Figure 2). The

Figure 2. Modeled and measured AOD (at 550 nm and 825 nm) as a function of longitude during the ICARB cruise period

343

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

possible reasons for underestimation of AOD at most of the points (over Bay of Bengal) under investigation are air transport, SO4 chemistry scheme, wet deposition or missing sources (Figure 2). The model derived aerosol optical depth and absorption optical depth at 550 nm were validated against satellite observations (MODIS, OMI, and AERONET). The paper will address the contribution of different aerosol species to AOD for understanding the spatial and temporal contrast in aerosol sources over Indian ocean region during ICARB period.

References

Adhikary, B., G. R. Carmichael, Y. Tang, L. R. Leung, Y. Qian, J. J. Schauer, E. A. Stone, V. Ramanathan, and M. V.Ramana (2007), Characterization of the seasonal cycle of south Asian aerosols: A regional-scale modeling analysis, J. Geophys. Res., 112, D22S22, doi:10.1029/2006JD008143

Chung, C. E., V. Ramanathan, D. Kim, and I. A. Podgorny (2005), Global anthropogenic aerosol direct forcing derived from satellite and ground based observations, J. Geophys. Res., 110, D24207, doi: 10.1029/ 2005JD006356.

Dentener, F. et al., (2006), Emissions of primary aerosol and precursor gases in the years 2000 and 1750, prescribed data-sets for AeroCom, Atmos. Chem. Phys., 6, 4321-4344.

Kedia, S., and S. Ramachandran (2008), Features of aerosol optical depths over the Bay of Bengal and the Arabian Sea during premonsoon season: variabilities and anthropogenic influence, J. Geophys. Res., 113, D11201, doi:10.1029/2007 JD009070.

Menon, S., J. Hansen, and L. Nazarenko (2002), Climate effects of black carbon aerosols in China and India, Science, 297, 2250-2253.

Moorthy, K. K., S.K. Satheesh, S.S. Babu, and C.B.S. Dutt (2008), Integrated Campaign for Aerosols, gases and Radiation Budget (ICARB): An overview, J. Earth Syst. Sci., 117, 243-262.

Reddy, M. S., and C. Venkataraman (2002), A 0.25o x 0.25o inventory of aerosol and sulphur dioxide emissions from India: I. Fossil fuel combustion, Atmos. Environ., 36, 677- 697.

Roeckner, E. et al., (2003), The atmospheric general circulation model ECHAM5. PART I: Model description, Report 349, Max Planck Institute for Meteorology, Hamburg, Germany.

Stier, P., Feichter, J., Kinne, S., Kloster, S., Vignati, E., Wilson, J., Ganzeveld, L., Tegen, I., Werner, M., Schulz, M., Balkanski, Y., Boucher, O., Minikin, A., and Petzold, A. (2005), The aerosol climate model ECHAM5- HAM, Atmos. Chem. Phys., 5, 1125–1165.

Venkataraman, C., G. Habib, D. Kadamba, M. Shrivastava, J.-F. Leon, B. Crouzille, O. Boucher, and D. G. Streets (2006), Emissions from open biomass burning in India: Integrating the inventory approach with high-resolution Moderate Resolution Imaging Spectroradiometer (MODIS) active-fire and land cover data, Global Biogeochem. Cycles, 20, GB2013, doi:10.1029/2005GB002547.

344

IASTA-2010

D–O–8

Validation of Various Formulations for Effective Radius of Droplets in Warm Clouds using CAIPEEX Aircraft Observations

Sathy Nair, J. Sanjay, G.Pandithurai, R.S. Maheshkumar

and J.R. Kulkarni

Indian Institute of Tropical Meteorology, Pune 411008

Introduction

The optical properties of the boundary layer clouds are known to be sensitive to changes in cloud droplet size distribution and concentration (Martin et al. 1994; hereafter M94). The cloud optical properties depend mainly on an effective radius (re). Thus, the effective radius, which is defined as the ratio of the third to the second moment of the cloud spectrum, is a substitute to the spectra in radiation transfer models. In most weather and climate models, re for warm clouds is either assumed to be a constant or parameterized as a function of cloud liquid water content (LWC) and droplet concentration (Nc) (e.g. Kiehl et al. 1994; Liu et al. 2009; Shaffrey et al. 2009 and others). In this study the data obtained from aircraft passes through summertime continental cumulus clouds in the Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX) around Hyderabad, India during June 2009 are used to validate different parameterizations of re. We hope that this research contribution will be useful in weather and climate models used over India.

Data and method

The cloud droplet size distribution was measured with the Cloud Droplet Probe (CDP) in the range of 3 to 50 m diameter using 30 bins of 1 to 2 m intervals. Other parameters calculated by CDP include the effective radius and the cloud droplet number concentration. The liquid water content is also derived from CDP cloud droplet size spectra. These measurements were made at an interval of 1 s.

To define a cloud record, three criteria are used following Deng et al. (2009; hereafter D09): (i) The records with Nc less than 10 cm”3 or LWC less than 0.001 g m”3 are rejected. (ii) To analyse warm clouds the records with the environment warmer than 0°C are only included. (iii) If at least 5 consecutive records are not obtained with the previous criteria, the cloud is excluded. From the daily flight performed in Hyderabad area during 11-22 June 2009, 11 days contained warm cloud records. No warm cloud was found on 13 June 2009. A total of 165 warm cloud traverses were obtained during this period totaling to 1638 one second records of cloud microphysical measurements. The predominant cloud type for each day was cumulus as noted from the observations of weather conditions obtained from the flight diary.

345

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Results and discussions

Table 1 shows the statistics of cloud microphysical parameters, such as maxima, averages and standard deviations of Nc, LWC, and re measured in the Hyderabad area during 11- 22 June 2009. The measured mean Nc of 329±210 cm”3 and maximum of 850 cm”3 are found to be lesser than the reported range of 384 to 884 cm”3 by Paul et al. (1985) for isolated warm cumuli formed during summer months of June to September in the Indian region. The measured mean LWC of 0.27 g m-3 is within the range of

Table 1. Statistics of the microphysical parameters of warm clouds

0.132 to 0.536 g m-3 reported by Paul et al. (1985). But the measured maximum LWC of 2.93 g m-3 is too high. It is seen that the measured maximum re of 13.2 ìm is higher than observed for cumulus clouds over Beijing, China (7.7 ìm) by D09. But the measured average re (5.5±1.8 m) is comparable with the overall mean re (5.4 m) given in Miles et al. (2000).

A relationship between LWC, Nc and mean volume radius rv exists as given in M94:

(1)

where is the density of liquid water. Bower et al. (1994; hereafter B94) suggested that

for both marine and continental regions re = rv is a good approximation for cumulus clouds where the LWC approximates the adiabatic value.

For stratocumulus clouds formed from continental air masses, M94 suggested that rather than assuming re = rv, it is more reasonable to assume that re and rv are proportional. They found

that there is a linear relationship between and

given as , where k is a constant. Figure 1 shows that for the CAIPEEX data although a linear relationship is found, is consistently lower than. However a straight line passing through the origin is found with a smaller gradient (k=0.698), which is closer to that obtained by M94 for continental air masses (0.67).

Previous studies (B94; M94; D09) and our observation (Fig. 2b) show that re = rv. A constant prefactor of 1.22 was suggested by D09 as a best

fit to the observations over Beijing area in China. The use of this equation, re=1.22 rv, for CAIPEEX data is found to overestimate parameterized re for values more than 4 m

346

IASTA-2010

this for (RMSE) error square mean root the data our for that shows 2 Table .2a) .(Fig indicates 2a .Fig in line (dashed) green The .large very not is D09 following re computed scheme radiation CAM WRF the in used ìm 8 = r of value fixed the e

radius effective of parameterizations the of Error square-mean-Root 2: Table

347

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

As discussed in the previous section, for cloud liquid water droplets, B94 found that a 1/3 power universal law represents the relationship of re ( m) to LWC. According to Bower’s formula, re is proportional to the ratio of LWC (g m-3) to Nc (cm-3):

(2)

where Nc = 150 over ocean and Nc = 400-600 over land. Figure 2b shows the scatterplot of observed re versus the parameterized re using Eq. 2 for the CAIPEEX data (black dots).

The best linear fit shown as a dashed line clearly indicates that this parameterization consistently underestimates re. Also shown in Fig. 2b are the parameterized re using Eq. 2 with fixed values for Nc (200, 400 and 600 cm-3) used by L09 for optimizing the Fu–Liou radiative transfer parameterization scheme in COAMPS model. It is seen that the scatter is large when fixed values for Nc are used. Table 2 (column 3) shows that for CAIPEEX data the RMSE with Nc = 600 cm-3 recommended by B94 for continental shallow cumulus clouds is not the best choice. The least RMSE is obtained with Nc = 400 cm-3, which was also found to be the best choice by L09.

Using the gradient k=0.698, Fig. 2c shows the scatterplot of observed re versus the parameterized re using Eq. 3. The computed re with the measured Nc (black dots) appears to be a good fit to the measured re with small RMSE (0.25 ìm). With fixed values for Nc, the scatter in the plot and the RMSE are found to be higher. Figure 2d shows the scatterplot of observed re versus the best estimate of

re = 0.96 rv + 0.763 (4)

obtained by linear fit in Fig. 2b. Again the computed re with the measured Nc is found to be a near perfect fit for the entire range of measured re with least RMSE (0.21 m). The measured aerosol concentration is not analysed in this study. However, the RMSE with fixed values of Nc in Eq. 4 are also found to be lower than the corresponding values for D09 and L09 re parameterizations, with Nc = 400 cm-3 indicating the least RMSE of 1.19 ìm. Thus we suggest the best parameterization of re for the CAIPEEX data in Hyderabad area is Eq. 4 with Nc = 400 cm-3.

