 First of all, VBMP and BWSP, they were very helpful in providing the data and it helped that Indian Institute of Science, Indian Institute of Management, they're all Indian institutes and very well respected. The PIs, my co-PIs are so well respected, they didn't have a problem. The only thing was that despite all these positive things, actually getting the data always took a lot of time and effort. We didn't have much idea of quality control, for example. If there was a GIS or spatial unit generating some data that might not be in sync with the time series or Excel data, so you try to fill in and you don't know really who to talk to, who might know how to verify the data that we've been provided, or even what the units were in some cases. So those are endemic problems, I think. I'm not sure when those will be solved but a lot of the time is spent in trying to get hold of the data and understand what it is actually. Then when it came to the primary data collection, so the survey that was run by Deepak, that survey I think took about a year to finish. It took several months to design and to implement and we're just about getting the results now, so it was a very long drawn out process. And then when it comes to the groundwater monitoring part, our field assistants have shared that it is so difficult to go around the city every month to 150 different locations. They are so efficient that it takes only minutes at each location but they can only do five to seven a day because that takes up about eight or ten hours in traffic and so on. So it's been a very intense effort from both the supply and demand side of our data collection and that I think gives you a good idea of the kinds of challenges we generally face when we're trying to put up this kind of information. So why is data important? If we want to manage a system, you have to understand its components, you have to be able to quantify things. And in the Indian context, it's well known that we don't know much about the demand side, we don't know much about the supply side, especially when it comes to water. It's one of the most difficult things to tackle. So the best way to understand how data can be useful in finding solutions to problems on the one hand and also in planning ahead, I'll give you an example of each. The first example I can think of is from Bangkok. Bangkok is a city about the same size as Bangalore today and before the utility used to supply just groundwater. This utility is called MWA and in the 1950s onwards, groundwater use started increasing and over the next decade or two, groundwater levels started dropping. Now, the interesting thing about Bangkok though is that they had a lot of knowledge and data over several decades about groundwater levels, groundwater quality, what kind of aquifers there were, how many aquifers, seawater intrusion, all kinds of things. Researchers from Thailand as well as outside of Thailand, many countries, they know a lot about the biophysical system of Bangkok. So even from the 70s, they had about 60 monitoring stations and so on. With the help of this data, once they reach this crisis, the water levels are falling to about 40 meters by the 1990s in some areas. There was ground subsidence because the aquifers were compacting from extracting too much water and so the national government came in and they started instituting new groundwater regulations. They started with the help of the data network, monitoring network, they identified critical zones in different parts of the city where they imposed restrictions on groundwater pumping. Over about two decades or so, the MWA stopped supplying groundwater to its customers and they brought in an alternative water source. So that's of course an important part of the story was that if you stop pumping groundwater, then somewhere else they have to supply water. And what you started seeing was they also started imposing groundwater taxes, groundwater use taxes, groundwater conservation charges. And you can see in a 30-year trace of groundwater wells that I've got that in different zones when they implemented these policies, the water levels start bouncing back up, the groundwater levels. It's quite a beautiful graphic to see that you have the timeline of these policies over the past 20 years and then the groundwater levels are actually responding back. So in many parts of the city, groundwater levels have actually recovered after all this time. That's one example I can give you of how data and knowledge can help remedy a situation, improve things, find solutions. And in a planning context, I've got a different example. And this is in the context of the fact that even if you collect a lot of data, you still need an interface, like the science policy interface. You need some sort of institutional interface so that that data and knowledge is actually going to implement policy, is going to help implement sound policy. You still need an institutional structure. So the example I've got is known as the public interest in energy research program called the peer program. And it comes out of California. Now, California has a very vibrant, proactive climate change policy. And for many years they had this peer program. It was funded with a small surcharge on every California residents energy bills. This small surcharge came to about a millions of dollars a year. This funded the research. The research was competitive. You had to apply competitively for these grants every year. And if you got it, you were directed to do research on something like six sectors related