Punjab and the 1% Flood
The land of five rivers, floored flat — a shallow flood over a third of the state, made worse the day the dams let go
In the second week of July 2023, the monsoon parked itself over the Shivalik foothills and Punjab’s rivers turned on the plain below them. The Sutlej, Beas and Ravi rose together, the seasonal Ghaggar breached its banks, and upstream the Bhakra and Pong dams — reservoirs already near the top of their rule curves — began releasing water into rivers that were already full. By the time the water pulled back, 41 people were dead, more than 1,600 were living in 173 relief camps, over 27,000 had been evacuated, and the state counted roughly ₹1,000 crore in damage. The worst of it fell on Rupnagar, Mohali (SAS Nagar), Patiala and Sangrur — Patiala drowned not by a great Himalayan river but by the humble Ghaggar, a seasonal stream that carries no water for most of the year and then, for a fortnight, carries far too much.
Punjab’s flooding is not Kerala’s deep hill-and-dam torrent, nor Andhra’s cyclone-fed delta sheet. It is the flood of a flat, canal-dense alluvial plain — the land of five rivers, floored so level that when the water comes, it spreads wide and thin across an enormous area rather than stacking deep in any one place. This piece uses the India Flood Atlas (a modeled 1% annual-chance flood layer) to show the scale, explains why so much of Punjab sits in harm’s way, and lays out a realistic, phased path to resilience.
The scale: the widest, shallowest exposure in the atlas
The 2023 deluge is observed history. The atlas adds the modeled 1% annual-chance (100-year) flood — the severe event that has a 1% chance of arriving in any given year, mapped consistently across the state.
Across Punjab, that event reaches:
- ~3.00 million building footprints — 31.1% of the state’s 9.63 million buildings — sit where the modeled 100-year flood would arrive. Nearly a third of every building in Punjab is in the floodplain — one of the highest shares in the atlas.
- The mean depth over those exposed buildings is just 0.44 m — the shallowest exposure of any state in the atlas. This is the signature of a dead-flat plain: water everywhere, deep almost nowhere. Of the exposed buildings, roughly 90% would face water under 1 m, about 10% between 1 and 3 m, and effectively none deeper than 3 m (only around 2% over 2 m).
- ~69,000 km of road (about 20.5% of the state network) and ~1,148 km of railway (19.3% of the track) lie in the modeled floodplain — a fifth of the arteries of India’s grain state.
That combination — very broad, very shallow — is the defining fact about Punjab. Where Kerala and Telangana concentrate a smaller share of buildings into a deep, dangerous tail, Punjab does the opposite: it puts an unusually large share of its buildings in the water, but at an unusually shallow depth. The five rivers and the vast canal networks have graded the land so level that a flood does not gather into a column; it slides out sideways across fields and villages in a broad half-metre sheet.
Shallow does not mean harmless. A half-metre of moving water still drowns a paddy crop, cuts a village road, contaminates a hand-pump, collapses a mud-brick wall, and strands thousands. In a state where farming is the economy, a shallow flood over a third of the land is a first-order disaster measured in ruined harvests, not just ruined houses.
A note on what this is: these are modeled figures at ~30 m resolution — regional hazard screening, not a parcel-level guarantee, and they count building footprints, not people. The good news for Punjab is that a flat alluvial plain is exactly the terrain where this kind of modelling is most reliable: there are no steep gorges or DEM blind spots to distort the depths. The state’s very shallow mean (0.44 m) should be read as genuinely characteristic, not an artefact.
Where it concentrates: the doab and the border rivers
Risk pools where the rivers run — along the Sutlej, Beas and Ravi and through the fertile doabs between them, with a heavy concentration in the border districts of the Majha region. The most-exposed districts by share of their buildings in the floodplain:
| District | % of buildings exposed | Exposed buildings |
|---|---|---|
| Kapurthala | 58.5% | 158,656 |
| Amritsar | 56.4% | 350,481 |
| Tarn Taran | 44.6% | 200,764 |
| Ferozepur | 38.3% | 135,229 |
| Gurdaspur | 37.0% | 198,132 |
| Fazilka | 36.1% | 144,930 |
| Patiala | 35.3% | 215,346 |
| Jalandhar | 32.3% | 198,261 |
The numbers are startling. Kapurthala (58.5%) and Amritsar (56.4%) each have more than half of their buildings in the modeled floodplain — Kapurthala because it sits low in the Beas–Sutlej doab where the two rivers converge, Amritsar because the Ravi hugs the Pakistan border along its western edge. Tarn Taran, Ferozepur, Gurdaspur and Fazilka are all border districts pinned against the Ravi and the lower Sutlej. Patiala (35.3%) is the Ghaggar’s victim — the district the seasonal river devastated in 2023 — and Jalandhar (32.3%), in the heart of the doab, floods when the Sutlej breaches its embankments near Shahkot. This is not a narrow risk core but a broad band across the state’s north and west, which is why Punjab’s overall exposed share runs so high.