Conclusions

The main conclusions of this study by analyzing warm, non-precipitating cloud observations around the Hyderabad area, India during 11-22 June 2009 using 1638 samples obtained from 165 warm cloud traverses can be summarized as follows. The measured mean and maximum Nc are found to be lesser than the reported range for isolated warm cumuli formed during summer months in the Indian region. Although the measured mean LWC and re are within the range reported in previous studies, the maximum appears to be higher than observed for cumulus clouds. Different parameterizations of re are examined with the in-situ data in this area. It is found that re=1.22 rv overestimates re at the large sizes. The parameterization for re suggested by M94 using the measured Nc

348

IASTA-2010

appears to be a good fit to the measured re. A practical scheme with fixed value of Nc = 400 cm-3 is found to be suitable for this CAIPEEX observations. However it is suggested that more accuracy can be achieved by using the best estimate of re = 0.96 rv + 0.763 obtained by linear fit of the Hyderabad data with Nc=400 cm-3.

We propose to extend this study by analyzing the CAIPEEX aircraft observations taken in different regions over India to understand whether these conclusions are region specific.

Acknowledgements

Indian Institute of Tropical Meteorology is funded by Ministry of Earth Sciences, Govt. India. We would like to thank the CAIPEEX scientists and AGNI Aviation crew, without whose dedicated support this work would not have been possible.

References

Bower et al.,1994: J. Atmos. Sci., 51, 2722–2732.

Deng et al., 2009: Atmospheric Research, 93, 888–896.

Kiehl et al., 1994: J. Geophys. Res., 99, 20815-20827.

Liu et al., 2009: Wea. Forecasting, 24, 286–305.

Martin et al., 1994: J. Atmos. Sci., 51, 1823-1842.

Paul et al., 1985: PAGEOPH, 123, 930-940.

Shaffrey et al., 2009: J. Climate, 22, 1861-1896

349

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–O–9

Vertical & Spatial Distribution of Aerosols at Different Environments over India During CAIPEEX

B. Padma Kumari, R. S. Mahesh Kumar, J. R. Kulkarni and

B. N. Goswami

Indian Institute of Tropical Meteorology, Pune

ABSTRACT : Cloud Aerosol Interactions and Precipitation Enhancement Experiment (CAIPEEX) is an Indian National program conducted by the Indian Institute of Tropical Meteorology (IITM), Pune, during summer monsoon season May-September 2009. Under this program PCASP (Passive Cavity aerosol Spectro Photometer) which measures aerosol concentrations in the range 0.1 to 3 microns is being operated in the aircraft over different parts of India viz., Pathankot (May), Hyderabad (June), Bengaluru (June-July), Bareilly (July-August), Guwahati (August-September) and Pune (September). The preliminary results suggest that aerosol vertical distribution is observed up to a maximum of 7.0 - 7.5 km, with high surface concentrations and enhanced layers at higher altitudes.

Introduction

Atmospheric aerosols play a major role in climate change by directly scattering and absorbing the incoming and outgoing radiation as well as through modifying cloud properties, such as droplet size distribution and cloud lifetime (Twomey, 1974; Kaufman et al., 2005). However, aerosol measurements, particularly their vertical distribution, are less and unevenly distributed around the globe. Over India, the vertical distribution of aerosols has been studied mostly from ground-based lidar measurements (Devara et al., 1995; Jayaraman et al., 1995), twilight photometric measurements (Padma Kumari et al., 2004), in situ probing using rocket and balloon-borne instrumentations (Jayaraman et al., 1987; Ramachandran and Jayaraman, 2003) and spatial distribution from satellites. Aircraft measurements of vertical distribution of aerosol black carbon to a maximum height of 2 km were carried out at Kanpur, in IG plains (Tripathi et al., 2005) and south central India (Moorthy et al., 2004).

Under CAIPEEX program, conducted by IITM, an instrumented aircraft has been used to study background aerosol along with cloud microphysical properties and their interactions over different parts of India, during monsoon season May to September 2009. This experiment has been carried out for the first time in India. In this paper, the preliminary results of vertical distribution of aerosols are discussed.

Instrumentation

The aircraft used for CAIPEEX program is N361 JC Piper PA-31T from SOAR company USA. The instrument used onboard for aerosol measurement is PCASP (Passive Cavity aerosol Spectro Photometer). PCASP measures aerosol particles in the size range 0.1 to 3 μm. The air is collected through the cone at the front side of the instrument. All particles entrained in the sample flow are aerodynamically focused to interact with the laser beam in the controlled sampling zone location at the

350

IASTA-2010

center of the beam. The light scattered by a particle entering the laser beam is collected and focused on to a photo detector.

Results & discussions

PCASP is operated in the aircraft over different parts of India viz., Pathankot (May), Hyderabad (June), Bengaluru (June-July), Bareilly (July-August), Guwahati (August- September) and Pune (September). The data has been cleaned for clouds. The vertical distribution of aerosol concentrations

Figure 1. Vertical distribution of aerosols over different locations during CAIPEEX

351

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

at all the above stations is shown in figure 1. Pathankot shows high concentrations at all the levels with layered structures. The concentration at the ground is about 3000 particles/cc. The high concentrations at Pathankot as compared to other stations are due to dust loading and biomass burning. At Hyderabad the concentrations are ~1000 particles/cc at the ground and below ~1000 particles/cc up to 5km with layers peaking at 1.8 km and 3.2 km. Between 5-6 km the concentrations are very high due to haze. At Bengaluru the concentrations during July are very low. At Bareilly the surface concentrations are above ~1000 particles/cc with steep decrease with height. Three prominent aerosol layers are seen at 2-3 km, 4-5 km and 6-7 km. At Guwahati, a NE station, high concentrations are seen at the surface with steep decrease with height and a prominent layer at 5-6 km. At Pune the surface concentration is about ~1000 particles/cc with a layer at 4-5 km.

Conclusions

Aerosol data from PCASP is collected from in situ aircraft measurements during CAIPEEX program. The aircraft measurements carried out during summer monsoon season, revealed aerosol vertical distribution at different environments with high surface concentrations and enhanced layers at higher altitudes. In-depth analysis of CAIPEEX data gives a clear picture of aerosol distribution over different parts of India during monsoon season.

Acknowledgement

The authors wish to thank Director IITM and Ministry of Earth Sciences.

References

Devara, P. C. S., P. E. Raj and G. Pandithurai, G.: Aerosol-profile measurements in the lower troposphere with four-wavelength bistatic argon-ion lidar, Appl. Opt. 34(21), 4416–4425, 1995.

Jayaraman, A., S. Ramachandran, Y. B. Acharya, and B. H. Subbaraya: Pinatubo volcanic aerosol layer decay observed at Ahmedabad (23 N) India using Nd:YAG backscatter lidar, J. Geophys. Res., 100, 23 209–23 214, 1995.

Jayaraman, A., B. H. Subbaraya, and Y. B. Acharya: The vertical distribution of aerosol concentration and their size distribution function over the tropics and their role in radiation transfer, Physica Scripta, 36, 358– 361, 1987.

Kaufman, Y. J., I. Koren, L. A. Remer, D. Rosenfeld, and Y. Rudich, Y.: The effect of smoke, dust and pollution aerosol on shallow cloud development over the Atlantic Ocean, Proc. Nation. Acad. Sci. (USA), 102(32), 11 207–11 212, doi:10.1073/pnas.0505191102, 2005.

Moorthy, K. K., S. S. Babu, S. V. sunilkumar, P. K. Gupta, and B. S. Gera (2004), Altitude profiles of aerosol BC, derived from aircraft measurements over an inland urban location in India, Geophys. Res. Letts., 31, L22103.

B. Padma Kumari, A. L. Londhe, H. K. Trimbake & D. B. Jadhav: Comparison of aerosol vertical profiles derived by a passive and active remote sensing techniques – a case study, Atmospheric Environment, 38, 6679-6685, 2004.

Ramachandran, S. and A. Jayaraman: Balloon-borne study of the upper tropospheric and stratospheric aerosols over a tropical station in India, Tellus, 55(3)B, 820–836, 2003.

Tripathi, S. N., Sagnik Dey, Vinod Tare, S. K. Satheesh, Shyam Lal and S. Venkataramani (2005), Enhanced layer of black carbon in a north Indian industrial city, Geophy. Res. Letts, 32, L12802, doi:10.1029/ 2005GL022564.

Twomey, S.: Pollution and the planetary albedo, Atmos. Environ., 8, 1251–1256, 1974.

352

IASTA-2010

D–O–-10

Relation between CCN and Cloud Droplet Concentration and their Effect on Cloud Microphysics

Rohini Bhalwankar, C. G. Deshpande and J. R. Kulkarni

Indian Institute of Tropical Meteorology, Pune

ABSTRACT : A field campaign CAIPEEX-09 (Cloud Aerosol Interaction and Precipitation Enhancement Experiment, Phase I) was conducted to measure the cloud microphysical parameters to investigate the interaction between cloud and aerosols at various altitudes over the Indian region during pre-monsoon, active and break monsoon conditions. It is well known that the atmospheric aerosol particles serving as cloud condensation nuclei (CCN) play an important role in influencing the cloud droplet concentration. The conversion of CCN into the cloud droplets and their concentrations is one of the important aspect in rain formation process. The correlation between the CCN concentrations and the in-cloud droplet concentrations before and after each cloud pass at the same altitude has been computed from one day flight data of 3 July 2009 during CAIPEEX-09. The preliminary results show linear relationship between the CCN concentration and in-situ cloud droplet concentration. Aerosols impact on cloud microphysics has been quantified by the regression equations.