to climate change, like agriculture, water supply, infrastructure, etc., etc. This program was a very successful program and all the outputs of it were peer reviewed. You had to be published in top class journals like climatic change and so on. So the scientific rigor was verified and peer reviewed. You couldn't just do anything and get away with it. And then they set up a virtual California Climate Change Center. And it really helped inform California's very advanced climate change policies and still does to this day. It helped set standards for building standards, efficiency standards, renewable energy standards. It helped a lot in actually having standards, mandates, policies, laws being passed. So I gave the example of this peer program. It was this small surcharge on the energy bill. Now, let's think about this. I think that in Bangalore, the number of domestic water connections might be about one million. Let's say you would take. So if you had every connection, just domestic connection, pitching 10 rupees per connection per month, that comes to about 12 crores per year, you would raise in funds. And now if you related to what Dr. Shekhar said would be the cost of an automated monitoring network at the same 150 locations we are doing manual sampling of groundwater levels. He had estimated 20 million a year, which is about 2 crores a year. So you can see that even this little surcharge on the water bill could easily fund an automated monitoring network. So you now start seeing the possibilities, right? And if you take it further, let's say this is a program that goes on year after year, right? So the first year could be focusing on monitoring, correct? Second year could be a program that is maybe voluntary based, where people want to exchange their inefficient old Borewell pumps with new efficient pumps. You'll save on energy costs, you'll save on energy, you'll save on emissions. Maybe that could be funded through some carbon program, you know? And the possibilities are really right there if you have some kind of tight linkage like this and this could be funded because Bangalore is a large city of 10 million or so. So we should be able to generate funds. It's a Silicon Valley in all of India. We should be able to easily generate the funds. I don't think that it would be difficult. We could really try to see how we could emulate programs like PURE in India. So the lessons learned from this exercise is that without going for an automated network, without having our own captive Borewells drilled by under the project as a piezometer, we managed to show that we could get a lot of information about groundwater from existing Borewell networks in a city and which were not under use. Plus also use a human engineered way of monitoring and through the lesson we know that it's difficult but it's possible. So coming forward from this we need to, going forward, we should keep some amount of manual monitoring and some amount of automated monitoring. The amount of monitoring which we achieved in this project appeared to be good enough at a density of one per five square kilometer seems to be a good network in able to understand that we are able to capture the variabilities. The temporal frequency is something because we are doing humanly monitored. We were only able to do once in a month. It was not viable to do even in thought in our sense. We should do much finer sampling which would require much more number of people and much more travel in the city. It's not about the man hours involved or the man power involved. It's about the travel time taken to visit these spots and come back which is more all to us. I think at some point difficult places can be put up with automated networks and some can be manually managed so that we have a blend of these two creating a robustness in the network. And we are initiating to do this through the future projects. In the future projects we are looking for continuing this saga by blending them with additional automated networks. We can continue what we are doing especially in areas which are hot spots of the city or to re-investigate something which we have missed out in the current investigation. So there are many host of other things which we have not investigated in this two-year period. That can become a new mandate of investigation or new hypothesis to be investigated which can be improved through the smart network. Among the lessons learned many cities should have this kind of a network. We just started with the Bangalore as an example. The usefulness of this kind of information about groundwater could be brought into some kind of a participatory management in other cities. So that way they will become more robust than institutionalized or a project-based monitoring networks. If that is able to be achieved in many cities then there is a huge opportunity of creating a crowdsourced data at a high granularity which can be used for analyzing the systems. The groundwater throws up a lot of information about the state and health of water within a city as a resource. We have spent some time talking about what Bangalore looks like and the intimate connection between human systems and the background biophysical system. I now want to turn to questions institutional, questions policy and questions governance. So what do we do with this? There are multiple ways to address this question. Often times I get asked after I paint this reasonably bleak picture of where Bangalore is headed in terms of how it uses water, so what? My usual response is even while I am not optimistic