Why Punjab floods
Several forces stack on top of one another:
- Five rivers on a flat plain. Punjab — literally panj-ab, “five waters” — is defined by the Sutlej, Beas, Ravi, and historically the Chenab and Jhelum, plus the seasonal Ghaggar. They descend fast from the Himalaya and Shivaliks and then hit a plain with almost no gradient. Water that arrived as a fast pulse spreads out into a slow, wide sheet with nowhere steep to drain.
- Dam releases that arrive with the flood, not instead of it. The Bhakra dam on the Sutlej and the Pong dam on the Beas are meant to moderate floods — but they are operated primarily for irrigation and power, so their reservoirs are often kept full through the monsoon. When an extreme spell arrives on an already-full reservoir, the dams must release large volumes into rivers that are already in spate, compounding the peak downstream. In September 1988, four days of cloudburst (~634 mm in the Bhakra catchment) forced emergency releases from Bhakra and Pong; the resulting floods struck around 9,000 of Punjab’s ~13,000 villages, disrupted the lives of ~3.4 million people, and killed several hundred (estimates run from ~600 upward) — still the benchmark catastrophe. The 2023 and 2019 floods repeated the pattern in miniature.
- The Ghaggar — a seasonal river that punches above its weight. The Ghaggar carries little water most of the year, so its floodplain in Patiala and Sangrur has been farmed and settled right up to the channel, and its embankments are weak. When the monsoon fills it, it breaches repeatedly — flooding over 100 villages in Patiala and Sangrur in 2019, and again in 2023 and 2025. It is the archetype of a made flood risk: a small river given no room.
- Embankments that fail — the Dhussi bundh. Much of Punjab’s flood defence rests on earthen embankments (bundhs, notably the Dhussi bundh along the Sutlej near Shahkot/Gidderpindi). Poorly maintained, rat-riddled and under-designed, these bundhs breach under pressure — in 2019 a breach at Sultanpur Lodhi inundated dozens of villages for weeks; in 2023 and 2025 the Dhussi bundh failed again. A breached embankment turns a contained river into a sudden inland sea.
- A canal-dense, built-over plain. Punjab is laced with one of the world’s densest irrigation-canal networks. Canals, roads and rail embankments criss-cross the flat land, and where they obstruct natural drainage they pond water — turning a shallow flood into a long-standing one.
- A shared, contested river geography. The Ravi and Sutlej cross into Pakistan, and the Bhakra Beas Management Board (BBMB) operates the big dams for multiple states (Punjab, Haryana, Rajasthan). Flood decisions are therefore inter-state and cross-border by nature — release timing that protects one stakeholder can flood another, a tension that surfaced sharply in 2023 and 2025.
- Climate change is intensifying the short, extreme rainfall bursts over the Himalayan foothills that fill the reservoirs and rivers fastest — pushing today’s “1% event” toward the more-frequent event of 2050 and 2100.
Two years after 2023, in August–September 2025, all of these forces combined into Punjab’s worst flood since 1988: the Sutlej, Beas and Ravi in spate together, large releases from Bhakra, Pong and Ranjit Sagar, the Dhussi bundh and Ghaggar embankments breached, and eventually more than 1,400 villages and all 23 districts affected, over 350,000 people hit, roughly 55 deaths, and hundreds of thousands of acres of standing crops lost. The shallow-but-vast exposure the atlas describes is not theoretical.
A strategic approach: give the rivers room, and fix the releases
Because Punjab floods wide and shallow, the levers are different from a deep-flood state. The enemy here is extent, drainage and dam timing — not a single deep torrent to wall off. Much of what is needed is operational and institutional rather than concrete-heavy.
Pillar 1 — Know the risk, and warn early. Statewide hazard mapping (like this atlas) plus real-time rainfall, river-gauge and reservoir-inflow forecasting. On a flat plain the flood travels relatively slowly, so good forecasting buys real evacuation time. The highest-leverage life-saver is a Shivalik-to-plain warning chain tied directly to BBMB dam operations.
Pillar 2 — Operate Bhakra and Pong for floods, not just for water. The single biggest change available to Punjab is dam rule-curve reform: pre-monsoon drawdown, forecast-informed releases, and a hard commitment to early, gradual releases rather than late, large ones. A reservoir that spills its surplus in July at a river’s safe capacity prevents the panicked August release that arrives on top of a flooding river. This is a governance fix, not a construction project.