Introduction

A large number of anthropogenic aerosols are added into the atmosphere due to increased industrial and urban air pollution, smoke from biomass burning etc., which leads to increase in CCN concentrations. It is well known that, concentration and spectra of CCN along with the updraft velocity determine the initial cloud droplet concentrations. Several in-cloud measurements conducted in recent years indicate that for constant liquid water content, increase in CCN concentrations increases the cloud droplet concentration and decreases the mean cloud droplet size. Therefore, formation of precipitation in such clouds by collision-coalescence is inhibited since the collision efficiency is small for smaller droplets [Rosenfeld, 2000, 2008]. Thus, aerosols have a direct effect on in-cloud microphysics. Therefore, distribution of CCN in and around the cloud is an important factor in growth and formation of precipitation [Hudson and Yum, 2002; Hudson and Mishra , 2007].

A field campaign CAIPEEX -09 (Cloud Aerosol Interaction and Precipitation Enhancement Experiment, Phase I) was conducted mainly to understand the cloud microphysics and to study the direct and/or indirect effect of aerosols on cloud processes over Indian region. The campaign consisted of in situ aircraft measurements of the aerosols, CCN and cloud droplet concentrations and their size distribution along with many other cloud and meteorological parameters during pre-monsoon, active and break monsoon conditions. The campaign was conducted during 14 May to 30 September, 2009 over Indian subcontinent. To investigate the effect of CCN concentration on in-cloud droplet concentration, we have analyzed the flight data of CCN concentrations and cloud drop concentrations measured between 2 -5 km altitude on 3 July, 2009. The aircraft took off from Bangalore base and cruised over Arabian Sea coast and Western ghats region near Mangalore before returning to base.

353

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

The dataset is grouped into two super-saturations (SS) values with 0.2 and 0.4 % SS.

Instrumentation

During aircraft observations of CAIPEEX–phase I campaign, the measurements of CCN were made with DMT (Droplet measurement Technologies) CCN counter which operates at a single super-saturation. The super-saturation can be varied between 0.1 to 2 % and the particle sizing range is from 0.75 to 10 microns in 20 bins. The cloud droplet concentration was measured with cloud droplet probe (CDP). The details of the instrument are given in Kulkarni and Maheshkumar [2009]. The data is obtained at 1-sec intervals.

Data Analysis

As a preliminary exercise, one day data obtained on 3 July, 2009 – Bangalore base is analysed and relation between CCN

concentration and cloud droplet concentration is illustrated for this data as a case study. This flight consists of 20 cloud passes and CCN measurements are taken for super-saturations of (SS) 0.2, 0.4 and 0.6%. We confine this presentation to 0.2 and 0.4% SS only. From these observations the values of mean and maximum in-cloud droplet concentration and CCN concentrations - before and after each cloud pass are considered for analysis. The CCN concentrations are taken from the immediate clear air vicinity of before and after the cloud passes.

Results

Mean and maximum values of in-situ cloud droplet concentrations (CDP Concentrations) are plotted as a function of the CCN concentration – before and after the cloud penetrations at the same altitude for SS = 0.2% in Figure 1 and for SS = 0.4% in Figure 2. Each data point plotted in both the figures represents the mean (Triangles) and maximum (Circles) droplet concentrations for one cloud pass. The lines show regression analysis of the scatter plot. Continuous line corresponds to best fit for mean CDP concentration and dashed line corresponds to best fit

354

IASTA-2010

for maximum CDP concentration with CCN concentration. In conformity with the previous results observed by Hudson and Mishra [2007] and Rosenfeld et al. [2008], we do observe linear relationship between CCN concentration and cloud droplet concentration. Both the best-fit lines in Figure 1 and 2 show increase in in-cloud droplet concentration as the CCN concentration around the cloud increases. The relationships are quantified with regression equations given in the figures. The correlation coefficient (R) for the flight data of 3 July, 2009 in Figure 1 and Figure 2 are 0.53 and 0.46 at SS= 0.2% and 0.79 and 0.86 at SS= 0.4% respectively.

Conclusions

Preliminary results of the aircraft measurements of CCN and CDP concentration in and around the cloud pass shows that there is a linear relationship between in-cloud drop concentration and CCN concentrations before and after the cloud penetration at the same altitude. Moreover, it illustrates that the greater the CCN concentration around the cloud - the greater the in-cloud droplet concentrations. Further, increased CDP concentration with increase in CCN concentration may alter the in-cloud size distribution of cloud droplets and affect the formation of precipitation process. Thus, aerosols would appear to have a direct effect on in-cloud microphysics. Although, the concentration and spectra of CCN are the dominating parameters in determining the cloud droplet concentrations and their size distribution, the updraft velocity at cloud base also influences the CDP concentrations and mean diameter. Therefore, combined effect of both CCN concentration and updraft velocity may give better correlation between CCN and droplet concentration. Data obtained over different regions and under different meteorological and synoptic conditions are being analysed to find out the relationship between CCN and cloud droplets concentrations.

More analysis is required to confirm these results.

References

Hudson, H. G., S. S. Yum (2002),Cloud condensation nuclei spectra and polluted and clean clouds over the Indian Ocean, J. Geophys. Res., 107, D19, 8022 doi:10.1029/2001JD000829.

Hudson, J. G., and S. Mishra (2007), Relationships between CCN and cloud microphysics variations in clean maritime air, Geophys. Res. Lett., 34, L16804, doi:10.1029/2007GL030044.

Kulkarni, J. R. and R. S. Maheshkumar (2009), Lecture notes on aircraft instruments, CAIPEEX Tech. Notes, 1- 75.

Rosenfeld, D. (2000), Suppression of rain and snow by urban and industrial air pollution, Science, 287(5459), 1793– 1796.

Rosenfeld, D., W. L. Woodley, D. Axisa, Eyal Freud, J. G. Hudson, and A. Givati (2008), Aircraft measurements of the impacts of pollution aerosols on clouds and precipitation over the Sierra Nevada, J. Geophys. Res., 113, D15203, doi:10.1029/2007JD009544.

355

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–P–1

Size Distribution of Cloud Condensation Nuclei Observed during Cloud Passage (CAIPEEX-2009)

C. G. Deshpande, Rohini Bhalwankar and J. R. Kulkarni

Indian Institute of Tropical Meteorology, Pune, India

ABSTRACT : Atmospheric aerosol particles serving as cloud condensation nuclei (CCN) are key elements of the hydrological cycle and climate. Elevated concentrations of CCN tend to increase the concentration and decrease the size of droplets in cloud. A field campaign CAIPEEX- 09 was conducted to measure CCN and other cloud microphysical parameters to investigate the interaction between cloud and aerosols at various altitudes over the Indian region. The CCN size distribution measured during cloud passage at constant altitude on 3 July 2009, during CAIPEEX-09 was analysed. The preliminary results show single mode size distribution with modal maximum at 2 μm diameter. The systematic variations are observed in size distribution during cloud passage. Various processes taking place inside cloud govern the CCN size distribution variations. The abundance of CCN particles observed over Indian west coast is attributed to sea salt particles of marine origin with favorable westerly winds from Arabian sea.

Introduction

The CCN concentration and their size distribution over continent is modified because of increase in anthropogenic aerosol due to increased industrial and urban air pollution, smoke from biomass burning etc. The chemical nature of aerosols affects the hygroscopic parameters which govern the CCN activity of aerosols. Several planned aircraft measurements were conducted in different parts of world in last decade to understand cloud-aerosol interaction and processes involved [Rosenfeld et al. 2008, references there in]. These in-cloud measurements indicated the dependence of in-cloud droplet concentration and their size on CCN concentration around the cloud. Their results have shown that higher CCN concentrations increase the cloud droplet concentrations with reduction of effective radius of the droplets and suppress the precipitation in the clouds. Thus, size distribution of CCN in and around the cloud is of great interest in the study of microphysical properties inside the cloud.

The characteristics of tropical clouds are very different than the extra-tropical one. Thus to understand the cloud microphysics and to study the effect of aerosols on cloud processes over Indian region a field campaign of Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX) Phase I was conducted during 14 May to 30 September 2009 over different parts of India. In situ aircraft measurements of the concentrations and size distribution of aerosols, CCN and cloud droplets along with various meteorological parameters during monsoon season were conducted in this field experiment. In the present study, the CCN size distribution spectra are analyzed using flight data of 3 July 2009, Bangalore base measured at constant altitude for super-saturation (SS) with 0.2 and 0.4 %.

356

IASTA-2010

Instrumentation and Data Analysis

During aircraft observations of CAIPEEX–phase I campaign, the measurements of CCN were made with CCN counter (DMT, Droplet measurement Technologies, USA) operating at a single super-saturation. The super-saturation can be varied between 0.1 to 2 % and the particle sizing range is from 0.75 to 10 microns in 20 bins. The details of the instrument are given in Kulkarni and Maheshkumar [2009].

