top of page

Tech4Nature MEXICO

THE ROAD TO TECH4NATURE MEXICO'S PHASE 2

READ OUR REPORT!

Cover_reportF1.png

Nestled in the heart of the Yucatán Peninsula, the Dzilam State Reserve is a biodiversity hotspot. Enter Tech4Nature Mexico, the result of a collaboration between tech experts, conservationists, and local communities. This project is all about using AI for good – monitoring, understanding, and safeguarding biodiversity like never before.

Tech4Nature Mexico showcases a unique blend of community-driven methodologies, machine learning techniques, and multi-sector partnerships. This project underscores the significance of responsible and ethical development of advanced technologies, ensuring proactive mitigation of social and environmental risks.

Access the webinar The Road to Tech4Nature Mexico Phase 2 to meet the people and learn the stories behind the project, and don't miss our newly launched report which not only presents the methodology and achievements of this initiative, but also the opportunities and strategies to scale and replicate initiatives like these that leverage AI systems for conservation.

OUR INITIATIVE

Tech4Nature Mexico aims to accelerate the effective conservation and regeneration of biodiversity and ecosystem health by strengthening the monitoring, conservation, and understanding of the effects of climate change on priority ecosystems and species in the mangrove zone of the Yucatan Peninsula. 

 

This is achieved by employing the power of community-centered approaches, AI systems, and multi-sectoral collaborations in the Dzilam de Bravo reserve.

OUR ACTIONS

WhatsApp Image 2022-06-20 at 1.33.25 PM(1).jpeg

Biodiverisity Monitoring

WhatsApp Image 2022-06-20 at 1.33.26 PM(1).jpeg

Regional CNNs for species detection and identification

Image by Luca Bravo

AI Ethics Protocols

Jaguar (Panthera onca.JPG

Data-informed conservation actions

THE RESERVE

Our mission at Tech4Nature centers on understanding, preserving and restoring the Dzilam State Reserve to provide a secure sanctuary for a rich variety of plant and animal species.

Nestled in the northeastern region of Yucatan, the Dzilam State Reserve is a natural protected area with over 69,000 hectares that belongs to the municipalities of Dzilam de Bravo and San Felipe.

This reserve holds a special status as a critical wetland conservation site, boasting nearly 290 species of fauna intricately linked with over 300 flora species. It spans five distinct vegetation types, including coastal dunes, mangroves, petenes, along with vibrant aquatic flora in coastal lagoons. 

HIGHLIGHTED RESULTS

+80,000
IMAGES

Analyzed with the 

ModelArts AI platform from Huawei Cloud

+600,000
AUDIO RECORDINGS

  Analyzed with the 

Arbimon AI platform, from Rainforest Connection

Sin título (1920 x 1920 px).png

+120 SPECIES IDENTIFIED

These species are part of a Convolutional Neural Network (CNN) for automatic identification

Diseño sin título (2).png

7 JAGUARS
IDENTIFIED

Automatically identified and validated by experts

OUR IMPACT

LOCAL IMPACT

  • The monitoring algorithms provided valuable information that validated the presence of threatened species in the Reserve in a way never before verified.

  • The data provided tools to strengthen knowledge of the biodiversity in the area and local involvement to reinforce conservation.

  • Forged a powerful alliance between the automatic jaguar identification model and esteemed biologists specializing in jaguar studies.

  • Human threats are and will continue to be reduced thanks to the continuous monitoring and presence of authorities in the area, strengthening the protection of jaguars in the area.

  • Local partners and the local community have been trained in the use of these technologies for the protection of nature.

INTERNATIONAL IMPACT

  • The information generated is supporting the process of including the Reserve in the IUCN Green List of Protected and Conserved Areas.

  • Inputs for the replicability of AI tools and resources for the protection of biodiversity for Latin America were generated.

  • Creation of recommendations for good practices in the ethical use of AI for conservation.

  • Pilot positioning as an international case study.

IMG_6620 (1)

IMG_6620 (1)

implementacion-programa-busca-promover-soluciones

implementacion-programa-busca-promover-soluciones

IMG_4271

IMG_4271

c34a21d7-a898-4f41-8f2f-f8243a4c6430

c34a21d7-a898-4f41-8f2f-f8243a4c6430

1683843972543

1683843972543

WhatsApp Image 2023-07-11 at 6.38.24 PM

WhatsApp Image 2023-07-11 at 6.38.24 PM

20230803_162523

20230803_162523

20230803_184528

20230803_184528

07F401D4-526C-4A47-A67F-3D8CE63D7AE7

07F401D4-526C-4A47-A67F-3D8CE63D7AE7

WhatsApp Image 2022-05-08 at 8.57.38 PM (3)

WhatsApp Image 2022-05-08 at 8.57.38 PM (3)

IMG_7048

IMG_7048

WhatsApp Image 2022-06-20 at 1.33.58 PM

WhatsApp Image 2022-06-20 at 1.33.58 PM

WhatsApp Image 2022-06-20 at 1.33.26 PM

WhatsApp Image 2022-06-20 at 1.33.26 PM

WhatsApp Image 2022-06-20 at 1.33.25 PM(2)

WhatsApp Image 2022-06-20 at 1.33.25 PM(2)

WhatsApp Image 2022-08-31 at 2.14.52 PM

WhatsApp Image 2022-08-31 at 2.14.52 PM

THE BIODIVERSITY

Watch, listen, and see the biodiversity of the Dzilam de Bravo reserve!

IMG_0038
Crax curios
Ocelote Closer
Ocelote ears
IMG_0026_Trim
IMG_0071
IMG_0004 - copia
IMG_0029

DIGITAL COMPONENTS

PROGRAMME EXPLAINABILITY, TRANSPARENCY, INCLUSIVENESS​​, AND OTHER PRINCIPLES

AI Ethics is at the core of this project. Based on UNESCO's Recommendation on AI Ethics  (in which C Minds' founder had the role of leading the Environmental chapter), geographic locations and other sensitive information is protected. 

If you are a scientist and need access for scientific research and conservation purposes only, please email the project's data governance committee at regina@cminds.co

ARBIMON_RCFx.min-680x400.png

PATTERN-MATCHING ALGORITHMS FOR SPECIES IDENTIFICATION THROUGH SOUND

Audio analysis model through Airbimon, an audio analysis platform developed by Rainforest Connection (RFCx), to identify, classify and create patterns of biodiversity in Dzilam.

image6.png

IMAGE RECOGNITION ALGORITHMS FOR JAGUARS’ DETECTION AND IDENTIFICATION

Development of Convolutional Neural Networks to label images by the (non)presence of a jaguar, detect its location within an image, and determine jaguar individuality across multiple images.

OUR TEAM

Meet the driving force behind Tech4Nature! 

Our talented and passionate team is committed to preserving the environment and bringing diverse expertise to make a real impact.

 

Get acquainted with the faces behind our conservation efforts.

AS SEEN IN

1
2
bottom of page