THE Indus River is moving east in central Pakistan. Along an 85-kilometre stretch from upstream of Layyah city down to Taunsa Barrage, it is pressing hard against its eastern bank. In this year’s flood season, it destroyed Basti Kanjal, a riverine settlement of more than 500 houses some 15km northwest of Layyah. It did not stop there, washing away several more settlements. Since June, close to 5,000 acres of cultivated land have been lost; according to news reports, 100,000 people have been displaced. A close analysis of 38 years of satellite imagery shows that the river is not wandering. It is being pushed. Ask any elder in the riverside villages of Churkin or Basira which face the Vehowa hill torrent’s mouth 30km upstream of Layyah about the river of his childhood, and he will tell you the Indus was 10 miles (16km) wide. He will also tell you that the river gave as much as it took: land drowned on one bank surfaced on the other, and the patwari’s records followed the water back and forth. The satellites agree with him on the width. Between 1988 and 1997, the active belt below Churkin measured 16km to 18km, and above Taunsa Barrage nearly 24km. Today, it is reduced to a third. A river squeezed into a narrow gut runs deeper and faster, and expends its vigour on eroding vulnerable banks. Three engineering structures, built with public money and international loans, have together hemmed in the river on three sides. The first of these is the Chashma Right Bank Canal, financed by the Asian Development Bank. The canal runs through the Suleiman hill torrents. These are no small streams: the Vehowa alone drains a catchment of over 2,600 square kilometres and has recorded flood peaks above 35,000 cusecs. Before the canal, such floods spread across the Daman plain under the old rod-kohi system, watering the land and dropping their silt on the way. After 2003, the canal’s 68 cross-drainage structures began delivering every flood, with its full load of mountain sediment, to fixed points on the river’s right bank. Between 2011 and 2016, two great fans of new land rose out of the riverbed — one at the Vehowa’s mouth opposite Layyah, and one at the mouth of the Sanghar torrent near Taunsa city. Together, they hold some 37sq km of new permanent land, a metre and more above the old riverbed, and about 100 million cubic metres of sediment. Where the trained torrents deliver their load, walls have risen; in the intervening stretches, the river’s western edge hasn’t moved in 38 years. The canal itself has been closed since the 2022 floods, but its silt walls remain standing in the river. A close analysis of 38 years of satellite imagery shows that the Indus is not wandering. It is being pushed. The second factor is Taunsa Barrage. Built in 1958 to pass a million cusecs, its capacity fell by half within five decades as sediment accumulated. A wedge of silt today stands nearly 2m high near the pond and stretches 40km upstream, flattening the river’s slope and promoting further deposition. The World Bank’s 2005-08 rehabilitation project raised the crest and built a secondary weir downstream. But the embankments were never raised proportionately. Local communities protested and wrote to the Bank that the remodelling ignored the sedimentation, causing flood protection for structures to fail. In 2010, a flood peak of 960,000 cusecs breached the eastern marginal embankment upstream, allowing the river to outflank the structure and carve a new channel. Their warning proved correct. The third factor relates to the embankments. The marginal ones along the left bank, running some 40km from Daira Din Pannah up towards Layyah, froze the river’s eastern edge decades ago, and the floodplain strip between the river and bund has since been almost entirely converted into farms and settlements. In the 1990s, the river still reclaimed that strip in every high flood; today, 95 per cent of it has turned into permanent land. Meanwhile, the restricted riverbed is level with the protected villages outside — all that protects them now is the crest of an earthen bund. This is the fuller picture of the narrative that those affected recounted outside the Layyah Press Club this September, as they blamed two decades of Chashma Canal silt for their plight: sediment walls advancing from the west, a wedge of silt backed up behind the barrage, and, on the east, the embankments holding the door shut. A river so enclosed has nowhere to go but against the weakest point. In 2010, the weakest point was the marginal bund above Taunsa; in 2026, it was the training works at the Layyah bridge. The river has now cut to within 200m of the Lala, its old channel passing along the western edge of Layyah city. The city’s protection bund stops 3.5km short of that point. What happens if the river completes its eastward shift? The low floodplain east of the river holds hundreds of villages built on the assumption that the river would keep to its corridor. A river breaching the Lala line would place Layyah on the front line, and behind it Jaman Shah, Kot Sultan, and Paharpur. An eastward-shifted Indus would progressively outflank Taunsa Barrage itself. A barrage commands only the water that arrives at its gates; the canals of Dera Ghazi Khan and Muzaffargarh, irrigating millions of acres, depend on that command. The river would leave the barrage presiding over a dying channel. The communities of this region warned both lending institutions years ago, the ADB over the canal and the World Bank over the barrage, through formal complaints, inspection claims, and hunger strikes. I was among those who warned. I remember how patiently they were heard and how completely they were ignored. Both institutions noted lessons and closed their files. The river has now reopened them, writing its supplement in drowned villages and 5,000 acres of lost land. Reconstruction without the benefit of an honest analysis of these structural challenges would be tantamount to allowing a similar situation, or even worse, in the future. The analysis now exists, measured from orbit and verifiable by anyone. What remains is whether anyone with authority is willing to read it. The writer is a native of Taunsa and teaches at Quaid-i-Azam University, Islamabad. Published in Dawn, October 11th, 2026