The variation of size distribution of CCN particles at different altitudes and during cloud passage is illustrated in the present study. One day data from flight of 3 July, 2009 is analysed. This flight consists of total 20 cloud passes and CCN measurements are taken for super-saturations (SS) 0.2, 0.4 and 0.6%. The average CCN size distribution is computed and plotted at different altitudes. The size distribution variation before, during and after exit of cloud passage is considered for analysis.

Observations

Figure 1 shows the average CCN size distribution at different altitudes during different cloud passages on 3 July 2009 when aircraft took off from Bangalore base and travelled over west coast near Mangalore and return through Western ghat region. The CCN size distribution does not show any systematic trend with respect to the altitude during the flight. However, it is worth to point out that the modal diameter of 2 μm is maintained at all altitudes and maximum CCN particles up to 6 μm are observed during all cloud passage. The CCN concentration during cloud passage depends on meteorological conditions.

Figure 1. Average CCN size distribution observed on 3 July 2009 at Bangalore base

The CCN size distribution spectra during single cloud passage with 5 second before and after cloud passage on two different occasions is plotted in figure 2. The passage (a) and (b) were about 8 and 12 second duration at constant altitude of 3382 and 7046 m respectively over western coast region near

357

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Mangalore. The size distribution in case (a) shows systematic increase with minimum size distribution before entering into cloud, increases when entered into cloud and maximum size distribution after exiting from cloud. The increase is more pronounced for CCN particles < 2 μm diameter. It suggests growth of small CCN particles during cloud passage. However, in case (b) initial increase in concentration of CCN < 2 μm is observed followed by shift in peak of size distribution from 2 μm to 4 μm during passage of cloud. The presence of particles > 5 μm at the end of passage is noteworthy. The particle growth inside the cloud may cause shift in this peak. This figure also shows that the CCN size distribution observed at altitude of 7046 m is higher as compared to that at 3382 m. Moreover, reduction in CCN concentration is observed as aircraft traveled eastward (continent) from the west coast.

Figure 2. CCN size distribution observed during passage and 5 second before and after passage of (a) 8 second at 3382 m and (b) 12 second at 7046 m altitude on 3 July 2009 at Bangalore base

Discussions

The atmosphere contains significant concentration of particles of micron or submicron size which have an affinity for water and serves as condensation nuclei. The CCN particles available inside the cloud grow in size due to various processes to form cloud droplets of larger diameter. Also the stage of cloud development and vertical movement inside the cloud decides the conversion of CCN into cloud droplets. Such in-cloud processes may be responsible for the observed increase in CCN concentration of particles < 2 μm in our measurements. Moreover, the shift in peak size distribution reported during the in-cloud passage and after exit shows such growth of particles. This increase may contribute to form larger size cloud droplets. The abundance of sea salt aerosols at west coast during monsoon season is expected. We observe the higher CCN size distribution at 7046 m altitude than that at 3382 m on west coast which may be due to entrainment of these sea salt aerosols of marine origin from upper troposphere with westerly winds.

References

Kulkarni, J. R. and R. S. Maheshkumar (2009), Lecture notes on aircraft instruments, CAIPEEX Tech. Notes, 1- 75.

Rosenfeld, D., W. L. Woodley, D. Axisa, E. Freud, J. G. Hudson, and A. Givati (2008), Aircraft measurements of the impacts of pollution aerosols on clouds and precipitation over the Sierra Nevada, J. Geophys. Res., 113, D15203, doi:10.1029/2007JD009544.

358

IASTA-2010

D–P–2

Seasonal Variation of Extreme mean Sea Level Pressure over India in Relation to the Volcanic Aerosol, Warming Environment and El Nino Events

H. N. Singh, S. D. Patil and Nityanand Singh

Indian Institute of Tropical Meteorology, Pune-411008

E-mail : narendra@tropmet.res.in

Introduction

Extreme climate and weather events are increasingly being recognized as one of the key aspects of climate change and they adversely affect humans and ecosystems, with serious socio-economic consequences (Kondratyev and Varotsos 1995). They held a major responsible for the incessant alteration from sunny and calm to rainy and windy conditions. The most recent reports on the evaluation of climate change (IPCC 2002, 2007) highlighted the need to study the behaviour of the extreme events of the various components of the climate. India being mainly an agricultural country, the economy and further its growth purely depends on the vagaries of the weather and in particular the extreme weather events. Thus the need arises for the detailed investigation about the vulnerability of our societies and socio-economic structures with respect to the occurrence of extreme weather events or climatic conditions. Specific studies in the field of climate and climate change addressing at Mean Sea Level Pressure (MSLP) have usually considered the reconstruction of series (Barring et al. 1999), temporal changes (Zveryaev 1999) and their relationship to other climate variables (Kutiel and Paz 1998). Detection of human influence on MSLP has also been noticed (Nathan et al. 2003). Analysis of the behaviour of the extreme values of maximum and minimum daily atmospheric pressure has also been noticed (Labajo et al., 2009).

El Nino and La Nina are well known to be associated with significant monthly/ seasonal climate anomalies at many places around the globe. The impact of El Nino- southern oscillation on MSLP anomalies has been found (Gouirand and Moron 2003). Swolter et al. (1999) have shown that El Nino/La Nina events have a significant relationship with the relative frequency of climate extremes occurring around the globe. Kothawale and Rupa Kumar (2005) have reported that all-India mean annual temperature has shown significant warming trend of 0.05°C/10yr during the period 1901-2003, the recent period 1971-2003 has also seen a relatively accelerated warming trend of 0.22°C/10yr, which is largely due to unprecedented warming during the last decade. Similarly warming trend over the northern hemisphere has also been witnessed since late 1970’s (Singh, et al. 2005). Kelly et al. (1996) identified the spatial climate response to historic eruptions in the surface air temperatures and mean sea level pressure record and used the information to assess the impact of the Mt. Pinatubo eruption. The widespread cooling and change in the atmospheric circulation over northern hemisphere have also been found. On the other hand, aerosols are known to have a significant influence on the

359

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

climate system (IPCC, 2007) in one way or the other. They can exert a notable radiative forcing and hence also influence both the meridional temperature gradient and the latent heat content of the atmosphere. The analysis of the monthly and seasonal characteristics of the extremes in the MSLP over Indian region has never been reported earlier though plays an important role in modulating the weather systems.

In view of this, an attempt has been made in the present study to analyze the frequencies of the seasonal extremes (low/high) MSLP days as well as the lowest/highest MSLP values over the Indian region to understand if and to what extent aerosols loaded in the atmosphere, warming environment and El Nino events affect the weather systems and in turn the monsoon performance.

Data and analysis

In recent years, NCEP/NCAR reanalysis data are being used widely by various atmospheric scientists for climate change studies and its impact on our environment because of its global geographical coverage and near real time update in various temporal- scales. The 2.5×2.5 degree lat/long rasterized daily MSLP data available from the NCEP/ NCAR reanalysis (Kalnay et al. 1996) have been used in this study for the period 1951- 2007. The daily MSLP data for 57 years during winter (January and February), pre- monsoon (March, April and May), summer monsoon (June, July, August and September) and post monsoon (October, November and December) seasons over the Indian region have been analyzed to examine the trends and their characteristics in the frequencies of the extremes (low/high) MSLP days and the lowest/highest MSLP values. The statistical percentile threshold values computed based on the daily data for the period 1951-2007 for all the seasons determine the criteria whether the day is a low, high or normal MSLP day. The extreme high MSLP days are defined via the exceedance of the 90th percentile of the daily mean sea level pressure for the particular season and those days with MSLP values less than the 10th percentile are categorized as low MSLP days. The remaining days are grouped as normal MSLP days. The frequencies of the extreme low/high MSLP days and the lowest/highest MSLP values during winter, pre-monsoon, summer monsoon and post monsoon seasons are carried out and analysis have been performed over the Indian region for the period 1951-2007 to locate distinct atmospheric circulation patterns favoring for the initiation of the convergence processes over the region that yield a major outbreak of the weather and climate changes.

Results and discussions

The time series for the frequencies of extreme (low/high) MSLP days as well as the lowest/highest MSLP values are shown in figures 1 and 2 respectively along with their best linear fit over the Indian region for winter, pre-monsoon, summer monsoon and post monsoon seasons during the period 1951-2007. It is clear from the detailed analysis that the trends are found to be increasing (decreasing) for the frequencies of the high (low) MSLP days. Similarly, the trends in the lowest/highest MSLP values are also found towards the increasing tendency for all the seasons over the Indian region. Both the trends in low/ high days and lowest/highest MSLP values are highly significant at 0.01% and above. The trends in the frequencies of high (low) MSLP days are increasing (decreasing) indicate the reduction in the cyclonic activities that may affect the monsoon performance over the

360

IASTA-2010

Indian region. The seasonal lowest and highest MSLP values have also been computed to locate distinct atmospheric circulation patterns favoring monsoon initiation and its role in modulating the Indian summer monsoon performance. The increasing tendency of the seasonal lowest MSLP values leads to the reduction of the cyclonic circulation and ultimately affects the monsoon activities when it is integrated with increasing tendency of the frequency of the low MSLP days during summer monsoon season. Similarly the increasing tendency of the seasonal highest MSLP values signifies the weakening of convergence process over the summer monsoon region. Though the occurrence of seasons over India is regular in each year, there exists year-to-year variability in magnitude. The lowest MSLP values are higher (1006-1014 hPa) in winter and lower (995-1000 hPa) in summer monsoon season.

