Telangana and the 1% Flood
A smaller, sharper footprint — the Godavari belt and a lake-choked capital
On the night of 28 September 1908, after roughly 17 inches of rain in about 36 hours, the Musi river tore through Hyderabad. It rose some 60 feet, destroyed an estimated 80,000 houses, and left about a quarter of the city’s population homeless. The death toll has never been settled — it is commonly put around 15,000, though some accounts run as high as 50,000 (the true figure is genuinely uncertain). The Great Musi Flood so scarred the Nizam’s city that he engaged the engineer M. Visvesvaraya, whose plan gave Hyderabad the reservoirs that still define its water geography — Osman Sagar (~1920) and Himayat Sagar (~1927) on the Musi and Esi. More than a century later, the Musi still floods, only now it winds through a metropolis of ten million.
Telangana’s flooding is not the coast’s. It is a Deccan-plateau story: deep, confined river pulses down the Godavari and Krishna in the north and west, and a very different, urban-drainage disaster in Hyderabad. This piece uses the India Flood Atlas (a modeled 1% annual-chance flood layer) to show the scale, explains why the two kinds of flooding behave so differently, and lays out a realistic, phased path to resilience.
The scale: a smaller, sharply concentrated footprint
The 1908 flood and its modern successors are 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 Telangana, that event reaches:
- ~861,000 building footprints — 8.3% of the state’s 10.38 million buildings — sit where the modeled 100-year flood would arrive. That is the lowest share of buildings exposed among the atlas states.
- The mean flood depth over those exposed buildings is 1.26 m — mid-range, similar to Andhra Pradesh, with the risk concentrated in a handful of Godavari-belt districts. Of the exposed buildings, 67% face under 1 m, 24% face 1–3 m, and 9% face water deeper than 3 m (15% are over 2 m).
- ~23,300 km of road (about 12% of the state network) and ~390 km of railway (10% of the track) lie in the modeled floodplain — a smaller share than the coastal states, concentrated in the same Godavari-belt districts.
That is the defining contrast. Telangana’s statewide share exposed (8.3%) is markedly lower than the coastal states — Kerala at 16.7%, Tamil Nadu at 16.1% — not because it floods uniformly deep, but because the risk here is inland and concentrated rather than a broad coastal sheet. It pools in two places: the incised Godavari river valley of the north, where a confined river stacks water high, and the urban lake-and-nala system of Hyderabad. The rest of the plateau is largely dry in a 100-year event.
A note on what this is: these are modeled figures at ~30 m resolution — regional hazard screening, not a parcel-level guarantee. They are at their most reliable on flat valley floors and, crucially, least reliable in the steep, confined Godavari gorge — which is exactly where the deepest values sit (more on that below).
Where it concentrates: the Godavari belt and the Musi
Risk pools overwhelmingly in the Godavari belt of north Telangana, and secondarily in Hyderabad’s river-and-lake system. The most-exposed districts by share of their buildings in the floodplain (post-2016, 33-district Telangana):
| District | % of buildings exposed | Exposed buildings | Mean depth |
|---|---|---|---|
| Bhadradri Kothagudem (Bhadrachalam) | 22.0% | 83,895 | 4.22 m |
| Mulugu | 21.7% | 24,445 | 2.48 m |
| Hyderabad (GHMC core) | 15.7% | 56,965 | 1.20 m |
| Medchal–Malkajgiri | 15.1% | 102,615 | 0.83 m |
| Kumurambheem Asifabad | 13.1% | 27,281 | 1.83 m |
| Adilabad | 12.5% | 28,664 | 1.24 m |
| Ranga Reddy | 10.1% | 91,252 | 0.81 m |
| Hanumakonda | 9.1% | 24,162 | 0.64 m |
| Mancherial | 8.5% | 24,428 | 1.15 m |
| Karimnagar | 8.1% | 28,134 | 0.97 m |
| Warangal | 7.7% | 19,721 | 0.57 m |
Bhadradri Kothagudem — home to the temple town of Bhadrachalam — is in a class of its own: 22% of its buildings exposed, at a mean depth of 4.22 m. This is the deep-water core, where the Godavari runs confined and rises fast. Hyderabad’s 15.7% (mean 1.20 m) is a different animal: chiefly the Musi river and the lake–nala (cheruvu–stormwater) system threading through the city.
