A formal recommendation for saving lives through data transparency
During India's 2025 monsoon, dam discharges with inadequate advance warning caused catastrophic flooding across Punjab — over 100 lives lost, 1,900+ villages submerged, and an estimated ₹5,000–10,000 Crore in economic damage. International evidence shows that 72-hour early warning can prevent 80–90% of flood fatalities.
The technology for prediction exists today. The only barrier is data accessibility. This document recommends that the Government mandate real-time, machine-readable publication of dam water level data via open API — enabling automated early warning systems that can save hundreds of lives every monsoon season.
100+ lives lost in 2025 Punjab monsoon due to dam discharges with hardly any credible advance warning to downstream populations.
| Impact Category | 2025 Monsoon Estimate | Source |
|---|---|---|
| Human lives lost | 100+ | State disaster reports, media |
| Livestock drowned | Thousands | District damage assessments |
| Economic damage | ₹5,000 – ₹10,000 Crore | Crop + infrastructure + relief |
| Villages inundated | 1,900+ | Mongabay, Times of India, Indian Express |
| Advance warning given | Hardly any credible warning | Media reports (Indian Express, Tribune) |
Constitutional Imperative: Under Article 21 (Right to Life), citizens have a fundamental right to know when dam gates will open. This is life-or-death information that must reach downstream populations days in advance — not hours.
The barrier to effective flood prediction is not technology — it is data accessibility. Critical dam operations data exists in SCADA/telemetry systems at every major dam, but remains locked behind formats that prevent automated analysis:
| Data Source | Method | Frequency | Machine-Readable? | API? |
|---|---|---|---|---|
| CWC India-WRIS | HTML web page | Once daily | No | No |
| BBMB Daily Bulletin | HTML / PDF | Once daily | No | No |
| IMD Rainfall Forecast | HTML / Image maps | Twice daily | Partial | No |
| Punjab State (PSPCL) | Not published at all | N/A | No | No |
| Gate Opening Decisions | Press release / verbal | Ad-hoc | No | No |
Ranjit Sagar Dam (Thein Dam) on the River Ravi is the largest dam in Punjab, with a storage capacity of 3.28 billion cubic metres. It is managed by Punjab State Power Corporation Limited (PSPCL) and sits upstream of Pathankot, Gurdaspur, and densely populated Ravi floodplain villages.
Currently, Ranjit Sagar reservoir data is not published through any public channel. This represents an opportunity — by establishing even a basic hourly data exposure, Punjab State can bring Ranjit Sagar into the early warning network and protect downstream communities during monsoon season.
We propose that PSPCL publish the following data at minimum hourly intervals during monsoon season (June–September):
| Data Point | Minimum Frequency | Format |
|---|---|---|
| Reservoir level (ft above MSL) | Every 1 hour | JSON endpoint or public bulletin |
| Inflow rate (cusecs) | Every 1 hour | JSON endpoint or public bulletin |
| Outflow / discharge (cusecs) | Every 1 hour | JSON endpoint or public bulletin |
| Gate status (open/closed, count) | On change + hourly | JSON endpoint or push notification |
Why hourly? Dam levels can rise 2–5 feet within hours during peak monsoon inflow. Daily data (as currently published by BBMB for Bhakra/Pong) is a starting point, but hourly readings enable predictive models to detect danger 48–72 hours in advance — the difference between a successful evacuation and a disaster.
Our early warning system is fully built and tested for Ranjit Sagar — the data pipeline, analysis engine, flood simulation, and multi-language alerting are all deployed. The moment Punjab State exposes Ranjit Sagar data through any public channel (API, web page, or even a daily PDF bulletin), our system will activate monitoring within hours. The technology is ready; we only need the data feed.
We recommend the Government mandate a Real-Time Open Data API for all CWC and state-monitored reservoirs. The data already exists in SCADA/telemetry systems at each dam — only the publication layer needs to be built.
| Data Point | Frequency | Format | Latency |
|---|---|---|---|
| Current water level (ft above MSL) | Every 1 hour | JSON REST API | < 5 minutes |
| Inflow rate (cusecs) | Every 1 hour | JSON REST API | < 5 minutes |
| Outflow rate (cusecs) | Every 1 hour | JSON REST API | < 5 minutes |
| Gate opening schedule | As decided (push) | JSON + SMS push | Immediate |
| 72-hour inflow forecast | Every 6 hours | JSON REST API | < 30 minutes |
| Historical readings (90-day) | On-demand query | JSON REST API | < 10 seconds |
Key Point: No new sensor hardware is required. The data already exists in SCADA systems at every major dam. Only the publication layer (API endpoints) needs to be built — a straightforward engineering task achievable in weeks, not months.
International evidence (UNDRR, Bangladesh FFWC, European EFAS) consistently shows that warning lead time is the single most critical factor in flood survival:
| Warning Lead Time | Evacuation Rate | Deaths Prevented | Economic Savings |
|---|---|---|---|
| < 6 hours (current) | 20–35% | ~30% | Minimal |
| 12–24 hours | 60–75% | ~60% | ₹1,000–2,000 Cr |
| 48–72 hours (proposed) | 90–95% | 80–90% | ₹1,500–3,000 Cr |
Per major flood event (2025 Punjab scale):
• Lives saved: 60–90 (of ~100 fatalities)
• Livestock saved: 5,000–15,000 animals
• Economic savings: ₹1,500–3,000 Crore
Early warning saves lives, livestock, and movable property. Standing crops and fixed infrastructure cannot be moved — but people and animals can.