The frequencies of both the extremes in the mean sea level pressure over the Indian region show a peculiar change after the late 1970’s when the global surface air temperature start rises continuously. The frequencies of low MSLP days are much higher than that of the high MSLP days after the late 1970’s whereas the reverse was observed during the earlier half. The possible causes for these drastic changes in frequencies after the 1975 may be unprecedented warming trend due to greenhouse gases during recent decades, volcanic aerosols from major eruptions and strong El Nino events occurred during the period 1975-2007 over northern hemisphere as well as over the Indian region. An extensive study is initiated to investigate if and to what extent aerosols loaded in the atmosphere

361

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

affect cyclonic circulation and its characteristics for monitoring the rainfall events over India.

Acknowledgements

The authors are grateful to Prof. B. N. Goswami, Director, Indian Institute of Tropical Meteorology (IITM) Pune for providing the facilities and continuous encouragement during the course of the study.

References

Kondratyev and Varotsos, 1995, Geophysics and Space Physics, 18, 123-151. IPCC, 2002, Madrid : Spain, 36.

IPCC, 2007, WMO, Switzerland, 21.

Barring et al., 1999, Int. J. Climatology, 19, 1427-1443. Zveryaev, 1999, Int. J. Climatology, 11, 1177-1185.

Kutiel and Paz, 1998, Theoretical and Applied Climatology, 60, 93-109. Gouirand and Moron, 2003, Int. J. Climatology, 23, 1549-1566. Nathan et al., 2003, Nature, 422, 292-294.

Labajo et al., 2008, Int. J. Climatology, 28, 1227-1238. Labajo et al., 2008, Atmosfera, 22, 125-139.

Swolter et al., 1999, J. Climate, 12, 3255-3277.

Kothawale and Rupa Kumar, 2005, Geophysical Research Letter, 32, L18714.

Singh et al., 2005, Regional Hydrological Impacts of Climatic Change Hydroclimatic Variability, Red Book (Eds.), 296, 273-282.

Kelly et al., 1996, Int. J. Climatology, 16, 537-550.

Kalnay et al., 1996, Bull. American Met. Soc., 77, 437-471.

362

IASTA-2010

D–P–3

Aerosol Radiative Impacts over Indian CTCZ Region: Results from Pilot 2008 Aircraft Experiment

J. Jai Devi1, S. N. Tripathi1*, Tarun Gupta1, B. N Singh1 and

V. Gopalakrishnan2

1 Department of Civil Engineering, Indian Institute of Technology Kanpur, 208016, Kanpur, India

2Instruments and Observational Techniques Division, Indian Institute of Tropical Meteorology Pune, 411008, Pune, India

*Corresponding Author Email: snt@iitk.ac.in

Introduction

Aerosols can affect climate in several ways which is well explained in recent literature (Kiehl, 1999; Bréon, 2006). Modeling results give a good foresight on both the direct and indirect effects of aerosols on solar radiation. Monsoon rainfall over Indian subcontinent is the lifeline for more than hundreds of million people. Therefore, any climatological changes in its characteristics warrant investigation. Recent global modeling studies, albeit few only, that estimated the aerosol effects on monsoon show conflicting results- one group of investigators report enhancement of rainfall in the pre-monsoon season (May- June) and reduction during the main monsoon (July-September) phase (Lau et al., 2006); while others (Ramanathan et al., 2006) show an overall rainfall reduction in whole India during May-September months. Here, we present the computed heating rate over the entire I-CTCZ region based on measured profiles of aerosol optical and microphysical properties.

Study Area

The continental tropical convergence zone (CTCZ) over India has a complicated atmospheric structure due to uneven terrain which rises from sea level to 6 km. In the Indian CTCZ lies the Great Indian Desert in the west, heavily polluted Indo-Gangetic basin in the North-Central India, high altitude Himalayas in the North East and peninsular region towards South. Elevated terrains act both as preferred centres for local convection and barriers to the transcontinental free flow of air at lower levels (Hills, 1978).

Method

An extensive aircraft experiment over I-CTCZ was performed during pre-monsoon (May 2008) and late monsoon (September 2008) seasons to study aerosol microphysical and optical properties. During the study, based on the striking differences in physical, optical and chemical properties of aerosols present over I-CTCZ, the region was divided into five zones- The Himalayan foothills; The Indo Gangetic (IG) basin; the desert region; The Central India; and the marine region. Knowledge of the spatial and vertical structure of the aerosol layer will help to verify and constrain the modeled results for Indian monsoon.

363

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Results

Results on the vertical and spatial distribution of aerosol optical properties e.g. SSA obtained during the airborne measurements using an onboard Photoacoustic Soot Spectrometer (PASS-1) which was integrated with photomultiplier tube for measuring scattering coefficient will be presented together with black carbon mass fraction obtained using Aethalometer and Scanning Mobility Particle Sizer and Aerodynamic Particle Sizer.

The inhomogeneous distribution of aerosol both spatially and temporally with the perturbation of natural aerosol by anthropogenic activities adds to the uncertainty in the forcing estimates. Thus, the heating rate was calculated at all stations where vertical aircraft profiling was carried out using OPAC (Hess et al., 1998) and SBDART (Ricchiazzi et al., 1998) Model. OPAC was used for the calculation of optical properties by constraining with observed profiles of scattering and absorption coefficients, total particle concentration at different levels by fine tuning the number concentrations of constituents in the model. When modeled parameters and observed parameters agree with ±5% tolerance, optical properties such as SSA, asymmetry parameter, extinction coefficient were calculated for the new model. These along with AOD, asymmetry parameter, SSA, cloud parameters from MODIS (if present), wavelength range (shortwave- 0.3- 0.9 μm) and surface albedo, water vapor amount from were given as input to SBDART code to calculate the short wave net radiative flux from top and bottom of each atmospheric layer. The heating rate was then calculated using the standard equation.

Interesting results for other zones with strikingly different aerosol properties will also be presented to give overall variation in the optical properties of aerosols within and above the mixed layer during pre-monsoon and monsoon with vertical variation at selected stations.

Acknowledgement

This work is financially supported by DST ICRP and ISRO GBP and MT programmes. We are also thankful to the entire NRSC team for making the aircraft available for the experiment.

Reference

Breon, F.-M. (2006), Climate - How do aerosols affect cloudiness and climate? Science, 313, 5787, 623-624. Hills, R. C. (1979), Inst. Brit. Geogr., 4, 3, 329–352.

Kiehl, J. T. (1999), Climate change- Solving the aerosol puzzle, Science, 283(5406): 1273. Lau, K. M., Lau, M. K., and Kim, K.M (2006), Climate Dynamics, 26, 855-864. Ramanathan, V., Chung, C., Kim, D., Bettge, T., Buja, L., Kiehl, J.T., Washington, W. M., Fu, Q., Sikka, D.R., and Wild, M. (2006), Proc. Nat. Acad. Sci. 102(15), 5326-33.

364

IASTA-2010

D–P–4

Satellite Observations of Jaipur IOC Fire Incident – A Case

Study

K.V.S. Badarinath, Shailesh Kumar Kharol, Anu Rani Sharma and Biswadip Gharai

Atmospheric Science Section, National Remote Sensing Centre,

Dept. of Space-Govt. of India, Balanagar, Hyderabad – 500 625, India

Email: badrinath_kvs@nrsc.gov.in

A massive fire erupted at an Indian Oil Corporation (IOC) fuel depot in Sitapura Industrial area, about 20 km from Jaipur city on 29th October, 2009. The fire incident was so huge in intensity and could be seen even 10 kilometers from the site. Total 12 tanks containing 7000 kiloliters of petrol, 24000 kiloliters of diesel and 1000 kiloliters of kerosene, went on blowing one after another for three days left the entire area into darkness due to thick plumes of black smoke. The devastating fire in IOC fuel depot have taken 13 lives and caused an approximate loss of 300 crore to the IOC. In the present study, Terra MODIS and IRS-P6 AWiFS satellite data along with mesoscale model (MM5) derived winds were analyzed to evaluate the long-range transport of smoke plume from IOC fuel depot fire accident site. Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) data are also used to analyze the vertical distribution of pollutants associated with fire accident. Figure -1 (a & b) shows the Terra MODIS derived True color image on 30th October, 2009 and False color composite (FCC) of IRS-P6 AWiFS on 01st November, 2009. It is clear from the figure – 1 (a & b) that thick plume of black smoke was transporting towards the south – east direction coinciding with wind direction (figure-1(d)) and spread up to 266 kilometers in downwind direction on 01st November, 2009. Nearest CALIPSO pass over the site on 30th October, 2009 confirmed that the emissions were confined up to 3.0 km height (figure – 1c). Black carbon emission in to the atmosphere was estimated using the average emission factors found to be ~75 tones on a lower side. Considering the life time of BC of the order of 5 days to a week, such large emissions to the atmosphere will have impacts more on the human health.

365

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Figure 1. (a) True color image of Terra MODIS on 30th October, 2009, (b) False Color Composite (FCC) of IRS- P6 AWiFS on 1st November, 2009, (c) Vertical feature mask image on 30th October, 2009 over central India derived from CALIPSO and (d) MM5 model derived winds at 850 mb on 01 November, 2009

366

IASTA-2010

D–P–5

Onset of Warm Rain Process in Convective Cloud: as Observed in CAIPEEX Program

Mahen Konwar1, J. R. Kulkarni1, R. S. Maheshkumar1, B. N. Goswami1, Eyal Freud2 and D. Rosenfeld2

1Indian Institute of Tropical Meteorology, Pune, India

2Institute of Earth Sciences, Hebrew University of Jerusalem, Jerusalem, Israel

ABSTRACT : Initiation of warm rain process in a convective cloud is identified by the effective radius of cloud drop size distribution and further investigated by the images of the hydrometeors. Polluted air mass suppressed the warm rain process. In relatively less polluted air mass, coastal haze and polluted haze the warm rain process occur at much less cloud depth. The cloud microphysics was found to be governed by the amount of cloud condensation nuclei present and type of CCN.

