Telangana’s Politics Of Water, And A Problem Of Timelines
The Priyadarshini Jurala dam in the Krishna Basin, Mahbubnagar, Telangana. (Image: Water Resources Information System of India)

Telangana’s Politics Of Water, And A Problem Of Timelines


In Telangana, politics is water, and water defines its politics, in a connection that stretches back through the sea of history. The Kakatiya Dynasty, who ruled Telangana from 1163 AD to 1323 AD realised, like other kings of hilly, forested regions with temperamental, highly-seasonal rainfall, that water storage held the key to prosperity. To that end, they built tanks – thousands of them. Tanks functioned alongside temples – the ruler or a noble established the tank as an act of charity, and the maintenance was left to the people, often with highly codified community responsibilities and rewards. This equilibrium lasted, for the most part, for centuries.

How Telangana Lost Its Water

This began to change with the British entry into India. Written in 1861, “Nizam – His History and Relations with the British Government, describes the Hyderabad principality thus:

soil in general is extremely rich and fertile; and, except where the tanks have been allowed to fall into decay, the country is well watered.”

It appears the British did not appreciate the beauty of cascading tanks. Instead, they appear to consider tanks an extravagance:

“individuals have spent enormous sums of money in the construction of tanks, from the childish desire which the natives generally have of perpetuating their names for beneficence.”

The British philosophy was rooted in controlling natural resources, which manifested in command-and-control type of irrigation projects and massive deforestation. A series of devastating famines led to the construction of the Cotton Barrage, which made the Godavari navigable—and allowed easier transport of timber and cotton to the ports— while simultaneously creating a rice bowl in coastal Andhra Pradesh.

With the focus firmly on canal irrigation, tanks began to lose popularity, and Telangana, her water storage.
The Yerravagu Project in the Godavari Basin, Adilabad, Telangana. (Image: Water Resources Information System of India) 
The Yerravagu Project in the Godavari Basin, Adilabad, Telangana. (Image: Water Resources Information System of India) 

Even after the British left, Telangana continued to deforest rapidly, losing half its forest cover between 1930 and 2013.

A lack of trees and tree roots meant less silt was held back, and more silt flowed down the rivers, making tank maintenance harder. Meanwhile the government continued to emphasise on dams and canals over tanks.

This resulted in the shift of water resources from Telangana to other parts of Andhra Pradesh.

Take the case of the Krishna and Godavari rivers in erstwhile Andhra Pradesh.

  • While 69 percent of the Krishna’s catchment area lies within Telangana, only 35 percent of Krishna’s water was allocated to Telangana.
  • While coastal Andhra has 13 percent of the catchment area, it received 49 percent of the Krishna waters.
  • 79 percent of the catchment of the Godavari, the second longest river in India after the Ganga, lies in Telangana, but, here too, coastal Andhra became the major beneficiary.
The Priyadarshini Jurala dam in the Krishna Basin, Mahbubnagar, Telangana. (Image: Water Resources Information System of India)
The Priyadarshini Jurala dam in the Krishna Basin, Mahbubnagar, Telangana. (Image: Water Resources Information System of India)

Also read: How India Could Cut Irrigation Water By 33% – And Reduce Anaemia, Zinc Deficiency

There were other issues that shifted the equilibrium. In a cascading tank system, linkages are key; continual encroachment – both of the connecting channels and the tanks, together with a breakdown of the social systems lessened effective tank functioning: After all, why spend labour desilting a tank, when the water may not flow because a powerful upstream farmer decided to break the bunds, and take the water for himself?

Hungry for water, farmers discovered a timely, technological miracle: the borewell.

In a few decades, wells and borewells assumed primacy in the irrigation systems of Telangana, as seen below.

Telangana’s Politics Of Water, And A Problem Of Timelines

But it was not enough. Even today, 63 percent of the crop is rain-fed and exposed to the whims of the weather gods. A farmer with a 5-acre, rain-fed plot in Telangana looks to the sky and then downstream at the lush, irrigated paddy fields and bemoans his fate, and the priorities of his leaders. Given that 56 percent of the workforce is in agriculture, this envy translates to a powerful political force.