that the city can solve its problems, I remain very hopeful. Even without being hopeful I could not be doing this day in and day out. I hope that optimism are two very very different things. One can remain hopeful without being optimistic. Can some of that hope actually turn into optimism? Yes. I mean if we are actually able to get our institutions to think very differently from how we currently think through the water conundrum, there is actually hope when we can get a handle on our water conundrum. I and all my collaborators are very grateful to BWSSB or the public utility. Their contribution to our project has been immense. We could not have possibly done any of what we have done without BWSSB helping us with all forms of things and certainly with data. Let us ask a few institutional questions. Our public utility, the BWSSB, which is responsible for 1400 million liters a day of surface water coming into the city, what is the BWSSB thinking in terms of groundwater dynamics? Let me ask a very direct question. How many groundwater hydrologists does the public utility employ? The answer to that question is a zero, ZDRO zero. I mean some of these things have to change. I mean you need a mindset change that moves away from looking at water as an engineering technocratic problem to one that even in hydrological terms as a complex problem, this is not only merely an engineering problem. So from an engineering problem you have to recognize that this is actually a complex hydrology problem. From that point on you have to recognize that this is not only a hydrology problem, but actually a social hydrology problem. So once you are able to make these leaps in your imagination, then you have a realistic chance of grappling with the water conundrum. I mean thus far I have done nothing in terms of talking about what happens on the wastewater side. That is an important part of the problem. I mean scientists at Atree, I mentioned collaborating institution. I mean Atree scientists have been instrumental in helping us actually characterize this problem. For a number of years they have been working on understanding wastewater flows. I mean what is actually happening to water that is used, industrial effluents in our waste streams so on and so forth. Eventually we do need to take this problem of wastewater head on. One of the modules actually tried to understand people's attitude to be open to using wastewater. What we found was actually very revealing. I mean despite the obvious crisis that Bangalore faces with its water, less than a fourth of Bangalore's households are willing to consider using recycled water for any use at all, even non-potable uses. I mean this obviously has its roots in our social conditioning. But the point is that questions of sociology, questions of economics, questions of politics are fundamental to even understanding hydrology. I mean we need to internalize this. And there is no one question more important than asking simply what is water? Is it a private good or is it a shared commons? I mean today we absolutely treat water as a private good. And that's a fundamental category error. I mean water isn't a private commodity. I mean water is a shared commons. And once you start understanding water as a commons and understand that there are social, political, economic drivers of what we do hydrologically with this commons, then I think you have started to put yourself on a path to be able to understand the conundrum. And then you might, one hopes, that you might actually be able to grapple with it and also come up with interesting solutions that are equitable, sustainable and efficient. Well, you've been working for five years collecting a lot of data. So what does this all mean? As an academic I obviously love data. I mean data is crucially important. We understand very little about what's happening to our groundwater aquifers. So the, I mean I actually stick my neck out and describe it as a valiant effort led by Shaker and Vishal to try and get a preliminary sort of understanding of changing groundwater levels in the city. That's a part that needs to be institutionalized. I mean data there is clearly important. I mean unless we are able to continuously be aware of what's happening to water beneath us, I mean I don't see how we are going to be able to institutionally act. I mean when data is certainly knowledge, data is not wisdom. I mean let me illustrate this with a parable. Assume you're jumping out of an airplane. It is fantastic to be able to have the best altimeter available around. But the one thing that you really do want to have, you can do without an altimeter, but one thing that you really need to have is your parachute. So the argument that this project has been making is there are important gaps in terms of altimeters, but the real gaping hole is we don't have enough parachutes. So parachutes are basically in terms of what we've been talking, institutional changes, lasting institutional changes, lasting governance changes. I mean to basically fundamentally look at waters as a shared common. And then peel through layers of what it means to look at waters as common. Obviously there are important parts of the altimeter part of the puzzle that needs to fall in place, but I think that's easier. Altimeters are easier to conjure up than parachutes. So one of the things that I hope all of us will do on the project team is to try and look at how we might be able to come up with parachutes. We have some altimeter data now. Can we also use that to fashion sustainable parachutes?