Lokubanda Tillakaratne
This past November and December, Ditwah showed us how dark, eerie, and haunting cyclones can be. Past generations have suffered: per PWD records, in 1834, the Mahaweli rose 60 feet above its usual flood level at Gampola, and the Canberra Times reported in 1911 the great flood of Ceylon on December 30 of that year. It killed 200 people and left over 300,000 homeless. Half a century later, on December 25, 1957, a nameless cyclone brought severe rain to the North Central Province (NCP). The Nachchaduwa reservoir near Anuradhapura city breached, releasing its full volume into Malwatu Oya, a mid-level river flowing through the city, nearly washing away its colonial-era bridge near the Lion Tower. A cyclone paid a visit to the Eastern Coast of Sri Lanka on November 17-23, 1978.
Half a century later, Ditwah came with swagger.
River Quartet
These cyclones came spaced by a generation or two. During Cyclone Ditwah’s moment, it unleashed disaster and tragedy, terrorizing every breath of hundreds of thousands of people. How the Great River Quartet – Mahaweli, Kelani, Kalu, and Walawe, and their attendant mid-level rivers behaved before Ditwah masks the reality that they are not the loving and smiling beauties poets claim them to be. During the Ditwah visit, our river Quartet showed its true colors in plain sight when wave after wave of chocolate rage pushed and uprooted forests, creating islands of floating debris and crashing onto bridges, shattering their potency. These rivers are nothing more than a bunch of evil reincarnations cloaked in ruinous intentions.
The River Quartet and its mates woke up to the first thunder of Ditwah. They carried away villages, people, property, herds of cattle, and wild elephants to the depths of the Indian Ocean. While we continue to dig out the dead buried in muddy mountainsides, dislodged from their moorings during this flood of biblical proportions, how our rivers, streams, and, particularly, the village tanks handled the pressure on their own will will be the core of many future discussions.
The destruction and tragedy caused by this water hurt all of us in many ways. But we all wish they were only a fleeting dream. Sadly, though, the real-life sight of the pulverized railway bridge at Peradeniya is not a dream. Ditwah flood stripped the modesty of the bridge on the rail line here, laying it bare. It hung in the air, literally, like strands of an abandoned spider web on a wet Kandyan morning. It was a reminder to us that running water is a masked devil and should not be considered inviting. It can unleash the misery with a chilling ending that no one wants to experience in a lifetime.
Tank Cascade Systems (TCS)
Although the Ditwah cyclone covered Sri Lanka from top to bottom with equal fury, the mountainous areas and floodplains of our River Quartet surrendered soon. However, the village tanks in the Dry Zone – Northern, North Central, Northeast, and Eastern provinces – weathered that onslaught, sustaining only manageable damage. They collectively mitigated the damage caused by over 200 mm of rain that fell across the catchment areas they represented. Thus, the tank, the precious possession of the village, deserves to be titled as a real beauty.
Let me introduce the village tanks systems our engineering ancestors built with sophistication and ingenuity, a force like Ditwah hardly made a dent in groups of these tanks called Tank Cascade Systems (TDS). Many of the village tanks in the Dry Zone, covering 60% of Sri Lanka’s land area, stand in groups of TDS, separated as individual bodies of water but sharing water from one or more dedicated ephemeral streams. R.W. Ievers, the Government agent for North Central Province in the 1890s, noted that these tanks were the result of “one thousand years of experiment and experience,” and “ancient tank builders took advantage of the flat and undulating topography of the NCP to make chains of tanks in the valleys.” Colonial Irrigation Engineers of the early 20th century also recognized this uniqueness. Still, they could not connect the dots to provide a comprehensive definition for this major appurtenance of the village.
Although these tanks appear to be segregated ecosystems, a closer look at the peneplain topographic map of Sri Lanka shows that each stream feeding them ultimately flows into a larger reservoir or river, jointly or independently influencing the mechanics of regional water use and debouching patterns. This character is the spirit of the dictum of King Parakramabahu centuries earlier: “let not a single drop of water go to waste into the sea without being used by people.” Villagers knew that each tank in their meso-catchment area was linked to other tanks on the stream it was on, ensuring the maximized use of water.
With their embodied wisdom, our ancestors centuries ago configured the placement of individual tanks that shared water from a catchment area. But not until 1985, following a careful autopsy of the pattern of these small tanks in the Dry Zone, Professor Madduma Bandara noticed a distinctive intrinsic relationship within each group of tanks. He called a group of such tanks a Cascade of Tanks. He wrote, “a (tank) cascade is a connected series of tanks organized within a micro-catchment of the Dry Zone landscape, storing, conveying, and utilizing water from an ephemeral rivulet.” In short, it is a “series of tanks located in succession one below the other.” Dr. M.U.A. Tennakoon shared the name the villagers in Nuwarakalaviya use for this configuration of tanks: Ellangawa (එල්ලංගාව). On a map, these tanks appear as hanging (එල්ලෙන වැව) on a string. Thus, Ellangawa can be a portmanteau, a blend, of these two words.