This is the force that helped create the Telangana state and brought the Telangana Rashtra Samithi to power in 2014.

The state’s political map clearly shows that the TRS owes its ascendancy to Telangana’s farmers.

Telangana’s Politics Of Water, And A Problem Of Timelines

TRS Finds A Solution In The History Books

The farmer’s profitability depends crucially on his or her crop choice, crop yield and crop price. Choice and yield directly depend on water availability, making widening irrigation access a key thrust of the government.

Renegotiating higher shares from existing large scale irrigation structures is unlikely to succeed, while building new major irrigation projects is expensive, lengthy, and runs into populist and environmental roadblocks because somebody’s home is submerged.

As a result, major irrigation projects, while offering rent capturing opportunities, do not mesh well with shorter political timelines.

Enter Mission Kakatiya, the large-scale, revival of 46,531 tanks, to improve the irrigation access for Telangana’s farmers.
(Image: Mission Kakatiya Official Facebook page)
(Image: Mission Kakatiya Official Facebook page)

Also read: India Suffering Worst Water Crisis In History: NITI Aayog 

On paper, the scheme that hasbeen named after the 12th-century rulers, ticks several boxes:

  1. It is farmer-friendly.
  2. The timelines are far shorter—years versus decades—making it politically relevant.
  3. There is no submergence of forest or village, making it politically feasible. The removal of encroachments is another story though.
  4. Done right, it improves water storage.

The last point may seem a little technical, so let me explain.

India, with its highly seasonal water supply, is woefully short of water storage.

India has only about one-eighth the world average per capita water storage. This is why tanks are a great solution.

  • For one, they collect rainwater from the area they directly drain, and allow the rainwater a chance to percolate into the ground, rather than ‘run-off’.
  • Second, a subset of tanks, called system tanks, are connected to a network of other tanks and to the river through canals.
  • These system tanks are the beneficiaries of surplus non-local rainfall.
  • During the southwest monsoon, the Godavari and her system of rivers swell as they capture the rain from their upstream regions.
  • This surplus water flows through a set of channels to tanks, and as each tank overflows, downstream tanks, also connected via channels get filled.

Off To A Good Start

Initially, as per all accounts, the Kakatiya Mission worked well. Several international universities, including the University of Michigan and the University of Chicago started studying the project, and reported positive early findings. The IWMI-Tata Programme reported positive effects on farm economics and groundwater, with a more nuanced impact on irrigation access from Mission Kakatiya when well implemented. Other studies reported an explosion in fish production, benefiting the 19 lakh fishermen in the state.

Anecdotal evidence suggests that the mission was perceived as a bold reset on water management.
(Image: Mission Kakatiya Official Facebook page)
(Image: Mission Kakatiya Official Facebook page)

Also read: Changing Monsoon Patterns Worsening India’s Water Crisis, Conflicts: Study  

In 2016, a taxi driver, with roots in Mahbubnagar, was quoted as saying, “My father had tears in his eyes when he saw the tanks full. In his life, he hadn’t seen so much water. Some of us are even considering going back since there is now real potential in the fields”.

The patchy data from the Water Resources Information System (some blocks are missing data for some years) supports his statement, as the chart below shows.

After Mission Kakatiya’s implementation in early-2015, the water table has risen again.

Incidentally, seven of Mahbubnagar district’s fourteen assembly constituencies, had voted for the TRS in 2014.

Telangana’s Politics Of Water, And A Problem Of Timelines

In Warangal’s Parkal block, small cotton farmers, owning land between 1-4 acres, sound happy today. This year, there is water enough for a second crop. Parkal had close to normal rainfall, and they appreciated the silt from the tanks, which raised their yields.

Telangana’s Politics Of Water, And A Problem Of Timelines

Here too, the groundwater data supports their claims.

Telangana’s Politics Of Water, And A Problem Of Timelines

Both Warangal and Mahbubnagar, had poor rainfall in 2014 and 2015, and above normal rainfall in 2016. This highlights the effectiveness of the scheme in 2015, while making the 2016 impact harder to call. Keep in mind, agricultural production, specifically paddy production has gone up steeply since 2015, showing there has been better water access.

