Jalmitra AI logo

JALMITRA AI

Smart Fisherman Support System

About Jalmitra AI

Smart Data. Safe Seas. Strong Communities.

๐ŸŽ“ A Student Project โ€” St John's School, Anchal ยท Class 12 B

The Project

Jalmitra AI is a real-time support dashboard designed for the fishing community of Kerala. It brings together sea safety alerts, weather forecasts, and potential fishing zone (PFZ) information โ€” inspired by data from the India Meteorological Department (IMD) and INCOIS โ€” into one easy-to-read screen.

The dashboard currently covers 22 coastal fishing zones across all 9 coastal districts of Kerala, from Muthalapozhi in the south to Bekal in the north, with a bilingual (English/Malayalam) AI chat advisor, a live ocean map, a 7-day wave forecast, and an emergency SOS beacon.

22 zones across 9 coastal districts
Live wind & wave data (Open-Meteo)
Bilingual AI chat advisor (English/Malayalam)
Tide, sea-hazard, and current-flow monitor
Interactive ocean map with fishing zones
Emergency SOS distress beacon

Project Team

Jalmitra AI was designed and built by the students of St John's School, Anchal โ€” Class 12 B.

MF
Mohammed Fadil MF
RB
Rofin Binoy
AS
Ahana N Sumesh
JS
Jemi Anna Sherry

References & Data Sources

This project is a student demonstration, not an official IMD/INCOIS product. Here is exactly where each piece of data comes from:

Wind speed, wave height, sea temperatureLive โ€” Open-Meteo Marine & Weather APIs (open-meteo.com)
Harbour names, districts, coordinatesVerified from public geographic references
Fish species & abundanceSimulated / illustrative, based on general regional fishing patterns โ€” not a live feed
Potential Fishing Zones (PFZ)Concept inspired by INCOIS PFZ advisories โ€” not live INCOIS data
Tide levels & timingSimulated semidiurnal tide model โ€” not official tide tables (see Survey of India / INCOIS for real tide tables)
Sea hazard / safety levelCalculated from live wind + wave readings against our own safety thresholds
AI Catch Prediction (per-species)Weighted model over live chlorophyll/temp + simulated moon/tide, with species preference bands that are illustrative, based on general fisheries knowledge โ€” not measured biological data
Weather anomaly ("unusual today")Live โ€” compares today's reading against that harbour's own real past-7-day average (Open-Meteo historical data)
Confidence scoreCalculated from data completeness/freshness and personal sample size โ€” a transparency signal, not a statistical margin of error
Personalized Risk ProfileShifts the Route Safety Score's Safe/Caution/High-Risk cut-offs by your stated boat size/experience/tolerance โ€” stored on this device only, never changes the underlying weather data
Daily Briefing textAuto-generated from the same live/simulated data above โ€” a written summary, not a new data source
Zone Recommendation (Overall score)Fishing Potential x Safety, using the same two models above โ€” wave period, current speed, visibility, and an official storm feed are not scored, since there is no live channel for them here
Wave period, ocean current, visibilityLIVE โ€” same Open-Meteo Marine & Weather APIs as everything else above, now also requesting these three fields; a missing value stays missing, never assumed safe
Official cyclone/storm warnings (IMD)India Meteorological Department's public API (api.imd.gov.in) โ€” written and wired in, but not independently confirmed reachable from every browser due to possible CORS restrictions; see the project notes for a proxy option
Land/water route checkAn approximation built from Jalmitra's 12 verified harbour coordinates, not a certified nautical chart or coastline survey โ€” reliable for obvious cases, imprecise near river mouths and inlets
Restricted maritime zonesNo verified public dataset of Indian restricted zones is integrated โ€” the check exists in code but has nothing to check against yet, so it never falsely clears or falsely blocks an area
Open-Meteo (live weather) INCOIS (concept reference) IMD (concept reference) Survey of India (concept reference)

Model evaluation & feedback (developer / AI section)