UEES - Espiritu Santo University > Sustainable Biodiversity Program

About the Project

ManglArt AI is the computational framework developed within the the Sustainable Biodiversity Program (BIO-SOS, 2021–2050). through Unsupervised Machine Learning, the biosignals that dendrometer sensors capture from mangroves.

The Program provides the institutional, scientific, and territorial context; ManglArt AI provides the analytical layer that turns those data into knowledge for research, environmental education, and decision-making.

Mission

To interpret mangrove biosignals through Artificial Intelligence and high-resolution data, advancing environmental education and strengthening the foundations for a predictive conservation model for Ecuador's coastal ecosystems.

Vision

To position Ecuador as a reference in intelligent ecological monitoring, consolidating urban mangroves and natural reserves as the most important living laboratories in the region.

Why monitor mangroves?

Mangroves fulfill essential functions for coastal protection, biodiversity, and the well-being of communities. Monitoring them makes it possible to understand the changes they undergo and strengthen conservation efforts.

Coastal Protection

 

Reduces erosion and lowers the risk of flooding.

 

Carbon Capture

 

Stores up to four times more carbon than many terrestrial forests.

 

Biodiversity

 

Habitat and nursery for fish, crustaceans, and birds.

 

Food Security

 

Supports fisheries and local economies.

 

Climate Regulation

 

Increases coastal resilience to climate change.

 

How does ManglArt AI work?

Mangroves continuously respond to their environment by expanding and contracting their tissues throughout the day.

ManglArt AI analyzes thousands of records with Unsupervised Machine Learning to recognize the tree's predominant behavior and flag the days that deviate from that pattern for physiological and environmental analysis.

Unlike traditional methods, it does not rely on fixed thresholds: it learns directly from the ecosystem's behavior.

01 · The signal

High-precision dendrometric sensors record every 30 minutes:

  • The growth of the stem.
  • The temperature.

The measurements are taken in Rhizophora × harrisonii.

02 · The model: ManglArt AI

ManglArt AI analyzes the records and identifies patterns associated with:

  • Rehydration.
  • Perspiration.
  • Maximum contraction.

It also identifies atypical days that deviate from the prevailing pattern, so that they can be studied.

03 · Readable indicators

El procesamiento permite obtener indicadores como:

  • Duración de la máxima contracción.
  • Máxima contracción diaria — MDS.
  • Déficit Hídrico del Árbol — TWD.
  • Jornadas atípicas.

Estos indicadores orientan la investigación, la enseñanza y las acciones de conservación.

Scientific note

  • Important: an atypical day does not necessarily equate to deterioration or damage. It could correspond to a severe contraction, a favorable recovery, or a different environmental condition.

Associated research: ManglArt AI

This research presents the scientific and computational contribution of ManglArt AI within the Sustainable Biodiversity Program, applying unsupervised machine learning to the analysis of biosignals obtained from mangroves.

Scientific Research · ManglArt AI

Unsupervised computational framework for the dynamic inference of the water status of Rhizophora × harrisonii in Guayaquil Historical Park

  • AuthorLeonardo López Vallejo, developer of the ManglArt AI computational framework.
  • InstitutionEspíritu Santo University — Faculty of Engineering.
  • Especie monitoreadaRhizophora × harrisonii
  • Ubicación del estudioParque Histórico, Samborondón, Ecuador.
  • Frecuencia de mediciónCada 30 minutos, registrando el incremento del tallo y su temperatura.
  • Número de observaciones10.371 observaciones
  • Methodology
    • Clustering evaluado con Silhouette, Calinski–Harabasz y Davies–Bouldin.
    • Análisis de Componentes Principales — PCA.
    • DBSCAN para la identificación de jornadas atípicas.
    • Matrices de correlación de Spearman.
  • ResultadosSegmentación de estados fisiológicos y detección de jornadas atípicas a partir de un conjunto mínimo de variables:
    • Incremento del tallo.
    • Temperatura.
    • Variables derivadas del tiempo.
  • CongressIII Congreso Internacional de Manglares de las Américas Panamá, 2026.

Strategic Pillars

Technological Innovation

Development of Artificial Intelligence models that transform mangrove biosignals into indicators of ecosystem health.

Educational Ecosystem

Integration of real environmental monitoring data into training, research, and experiential learning: the mangrove as a living classroom.

Restoration and Community Engagement

Strengthening conservation through technological tools, environmental education, and active participation of coastal communities.

Impacto en Números

+0
Hectares reached through restoration, monitoring, and conservation activities between 2023 and 2024.
0
Provinces of Ecuador reached by the program's research, education, and conservation initiatives.
+0
Peer-reviewed publications indexed in Scopus and Web of Science produced within the Sustainable Biodiversity Program.
0
Open AI-based ecophysiological monitoring platform.

Hitos y Reconocimientos Globales

Arizona State University

Premio Internacional — SpaceHack for Sustainability 2026

First place. Our students used the database from our dendrometer sensors to solve global challenges, winning this world-class hackathon.

Panamá 2026

Investigación — III Congreso Internacional de Manglares de las Américas

Research accepted for presentation: a computational framework based on Unsupervised Machine Learning for the dynamic inference of the water status of Rhizophora × harrisonii.

Categoría Sustainability Education Action

Postulación — QS Reimagine Education Awards 2026

Submission to the QS Reimagine Education Awards 2026 in the Sustainability Education Action category.

Strategic Alliances

The Nature Conservancy

Universidad Nacional de Loja

Voices of the Project

PhD. Natalia Molina Moreira

Director of the BIO-SOS 2021-2050 Project

The Program not only met its scientific objectives but also transformed the perception of the mangrove: shifting it from a purely extractive resource to a fundamental pillar of climate and biotechnological stability for Ecuador.

Leonardo López Vallejo

Computational Framework Researcher and Developer

Here, artificial intelligence is not the end; it is the means. We have succeeded in giving the tree a voice in the face of climate threats, and in doing so, we are preparing the new generation to learn to listen to it and take action.

Sources, Publications and Evidence

  • Publicaciones científicas (Scopus / Web of Science) con DOI o enlace
  • Informes institucionales y memorias del Programa BIO-SOS
  • Documento o enlace del III Congreso Internacional de Manglares de las Américas
  • Fuente para la afirmación de captura de carbono de los manglares
  • Fuente para la población de Guayaquil (≈ 2,75 millones, 2022)
  • Repositorios académicos y materiales asociados a la postulación de QS