Telangana’s Politics Of Water, And A Problem Of Timelines

Parkal farmers mentioned another thing: The Rythu Bandu scheme, which by directly transferring Rs 4,000 per acre helps alleviate working capital problems for farmers who own their land.

These schemes (along with others like the drinking water scheme, important in regions plagued by Fluoride-rich groundwater) improve farmer resilience, which, as Telangana looks to be affected by climate change, makes this farmer-centricity a shrewd political move.

Also read: Telangana’s Rythu Bandhu Scheme Is Penny Wise, Pound Foolish

Losing Steam

Sadly, Mission Kakatiya appears to have lost steam. The strongest criticism/evidence comes from the Comptroller and Auditor General’s office.

The CAG report showed that while the completion rate of desilting tanks was a commendable 84 percent in Phase-I of the mission, completion rates had fallen to 0 percent by Phase-III.

This is supported by the Mission Kakatiya Dashboard.

A screen-shot of the Mission Kakatiya Dashboard. (Source: Mission Kakatiya official website)
A screen-shot of the Mission Kakatiya Dashboard. (Source: Mission Kakatiya official website)

The scathing CAG report states the mission could have:

  1. Coordinated better with MGNREGA,
  2. Planned the desilting of tanks during the non-rainy season,
  3. Estimated the amount of silt more accurately, to ensure adequate desilting. To quote, “Average shortfall was 33 per cent … Thus, it could not be ensured that the storage capacity of these tanks was restored as intended in absence of proper mechanism to assess the quantum of silt to be removed and shortfalls in execution.”
  4. Prioritised the tanks to be desilted better. Ideally, system tanks with a dependable flow should be prioritised, so that desilting translates to better irrigation. However, to quote, “The sampled divisions could not produce any records with regard to assessment of dependable flows in tanks for prioritisation…None of the sampled division furnished list of chain linked tanks to Audit.”

This last criticism is echoed by others: system tanks, as their name implies, work as a system. If one does not desilt the connecting channels, and prioritise upstream tanks, desilting downstream tanks will not result in a higher irrigated area. One irrigation expert opined: “If something is done in the entire catchment, that will be useful, if some of the links are not covered, this is less helpful.” This loss of steam was supported in some interviews. The driver from Mahbubnagar bemoans two years later, “Follow-up is poor, and implementation is patchy.” He is not planning to return now.

In Adilabad, where the TRS enjoyed an overwhelming victory in 2014, a young farmer growing 21 acres of cotton said “Ideas are simply super”, he said. “But it is not reaching the farmers. Many tanks are there, there are no connecting channels. These have been encroached”. The groundwater data, which is available only to 2016, could not be used to verify his statement.

Also read: How A Bit Of Imagination And Community Support Can Solve Maharashtra’s Water Problem

Why Is This Happening?

In January 2018, the Telangana government offered 24 hour free power to farmers. Given that over 80 percent of irrigated land in Telangana is irrigated by wells, this politically savvy move helps explain the slowdown on tank irrigation. Groundwater access through free electricity is a quicker short-term fix. Until, of course, the groundwater runs out.

In 2013, 30 percent of Telangana’s groundwater mandals were classified as unsafe.

What are shrewd tactics for the current election, may not build longer term resilience of the state.

The current manifesto shows short-term focus (farm loan waivers, free housing etc.) has prevailed over even medium-term resilience-building schemes. This does not bode well for the water or farmer-resiliency of India. Building resilience is a burning issue because Telangana ranks second in farmer suicides in India, and a recent study highlighted irrigation issues—including borewell failure—as a key cause.

Will the voter, the farmer, care? Maybe not, as their horizon, driven by high discounting rates, tends to be short. In truth, we must wait for the elections for the real answer.

Mridula Ramesh is the founder of the Sundaram Climate Institute, and the author of ‘The Climate Solution - India’s Climate Change Crisis and What We Can Do About It’.

The views expressed here are those of the author and do not necessarily represent the views of Bloomberg Quint or its editorial team.

BQ Install

Bloomberg Quint

Add BloombergQuint App to Home screen.