Bangalore Floods – A Call for Open Data
October 27, 2022 Vaidya R
A few days of rains in the first week of September and India’s IT capital, Bangalore, got the attention of the international media. The Wall Street Journal headline read – India’s Silicon Valley Flooded by Rains, Exposing Infrastructure Woes. BBC was more dramatic – Boats and tractors replace cars in India’s Silicon Valley.
A week prior to that, a few hours of rain early in the morning had buried Kochi city underwater. The flooding inundated neighborhoods, caused power outages, traffic blocks, and delayed trains for hours. It did not make news outside of Kochi. For Kochi residents, flooding is an annual event to be expected during the monsoons.
A Tale of Two Cities
Bangalore lies on a ridge on the Deccan Plateau and is 949 meters above sea level. Kochi is on the Malabar Coast at sea level. Bangalore is a city of lakes interconnected by streams. Kochi has a network of canals that drain into the Vembanad Lake and backwaters. Both are growing cities where urban infrastructure has lagged growth. The growth has over the years encroached the lakes, streams, canals, and the backwaters.
The reactions in the two cities couldn’t have been more different. In Bangalore, after an initial period of angst, fingers were pointed, social media went into overdrive to activate citizens to map flooded areas, JCBs were prominently shown on TV demolishing unauthorized structures, the state government announced setting up a committee of technical experts, and ordinary citizens became acquainted with hydrology and the hydraulics of stormwater drains.
For Kochi residents, it was déjà vu. They had survived the 2018 floods; stoic resignation had replaced angst. A technical committee under Metroman Dr. E. Sreedharan had been set up after the 2018 floods, and the Kochi Corporation had announced Operation Breakthrough to prevent future floods. Kochi residents had seen videos before of JCBs clearing encroachments and de-silting canals in the media under Operation Breakthrough.
When asked about Operation Breakthrough and why it hadn’t prevented the flooding, the Kochi mayor said that Operation Breakthrough was only targeted towards areas known for frequent inundation in the past and the intensity and duration of the rains had caught them unawares. In Bangalore too, the authorities had been caught unawares. Bangalore was forecasted to receive normal rainfall of around 3 cm in the first week of September, but it received 13 cm of rains.
A Lesson in Hydrology (Rainfall and Flooding)
Data models for predicting rainfall and identifying flood prone areas are not as reliable today. With global warming, every 1° C rise in temperature increases evaporation and results in more than 7% increase in a cloud’s moisture carrying capacity. We have to expect and plan for climatic extremes – cloudbursts and atmospheric rivers may be the new normal.
Any precipitation ends up soaking into the ground or as runoff. Water that soaks into the ground will eventually find its way into underground channels and replenish the groundwater. Water that falls on saturated soil or impervious surfaces ends up as runoff and must find its way into a city’s stormwater drains.
Increased urbanization affects the hydrologic processes. When the natural landscape is replaced by impervious surfaces like roads, buildings, parking lots, and pavements, it reduces the infiltration of water into the ground and increases runoff volumes. In overly built-up urban environments, water that soaks into the ground may find its path blocked by underground concrete structures (underground basements and parking lots, or concrete rainwater tanks). In these cases, the ground gets saturated and any extra precipitation will add to the runoff. Runoff that cannot flow into a city’s stormwater drain network causes flooding.
In natural landscapes and open areas, topography, soil composition, and gravity determine the path that the water takes. In urban built-up environments, the path of the runoff is determined by the presence and capacity of the stormwater drain network. A well-designed and maintained stormwater drain network prevents floods, recharges groundwater, and maintains clean water bodies.
Under Operation Breakthrough in Kochi, areas were identified that were known for frequent and severe inundation in the past, root causes were determined, and targeted interventions were done.
The Bruhat Bengaluru Mahanagara Palike (BBMP), the Bangalore city corporation, administers over 800 kilometers of stormwater drains and another 1500 kilometers of secondary drains. In 2020, BBMP went through a similar exercise and published a list of over 200 of the most flood prone areas in Bangalore. The list also identified the root causes for the vulnerability.
I extracted, cleansed, and ingested this data into the Hawkai Data platform and overlaid the predictions to the actuals. The blue dots are the predicted areas and the red indicates the actual inundated areas in the Mahadevapura Zone in Bangalore. The interactive map also indicates the cause for flooding and the expected radius in meters.
The BBMP report did not make available the raw data or specify the process by which they analyzed and identified the most flood prone areas. The mismatch between the predictions and the actuals is very likely due to incomplete or non-current data being used in the prediction model. The intensity of the rains should have only affected the flooding radius around the predicted areas. Of more interest, though, is what the data tells us about the root causes for flooding.