Introduction

Aircraft observations during Cloud Aerosol Interactions and Precipitation Enhancement Experiment (CAIPEEX) conducted from May to September 2009 over India are utilized for this study. Identification of the onset of warm rain process is important as the cloud microphysical processes are directly influenced by the cloud condensation nuclei (CCN). It is reported that cloud droplets formed from polluted surface air are usually too small to precipitate (Gunn and Phillips, 1957). The increase in CCN concentration from anthropogenic aerosols yield higher number of cloud droplets and can reduce the cloud droplet size that may inhibit precipitation and increase cloud lifetime (Albrecht 1989, Rosenfeld , 2000). The onset of warm rain process could be identified from the effective radius (Re) of the cloud drop size distribution of values ranging from 12 to 14 ¼m. This will be further investigated by the cloud image probe installed at the aircraft. Some preliminary results will be presented here.

Experiment and Instrumentation

Intensive cloud observations were taken during the CAIPEEX mission at different environmental conditions around the country which include polluted, less polluted, coastal haze and polluted haze. From the cloud base to the available permissible altitude profiles of the cloud were taken. The aircraft includes onboard scientific instruments such as Passive Cavity Aerosol Spectrometer Probe (0.1 to 3 m), Cloud Droplet Probe (3 to 50m), Cloud Image Probe (50 to 1550 m) and CCN counter. Other sensors provide temperature, liquid water content, relative humidity, dew point temperature, wind speed and directions.

Observations and Results

From the relative concentration of CCN the air masses were categorized as polluted, less polluted,

367

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

coastal haze and polluted haze. Profiles of cloud drop size distribution for the air masses are presented here. Figure 1 shows the cloud drop size distribution of a convective cloud in a polluted air mass. The coalescence process was pushed to higher altitude with the increasing CCN concentration which could not be achieved in this case even at 7 km. However for less polluted air mass the coalescence process started at the altitude of ~ 5 km. This shows that pollution effect the microphysical properties of cloud which can inhibit or delay the warm rain process. The narrow DSD at the cloud base (Figure 1a) for the polluted air mass shows that the number of cloud drop size concentration increases without broadening the shape of the DSD. This is opposite to the less polluted are mass where relatively broad spectra is found at the cloud base.

Figure 1(a-d). (a) Cloud DSD profile for an environment with polluted air mass, (b,c,d) show the CIP images at different altitudes. Notice that warm rain process could not be achieved even at ~ 7 km

Figure 2(a-d). Same as figure1 but for less polluted air mass. The warm rain process could be observed at ~ 5 km

Figure 3 and 4 show the DSD profiles for coastal haze and polluted haze layer respectively. Notice the broad DSD spectra at the cloud base and the early onset of warm rain process. The coastal air mass contains sea spray that act as giant CCN and encourages coalescence process at a very shallow cloud depth. It is important to note that the polluted haze layer at the atmospheric boundary layer may contain giant CCN which could be instrumental in the initiation of warm rain process.

368

IASTA-2010

The depth of the haze layer which extended up to 4 km may be important for the suppression of convective processes which we found in few cases.

Figure 3(a-d). Same as figure1 but for coastal haze layer. The warm rain process could be observed at ~ 3.1 km.

Figure 4 (a-d). Same as figure 1 but for polluted haze layer. Notice the onset of warm rain at ~4.8 km in panel

(c). The broad DSD spectra at the cloud base show the presence of large drops

Summary

The onset of warm rain process is effected by the concentration of CCN and giant CCN. The polluted air mass suppresses or pushes the depth of warm rain to higher altitude. The coastal haze and polluted haze layer at the boundary layer contains giant CCN and enhanced the warm rain process that may result shallow depth of warm rain.

369

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Acknowledgements

We thank all the CAIPEEX team members for successfully conducting the aircraft observations. The CAIPEEX program has been funded by the Ministry of Earth Sciences, Govt. of India.

References

Albrecht, B. A., Aerosols, Cloud Microphysics, and Fractional Cloudiness, 245, 1227 – 1230, 1989.

Gunn R. and Phillips B.B., An Experimental investigation of the effect of air pollution on the initiation of rain, Jour.of Metero., 14, 272-279, 1957.

Rosenfeld, D., Suppression of Rain and Snow by Urban and Industrial Air Pollution, 287, 1793- 1796, 2000.

370

IASTA-2010

D–P–6

Mode of Convection Associated with the Boundary Layer Clouds during CAIPEEX

Prabha T. V.* , G. Pandithurai, S. Dipu, B. N. Goswami, and J. R. Kulkarni

Indian Institute of Tropical Meteorology, Pune, India

*thara@tropmet.res.in

Boundary Layer Clouds (BLCs) are shallow clouds with their base on the top of the boundary layer. BLCs reflect a significant amount of radiation back to space and eventually lead to a cooling effect at the surface. The presence of aerosols can influence the formation and disappearance of BLCs through a variety of complex interaction. Observations during Cloud Aerosol Interaction and Precipitation Enhancement EXperiment (CAIPEEX) together with numerical simulations provide an opportunity to investigate the multifold interaction between the boundary layer dynamics, cloud microphysics and precipitation. CAIPEEX observations of cloud condensation nuclei are used in a set of cloud resolving simulations and to illustrate multifold interactions of BLCs on the radiative and thermodynamic characteristics of the surface and boundary layer.

Four different planetary boundary layer schemes in the Advanced Research Weather Forecasting model are used at a spatial resolution of 1.8 km, namely the non-local (A), local schemes (B and D) based on the Mellor Yamada scheme both use 2.5 level closure based on turbulent kinetic energy, however differ in their characteristic length scale formulations. Quasi-normal scale elimination approach for PBL is used in (C). Differences between the cloud patterns are a result of different boundary layer dynamics. Results indicate that mode of organized convection in the boundary layer (PBL), often in the form of Horizontal Convective Rolls and three dimensional thermals, plays a key role in the distribution and evolution of BLCs, which can be dictated by the boundary layer stability, moisture distribution, etc.

Figure 1. Number concentration of cloud water (x 1E9 kg-1) using four PBL physics

371

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Figure 1 gives simulated cloud water number concentration from four cloud resolving simulations with distinct differences in the PBL physics.

The environmental stability has contributed to these distinct patterns through a precipitation-dynamics-radiation-cloud interaction. Although precipitation associated with BLCs is less significant, it can change the radiative impact by influencing the mode of boundary layer convection in the subcloud layer. Although cloud optical depth was too low for these clouds, these shallow warm clouds have maximum impact on the radiation reaching the surface. The surface radiation was reduced to 40 % (Figure 2a) along these cloud streets. The surface layer fluxes in the model are influenced considerably due to these aspects. Latent heat flux (Figure 2b) showed significant differences between the simulations which were modulated by the roll circulations. The boundary layer stability inside a roll in the simulations (present in A, C, and D, however no rolls in C) is presented in Figure 2c. Updrafts rising from an unstable surface layer bring saturated warm air in the boundary layer where instability reduces. This also caused undulations in the boundary layer height (420-540 m) and cloud base is noted at this height. Simulation B has no clouds formed and continued to show comparatively stable conditions with height.

Figure 2. Short wave flux (a) and latent heat flux (b) along 20 oN showed systematic variations associated with roll circulations. Stability (Brunt–Väisälä frequency) conditions (c) inside a roll is presented

Results emphasize that mode of convection in the BLCs can be dictated by the microphysical and surface-boundary layer interactions. Investigation of these processes with observations and simulations is a necessary step, also in unraveling their role in cloud-aerosol interaction.

372

IASTA-2010

D–P–7

Large Scale Forcings for Cloud and Aerosol Variability over India during CAIPEEX Phase I Observational Period

S. G. Narkhedkar, M. S. Mujumdar, K. D. Salunke and J. R. Kulkarni

Indian Institute of Tropical meteorology, Pune

Introduction

The spatio-temporal variability of aerosols and cloud microphysics is governed by hierarchy of forcings ranging from turbulent scale to planetary scale. The synoptic scale or large scale forcing controls variability on the time scales of days to weeks. The strong convergence in the boundary layer generates upward vertical motion which lifts the moisture and pollution rich surface air to higher levels. Reverse occurs when the boundary layer is stable with subsidence. In that case, free atmosphere is decoupled from the boundary layer and we get the clouds retaining their original microphysical characters for longer time. The large scale circulation is governed by the diabatic heating of the atmosphere which in turn depends on the circulation. During CAIPEEX Phase I program, observations of aerosols and microphysics are taken from different bases on daily basis. Diagnosis of large scale forcings is necessary to understand the variability in aerosols and cloud microphysics. With this view, the large scale forcings during CAIPEEX observational periods are studied in this paper. The dynamical parameters viz. divergence, vorticity, vertical velocity and diabatic heating are used to quantify the forcings.

Data and Computations

The CAIPEEX missions were conducted from Pathankot 20- 30 May, Hyderabad 11-23 June and Bangalore 27 June to 8 July 2009. The dynamical parameters are computed at the grid point near to these stations to represent large scale circulation features. In the vertical the computations are done for standard levels viz. 850,700, 500,300 and 200 hPa levels. The grid point data at 50 km resolution from National Center of Medium Range Forecasting (NCMRWF) has been used in the study. The vertical velocity is computed using kinematic method.

