Crowdmapping Bengaluru’s Lakes

Here’s how we collated lake data from different sources to put together a comprehensive open-access dataset.

Rashmi Kulranjan and Shashank Palur

Illustration by Aparna Nambiar

How many lakes do you think there are in Bengaluru? You’re not going to find the right answer on any government website or document. For a city famed for its many water bodies, there is no comprehensive dataset that provides an accurate number. Until now.

With the help of citizen volunteers, CSEI’s Urban Lakes Initiative collated lake data from different sources to identify 1,350 lakes in Bengaluru. These lakes are spread across Bengaluru — 777 lakes in Bengaluru Urban and 573 in Bengaluru Rural. How did we arrive at such exact numbers? This blogpost documents that effort.

You can access the Bengaluru Lakes Masterlist here.

Bengaluru’s loss of water bodies to encroachment, degradation, shrinking or flooding has been well-documented. The lack of reliable records is having tangible impacts on the ground, as poor planning and enforcement bulldozes what’s left of the city’s water bodies. The key to protecting existing lakes is first documenting them, and then ensuring that this data remains openly accessible to the public.

We wrote about this in more detail in this article published by IndiaSpend: Crowdmapping Bengaluru’s Vanishing Lakes

Why was this necessary?

There are multiple data sources for Bengaluru’s water bodies because the city’s lakes are governed by multiple agencies (here’s a primer on the agencies to contact to address lake issues). But much of this data is incomplete, incorrect or outdated, which is why there is a need to verify their existence with historical and present images for each lake. This is a mammoth task to carry out for a large metropolitan city like Bengaluru. A more collaborative approach can considerably speed up the process — which is what we did.

But speed isn’t the only benefit of crowdsourcing this data. It made participants, all residents of Bengaluru and students based here, an active part of data gathering and helped them appreciate the complexity of the problem and the urgent need to protect existing lakes.

What we did: step-by-step

In the spirit of keeping this as open and accessible as possible, we did not want to confine this exercise only to those who were familiar with mapping techniques and GIS tools (Geographic Information Systems spatially creates and analyses data). So, we first needed to organise a training session.

Shashank Palur, one of the researchers who led this project (and an author of this blog post), conducted six GIS training sessions over six weeks online. Over 35 participants signed up and learned basic techniques such as projection systems and shapefile delineation. This was the process we followed to put together the database:

1. Data on lakes was available from multiple sources but each of them had a different issue. Our dataset was created using the list provided by the erstwhile KLCDA (now the Karnataka Tank Conservation and Development Authority), Google Earth, and toposheets sourced from the Survey of India.
2. The KLCDA list contained names and locations of only around 210 lakes within the BBMP boundary. On verification, it was found that the locations mentioned in the list for many of the lakes were marked wrongly. It was further found that many lakes in the list were either missing, had been encroached upon or had dried up.
3. Lakes where location details were available were then corrected — after which GPS points were edited and moved to the nearest water bodies. Using these points as a reference the lakes were marked on Google Earth. This was further verified using toposheets.
4. Relevant details like ward information, citizen lake groups involved in their management including their contact numbers and social media handles were also collected and added to the database.
5. The second source we used was geo-referenced toposheets, which we obtained directly from the Survey of India (these can be downloaded for free from here but are not geo-referenced). The dataset contained the location and boundary of both perennial and seasonal lakes in Bengaluru Urban and Rural districts.
6. The volunteers were then taught how to identify perennial and seasonal urban water bodies from geo-referenced toposheets as part of the training session. The identified lakes were named as per the names on the toposheet or according to the nearest settlement to the lake. Citizen volunteers mapped the lakes beyond the BBMP’s administrative area to cover all Bengaluru Urban as well as Rural districts.
7. Finally, the existence of these lakes was then verified by our team using Google Earth.

Challenges

• The volunteers came from different backgrounds (students, homemakers, IT professionals, activists) and had varied skill levels (beginners, intermediate, and experts).

We solved this by organising multiple troubleshooting sessions for the volunteers.

• There were many errors in the toposheets because they are hand-drawn maps.

We verified and corrected data using Google Earth.

• To make the dataset as comprehensive and useful as possible, we also wanted to include updated contact details of relevant authorities. Gathering this information for ward councillors in Urban Bangalore and Panchayat members in Rural Bangalore district took time.

We have not added these details for every lake; we’ll update as we find the most current information.

Mapping for better planning

One of the big problems hampering integrated urban water management is that different stakeholders are not working off the same base layers — be it urban planners, researchers, citizen lake groups. All of them need maps but they work off different data, often in isolation

Read | Bangalore’s lakes: What role do citizens play?

For example, urban planners need to know where to build sewage treatment plants, they need to know where flooding is likely to happen, conduct land suitability analysis before planning where to expand existing cities. Students want to do research; citizen groups want to save their local lakes, understand the cascading system and plan nature-based solutions. Anyone who needs to do this kind of assessment needs a comprehensive map of the region’s lake system, which is what our crowdsourced database sets out to do. It establishes a common layer that gives everyone a common language and framework. This allows the flow of information between government bodies and the public more easily.

Maps are a great starting point for monitoring, supervising and planning restoration projects. It can empower citizens with information about the current state of lakes and help them monitor changes in the lakes over time. This is key for fighting encroachment as well as getting a better understanding of the different factors that affect each lake’s health.

Read more about Bengaluru’s lakes and the way forward in the insight articles we have published here. In a follow up to this blog post, we will discuss the next phase of our journey of tracking the historical changes in lakes, like identifying encroached and vanished lakes and the change in the seasonality of lakes. You can also follow the Mira community on Instagram for more on nurturing Bengaluru lakes.

We would love to hear from you. To collaborate with us, write to csei.collab@atree.org. If you’re interested in contributing to this initiative, reach out to us at lakescourse2020@atree.org. You can also use this form to contribute data to this open source project.