Wednesday, June 26, 2013


Cloud seeding operations for drought mitigation- an experience of Andhra Pradesh

B.Venkateswara Rao1, S.S.Chandra Bharadwaz2, N.Srinivasa Rao3, B.V.S.Amathya4, P.Lakshminarayana5

Note:  This article prepared by the JNTU Cloud seeding center staff was obtained by Prof.T.Shivaji Rao to show how cloud seeding is bound to succeed in the hands of genuine experts who plan the programme on scientific lines and get it executed with dedicated workers.  Since Prof.Rao happen to be one of the advisors for this programme for 6 years that is 2004 - 2009.He had a unique opportunity in supervising the  planning, execution and assessing the results. For understanding the scientific basis for warm cloud seeding and cold cloud seeding please browse through the following articles on the websites below.


Cloud Seeding for Precipitation Enhancement has been carried out by Government of Andhra Pradesh, India in the Rain Shadow (<600 12="" andhra="" between="" latitudes="" mm="" of="" pradesh="" region="" sup="">0
36’  to 19008’ and longitudes 76038 to 80055’ covering an area of 1,35,680 km2 in the 12 districts during 2003-2009. Mostly the cloud seeding has occurred between the altitudes of 1200-2500 m above mean sea level with an updraft velocity of 0.5-3 m/s. This conforms that these clouds are warm clouds well below the freezing level (5800 m) in the seeding area. Most of the clouds are seeded at the base by releasing hygroscopic material (mostly Calcium Chloride) with the help of pressurized aircrafts fitted with hygroscopic flares. C-Band analogue weather RADARs with TITAN (Thunder Strom Identification Tracking and Now casting) software were used to identify the clouds and to analyze the cloud growth. On an average not more than 23% of the seedable clouds are seeded with the existing infrastructure.  It is observed from the RADAR data that there is an increase in Cloud Volume, Area, Reflectivity, Life and Lowering of Centroid after seeding indicating that the Clouds are responding to seeding material. It is also observed in general that,  cloud seeding is more effective if we get active monsoon clouds and it is less effective if we get weak monsoon clouds. Overall about 17% of total rainfall could be attributed to cloud seeding during the years 2005-2009.

1. Introduction
Thirteenth world meteorological congress in May 1999 noted that the weather modification for rainfall enhancement is one of the tools in the total gamut of water resources management (WMO, 1999). In the Indian context it is well established fact that even if we achieve ultimate irrigation potential, we can irrigate only 30% of cultivable land, remaining 70% is under rain fed conditions where erratic and low rainfall is the problem (IWRS, 1999). Research shows that even the watershed management activities in the rain fed agriculture may not yield good results if the rainfall deficiency is more than 20% (ICHWAM, 2002). In India the Meteorological drought is declared if the rainfall deficiency is 25% and above (Subramanya, 2005). If the cloud seeding programme could cover the above deficiency of rainfall it is worth taking up. This was the background under which cloud seeding operations in Andhra Pradesh were started in the year 2003 and has been continued till 2009.

2. What is cloud seeding for rainfall enhancement?
Cloud seeding for rainfall enhancement is a process of introducing Cloud Condensation Nuclei (CCN) of appropriate size of hygroscopic (for warm cloud) or glacio-genic (for cold cloud) nature in to the cloud to enhance the cloud’s ability to precipitate. Usually to produce a rain drop, a cloud condensation nuclei is required surrounding which water vapor condenses and forms as rain drop that falls on the earth under the gravity. The CCN is naturally produced by aerosols emanated from various mechanisms including evaporation from the oceans (Sodium Chloride) and continents, apart from wind blown dust and various gaseous emissions from industries, transport and burning of forests etc. Not all these aerosols contain required size and number of CCN to produce natural rain. Whenever such depletion of CCN is observed in the natural clouds we introduce required CCN in to the clouds artificially to enhance the rainfall known as cloud seeding for rainfall enhancement.