There are over 475 such cascading tank groups in the Dry Zone. On average, each cascade typically supports four tanks. One cascade, Toruwewa, near Kekirawa, has 12 tanks. According to Professor Madduma Bandara, a cascade of tanks held about 20-30% of the water falling on its catchment area. As I will show later in this essay, the tank cascades behave like buddies in good times and bad times. By undertaking to build a vascular structure to collect, conserve, and share water with communities along the stream path, our ancestors forewarned of the consequences of failing to undertake such micro-projects where they chose to live. The following are a villager’s thoughts on how to retool this concept to mitigate the potential for damage from excess water flow in a larger river system.
To villagers, their tank is royalty. Its water is their lapis lazuli. Therefore, they often embroidered the title of the village with the suffix wewa (tank) or kulam (tank, in Tamil), indicating the close connection between the two. It is the village’s foremost provider and is interdependent. That is why we have the saying, “the village is the tank, and the tank is the village.”
A study in 1954/55 found that there were 16,000 tanks in Sri Lanka, of which over 12,500 were operational. Out-of-commission tanks were those that fell into disuse after the original settlers abandoned them for a host of reasons, such as a breach in the bund, fear of plague or disease, or superstition. Collectively, they supply water to an area larger than the combined area of the fields served by the major irrigation reservoirs in the country at the time.
In some villages, an additional tank called olagama, with its own acreage of fields, receives water from the same stream or from another feeder stream which joins the principal stream above or below the main tank. In the event the main tank is disabled, the olagama tank can often serve as an alternate water source for their fields.
Cultural and Engineering
A tank cascade is also an engineering undertaking. But village tank builders were not engineers with gold-trimmed diplomas. They were ordinary folks, endowed with generations of collective wisdom, including titbits on the physics of water, its speed, and its cruelty. Village pioneers responsible for starting the construction of the tank bund, gam bendeema, placed the first lump of earth after marking off home sites, not immediately below the future bund, but slightly towards one end of it, in the area called gammedda, or the elevated area the bund links to, gamgoda.
Engineering of a tank cascade has a cultural underpinning. It is founded on the feeling of solidarity among the villages along an ephemeral stream. In practice, it was a wholesome area with small communities of kin below each tank sorting out their own affairs without much intervention of the ruling class. For example, during heavy rains, each village in the chain communicated with the villages below the volume in its tank and the projected flow of the stream. When the tank reached its capacity and water began to spill over the spillway, the village below must take measures to protect its tank bund. If it breached, villagers up and down the cascade helped each other repair it.
They were aware that an earthen dam was susceptible to failure, so they used their own town-planning ideas. They avoided building residential zones directly under the stream’s path, generally at the midpoint of the dam. Instead, they built their triumvirate of life – tank, field, and dagoba (stupa) – keeping safety and practicality in mind. Dagoba was always on a higher ground, never supported by beams on a stream bank like what Ditwah revealed recently. We now know what happens to dagobas built on sagging beams by deceptively serenading riverbanks when thunder waters and unworldly debris came down hand in hand.
From top to bottom, the Tank Cascade showed the engineering instinct of the builders and accessory parts that helped its smooth functioning. There was the Olagama and Kulu Wewa associated with a system. Tank builders had an idea of the volume of water a given stream would bring in a year. In conjunction with this, the bunds of the Olagama and Kulu Wewa are built small. In contrast, the bunds of the tanks that formed the lower rung of the cascade are relatively larger. The idea behind this was that, in the event of a breach in an upstream tank, the downstream tanks could withstand an unexpected influx of water.
During the Ditwah’s death dance, the Mahaweli River did not have this luxury as it marched downstream from Kotmale dam. There were not enough dams to tame this river, and its beastly nature was allowed to run wild until it was too late for many.
The embodied imprints of experience inherited from their ancestors’ helped villagers design the tank’s physical attributes. In general, a tank supplied by this stream had a dam of a size proportional to the amount of water it could store for the fields. Later, as the village added families and field acreage increased, villagers raised the bund and the spillway to meet increased storage capacity. This simple practice guarded against eventualities like uncontrollable floods between villages. Excess water was allowed to flow through the sluice gate and the spillway, reducing the pressure on the bund. Had we applied this fundamental practice on a proportional scale to a large stream, i.e., oya or river, it would have lessened the destruction during a major rainstorm, ilk of which Ditwah brought.
With my experience living in a village with its tank, part of a TCS of five tanks, I wish large rivers like the Mahaweli had a few small-scale dams or partial diversions mimicking a rudimentary TCS so that the Railway Bridge at Peradeniya could have avoided the wrath of hell and high-water bringing muck and debris along its 46 km descent from Kotmale, where its lone dam is. I am glad I have company here. Professor Madduma Bandara noted 40 years ago, “much water flows through drainage lines due mainly to the absence of a village tank-type storage system.” Mahaweli turned out to be that drainage line this past November, holding hands, sadly, though, jubilantly, with the designs of Ditwah. Recently, former Head of Geo-Engineering at Peradeniya University, Udeni Bandara Amarasinghe, highlighted the importance of building reservoirs on other rivers to control floods like those we experienced recently.