Why the deepest numbers deserve a caveat
The deep values in the Godavari-belt districts — Bhadradri’s 4.22 m mean, the 9% of exposed buildings over 3 m — deserve to be read carefully.
They are partly real. Where the Godavari is confined in a narrow, incised valley, the water genuinely stacks high: in July 2022 the river at Bhadrachalam peaked at about 70.5 feet, well above its 43-foot danger mark, and the road bridge was closed for the first time in 36 years. That is deep, dangerous flooding, not a modelling artefact.
But they are also partly uncertain. Terrain and DEM accuracy is weakest in exactly this kind of steep, confined terrain — so the deepest figures in the atlas are also the least certain figures in the atlas. We flag that rather than smooth it over: treat Bhadradri’s 4.22 m as “very deep, direction reliable, exact metres not.” Worth noting, too, that the very deepest Godavari-gorge water (~11 m) actually lies just across the river in Andhra Pradesh’s Polavaram district — a boundary correction from an earlier version of this analysis, which had mistakenly counted that gorge on the Telangana side. Recomputed against the authoritative 33-district Telangana boundary, that deep-water tail belongs to Andhra Pradesh, not here.
And there is a caveat pulling the other way in Hyderabad. The city’s real-world flood disaster is largely urban and pluvial — a matter of drainage, encroached lakes, and blocked stormwater nalas — which a riverine 1% model under-captures. So 15.7% is best read as a floor for Hyderabad, not the whole story. The Musi can be within its banks while whole colonies go under because the water has nowhere to drain. October 2020 was exactly that.
Why Telangana floods
Two distinct mechanisms sit on the same plateau:
- The Deccan’s two great rivers. Telangana is drained by the Godavari in the north and the Krishna in the south. When the upper catchments and dam releases coincide, a flood pulse runs down to towns pinned against the rivers — above all Bhadrachalam, where the Godavari is confined and rises with little warning.
- Incised valleys stack water high. Unlike a flat delta that spreads water thin, a narrow river valley concentrates the same volume into far greater depth. This is why Telangana’s overall share exposed is the lowest in the atlas, yet its worst districts — the Godavari-belt towns — carry a deeper tail than the flat coastal deltas.
- Hyderabad’s lost sponge. The city grew on an interconnected cascade of tanks and lakes (cheruvus) that once passed water gently downstream. That cascade is heavily encroached — Hyderabad is widely reported to have lost around 70% of its lakes (treat as approximate) — and the stormwater nalas that connect them have been narrowed, silted, and built over.
- Cloudburst-scale rainfall. In October 2020, a Bay of Bengal deep depression dumped a record ~324 mm on Ghatkesar in 24 hours. On a built-over catchment with nowhere to drain, that becomes an urban flood in hours.
- Encroachment on record. The Kirloskar Committee (2003) and a 2020 Voyants study flagged on the order of 28,000 encroachments across the HMDA region — a direct, documented cause of the city’s drainage failure.
- Climate change is intensifying the extreme-rainfall bursts that drive both the riverine pulses and the urban deluges.
The two mechanisms need two different responses — which is why the strategy has to be layered.
A strategic approach: tame the river, unclog the city
Telangana’s problem splits cleanly, and so should its strategy: manage the deep riverine risk on the Godavari and Krishna, and rebuild the drainage-and-lake capacity of Hyderabad.
Pillar 1 — Know the risk, and warn early. Statewide hazard mapping (like this atlas) plus real-time rainfall, river-gauge and reservoir-release forecasting. For the Godavari towns, the highest-leverage life-saver is upstream-to-Bhadrachalam early warning tied to dam operations — hours of lead time on a fast-rising, confined river. The TSDMA + IMD early-warning chain is the backbone here.
Pillar 2 — Unclog and reconnect Hyderabad’s drainage. The Strategic Nala Development Programme (SNDP) — widening, desilting and reconnecting the city’s stormwater nalas — attacks the pluvial problem directly. Every metre of nala restored to its design width is capacity the 2020-style deluge can use.