3. Some earlier Cloud Seeding Experiments in India
In India as early as 1951, TATA firm has used Silver Iodide through ground generators to seed the clouds over the Western Ghats (CAIPEEX Implementation Plan, 2009). The rain and cloud physics research unit under the aegis of Council of Scientific and Industrial Research has conducted long term cloud seeding programme over North India, using ground based salt generators during the period 1957-1966. The results showed an increase in rainfall by 20% (Ramana Murty et al, 1968). Indian Institute of Tropical Meteorology has conducted several cloud seeding experiments during 1973-74, 1976 and 1979-86. The results showed 24% increase in the rainfall (Murty et al, 2000). During the decade 1980-1990, the world has witnessed many advances in the airborne instrumentation, radars, flares and software’s. They got imported to India in new millennium. In the year 2003 the Government of Karnataka has initiated cloud seeding with modern gadgets like radars and aircrafts Government of Maharashtra followed a month later.

4. Cloud seeding operations of Andhra Pradesh.
The Government of Andhra Pradesh has declared nearly 555 Mandals (revenue and administrative divisions with in the district) as Rain Shadow Area mainly in the districts of Ranga Reddy, Nalgonda, Medak, Karimnagar, Mahaboobnagar, Guntur, Prakasam, Nellore, Kurnool, Anantapur, Kadapa and Chittoor. In these Mandals the average annual rainfall is less than 600 mm and below. Hence they were declared as Rain Shadow Area in the year 2005. The Cloud Seeding is aimed to enhance the rainfall as well as to narrow down the dry spells during the monsoon season. In Andhra Pradesh Cloud Seeding was started in the year 2003 and has been continued year after year till 2009 - perhaps one of the biggest and longest operational cloud seeding program in Southeast Asia with an average cost of around rupees 25crores per year. In 2003 the Cloud Seeding was monitored by Water And Land Management Training and Research Institute (WALAMTARI) while it was Panchayati Raj Department, Government of Andhra Pradesh in the year 2004. Later from the year 2005 onwards the Government of Andhra Pradesh has created Rain Shadow Areas Development (RSAD) Department and entrusted the monitoring of the Cloud Seeding to Jawaharlal Nehru Technological University Hyderabad (JNTUH) in view of the program’s Scientific and Technological nature. The JNTUH in turn has established a separate department namely Centre for Earth, Atmosphere and Weather Modification Technologies (CEA&WMT) and has been monitoring the Cloud Seeding Programme with experienced and expert Meteorologists since the year 2005.

5. Cloud seeding operations in the 2009 drought year
 Never before the state has witnessed both severe drought and floods as in the year 2009. The A.P Cloud Seeding Programme for the first time has faced the severe drought situation and tried to mitigate it as much as possible. The result is that the cloud seeding has helped in mitigating the drought to some extent but could not avert it since the number of seedable clouds available in the severe drought situation is far less when compared to normal situation. An analysis of the rainfall data of seeded mandals provided by the Bureau of Economics and Statistics, Government of Andhra Pradesh has thrown an interesting fact that it is only 41.38 % of excess rainfall mandals in the state lies in the Rain Shadow Area at the beginning of the Cloud Seeding Programme, while the same figure has increased to nearly 95.65% by the end of the Cloud Seeding Programme in the year 2009 (Table.1). This figure has been achieved even before the floods in the Krishna Basin. The seeding locations during the monsoon period for the year 2009 and the corresponding rainfall for the same period is shown in Fig.1. From the figure it can be observed that most of the excess rainfall areas are located in the intense cloud seeding zones of Andhra pradesh.