Description of the model at NCMRWF

A Global Forecast System (GFS) at T254L64 resolution has been implemented at NCMRWF. The horizontal resolution is roughly 0.5 x 0.5 ° latitude/longitude on the globe. The model has 64 levels in the vertical. The parameterized model physical processes are: the gravity wave drag, realistic orography, radiation, cumulus convection (simplified Arakawa Schubert), shallow convection, large-scale condensation, diagnostic clouds, PBL, air-sea interaction and land-surface processes. The assimilation system is a global 3- dimensional variational assimilation system based on NCEP’s SSI scheme. In the current implementation conventional data sets and satellite Atmospheric Motion Vectors (AMV) are assimilated (NCMRWF technical report, 2007).

373

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Diabatic heating rate can be estimated as a residual, in the thermodynamic energy equation (Chen and Baker 1986)

, vertical velocity in the pressure co-ordinates.

static stability.

Terms indicated as (A) contribute to the horizontal advection of sensible heat. Those in

(B) are adiabatic heating due to vertical motion. Over the tropics the local change of temperature as well as temperature advections are negligible. So vertical velocity term has to be compensated by the diabatic heating term on RHS of equation (1). This means that over tropical region there is heat balance with adiabatic cooling compensated by diabatic heating. So when the local change and advection of temperature in equation (1) are neglected, we get,

(2)

Synoptic Situations

During first two IOPs the heat wave conditions were prevailing over the Indian region. During IOP-I a Cyclonic Storm formed over the Head Bay of Bengal. During IOP-II no synoptic system was formed and Heat wave conditions persisted over most parts of the country. Advancement of monsoon was in progress and during this period it has not covered whole India. During the first IOP a low pressure area was formed over central Bay of Bengal on 23rd May. It concentrated into Depression on 24th May within half a degree of lat. 16.5°N/long.88°E about 450 kms southsoutheast of Paradip. It moved northwards and intensified into a Deep Depression on 24th itself. This deep depression intensified into Cyclonic Storm (AILA) on 25th.

Results

The time-height cross sections of divergence, vertical velocity, vorticity and diabatic heating for Pathankot are shown in the figure 1. On 21 May, the maximum convergence of magnitude -6 X 10-5 s-1 is at 750 hPa level (fig. 1 a). Weak divergence is seen in the boundary layer. The level of maximum convergence extends up to 400 hPa on to 24 May and confined to 700 hPa level on 29 May. The middle level convergence produces subsiding motion in the boundary layer (fig. 1 b). The maximum sinking motion of magnitude 200 X

374

IASTA-2010

Figure 2 and 3: Time height cross sections of (a) divergence (b) vertical velocity (c) vorticity and (d) diabatic heating at Hyderabad – 11 June to 23 June 2009, 00Z and at Bangalore - 27 June to 08 July 2009, 00Z

375

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

376

IASTA-2010

figure 4. The aerosol distribution shows large scale spatial and temporal variability. The vertical distribution shows maximum concentration at the levels near to surface at Bangalore and Hyderabad. At Pathankot, the highest concentration is observed at middle levels. This may be due to transport form the desert regions.

Conclusions

The large circulation during three CAIPEEX missions has been found to vary to a large extent. This has been mainly due to onset and establishment of planetary scale monsoon system. Typical heat low type of circulation with deep sinking motion has been prevailed over Pathankot during the period 20- 29 May. The subsidence in the boundary layer has to be taken in to account for understanding the vertical distribution of aerosols and clouds. This entire tropospheric subsidence was replaced by the deep convection on 25 May indicating strong short term variability of the circulation system.

The progression of monsoon 2009 was slowed down after onset. The monsoon circulation was not established over Hyderabad during the mission period. There is absence of deep convection as clearly seen from the vertical velocity field. The upward motions limited to boundary layer are due to intense solar heating at the surface. This is consistent with presence of heat wave conditions over area during that period.

The monsoon circulation was established during the Bangalore mission. However it is seen that there is absence of deep convection. The heating pattern is typically different that is present in the deep convection.

The upward motion in the boundary layer over Hyderabad and Bangalore produced lifting of moisture and polluted air to higher levels. This may be reason for observed high concentrations of aerosols in the upper levels. The high aerosol-concentration at Pathankot at higher levels in the absence of favorable upward motion in the boundary layer, may be due to due to combined effect of horizontal advection and slow and persistent orographic lifting on the southern slopes of Himalayas.

References

Chen T. C. and W. E. Baker (1986) Global Diabatic Heating during FGGE SOP-1 and SOP-2, Mon. Wea. Rev., 114, 2578-2589.

Kasahara A. (1982) Nonlinear normal mode initialization and the bounded derivative method. , Rev. Geophys. Space Phys., 20, 385-397.

Kasahara A. and Mizz A. P. (1985) Preliminary evaluation of diabatic heating distribution from FGGE level III-b analysis data. Proc. First national Workshop on the Global Weather Experiment, National Academy Press, Washington D.C., vol. 2, part I, 317-329.

NCMRWF, Technical Report NMRF/TR/1/2007 Implementation of T254L64 Global Forecast System at NCMRWF May 2007.

377

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

D–P–8

Pan-Arctic Sunphotometry during the ARCTAS-A Campaign of April 2008

A. Saha1, N.T. O’Neill1, E. Eloranta2, R. Stone3, T. Eck4, S. Zidane1, D. Daou1, A. Lupu5, G. Lesins6, M. Shiobara7 and L.J.B. McArthur8

1CARTEL, Université de Sherbrooke, Sherbrooke, PQ, Canada

2University of Wisconsin, Madison, WI, U.S.A.

3NOAA/ESRL,Boulder, CO,USA

4NASA / GSFC, Greenbelt, MD, USA

5CRESS, York University,Toronto, ON, Canada

6Dalhousie University, Halifax, NS, Canada

7National Institute of Polar Research, Tokyo, Japan

8Environment Canada,Toronto, ON, Canada

Aerosol optical depth (AOD) measurements were acquired at six Arctic sunphotometer sites during the Arctic Research of the Composition of the Troposphere from Aircraft and Satellites (ARCTAS-A) campaign in April, 2008. Numerous smoke events were identified and related to massive forest and agricultural fires in eastern Russia and northern Kazakhstan/south-western Russia respectively. An analysis of the fine (sub-micron) optical depths from the six stations indicated the presence of underlying low frequency trends which were coherent with general meteorological considerations, source information, model estimates and remote sensing information. Low frequency (diurnal) coarse-mode optical depth events were observed at a number of the stations; these singular events are likely due to ice particles whose nucleation may have been associated with the presence of smoke, or possibly dust.

378

IASTA-2010

D–P–9

Background Large Scale Convection during CAIPEEX Period

V.R. Mujumdar and J.R. Kulkarni

Indian Institute of Tropical Meteorology,Pune

ABSTRACT : Time series of Outgoing Longwave Radiation is analysed over the region of 5°X 5° latitude/longitude, encompassing the six base stations where the first phase of Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX) has been conducted. With the prime consideration of large scale convection, the climatology of the OLR over these areas indicates the appropriate selection of the base satiations and the periods for which the experiment over the region was conducted. The temporal variation of the convection over the region, particularly during the period of the experiment, is discussed by comparing seasonal (May through September) time series of the area averaged OLR climatology and during the year 2009. It has been observed that during monsoon 2009 on most of the occasions the OLR values remained higher than the climatology, indicating overall subdued convection. The power spectral analyses of OLR over different base regions indicate the total absence the most dominant intra-seasonal Madden and Julian oscillations (MJO) during the monsoon season of 2009.

Introduction

First phase of Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX), conducted during 15 May through 30 September 2009, was primarily based on the objective to understand the microphysical processes involved in the formation of convective clouds and its interaction with aerosols leading to the precipitation, particularly in absence of large scale atmospheric forcing. As a primary requirement of the experiment the climatology of the convection during monsoon over India was taken in to consideration as a one of the major deciding criterion for the selection of Base stations and period for the operational missions. The experiment was executed by conducting special missions over 6 well distributed base stations of India viz, Pune (14 -22 May), Pathankot (23-29May), Bangaluru (1-10May, 26-30Jun and 1-15 July), Hyderabad (11-23 JUN, 14-18 August and 27-29 September), Bareilly (16-24 July, 19-27 August) and Guwahati (29August-6September). The instrumented Aircraft was flown over these base stations within a periphery of about 200-250 km radius and the observations regarding cloud parameters and aerosols were made. While selecting the station and the period for the mission the progress, establishment and withdrawal of monsoon, availability of convection and the probable advection of the aerosol over the region was also considered.

Satellite derived Outgoing Longwave Radiation (OLR) data estimates the depth of convective clouds over tropical and warm sub-tropical surfaces in a temporally and spatially extensive manner, at daily to seasonal scales (Lau and Chan, 1986). It has become a standard practice to use OLR analysis to diagnose tropical convection (Arkin and Ardanuy 1989). The OLR values less than the threshold of 240 Wm-2 generally represent the beginning of convective activity that leads the formation clouds and rainfall (Matsumoto,1992). Moron (1995), used a threshold of 235Wm-2 to delineate the African convective centre while Gunn et al. (1989) used monthly OLR values below 220 Wm-2 to identify locations

379

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

of intense or sustained convective activity over tropical regions. In this study, daily OLR data is used to track the climatology of the convection over the mission area and also to identify the convection experienced during the CAIPEEX-09 period.

