200,000+
Plots & polygons mapped
Georeferenced smallholder parcels and restoration boundaries across Africa and Southeast Asia.
Open-source geospatial platform
Map-first data collection for communities protecting the planet
Ground is an open-source, offline-first field data collection platform built by engineers at Google and maintained by the Food and Agriculture Organization of the United Nations (FAO) under the Open Foris initiative. Designed for local stakeholders, Ground connects frontline community observations with cloud-based mapping and sustainability analytics.
Deployed alongside international agencies, research institutes, and farmer cooperatives
Web survey console: Organizers design surveys, manage team access, and inspect geotagged field submissions over high-resolution satellite imagery.
30,000+
Coffee plots in Burundi
Registered under extreme rural connectivity constraints with ADISCO and the International Trade Centre.
21,000+
Cocoa polygons in Ghana
Collected with CERSGIS, SERVIR West Africa, WRI, and YouthMappers to train open climate and crop models.
15,000+
Plots in Ondo State, Nigeria
Cocoa and oil palm polygons mapped in Forest Reserves to issue farmer permits and provide agroforestry seedlings.
Current goals & strategic focus
Closing the gap between satellite observation and reality on the ground
Remote sensing and planetary models can detect landscape change from orbit, but protecting forests, restoring ecosystems, and securing smallholder livelihoods require trusted verification on the ground. Ground provides a "just right" geospatial platform so frontline communities and institutions can collect rigorous field data without specialized GIS training.
Enable deforestation-free, traceable smallholder supply chains
Global regulations such as the EU Deforestation Regulation (EUDR) and voluntary forest conservation frameworks (jREDD, Tropical Forest Forever Facility) require plot-level georeferencing for forest-risk commodities like cocoa, coffee, rubber, and timber. Ground enables cooperatives and field teams to map farm perimeters and points offline so smallholders aren't excluded from international markets.
- EUDR smallholder compliance
- Timber & commodity traceability
- Open Foris WHISP integration
Empower Indigenous Peoples, local communities & cooperatives
Instead of relying on external GIS contractors, Ground is designed so farmer cooperatives, Indigenous Peoples, and local community organizers can design and manage their own mapping campaigns. By putting spatial data ownership directly in community hands, local producers strengthen land stewardship and unlock new economic opportunities.
- Cooperative-led campaigns
- Community land management
- Zero GIS training required
Monitor ecosystem restoration & conservation outcomes
Conservation practitioners and national ministries use Ground to delineate restoration areas, record IUCN ecosystem biome classifications, and capture geotagged photo evidence over time—supporting transparent reporting to the UN Decade on Ecosystem Restoration Monitoring (FERM) and national forestry initiatives.
- UN FERM restoration reporting
- IUCN biome documentation
- Geotagged field photo evidence
Generate high-quality training & validation data for Earth AI
Accurate land-use and land-cover maps depend on verified ground truth—especially for shaded agroforestry systems and smallholder mosaics that look ambiguous from satellite imagery alone. Ground collects GPS-verified polygons with explicit location accuracy and altitude metadata to train and validate machine learning models for the Forest Data Partnership and FAO geospatial teams.
- ML training & validation sets
- Forest Data Partnership models
- CC-BY 4.0 open science datasets
Positive impact in the field
Proven at scale with farmers, cooperatives, and conservation teams
From national smallholder registries in East Africa to open-science agroforestry campaigns in West Africa and multi-commodity traceability pilots in Southeast Asia, Ground is delivering measurable environmental and economic impact today.
Burundi Coffee Registry: National-scale offline mapping for market access
Led by the Association pour l'Entraide et le Développement Communautaire (ADISCO) and the International Trade Centre (ITC), field teams deployed Ground across rural coffee-growing provinces operating under severe connectivity constraints. Reliable offline data capture and background synchronization enabled rapid registration of smallholder growers to safeguard their access to global export markets.
30,000+
Smallholder coffee plots mapped
100%
Offline field operation capability
Partners: ADISCO, International Trade Centre (ITC), Open Foris / FAO
Ghana Cocoa Campaign: Reference polygons for climate & forest AI models
During the Lacuna Fund Climate & Forests project, a consortium of ~30 enumerators—including CERSGIS staff, University of Ghana students, and YouthMappers from the University of Ghana and University of Cape Coast—mapped cocoa farms alongside shade-tree and management attributes. The data trained and validated Forest Data Partnership (FDaP) cocoa models, with an open CC-BY 4.0 subset published on Zenodo.