6. Preliminary Results of the CAIPEEX Experiment by Government of India
The government of India has taken up a national experiment namely “Cloud Aerosol Interaction and Precipitation Enhancement Experiment (CAIPEEX)” through Indian Institute of Tropical Meteorology (IITM). The main aim of this experiment is to understand microphysical properties of the clouds in the interaction between aerosol and the cloud apart from conducting cloud seeding experiments to enhance the rainfall in various parts of the country. This experiment was started in the year 2009 and will be continued till the year 2012. As part of the first phase, they have measured the microphysical properties of the clouds from May to September in the year 2009 at various parts of the country by flying with specially instrumented Cloud Physics aircraft for the purpose (CAIPEEX Implementation Plan, 2009). The preliminary results of these measurements are indicating that the clouds over Telangana and Rayalaseema regions are seedable to enhance the rainfall and observed that the natural Cloud Condensation Nuclei (CCN) over these regions is of the order of 0.5 microns (CAIPEEX Preliminary results Phase-I, 2009). In the cloud seeding programme it is the general practice that we introduce giant nuclei artificially which is having the size more than the natural CCN size to enhance the rainfall.

Meanwhile the CEA&WMT of JNTUH has tested the flares used in the Cloud Seeding programme of A.P by utilizing the above Cloud Physics Aircraft. In these tests it came to know that the flares being used are capable of producing the CCN size of the order of one micron and above which is more than the natural CCN size as mentioned above (Table.2). Infact these flares were tested on the ground, the actual particle size near the cloud base during the operations will be more as per the cloud seeding Experts. According to existing literature the artificial CCN size has to be further increased to about 2 to 3 microns to get the better results. Infact the CAIPEEX programme will examine the resulting rainfalls with varying CCN materials and sizes during the years 2010 and 2011. We have to wait for the results at least for one more year. In view of these findings it can be said that the A.P. Cloud Seeding programme has been progressing well with the latest scientific and technological advancements.

Table.1 weekly Cumulative rainfall status of Andhra pradesh for the monsoon period of year 2009.

Similarly such experiments can be attempted in other parts of the country particularly low rainfall regions of western, northern and central India where natural CCN depletion is observed.

7. Infrastructure & equipment for cloud seeding Operations.
The cloud seeding operations over Andhra Pradesh were conducted with the aid of two C-Band Weather RADARs located at Hyderabad and Anantapur (Fig.2) and two Cessna Pressurized Aircrafts operated from Hyderabad and Bangalore. Most of the seeding was done with the Hygroscopic material (Calcium Chloride) which is burnt below the cloud base where the
Table.2. size of the CCN in the Natural clouds and Hygroscopic Flares.
Particles Size (Microns)
% of Total
Particles Size (Microns)
% of Total
Particles Size (Microns)
% of Total

updrafts are maximum.  Generally the seeding activity was covered in the districts of Southern Telangana and Rayalaseema region with latitudes of 12036’ to 19008’ and longitudes of 76038 to 80055’. The raw data of RADAR has been processed with the help of TITAN (Thunderstorm Identification, Tracking, Analysis and Now casting) software (Dixon et al, 1993) and used to vector the aircrafts towards the seedable clouds. Usually the seedable clouds are having the reflectivity of at least 20 dBZ and appeared mostly in the monsoon season between the months of June to November covering both South West and North East monsoon.

The success of the cloud seeding experiment is evaluated by analyzing the TITAN processed RADAR data with Target and Control Method, Double Ratio Method and Chemical Analysis of Rainwater samples. Moreover the altitudes of the seeding operations, associated updrafts at the cloud base and the availability of clouds in a given season were also analyzed.

8. Target and Control Method
In this method the target means the seeded cloud, while the control means the unseeded cloud, which are selected in such a ways that they are 50-60 km apart in cross wind direction, appearing at same time within in the 90 km radius of the RADAR vicinity and they should possess the similar magnitude of microphysical properties up to the time of seeding. The results are analyzed by plotting the temporal variation in the parameters such as Volume, Area, Mass, Vertical Integrated Liquid Content, Precipitation Flux, Height of Maximum Reflection, Cloud Base Height, Cloud Top Height. However only four parameters are discussed as shown (Fig.3) for a case study on August 11, 2008, 11:20GMT.