Pillar 3 — Fix the embankments, permanently. The Dhussi bundh and the Ghaggar embankments fail because they are under-maintained. A standing programme to strengthen, raise and annually inspect the bundhs — before, not after, each monsoon — is the highest-return civil work in the state. Give the Ghaggar, in particular, room and firm banks.
Pillar 4 — Unblock drainage, and stop ponding. Because canals, roads and rail embankments trap water on the flat plain, the fix is cross-drainage — culverts, siphons and cleared natural drains (choes and nalas) — so a shallow flood drains out in days rather than sitting on crops for weeks.
Pillar 5 — Govern land use. The cheapest flood control is not farming and building in the active floodplain — least of all along the Ghaggar. Enforce floodplain zoning and keep settlement off the bundh-side of embankments.
Pillar 6 — Make buildings and communities resilient. For the three million buildings already exposed, the shallow depths are an advantage: raised plinths of half a metre to a metre take most of Punjab’s exposed buildings out of the water. Pair that with flood-resilient materials, rehearsed evacuation, and flood-inclusive home and crop insurance (covered under “STFI” perils in standard Indian home cover such as Bharat Griha Raksha, yet rarely held).
A realistic timeline
Flood resilience is a 15–25 year programme, not a five-year project — but Punjab’s shallow-flood profile means several high-value moves are cheap and fast.
Phase 0 — Foundation & quick wins (0–2 years). Low cost, high impact. Statewide hazard mapping and exposure prioritisation; forecast-tied dam release protocols for Bhakra and Pong; pre-monsoon inspection and repair of the Dhussi bundh and Ghaggar embankments; and evacuation planning for Kapurthala, Amritsar, Tarn Taran and Patiala. These save lives in the next monsoon.
Phase 1 — Defend & drain (2–5 years). Systematically strengthen and raise the embankment network; add cross-drainage where canals and roads pond water; clear the choes and natural drains; enact and enforce floodplain zoning along the Ghaggar and border rivers; stand up a permanent annual maintenance budget so cleared drains and repaired bundhs stay that way.
Phase 2 — Basin-scale resilience (5–10 years). Move from district-by-district fixes to whole-basin, inter-state management of the Sutlej, Beas and Ravi through a reformed BBMB — coordinated reservoir operation, upstream storage in the Shivaliks, and cross-border data-sharing on the Ravi and Sutlej. Retrofit exposed public assets (hospitals, schools, substations, grain silos, rail) across the high-exposure doabs.
Phase 3 — Climate-proof & sustain (10–20+ years). Design to future climate — size embankments, drainage and dam rule-curves for 2050/2100 monsoon extremes (the SSP scenarios) — and lock in maintenance, monitoring and updating as permanent functions.
The honest bottom line: Punjab carries one of the atlas’s largest exposed shares — nearly a third of all its buildings sit in the floodplain — but at its shallowest depths, which flips the strategy. The decisive levers are not big new dams but how the existing dams are operated, whether the embankments hold, and how fast a flat plain can be drained — and, for the millions already exposed, a raised plinth of half a metre does most of the work. The risk is broad but legible, concentrated in the doabs and border rivers of the north and west. Knowing exactly where the shallow sheet spreads is what lets a limited budget protect the most people and the most cropland.
Sources
- July 2023 Punjab floods — The Tribune: 41 killed, 1,600 in relief camps · Sphere India Situation Report (26 July 2023) · NRSC near-real-time flood mapping (July 2023)
- August 2023 floods & the role of dam releases — Mongabay India: July rainfall worsened Punjab’s 2023 floods · Down To Earth: Punjab’s floods are not purely natural events
- 1988 Punjab floods — Wikipedia
- 2019 floods, Sultanpur Lodhi & Dhussi bundh breach — The Tribune: 50 villages flooded in Jalandhar–Sultanpur · The Tribune: one year after breach, flood fear returns
- Ghaggar floods (Patiala, Sangrur) — The Tribune: Ghaggar breach destroys paddy
- 2025 Punjab floods (worst since 1988) & dam-board governance — 2025 Punjab floods (Wikipedia) · Mongabay India: worst flood in decades · The Caravan: the dam board and the 2025 floods
- Exposure statistics — India Flood Atlas (modeled 100-year / 1% annual-chance event), NEER.
Exposure figures are modeled estimates (~30 m) for a 1% annual-chance flood, intended for awareness and prioritisation — not parcel-level certainty.