Pillar 3 — Restore the lakes as living storage. HYDRAA (the Hyderabad Disaster Response and Assets Protection Agency, formed July 2024) is removing lake and nala encroachments and restoring water bodies. Statewide, Mission Kakatiya — the restoration of roughly 46,531 tanks at around ₹22,000 crore — was built for irrigation, but revived tanks also buffer floods by holding the peak. Protecting and reconnecting the cheruvu cascade is Hyderabad’s cheapest flood storage.
Pillar 4 — Govern land use, hardest in the city. The cheapest flood control is not building in the floodway. Enforce the Full Tank Level (FTL) and buffer-zone rules around lakes and nalas, stop fresh encroachment, and require flood-safe design in exposed zones — a policy-and-enforcement problem more than an engineering one, and the one HYDRAA exists to tackle.
Pillar 5 — Make buildings and communities resilient. For those already exposed — especially in the deep-water Godavari towns — raised plinths, flood-resilient materials, protected utilities, and rehearsed evacuation. In Bhadrachalam, where mean modeled depth is measured in metres, evacuation planning matters more than any wall. Pair it with flood-inclusive home insurance (covered under “STFI” perils in standard Indian home cover such as Bharat Griha Raksha — yet rarely held).
Pillar 6 — Institutions and money that last. HYDRAA, SNDP and Mission Kakatiya give Telangana real institutional footholds. The task now is to keep them funded, coordinated, and maintained for decades — with a standing maintenance budget, not just construction capital.
A realistic timeline
Flood resilience is a 15–25 year programme, not a five-year project. Telangana has recent momentum — SNDP, HYDRAA, Mission Kakatiya — but the work of undoing a century of encroachment and building a river-warning system is measured in decades.
Phase 0 — Foundation & quick wins (0–2 years). Low cost, high impact. Statewide hazard mapping and exposure prioritisation; Godavari early-warning tied to dam releases for Bhadrachalam and the northern towns (the fastest life-saver); desilt and de-encroach Hyderabad’s nalas before each monsoon; evacuation planning for the deep-water Godavari belt.
Phase 1 — Unclog & restore (2–5 years). Scale SNDP to reconnect the city’s nala network; let HYDRAA restore priority lakes and enforce FTL buffers; extend Mission Kakatiya tank restoration in the flood-buffering catchments; enact and enforce lake-and-floodplain zoning; stand up a permanent annual maintenance budget.
Phase 2 — Basin-scale resilience (5–10 years). Manage the Godavari and Krishna basin-by-basin — coordinated reservoir operation, upstream storage, and retrofit of exposed public assets (hospitals, schools, substations, rail) in the deep-water towns. In Hyderabad, move from nala-by-nala fixes to whole-catchment drainage. Extend the model to every high-exposure district.
Phase 3 — Climate-proof & sustain (10–20+ years). Design to future climate — size drainage, dam rules and buffers for 2050/2100 rainfall (the SSP scenarios) — and lock in maintenance, monitoring and updating as permanent functions.
The honest bottom line: Telangana exposes a smaller share of its buildings than the coastal states, but a larger share of that risk is deep and dangerous — concentrated in a handful of Godavari-valley towns where water stacks high, and in a Hyderabad whose lost lakes and clogged nalas mean the real urban risk runs above what a riverine model shows. Two problems, two fixes: warn and evacuate the deep river towns, and unclog and restore the city. Knowing exactly where the deepest risk sits — and being honest that the deepest numbers are also the least certain — is what lets a limited budget protect the most people.
Sources
- Great Musi Flood (1908) — Wikipedia · Living Waters Museum: the Musi in Hyderabad
- 2020 Hyderabad floods — Wikipedia
- Godavari / Bhadrachalam (July 2022) — Telangana Today: Bhadrachalam flood situation · Telangana Today: ₹1,400 crore flood loss
- Resilience efforts — Mission Kakatiya (Wikipedia) · HYDRAA (Wikipedia)
- 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, and least reliable in the steepest terrain.