From the Fig.3 it can be observed that, all the four parameters of the seeded cloud have shown increase in magnitude, though they are in decreasing stage at the time of seeding, while the same is not happened for the unseeded cloud. The increase of dBZ value after seeding, followed by increase in Vertical Integrated Liquid content indicates that there is a formation of liquid water droplets. This in turn is well supported by increase in Precipitation flux, indicating that there is an increase of mass at the base of the cloud. Similarly the growth of the Volume of the cloud after seeding may confirm that there is a release of heat energy due to the formation of water droplets in the cloud. Moreover the seeded cloud has lasted for longer duration, compared to the unseeded cloud.

In order to remove the bias in selection of the individual clouds for comparison purpose, 20 seeded and 20 unseeded clouds are selected from the same region, in such a way that they have the life time of 50 to 60 minutes and seeding was occurred with in the first 30 minutes of the cloud genesis. The graphs (Fig.4) are drawn for the temporal variation of the total rain mass of the seeded and unseeded clouds. From the graphs it can be observed that, initially both seeded and unseeded clouds have shown intense growth during first 15-20 minutes, however the seeded clouds have shown higher magnitude of rain mass and lasted longer than the unseeded clouds.

9. Double Ratio Method
It is the ratio of two ratios and is given as follows (Ali Umran, 2008).

                           ---- (1)
Where IC = Impact Coefficient, S = Total rainfall in the target area during seeding period, NS = Total rainfall in the control area during seeding period, H = Total rainfall in the target area during 10 unseeded years, K= Total rainfall in the control area during 10 unseeded years.

If I.C >1, then it indicates the influence of seeding. In the present paper, target area and control area are selected in such a way that, they are from the same climatic zone and the correlation coefficient between rainfalls of target and control areas for the 10 unseeded years is maximum. The results are as shown in Table.3.  From the Table.3 it can be observed that, Nalgonda district is taken as target and the neighboring districts are taken as controls. Nalgonda and Mahaboobnagar are highly correlated in their natural rainfall occurrence with a correlation coefficient of 0.78. The Impact Coefficient of 1.57 between these districts indicates that there is a 57% change in the rainfall in the target district during the seeding days, indicating the influence of seeding by way of rainfall enhancement up to 57% in the target district. 

Table.3 Calculation of Impact Coefficients using Double Ratio Method for Nalgonda (Target) and neighboring districts (Control).

Correlations coefficients for  Nalgonda district vs. Control districts
↓Target /Control districts→
Ranga Reddy
Impact Coefficients for the year2008 for 17 Days of Cloud Seeding operations

A = S/NS
b  =  H/K
I.C=  a/ b
Change %
Nalgonda vs. Mahaboobnagar
Nalgonda vs. Ranga Reddy
Nalgonda vs. Medak
Impact Coefficients for the year 2007 for 21 Days of Cloud Seeding operations

a  = H/K
b= S/NS
I.C= b/a
Change %
Nalgonda vs. Mahaboobnagar
Nalgonda vs. Ranga Reddy
Nalgonda vs. Medak

10. Longterm Rainfall analysis in Nalgonda district
The monsoon rainfall data of the Nalgonda district has been collected for the period of 1990-2009 for both seeded and unseeded areas in the district to verify whether the seeded areas have improved their rainfall or not when compared to the unseeded areas in the district. A graph has been drawn (Fig.5) to depict the yearly variation of rainfall and its 3 years moving average. It can be observed from this graph that the 3 years moving average curve has shown improving trend of rainfall in seeded area when compared to the unseeded areas during the period  of cloud seeding. This is an indication of success of cloud seeding in terms of long term rainfall improvement.

 11. Chemical Analysis of Rain Water Samples
Rain water samples for the years 2007 and 2008, from Nalgonda and Chittoor districts were collected during seeded and unseeded days and analyzed for the concentrations of Calcium and Chloride - the prime cation and anion present in the Hygroscopic material used for seeding. The results are as shown in Fig.6. It can be observed from the Fig.6

that, in the majority of the cases the magnitude of the calcium and chloride is more in rainwater samples collected from the seeded clouds than that of the unseeded clouds, confirming the fact that the Hygroscopic material is released at the base of the clouds is acting as Cloud Condensation Nuclei.

