Indian Meteorological Department

IMD estimates rainwater harvesting potential of Maharashtra

India Meteorological Department of Pune (IMD) functioning under Ministry of Earth Sciences has recently (Nov 2016) published a study titled “Rain water harvesting potential for different locations in state of Maharashtra”.[i] The study was undertaken to assess the potential of roof top rain water harvesting based on mean annual rainfall. The need for the study was prompted by requests from a number of NGOs for information on how much rain can be harvested in Maharashtra during a good rainfall.[ii] The study basically estimates volume of rainwater (in liters) that could be harvested per sqft of roof area at 326 locations across Maharashtra where IMD stations are located.

Though this study is an important step, there are a number of ways in which such a study, which IMD plans to eventually conduct for other states as well, can be improved, in addition to a number of limitations that the study has, as elaborated subsequently.

The study highlights that “rain is prime source of water” and suggests rainwater harvesting through rooftops as one of the crucial options. Storing rainwater locally is one option for making water available for future use, particularly in drought prone, hilly, urban and coastal areas. Due to short spells and high intensity water tends to quickly flow away.

Roof top rainwater harvesting potential

Rooftop Rainwater harvesting is the technique through which rain water is captured from the roof catchments and stored in reservoirs for future use. Harvested rain can also be recharged to groundwater.

To estimate rainwater potential (in liters) the study uses following formula

Amount of Rainfall X Area of Catchment X Runoff Coefficient = Rainwater Potential

Mean annual rainfall data of 326 rain gauge stations distributed across Maharashtra state for the period 1951-2000 (not clear why subsequent years’ rainfall figures are not used) has been used to arrive at rainfall figures. The data has been generated by Climate Application Group, Climate Division, IMD, Pune.

Rainwater potential has been calculated (in liters) first for a unit area of one square foot and subsequently for rooftops with a roof foot print of 1000 sqft (small house), 2000 sqft (medium house) and 3000 sqft (large house). Calculations for annual rainwater potential and for southwest monsoon season have been done separately.

Runoff coefficient has been used as 0.85 which is valid for concrete surfaces and tiles. It means that 85% of the water flowing over these surfaces can be harvested. (Runoff coefficient is considerably low for vegetative soil surface where runoff is affected by interception, evaporation, infiltration, slope, soil types, etc.) The minimum volume of water potential in small size house is around 17135 litres in Pusesauli (District – Satara) while maximum volume of water is around 436684 litres in Mahabaleshwar obsy (District – Satara) during southwest monsoon season. These figures are for tiled roofs.

Results have been presented in a tabulated form for each of the IMD station and also in form of map.

Maps with a wrong nomenclature

For the maps however the report seems to have interchanged the titles. For SW monsoon the min RWH potential calculated is 17.1 lt but map starts at 23-50 range. And for annual RWH potential the min figure is 23.6 lt but map starts at 17-50 range. As a result for a number of places, the Rain water potential shown for SW Monsoon season in below map is seen HIGHER than rainwater potential for the same location on annual basis, whereas it should be the other way round.

Wrong nomenclature for the maps in IMD report
Wrong nomenclature for the maps in IMD report

Rainwater harvesting calculations simplified

Such a study could come handy for those working for creating awareness about RWH. RWH potential calculated by IMD can be processed in number of ways to show much water can be saved.

For example ­ RWH potential for Latur (city) has been calculated at 48.5 lt/sqft. In Latur city, which was one of the worst drought-affected cities of India during 2012-15, if a family of 5 staying in house of 1000 sqft area stores and uses its roof top rainwater it could meet 20% of their water demand (calculated at 135 lpcd). And when per capita water availability is only 60 lpcd[iii], it could mean additional 26 lpcd of water.

Water scarcity in Latur in drought of 2015 (Photo: SANDRP)
Water scarcity in Latur in drought of 2015 (Photo: SANDRP)

Mumbai Metropolitan Region has proposed a number of dams to supply for the cities in the MMR. But not much rainwater harvesting is happening in these cities, in spite of huge potential (185.8 litres per sq ft). Rainwater to the extent of 186000 litres can be harvested by small house (1000 sq ft area) here.  

On the other hand for Pune blessed with four upstream dams supplying average 194 lpcd water to the city, RWH potential has been calculated at 44.7 lt/sqft (strangely lower than Latur!). There are many Housing societies of 3000 sqft of roof foot print. If 1000 such societies store and use their roof top rainwater 134 million liters of water could be saved annually. Pune city currently consumes 16 TMC (thousand million cubic feet) of water against allocation of 11.5 TMC water from Khadakwasla Dam.[iv] Pune Municipal Corporation does not consider RWH as a mainstream water supply option. To quench city’s growing thirst PMC is eying more water allocation from Khadakwasla and Bhama Askhed Dams, which can be avoided using options like RWH.

For Nagpur RWH potential has been calculated at 75.2 lt/sqft. In Nagpur if 100 houses of 200 sqft harvest their roof top rainwater it could save 15 million liters of water annually. Nagpur is looking for more water augmentation from Kanhan River.[v]

Rooftop Rainwater Harvesting at a residence in Pune (Photo: SANDRP)
Rooftop Rainwater Harvesting at a residence in Pune (Photo: SANDRP)

The present study could be improved in number of ways

Head of IMD’s Climate Application Group, Dr Pulak Guhathakurta has said that they plan to do such a study for all the other states as well.[vi] They also plan to make small booklets that can be freely available for anyone to use.