Data

A 33-year (1975 through 2008, Excluding the year 1978 for which data is not available) interpolated data set of daily OLR (2.5°X2.5°) from the National Oceanic and Atmospheric Administration (NOAA) polar orbiting satellites [Gruber and Krueger, 1984] is used to establish the climatology of convective episodes during the pre-monsoon and monsoon months of May to September over the six operational regions of CAIPEEX. For May to September 2009, the uninterpolated daily OLR data, available on near real time basis, from the same source (http://www.cdc.noaa.gov/) is utilized. A climatological time series for May through September is constructed by averaging OLR values over a the regions of 5°X5° lat./long. grid surrounding the six base stations where the CAIPEEX-09 missions has been conducted.

Results and Discussions

The temporal variation of mean OLR over six selected regions of 5°x5° lat./long. associated with corresponding base stations (fig. 1) have been analysed for the CAIPEEX period and compared with the long term (33 years) normal (climatology). The OLR value of 220 Wm-2 has been marked in the figures 2a and 2b to indicate threshold for deep convection. Large scale convection over Pathankot region, located almost near the extreme northern parts of India, climatologically does not show much deep convection during entire monsoon season (upper panel of fig.2a, dash line). The OLR value just crosses the convective threshold of 240Wm-2 for precipitation during peak monsoon period of second and third week of July. This part of India is generally dominated by mid-latitude disturbances which are short lived and

does not show its signature in the long term averages. It is hypothesized that the radiative forcing due to aerosol advection over this region from African desert gives rise to an elevated heat source and influence the onset of monsoon. To validate this phenomenon, the CAIPEEX mission was scheduled over Phankot during pre- monsoon period of 14 to 22 May, even though the climatology of OLR suggests the absence of convection during the period. However, the low OLR values during the actual mission (solid line) reflect the convection association with transient eastward moving disturbances. This gave an opportunity to acquire good amount of data regarding cloud aerosol interaction. During the monsoon season of 2009 OLR over the region fluctuated

380

IASTA-2010

above the climatological values on most of the occasions, suggesting the weak convection.

The base region associated with the station Bareilly, mostly comes under the influence of monsoon trough. Over this area, the monsoon arrives during first week of July. However, according to the climatology the convection starts building up since third week of June (OLR<240 Wm-2) and further deepens (OLR<220 Wm-2) in the second week of July till third week of August (fig. 2a, middle panel). The mission over this station was conducted during third week of June when building up of the convection was expected. But, existence of very high OLR values (solid line) during the period of mission indicates absence of convection and cloud free skies over the region. The second mission was planned during peak monsoon period with the expectation of deep convection as per climatology. However, during the mission only few cumulous continental type clouds were encountered due to the presence of weak convection as suggested by high OLR values.

Monsoon arrives over the north eastern part of India in the first week of June. The area being surrounded by hills, the circulation here is dominated by orographically forced convection. The climatology of the OLR over the region, with the base station as Guwahati, (Fig. 2a, bottom panel, dashed line) shows a persistent deep convection (OLR< 220Wm-2) throughout monsoon period from June to middle of September. Here deep convection and the hilly terrain leads to the interaction of orographycally forced anthropogenic continental aerosols with the clouds and gives copious rainfall. The mission was conducted in this region during 29 August to 6 September. The moderate convection (220<OLR<240) during this period resulted in to good amount of cloudiness towards the late afternoon hours when the aircraft could make safe sorties in the difficult hilly terrain.

Figure 2. Time series of OLR climatology (dished line) and actual OLR values during May to September 2009 averaged over the operational regions associated with the corresponding the base stations. (The periods of mission are enclosed in the solid rectangles)

381

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

CAIPEEX mission took off from the northwestern part of peninsular India with the base station at Pune during 15 to 20 May. The period is characterized by occasional pre- monsoon convective clouds and thundershowers, even though climatology shows high OLR values over the region (fig. 2b, upper panel). A sudden deep in the OLR values during the mission period suggest such convective activity. Similar kind of weather is also experienced in the month of September during withdrawal phase of the monsoon when climatologically very week convection exists. The mission was again conducted over this region during 16 to 25 September when a sudden fall in the actual OLR values suggests such an activity associated with intense convection.

The characteristic temporal variation of OLR over Hyderabad and Bengaluru regions are nearly similar to that of Pune region with the exception that the pre-monsoon conditions over Hyderabad are more drier with clear skies, suggested by high OLR values during May (fig. 2b, middle and lower panel). Three missions were conducted over Hyderabad region viz. 1) 11-23 June, 2) 14-18 August and 3) 27-29 September. The weak convection dominated the first period while deep convection with cloudy atmosphere prevailed during second and third period. The mission over the Bengaluru region, which covers the southern part of Peninsular India, was conducted keeping in view of the normal onset and establish phase of monsoon during first week of June and from last week of June to second week of July

respectively. The persistence of very low values of OLR during third week of May suggest the early onset of monsoon 2009 (Fig 2b, lower panel). Increase in OLR after the last week of May indicates absence of convection during stagnation in the progress of monsoon. During first mission of the CAIPEEX over Bengaluru, the OLR remained high with a short pulse of deep convection on 4 and 5 June. The deep convection prevailed over the area for the first week of July during the second mission over the region.

382

IASTA-2010

studies have investigated 30–60-day oscillations, popularly known as Madden-Julian Oscillation (MJO) (Madden and Julian 1972), and the relationships between the MJO and the active/ break periods of the Asian summer monsoon (e.g., Yasunari 1979, 1980, 1981; Murakami 1984; Annamalai and Slingo 2001), including the genesis of synoptic systems (Goswami et al. 2003). Now it is well established that the MJO provides the most dominant planetary scale forcing for the Asian monsoon. The OLR being the proxy for the convective activity, the power spectrum analysis of the OLR time series over the six base regions of CAIPEEX mission, for the monsoon season of 2009, is carried out. The spectrum shows total absence of significant dominance of any low frequency oscillation during the season (Fig. 3). The existence of high frequency oscillations of 3 to 5 days suggests short lived convective activity throughout the country. As the regions are well distributed over India, we note a total absence of MJO over the entire continent during monsoon of 2009. This seems to be one of the prime reason that no synoptic scale system could sustained long enough and intensify during the season leading to below normal rainfall (23% below long term seasonal normal) and drought condition all over India.

Conclusion

The climatology of the convective episodes over different parts of India during monsoon season is well reflected in the time variation of long term mean OLR over the region and the same is found to be useful for the initial planning and successful execution of the first phase of CAIPEEX during monsoon 2009. The general prevalence of high OLR values and the total absence of MJO during the CAIPEEX mission suggest lack of strong large-scale forcing, resulting into the weak synoptic scale systems that could not sustained long enough. Generally, the cloud micro-physics has very less role to play when a large scale planetary or synoptic scale forcing dominates the atmosphere. The clouds formed during the mission were under the background of subdued large scale forcing and the micro physical parameters such as local instability, aerosol concentration, condensation, CAPE (convective available potential energy) etc. played a dominant role in cloud processes. Therefore the data acquired during the CAIPEEX mission is more significant in view to study the cloud micro-physical processes.

References

Annamalai, H. and J.M. Slingo, 2001: Active/break cycles: diagnosis of the intraseasonal variability of the Asian summer monsoon. Clim. Dyn., 18, 85–102.

Arkin and Ardanuy, 1989 : Estimating Climatic-Scale Precipitation from Space: A Review, Jr. of climate, vol. 2, pp 1229,1288

Goswami, B. N., R. S. Ajayamohan, P. K. Xavier, and D. Sengupta, 2003: Clustering of synoptic activity by Indian summer monsoon intraseasonal oscillations. Geo. Res. Lett., 30, 1431, doi:10.1029/2002GL016734.

Gruber, A., and A. F. Krueger, 1984: The status of the NOAA outgoing longwave radiation dataset, Bull. Am. Meteorol. Soc., 65, 1675– 1680, 1984.

Gunn, B. W., McBride, J. L., Holland, G. J., Keenan, T. D.,Davidson, N. E., Hendon, H. H., 1989: The Australian Summer monsoon circulation during AMEX Phase II.Mon. Wea. Rev., 117, 2554-2574.

Lau, K. M., Chart, R H., 1986: Aspects of the 40-50 day oscillation during northern summer as inferred from outgoing longwave radiation. Mon. Wea. Rev., 114, 1354-1367.

Madden R. A., and P. R. Julian, 1972: Description of global scale circulation cells in the Tropics with 40–50 day period. J. Atmos. Sci., 29, 1109–1123.

Moron, V., 1995: Variability of the African Convection Centre as viewed by Outgoing Longwave Radiation records and relationships with surface temperature patterns. J. Climatol., 15, 25-34.

383

AEROSOLS & CLOUDS : CLIMATE CHANGE PERSPECTIVES

Matsumoto, J., 1992: The seasonal changes in Asian and Australian monsoon regions. J. Meteor. Soc. Japan., 70, 257-273.

Yasunari, T., 1979: Cloudiness fluctuations associated with the Northern Hemisphere summer monsoon. J. Meteor. Soc. Japan, 57, 227–242.

Yasunari, T., 1980: A quasi-stationary appearance of 30 to 40 day period in the cloudiness fluctuations during the summer monsoon over India. J. Meteor. Soc. Japan, 58, 225–229.

Yasunari, T., 1981: Structure of an Indian summer monsoon system with around 40-day period. J. Meteor. Soc. Japan, 59, 336–354.

384