From the above three analyses it can be reasonably inferred that, the clouds in this region are responding to the seeding. And the magnitude of the growth is quite substantial. In order to further probe in to the seeding details the updrafts and seeding altitudes are also analyzed.  

12. Seeding Altitudes and Updrafts
The altitudes and updrafts at which the seeding occurred over the years (2005-2008) were recorded from the flight logs and are shown in Fig.7 and 8.  From the Fig.7 it can be observed that most of the seedings were performed at an altitude range of 1 to 2 km, indicating that warm cloud seeding is predominant. Moreover during the initial years like 2005 and 2006, cold cloud seeding too (above 4 to 5 km) was also experimented, but from the operational point of view it was observed that it is better to seed warm clouds in view of their large number of availability over the region.  

From the Fig.8 it can be observed that the updrafts observed at the base of the cloud during the time of seeding is varying between 1 m/sec to 5 m/sec in most of the times, but occasionally it is reaching up to a maximum of 10 m/sec. Since most of the clouds were seeded within the first 30 minutes after their genesis, the updrafts indicated above gives suitable updraft range for effective seeding in this region and prevalence of convective activity in the cloud during this time.

13. Influence of Active and Weak monsoon over cloud seeding operations
During 2009 cloud seeding operations there was a continuous weak spell followed by active spell in the South West monsoon. The responses of the clouds due to seeding in these spells were observed by analyzing various microphysical properties of the seeded clouds.  Based on the growth observed in the properties of the cloud after seeding, the responsiveness of the clouds was determined and is depicted in the Fig.9.

From the Fig.9 it can be observed that, the more the active the monsoon, the more is the number of clouds available for seeding and the more is the responsiveness of clouds for seeding. It means that rainfall enhancement is more during the active monsoon due to increased moisture feeding.

14. Success rate and number of clouds seeded
For the year 2009 an analysis has been carried out to calculate number of clouds seeded and their success rate as identified from the RADAR data of Hyderabad and Anantapur stations and is shown in Table.4.

From this table it can be observed that the overall success rate of the cloud seeding programme is of the order of 76%. It means that nearly 25% of the clouds are not responding to the seeding operations and the reasons for the same could be attributed to variations in the microphysical properties of the clouds at the time of seeding in the same synoptic situation. Therefore in-depth observations of microphysical properties of the clouds before and after seeding are necessary to enhance the success rate for a given seeding material. Similarly there is a necessity of observing the microphysical properties of the clouds for different seeding materials by varying the particle size and chemical composition. More over it can also be observed from the Table.4 that we could able to seed hardly 23% of the available seedable clouds on a seeding day leaving nearly 77% clouds unseeded. This indicates that there is huge scope of development of infrastructures such as RADARs and aircrafts for cloud seeding operations in this region. 

15. Computation of Rainfall Enhancement through actual ground truth observations
Ground truth is often difficult to establish as the existing Rain Gauge network is insufficient. For such an experiment more Radio Reporting Rain Gauges are required to compliment the operations. However with existing Rain Gauge Network (i.e. one rain gauge per mandal covering an average area of 268.62 km2), the rainfall analysis is carried out. In each Rain Gauge, previous 24 hours rainfall is recorded and reported daily at 03:00 GMT. The total rainfall occurred on a particular day in a Rain Gauge is taken as the rainfall resulting out of seeding, if that Rain Gauge falls under zone of influence of the seeding. The zone of influence is taken as the area covered by the circle with a radius of 50 km from the point of seeding. Though exact calculation of enhancement of rainfall due to seeding is not possible with the present day rainfall measurements, but it has been estimated that on an average about 17% of rainfall may be attributed due to cloud seeding operations over Rain Shadow Region of Andhra Pradesh during the period 2005 to 2009 (Table.5). Out of the five years 2007 is a active monsoon year and 2009 is a weak monsoon year. From the Table.5, it can be found that the more the vigorous the monsoon the higher is the rainfall enhancement due to cloud seeding.  

Table.5 Summary of rainfall enhancement estimation over Andhra Pradesh during 2005-2009.