Going by the BBMP’s own data, encroachments account for only 4 – 5 % of the problem. Encroachments and buildings torn down by JCBs makes for good TV and plays to the public’s sentiments, but in reality addresses only a small fraction of the problem.
Over 60 % of the identified vulnerabilities are due to silting in the drains and rainwater storage and recharge wells. Silting reduces the carrying capacity of the drains, canals, and lakes, and will eventually lead to blocked drains and floods. Annual maintenance contracts are tendered for de-silting the drains and canals. De-silting contracts rarely specify where the silt and waste should be disposed. Without oversight and enforcement, it becomes a game of whack-a-mole where a de-silted drain in one area leads to a silted canal or lake in some other area.
Cities must make areas available where the silt and waste from the drains can be disposed. This is in the city’s interest – the city gets valuable data in terms of what is getting dumped into the drains, how much, and where it is getting dumped. This data can be used to increase public awareness and targeted enforcements to prevent dumping of waste in the future. Only when the stormwater is free of waste and pollutants can it be released to replenish groundwater.
Tragedy of the Commons
In 1968, in an influential essay published in the journal Science, Garrett Hardin introduced the term “The Tragedy of the Commons”. The Tragedy of the Commons is a social problem when an individual is incentivized to act on a shared resource in a way that is ultimately harmful to all individuals.
A few months back, I was remodeling my apartment in Kochi. The remodeling included demolishing three bathrooms to upgrade the plumbing. At the end of the week, I had bags full of concrete, broken ceramic, and tiles that I had to dispose. I googled “kochi kerala construction waste disposal” and learnt that the city did not have a facility or even a formal process for disposing construction waste. I asked around and was given two options. My apartment is right by the backwaters – I could hire a daily wage laborer to carry the bags and dump the contents into the backwaters late evening. Or, I could call a construction debris waste removal company to truck it and make it their problem where to dump it. Construction waste disposal companies routinely dump debris on wetlands to reclaim and encroach.
De-silting the drains can be effective only when we residents stop dumping waste into the canals and drains. For residents to stop dumping waste and garbage into the canals and drains, the city must make it easy to dispose waste sustainably. When a city proposes sustainable waste and sewage disposal, the NIMBYists oppose it. It is a perfect Catch-22.
Take a walk in any rapidly urbanizing or gentrifying area in India today. It is not unusual to see restaurants dump cooking grease down stormwater drain inlets. With the number of cars exceeding parking slots in most apartment buildings, it is not unusual to see cars parked on residential streets blocking the stormwater drain grates and the flow of water. It is not unusual to see new home construction sites where trucks dump sand and gravel on the road and have it washed off into the drains in the next storm. It is just the tragedy of the commons playing out.
Public awareness and citizen engagement are key to managing the shared commons. When a city resident is made aware of the impact of their actions on the commons, they will take more responsibility towards it. The images below show stormwater drain grates from Boston and San Jose.
Stormwater drain grates are clearly marked with a warning, and tell you the water body that it connects to. This raises public awareness about the connection between the stormwater drain and the local water body. When a fish-loving Kochite internalizes the link between the stormwater drain inlet and the Karimeen (Pearlspot) that they fish and eat from the Vembanad Lake, it will give them pause. Over time the signage will activate behavioral change in the community.
Garrett Hardin later retracted and admitted that “The Tragedy of the Unmanaged Commons” was a more appropriate title for his thesis. He asserted that when shared resources and common areas are managed and their use regulated, it should not end up in tragedy.
Kerala Coastal Zone Management Authority (Kerala CZMA) is the regulating authority for the Kochi backwaters and wetlands. Their website (http://keralaczma.gov.in/) has a 1 and a half star Google review rating, is extremely difficult to navigate, and has coastal zone markups from 2011 that can be downloaded as PDFs. In rapidly urbanizing cities like Kochi, a 10-year old map is pretty much useless.
BBMP also provides data and maps as PDFs. In a PDF, you cannot easily separate the data from the underlying map layer. Coastal zone markup data or the stormwater drain network markup data is only useful when shown on updated and current map tiles.
Closed data empowers the authorities. Unlocking the data and making it open and accessible, empowers the community. When data is separate from the map, the data can be easily integrated into other applications and services. It allows for data to be overlaid on any kind of map tile for analysis. You can use a street view, topography, satellite, cadastral, or some other type of map that is most appropriate for the analysis. The figure below shows how the base map tiles can be easily changed in the Hawkai Data platform.