There are a number of ways in which the present study could be improved. When SANDRP talked to a few professionals working in the field of RWH, a number of interesting suggestions emerged.

Amol Gowande a latur based civil engineer, who has been instrumental in running a campaign for rooftop rainwater harvesting during 2004 -2007, says “This data seems fairly generic. Mean annual rainfall is a figure which people are widely acquainted with. Arriving at a per sqft rainwater potential is thus not that difficult.” He further states “As a professional working in field of RWH I would rather like to have more accurate data on rainfall intensity. In my observation rainfall intensity and number of rainy days for Latur has changed drastically. I remember during 2006-07 I used 25mm per hour as a maximum rainfall intensity in while designing RWH systems. This year the figure seemed like a new normal. For almost all rainy hours it rained 25mm/hr. Even if the mean annual rainfall remained the same we received the entire rainfall in just about 10 to 12 days. As a result we observed that at several places pipes installed to cater for 3000 lt/hr of rainwater proved inadequate for 10,000 lt/hr of rainwater. This is why for designing RWH systems – for arriving at thickness of pipes, design of filters – we need more precise information about amount of water as well as rainfall intensity. If a 2 inch pipe is replaced by 3 inch pipe, it changes the entire estimate.”

This is important, also, as this IMD document notes, “In the context of climate change several research studies including scientists of India Meteorological Department have reported significant increasing trends in the frequency of higher intensities rainfall events and decreasing trend of light to moderate intensity rainfall events.”

According to him accurate information on number of rainy days and distribution of rainfall over 24 hours for each day will be a very useful piece of information. This however might be difficult to produce he says. “IMD takes reading at every rain gauge once daily, which gives them aggregate of rainfall in 24 hours. That is not enough to generate data on hourly rainfall intensity.” 

Niranjan Upasani a Pune based RWH expert highlights need for another piece of information: “It would be really good if they can also give basic information about quality of the rainwater collected. E.g. rain at most of the areas in Pune is acidic. If I get to know what is the level of acidity, turbidity etc. I can accordingly take precautions in my RWH schemes.”

Another information that could be included in such a report is a map of areas where groundwater is contaminated. (Even though this is not an IMD expertise it might be worth including this info from other authentic sources) Study can thus highlight such areas saying that sump creation to store the rainwater is more feasible option for these areas than groundwater recharge.

Need for public participation in IMD data collection

If IMD wants to come up with more nuanced data useful for RWH, it may be worth creating more space for public participation in the present weather network. Presently there does not seem any space for public participation at the weather stations. Amol Gowande says “I would myself like to participate in monitoring rainfall and weather data. I have been trying for several years now. But IMD has no platform where commoners can participate. If they open up such space the accuracy of the information will increase multifold. Perhaps even real-time data can be made available for rainfall.” Irfan Sheikh of Pani Foundation told SANDRP “At the weather stations located in remote talukas rainfall reading is taken only once in the morning. Campus is locked otherwise. We never get to know when the reading was taken. Rainfall hours are notionally recorded by the weather station in charge. There is a lot a room for improving the data quality.”

Limitations of the present study

The study only gives roof top rainwater harvesting potential for buildings, that too at sq ft level. It does not provide the rainwater harvesting potential of Maharashtra. The title of the report, “Rain water harvesting potential for different locations in state of Maharashtra”, is misleading, it should be “Roof-top Rain water harvesting potential guide for different locations in state of Maharashtra”.  Secondly, it only provides figures for roof top rainwater harvesting potential for only certain kind, mostly urban or cement concrete kind of houses. Thirdly, rainwater harvesting is possible in many different ways, besides the roof top rainwater harvesting. But the report does not mention any of them, nor assess their potential. Fourthly, the report mentions depleting groundwater level and reducing groundwater recharge in changing rainfall pattern, but the report does not provide guidelines how to achieve increased groundwater recharge either through roof top rainwater harvesting or through other ways. Fifthly, the figures are given only for 326 locations for which the rainfall figures are available. However, it only calculates based on South West Monsoon or Annual Rainfall figures for the years 1951-2000, but for a report published in Nov 2016, it could have included the latest rainfall figures to arrive at more reliable figures. These limitations are in addition to the suggestions mentioned by some of the experts, as listed earlier. Report preparation also has a lot of scope for improvement. Nomenclature for the maps has gone wrong in the current report. Much of the text in preface, abstract and introduction is repetitive. Such errors could be avoided in future.

In conclusion: We hope IMD will pay heed to the limitations and suggestions mentioned here and make necessary changes. We also hope it will bring this information out in local language and take it to the panchayats, gram sabhas, ward sabhas and communities. At the moment it is available only in English, as the report says.

Amruta Pradhan, SANDRP,

(With inputs from Himanshu Thakkar)


[i] Authors: P. Guhathakurta, K. J. Ramesh and S. D. Sangale, see:






One thought on “IMD estimates rainwater harvesting potential of Maharashtra

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.