← All research projects

2024 · Co-creator

FIRENET

A platform for forest-fire management that integrates geospatial technologies, artificial intelligence, open meteorological data, and environmental-crime data. Its neural models forecast monthly temperature and precipitation to help prioritize vulnerable areas.

01 / Problem

What is the scientific problem?

Forest fires threaten communities and ecosystems in Santander, while the environmental and meteorological indicators needed for prevention are distributed across separate public datasets.

02 / Motivation

Why does it matter?

Integrating open environmental data can help decision-makers identify vulnerable areas before seasonal conditions intensify fire risk.

03 / Method

How the problem is approached

FIRENET combines geospatial technology with open precipitation, temperature, and environmental-crime data. Convolutional and recurrent neural networks are used to forecast monthly temperature and precipitation, supporting the prioritization of areas during dry periods.

Geospatial AIConvolutional neural networksRecurrent neural networksOpen data

04 / Results

What the work establishes

The platform was awarded second place nationally in Colombia’s Datos a la U competition. Public reporting states that it can identify areas susceptible to forest fires and support regional prioritization.

06 / Future work

Where the project goes next

A verified technical account of this aspect is not yet available in the public record.

Research artifacts

Datasets

2 sources documented above.