Cities must unlock and open their data for public use. Public awareness, citizen engagement, and open data will allow communities to manage their shared physical and digital commons.
A Call for Open Data
Open data is data that anyone can access, use, enhance, and redistribute. Open data provides transparency and accountability to government policy. It allows governments at all levels to target interventions and direct investments to maximize impact. Once the interventions are done, open data provides the ability to measure the impact of the intervention and the return on the investments.
Many countries today have some form of Freedom of Information legislation. India has its own Right to Information act. These laws were formulated on the principle that an informed citizen is an engaged citizen; and an engaged citizen is vital to the functioning of a democratic society. The UN Sustainable Development Goal 16 has a target to ensure public access to information as a means towards effective, accountable, and inclusive institutions.
Many cities today have an Open Data initiative that makes city data available to the public. The data includes locations of city assets, transit data, 911 call data, reported road accidents, etc.
The city of San Francisco makes public the locations and type of each tree that the city maintains. Urban trees reduce urban heat and also slow rainfall through its canopy and can reduce the runoff. The species and gender of a tree can have an impact on the air quality in an urban area and may also contribute to localized flooding. Domination of male trees leads to increased pollen counts, which leads to incidents of urban pollen allergies. The Sweetgum (Liquidamber) tree produces vast amounts of spiked spherical fruits that often clog stormwater drains and cause localized flooding. Open data allows city planners to integrate trees into stormwater management design.
The city of Los Angeles makes public their complete stormwater drain network. A property owner or a buyer can easily determine whether a property has an easement. The figure below shows the stormwater drain network across Los Angeles. The map can be drilled down to the level of a property.
The city of Berkeley, California, provides the locations of all the manholes in the city as open data. This data was ingested into the Hawkai Data platform and the figure below shows the locations of manholes in a Berkeley neighborhood. The locations are provided as a latitude longitude pair with at least 7 decimal digits, which gives centimeter level accuracy.
Stormwater elements include inlets, manholes, drains, culverts, canals, rainfall recharge tanks, outlets, and water bodies. Trees, swales, and wetlands also play their part in reducing runoffs. Digitalizing the locations and attributes of these elements allows for integrated stormwater management that can help to mitigate flooding and achieve water quality goals for a city’s water bodies and groundwater.
Real value comes when solutions are integrated around a shared data and technology infrastructure, which allows for different departments and stakeholders to collaborate and solve problems. Open Data initiatives require a technology infrastructure that removes organization silos, that uses open data schemas to avoid vendor lock-ins, and that makes data available and accessible to government, citizens, and services.
Modeling City Assets
Open data has structure. The structure can be explicit by making a data dictionary available with the data, or it can be implicit when it is provided in formats like JSON, GeoJSON, KML, etc. This structure, or schema, ensures data integrity and allows applications and digital services to integrate the data and use in visualizations and analytics.
Any city asset can be mapped to one of three types of data schemas – a point, a shape, or an event. A city asset can be a point of interest, a shape that defines a boundary or outline of how the asset is laid out, or an event of interest. The point and shape data drive the spatial analytics and the event data drives the temporal analytics.
- Point Data Schema – The Point Data schema is used for city assets that are points of interest. The Point Data schema will include a location attribute (a latitude and longitude that identifies the location or an address that can be geocoded to a latitude and longitude) and a set of attributes that describe the asset. Examples of points of interest include, inlets, manholes, outlets, rainwater storage tanks, sewage treatment plants, etc.
- Shape Data Schema – The Shape Data schema is used to represent data about geographic features, their properties, and their spatial extents. City assets like parks, water bodies, water and sewage networks, coastal zones, etc. can be modeled using the Shape Data schema. The Shape Data schema includes a shape attribute (a GeoJSON representation of its spatial attributes), and other attributes that describe the asset.
- Event Data Schema – The Event Data schema is used to represent time-series events. The Event Data schema will have a time of day attribute, a location attribute (or some kind of identification that can be converted to a location), and other attributes that describe the event. The Event Data schema is used to represent reported issues like leaks, floods, etc. Any IoT sensor reported data would be of this type. The Event data typically provides the attributes that can be used to reflect the state of an asset at a particular time or over a certain period of time.
Any city utility or function can be modeled based on these three types of data schemas. To model more complex systems might require all three types of data schemas. For instance, a city’s water supply consists of sources of water that may include lakes, rivers, groundwater, etc., water purification facilities, water storage facilities like reservoirs, water tanks and water towers, pumping stations and a pipe network for piping water to the consumers and other usage points like fire hydrants.