21,000+
Cocoa farm polygons collected
2,000+
Zenodo open dataset views (280+ downloads)
Partners: CERSGIS, SERVIR West Africa, University of Ghana,
University of Cape Coast YouthMappers, World Resources Institute (WRI), Lacuna Fund
Ondo State Agroforestry: State-led cocoa & oil palm geolocation in forest reserves
Following a 2025 stakeholder meeting involving the governments of Ondo, Ekiti, Abia, and Cross River States alongside GIZ, TRACE, and FAO GEF7, the Government of Ondo State independently launched a geolocation campaign using Open Foris Ground with technical support from FAO GEF7. Working across Forest Reserves and Protected Areas, teams geolocate cocoa and oil palm plots to identify farmers, issue temporary permits, and provide tree seedlings for agroforestry.
15,000+
Polygons geolocated as of Sept 2026
+6,000–7,000
Additional plots projected by end of 2026
Partners: Government of Ondo State, FAO GEF7 (following multi-state
stakeholder engagement with Ekiti, Abia & Cross River States, GIZ, and TRACE)
Kenya Coffee Cooperatives: Scaling farmer-led cooperative mapping
Following initial pilots with Agrisolutions Enterprises Ltd (ASEL), Palladium Group, and ITC that mapped 6,000 coffee plots, local cooperative coordinators in Nandi County designed and managed their own Ground surveys. Within two months, four cooperative coordinators independently mapped over 2,000 additional plots—reaching ~60% coverage of Toroton (1,400/2,100 plots) and Meteitei Set Kobor (400/700 plots) Farmers Cooperative Societies.
8,000+
Total coffee plots mapped in Kenya
~60%
Cooperative coverage achieved in <2 months
Partners: Toroton FCS, Meteitei Set Kobor FCS, FAO, ASEL, Palladium
Group, International Trade Centre (ITC)
Viet Nam multi-commodity pilots & Open Foris WHISP risk analysis
In Viet Nam, FAO, local forest rangers, enterprises (Duc Hien Co., Quang Tri Trading Corp.), and cooperatives (Bon Phuong, Phat Loc, Hiep Phat) are running georeferencing campaigns across coffee (41 plots), timber (38 plots), and rubber (25 plots). Globally, field geometries collected in Ground feed directly into Open Foris WHISP to evaluate deforestation risk across millions of smallholder parcels.
3 commodities
Coffee, rubber & timber georeferenced
4.1M+ plots
Analyzed across the Open Foris WHISP ecosystem
Partners: FAO Viet Nam, Forest Rangers, Bon Phuong, Phat Loc &
Hiep Phat Cooperatives, Open Foris WHISP
Platform overview
Designed for the realities of disconnected fieldwork
Ground combines a web application for survey creation and data management with an Android app built for fast, accurate map-based data collection in remote environments.
- Pre-defined & opportunistic sites Import target collection locations in advance or let collectors add new sites in the field.
- Multi-language web console Switch interface languages directly from the web header to collaborate across global teams.
- Offline Sentinel-2 basemaps Cache cloud-free satellite imagery tiles to navigate accurately under remote canopy.
- GPS points & walked polygons Capture hand-drawn or GPS-walked boundaries alongside photos and structured form inputs.
- Accuracy & altitude exports CSV exports include dedicated GPS accuracy and altitude columns for scientific quality control.
- Resilient background sync Local data persists safely across app restarts and uploads automatically when back online.
Geospatial & structured data
First-class support for GPS points and walked or hand-drawn polygons paired with text, numeric, multiple-choice, date/time, and photo capture tasks.
Built on Material Design 3
Crafted using Google's Material Design 3 system for high-contrast legibility in bright sunlight and intuitive navigation with minimal training.
Open standards & self-hostable
Export clean GeoJSON and CSV datasets for downstream GIS, WHISP, or Earth Engine workflows, and deploy on your own Firebase and Google Cloud project.
Open source & community
Built in the open for global sustainability impact
Ground is released under the Apache 2.0 License as part of the Open Foris initiative. We welcome contributions from developers, researchers, and conservation practitioners worldwide.
Web & Cloud Backend
Explore google/ground-platform →
ground-platform
Contains Ground's hosted components, including the Angular web survey console, Firebase Cloud Functions, Firestore security rules, and shared TypeScript and Protocol Buffer schemas.
Android Client
Explore google/ground-android →
ground-android
Contains the native Android application for offline-first map navigation, GPS point and polygon boundary capture, offline satellite tile caching, and background synchronization.
Community & Updates
Stay informed & contribute
Join the Ground announcements mailing list for updates on field deployments and releases, or read the Developer's Guide to set up a local emulator and contribute pull requests.