TemanNelayan (Manela)

TemanNelayan (Manela)

A satellite-data-driven fisheries platform developed for GEMASTIK XVIII to help fishermen identify potential fishing zones and fish types using oceanographic analysis and AI-assisted insights.

Topics

#Side Project #Research Project #Mobile App #AI Platform #Geospatial Technology #GEMASTIK XVIII 2025

Stack

Javascript Flutter Node.js Express.js Python Google Earth Engine AI Gemini SQLite
Disclaimer:

DISCLAIMER: You can see a whole article talking about the competition HERE, but I will focus on technical side here.


Overview

TemanNelayan (Manela) is a GEMASTIK XVIII project developed collaboratively with Alwan Athallah Mumtaz, achieving 4th place in the competition.

The platform helps fishermen identify potential fishing zones and probable fish species using satellite-derived oceanographic data, analytical processing, and AI-assisted recommendations.


Responsibilities

  • Conducted collaborative research on methodologies for predicting fish locations and species based on oceanographic indicators.
  • Collected, processed, and analyzed::
  1. Sea Surface Temperature (SST)
  2. Chlorophyll-a (CHL-A)
  3. Bathymetry data
  4. Fish Zone Prediction
Satellite Data
  • Utilized Google Earth Engine for large-scale satellite data acquisition.
  • Processed and structured datasets using Python for analytical workflows.
  • Built backend services using Node.js and Express.js.
  • Designed and implemented offline-first system strategies.
  • Implemented partial rendering techniques to improve usability on low-end devices.
  • Participated in field testing and validation directly with fishermen.

Tech Stack

  • Flutter
  • Node.js
  • Express.js
  • Python
  • Google Earth Engine

App User Interface

Final App Mockup
  1. Loading Screen
  2. Home Screen (Weather condition and closest fishing spot)
  3. Fishing Zone Prediction Map
  4. Zone Detail (Zone prediction, distance, sea depth, temperature, wind speed and direction)
  5. Zone Detail (Fish type probability, Ship fuel calculation)
  6. Zone Detail (AI Recommendation)
  7. Map
  8. Fish Category Library
  9. Fish Library
  10. Fish Information

Impact

The project demonstrated how satellite data and lightweight AI-assisted systems could support practical fisheries decision-making in real-world maritime conditions.

Key outcomes included:

  • Supporting fishermen with accessible fishing zone insights
  • Enabling usability in low-connectivity environments
  • Optimizing performance for lower-end mobile devices
  • Achieved approximately 80% validation accuracy during field testing for core predictions and functionalities
  • Securing 4th place at GEMASTIK XVIII 2025
User Testing

What I Learn From this Project

This project strengthened experience in:

  • Satellite and oceanographic data processing
  • Studying fish habitat prediction using oceanographic data
  • Offline-first system architecture
  • Backend engineering for analytical platforms
  • Real-world field validation and user testing
  • Building technology solutions for underserved operational environments

Pitchdeck