When modeling a city’s water supply, the sources of water would be structured as a Point Data schema, the distribution network as a Shape Data schema, and customer reported issues and work orders as Event Data schema. In addition, water consumption points like fire hydrants and utility customers would be structured as Point Data schema records.
To properly analyze, find gaps and root causes, predict and target interventions, the data must be good. Good data is data that is consistent, correct, complete, and current. Good data will always trump more data. Good data will also trump the best model. You can have the best prediction algorithm, but the quality of the results will depend on the quality of the data.
Activating the Data
With a well-defined and published schema, open data can be analyzed and visualized. It can also be activated. Activation is when the same data is used to create new customer experiences, applications, and services.
A New Jersey utility has mapped all its underground utility lines and built a digital service that allows field technicians using smart glasses to ‘see’ objects that are normally concealed; underground cables, pipes and sewers, wires hidden behind natural and other obstacles. The figure below shows how augmented reality allows utility workers to visualize underground cables and pipes; lines are color-coded where electric lines are red, water blue, sewer and drain lines green, gas lines yellow.
The utility made sure that the field data was good and with the data infrastructure in place, turned their focus to providing their field workers tools where they could see the hidden utility lines while remaining spatially and situationally aware of their surroundings. Since each of these lines is typically managed by a different agency, this shared interface provides a shared vision for multiple city agencies to collaborate seamlessly when executing fieldwork.
In the city of South Bend, Indiana, even the sewer system is smart. Sensors mounted under manhole covers can sense flows through its sewers and canals, and divert water to prevent street flooding and prevent sewage and untreated water from entering its rivers. The city has saved millions of dollars in what it would have cost to build tanks and bigger sewers by optimizing the flow through its existing sewer network.
The data collected over the years by the smart sewer system is used to optimize the cost of future sewer upgrades to keep up with regulations. This data has also allowed the city to implement a data-driven maintenance and cleaning program that has increased the performance of the sewer system. The sensors can also detect areas that are outside the sewer system’s observed footprint to pinpoint and troubleshoot issues that affect the carrying capacity of the sewers.
Public awareness, citizen engagement, and open data and open maps are key to effectively managing the shared physical and digital commons.
Most cities monitor their assets, sensors, and systems separately, usually using different applications and vendors. To be truly effective, it is necessary to create a citywide data platform that can share data, combine and analyze data from different sources, and be able to not just respond to but predict events in advance.
When services and solutions are integrated around a common data platform, the city will be able to manage its resources efficiently, deliver services faster, and provide transparency in governance, allowing for bottom-up citizen and private enterprise driven initiatives that improves the ease of living for its citizens and visitors.
Hawkai Data provides a data-driven platform that leads with the customer experience. The Hawkai Data Customer Experience Platform (CXP) simplifies and accelerates your digital transformations. If you have any questions, talk to us at firstname.lastname@example.org, or follow us on LinkedIn at https://www.linkedin.com/company/hawkai-data/, or connect with us at https://hawkai.net.
- Flash Floods in the wake of unrelenting showers sink Kochi, Staff Reporter, The Hindu, 30 August 2022
- Bengaluru: 58 places identified as highly prone to flooding, Niranjan Kaggere, Deccan Herald, 4 June 2020
- Explainer: What climate models tell us about future rainfall, Zeke Hausfather, CarbonBrief, 19 January 2018
- How open data, wastewater reuse, and other measures can prevent flooding in Bengaluru, Shreya Nath, Citizen Matters, 25 September 2022
- No scientific process yet to handle construction and demolition waste in Ernakulam district, Staff Reporter, The Hindu, 24 June 2022
- Where do our storm drains lead to? Read the plaques, Samuele Petruccelli, The Dorchester Reporter, 4 February 2022
- Authorities design more sewerage projects for Kochi, MK Sunil Kumar, The Times of India, 8 July 2022
- New abattoir gets Kerala Infrastructure Investment Fund Board nod, MK Sunil Kumar, The Times of India, 25 September 2022
- Mixing the Real with the Virtual, ArcNews, 2018
- These Smart Sewers Are Part of A Growing Trend Connecting Infrastructure to Internet, Jennifer Weingart, NPR, 8 May 2018
* Header background image photo credit: Dipankar Roy/Unsplash
* Boston Harbor stormwater drain image credit: Robin Lubbock/WBUR
* Augmented Reality image credit: ArcNews
(This story has been republished from LinkedIn with permission from the author Ranjit John, Founder of Hawkai Data. The original article can be found here.)