Check Dams & Macroscopic Control
Within the TCS, the check dams, Kulu Wewa or Kele Wewa – forest tanks above a working tank held back sediments generated by upstream denudation. They controlled the volume and water entering the main tank. Kulu Wewa provided water for wild animals and checked their tendency to raid crops below the main tank. The difference between Kulu Wewa and Olagama was that, because of its topographical location, Kulu Wewa was occasionally used as a source of water for crops when the main tank below it became inoperable due to a breach or was undergoing repairs or used up its water early.
Based on these definitions, each working tank in the TCS also acted like a check dam for the one below it. Furthermore, if a tank in the cascade ran out of water, other tanks in the cascade stepped in. They linked up with the tanks above through temporary canals made by extending an existing minor canal, wella, or the wagala, excess water pan, of an upstream field.
The tank bund tamed and kept in check the three attributes of a stream – water velocity, volume, and its destructive power. By damming the stream, the villagers broke fueling momentum of it. They rerouted it via the spillway at the end of the bund, a form of recycling. Water from some spillways is diverted along a large niyara-like (field ridge) lesser dam, built along the wanatha (flanks) of the field, until it empties into the atrophied stream below the field.
Simultaneously, by controlling the release of water through two sluice gates on the bund, goda and mada horowwa, and directing it to the two flanks of the field, ihala and pahala wanatha, villagers succeeded in tamping down the pressure on the bund. Water from the neutered stream is thus redirected from all three exit points. It must now continue its journey along the wagala, to which field units (liyadi) also empty their excess water. This water is called wel pahu watura.
After going through this process, the momentum of the ephemeral stream water is passive by the time it reaches the tanks in the lower parts of the cascade, often a kilometer or two downstream. This way, a line of tanks along the stream’s axis now shares the responsibility of holding back its full potential, limiting its ability to cause damage.
Such a break of momentum was lacking in the Four Great River Quartet and their lesser cousins. For the long-term solution to prevent damage from future cousins of Ditwah, we must consider this ingenious water-control method for rivers on a macroscopical scale.
Reservoirs
As Ditwah-type floods occurred in 1911, 1957, 1978, and 2025, with a bit of luck, we can expect to have a few more decades of recess to work on cascading edifices along rivers, such as dams or diversions, before the next flood comes with roguish intentions. The Accelerated Mahaweli Diversion Program (AMDP), started in 1978, took 30 years to complete and now has over a dozen reservoirs between Kandy and the Dry Zone coastal belt, holding back its might. These reservoirs held their ground while Ditwah rained hell, so consulting the TCS’s ingenuity, which seems antiquated, is a good investment.
As soon as Cyclone Ditwah began to make noise, word spread that releasing water from a few of them on the Mahaweli and Kelani rivers could have made a difference. The problem with the Kelani River basin in Western Province and the Mahaweli basin in Central Province above Kandy is that, despite their combined population being nine times that of the NCP, they only have six reservoirs. On the contrary, the NCP has twice as much in the lower Mahaweli River basin, built under the AMDP. Furthermore, the NCP also has many ancient reservoirs that it inherited from our ancestors. A string (cascade) of large reservoirs or minor dams in the hill country could have helped break the river’s energy, which it accumulated along the way. G.T. Dharmasena, an irrigation engineer, had already raised the idea of “reorienting the operational approach of major reservoirs operators under extreme events, where flood control becomes a vital function.”
Unique Epitaphs for the Cyclones
The processes discussed above could have prevented the destruction of the railway track at the Peradeniya bridge, the image of which now stands like a pictorial epitaph to the malicious visit of the Ditwah and a reminder to us, “what if…?” or “what next…?”
As mentioned at the beginning of this essay, when the 1957 Cyclone dropped heavy rain on the NCP, a Railway Department employee at Anuradhapura made an exceptional effort to preserve the memory of that saga for posterity, leaving an epitaph still visible 70 years later. This person memorialized his near escape from the Malwatu Oya flood. As the river roared past over the railing of the bridge near the Lion Pillar roundabout, this employee, probably trapped in his two-storied house near the roundabout, day-stamped the visit of the flood with a red line on the wall of his house to mark the height it reached to trap him.
Three meters from the ground, right between two archtop windows facing the road to Sri Maha Bodhi, he wrote in Sinhala, Tamil, and English, “ගංවතුර මට්ටම – வெள்ள நிலை – Flood Level.” Right below it, at the end of the faded line, he added, “1957-12-25.”
As Ditwah came along, the current resident of the house would not break this seven-decade-old tradition. After the flood receded this time, this duty-bound resident drew a line in blue ink and wrote at its end, ‘2025-11-28’, his contribution serving as an epitaph, reminding us of the infamous day Ditwah showed her might by driving the river off its banks.
He added a coda to his epitaph – the numeral “8” in 28 is written in bold.
1957 and 2025 Cyclones Flood Marks written above window and below on the wall of a house by the banks of the Malwatu Oya in Anuradhapura.
Lokubanda Tillakaratne is the author of Rata Sabhawa of Nuwarakalaviya: Judicature in a Princely Province – An Ethnographical and Historical Reading
