What challenges did National Geographic Society face before AI implementation?

Difficulty measuring environmental changes (glacier melt, land use, migration) at global scale. Traditional field methods too costly, slow, or imprecise for wildlife tracking and ecosystem monitoring. Limited tools to predict and respond proactively to climate risks, such as harmful algal blooms and water shortages. Silos and resource constraints affecting cross-border species conservation (e.g., lions) and agricultural resource management.

What was the impact of the AI solution?

Improved global monitoring and protection for threatened species (penguins, birds, lions, insects). Made real-time land, water, and species data accessible to decision-makers and the public, informing policy on conservation and sustainability.

Source

MicrosoftPilotEvidence: Medium50/100

AI-Powered Environmental Solutions Accelerate Global Sustainability

Use case typeAgriculture optimizationUpdated Jun 13, 2026

Microsoft, in partnership with the National Geographic Society, awarded AI for Earth Innovation Grants to support 11 projects addressing climate change, biodiversity, water, and sustainable agriculture. Grantees applied AI and Azure technologies to real-world issues: monitoring melting glaciers using drone and satellite imagery, predicting climate-driven migration by digitizing decades of aerial photos, and employing computer vision for penguin population tracking. Grant projects included using AI-powered acoustics to identify bird songs for species monitoring, audio recognition for insect populations in rainforests, and collaborative AI platforms to track lion movements across reserves. In agriculture, advanced machine learning models merged meteorological, satellite, and optical data to optimize irrigation practices in Uganda and map groundwater usage, helping shape future policy for efficient water use. Land cover mapping in Murchison Falls detected ongoing environmental changes and impacts from competing land use priorities with supervised learning. AI-powered satellite analyses provided early warnings for harmful algal blooms in Guatemala and open-source models detected and mapped thousands of small, unmapped dams and reservoirs worldwide. Results empowered researchers, policymakers, and conservationists globally with accessible data and predictive insights for decision-making.

Organization: National Geographic Society
Industry: Agriculture
Location: Global
Published: December 2018

Planned next steps

The source says the pilot could deliver: Optimized irrigation to reduce water use.

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Quantified impact: 30% decrease

Microsoft BlogsDec 11, 2018Blog postInferred claimMedium evidence strength

Reduced agricultural water use via optimized irrigation by up to 30% in pilot regions.

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: National Geographic Society
Provider: Microsoft
Maturity: Pilot
Linked source: Microsoft Blogs

Reduced agricultural water use via optimized irrigation by up to 30% in pilot regions

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 5

1AI-based glacier monitoring with satellite and drone imagery
2Prediction of climate-induced migration using digitized aerial photos
3Automated animal population tracking with computer vision and acoustic analysis

Difficulty measuring environmental changes (glacier melt, land use, migration) at global scale.
Traditional field methods too costly, slow, or imprecise for wildlife tracking and ecosystem monitoring.
Limited tools to predict and respond proactively to climate risks, such as harmful algal blooms and water shortages.
Silos and resource constraints affecting cross-border species conservation (e.g., lions) and agricultural resource management.

Machine learning and computer vision applied to diverse satellite, drone, and acoustic data sources for rapid, scalable analysis.
Development of open-source AI algorithms for specific conservation and agricultural needs (e.g., irrigation optimization, dam mapping).
AI-powered platforms and collaborative databases for tracking and sharing biodiversity data in real time.
Forecasting models for climate migration and proactive ecosystem management via predictive analytics.

Technologies

Azure Machine learning AI

Improved global monitoring and protection for threatened species (penguins, birds, lions, insects).
Made real-time land, water, and species data accessible to decision-makers and the public, informing policy on conservation and sustainability.

Sources & evidence1

Evidence: Medium50/100Evidence strength