No. of Mandals

Cumulative rainfall during seeding period

Cumulative rainfall under zone of influence due to seeding

Average Rainfall per Mandal due to seeding

Average Rainfall per Mandal during seeding period

Rainfall attributed to cloud seeding operations in%


16. Other applications of weather modification
The Cloud Seeding Operations in Andhra Pradesh though mainly applied for enhancing the rainfall by controlled seeding, the other weather modification techniques includes suppression of hail storms during pre monsoon season, dissipation of Clouds during flooding season by over seeding and dissipation of fog during winter season by controlled seeding techniques. These applications are already being in use in many other countries. The biggest example is that during Olympics in China, they have stopped the rain during opening and closing ceremonies.

 17. Conclusions
Cloud Seeding operations for rainfall enhancement over Rain Shadow Region of Andhra Pradesh have clearly demonstrated that the clouds are responding substantially to the Hygroscopic seeding as has been observed from the RADAR data with respect to growth in cloud Area, Volume, Mass, Vertical Integrated Liquid Content and Precipitation Flux after the seeding. The same is confirmed from the presence of Calcium and Chloride- the basic Hygroscopic material used for seeding in the rain water samples. Similarly the ground truth rainfall analysis has also shown that on an average, 17% of rainfall could be attributed to cloud seeding in the Rain Shadow Region. It has been found that most of the seedable clouds in the area are warm clouds occurring at an average altitude of 1 to 2 km with an average updraft of 2 to 5 m/sec. it was also observed that the more the vigorous the monsoon the more is the success of the cloud seeding operations in enhancement of the rainfall. Hardly 23% of the clouds are seeded with the available infrastructure leaving a huge scope for infrastructural improvement in terms of deployment of additional RADARs and Aircrafts.

1. Ali Umran Komuscu, 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”, Journal of Weather Modification, Volume 40, pp.17- 

2. Cloud Aerosol Interaction and Precipitation Enhancement Experiment, 2009.   
    “Implementation Plan”, IITM. 46p.

3. Cloud Aerosol Interaction and Precipitation Enhancement Experiment, 2009.   
    “Preliminary Results Phase-I”, IITM. 46p.

4. Dixon, M. and Wiener,G., 1993. “TITAN: Thunderstorm identification, tracking, analysis
    and nowcasting - a radar-based methodology”, J. Atmos. Ocean. Tech, 10 (6), pp.785-797.

5. IWRS, 1999. “Theme Paper on Water: Vision 2050”,. 74p.

6. Subramanya, K., 2005.  Engineering Hydrology (2nd Edition), Tata Mc Graw-Hil. 392p.

7. Murty A.S.R. et al.,2000. “11-year warm cloud seeding experiments in Maharashtra”,
    J.Wea Modification, 32, pp.10-20.

8. Proceedings of International Conference on Hydrology and Watershed   Management,
    2002.B.S.Publications,Volume.1, 759p.

9. Ramana Murty Bh.V., Biswas K.R., 1968.” Weather modification in India”, J.
    Meteorological Society of  Japan, 46, 160-165.

 10.WMO, 1999. “Programme on physics and chemistry of clouds and weather modification  
      Research”, Report no. 35,pp.1-68.

About Me

My photo
Born in 1932 at Mudinepalli, near Gudivada, Krishna Dist. Andhra Pradesh, received Bachelors degree in Civil Engg., from Viswesaraiah Engineering College, Banglore (1956) and Masters Degree in Environmental Engineering from Rice university, Houston, Texas, (USA) (1962), Ph.D (Hony). Former Head of the Department of Civil Engineering and principal of College of Engineering, Andhra university.Formerly Hony.Professor in Andhra University,Manonmanian Sundarnar University,JNT University. Fellow of the Institution of Engineers,India Recipient of the University Grants Commissions National Award "Swami Pranavananda Award on Ecology and Environmental Sciences" for the year 1991. Recipient of Sivananda Eminent Citizen Award for 2002 by Sanathana Dharma Charitable Trust, Andhra Pradesh state. Presently Working as Director, centre for Environmental Studies, GITAM University,