Shrinking shape of water

Category: Best data-driven reporting (small and large newsrooms)

Country/area: India

Organisation: The Hindu

Organisation size: Big

Publication date: 30/06/2019

Credit: Vignesh Radhakrishnan, Raj Bhagat Palanichamy, K. Lakshmi

Project description:

In 2015, Chennai – the capital of Tamil Nadu, the southern most State of India – faced a devastating flood which killed hundreds and displaced many more. However, in 2019 it faced a water scarcity. While climate change is partly to blame, government apathy combined with public disregard for water resources resulted in bone-dry reservoirs, sewage-polluted rivers, lakes shrinking due to construction. While the government needed to be reminded of its responsibility towards its citizens, people need to know the extent of damage inflicted upon the water bodies due to their callous nature. In two parts, we achieved that. 

Impact reached:

The story was shared widely on twitter and facebook. It was also noted and appreciated by opposition members of the State assembly. However, the government was as usual silent.

Interestingly, north-west monsoon rains in 2019 exceeded the expected levels in Chennai. This time around, the residents were ready. Many apartments in the city had rain-harvesting technology in place. Reports of individual houses and apartments saving lakhs of liters of rain water and using it for daily needs were heart-warming. However, what role the two-part story played in this new beginning cannot be measured. As The Hindu, the paper I work for, is the number one daily in terms of circulation in Chennai, i think we can safely assume that it did play a significant role. The groundwater levels too improved at the end of 2019 compared to the levels seen in the previous years. 


Techniques/technologies used:

Satellite images from Copernicus Sentinel, LandSat (USGS/NASA) were procured for various time periods by zooming into various lakes and water bodies in Chennai. Using the maps, water surface extent was calculated and compared across time periods. While the former method was used to visually convey the decrease in water extent, the latter was used to convey the same using hard data. 

In the second part of the story, we mapped how construction projects by both the government and the civilians reduced the city’s lakes in size. To do this, the old maps of the lakes were snapshot from Digital Globe and Landsat and processed using google engine. Comparisons with lake extent in 1954 (sourced from a map developed by the US govt at that time) and satellite maps shot in 1973 were carried out. 

Similarly, the small rivers which criss-cross the city were mapped and all the sewer canals which drained into them were identified and plotted. The authorities can now precisely know the meeting point of sewer drains and rivers and curb the menace. 



What was the hardest part of this project?

The hardest part was collection of old maps and making sense of them. Also, it was tough to set the exact zoom levels thereby enabling comparison of maps. Cloud cover was a challenge, because similar dates must be maintained as much as possible to make the comparisons valid (if the dates are not similar then  comparison becomes a farce as seasons vary wildly in India). Digitising old maps was difficult, especially mapping the lake extent from old maps was a challenge. If it was not done accurately, then it cannot be compared to present day maps to show the level of reduction in lake levels due to construction of buildings. Rivers that criss cross Chennai had to be followed inch by inch to see where the sewer canals meet them. It’s a painstaking process given that sewer canals are extremely narrow and thus wont get highlighted in google maps often. 

What can others learn from this project?

Random collection of old maps showing the extent of river and lakes may sound futile, however that’s precisely what helped in making the core of our argument. Collect old files whenever you come across and archive them using referenceable labels, one day you will find a use in them. 

Project links: