Engines
Engines, platforms and models that twins are built with.
Knowledge graph of urban climate adaptation plans from the Urban Tech Hub at Cornell Tech; AI agents read city plans and extract actions, hazards, outcomes and costs, each traced to its source sentence.
ECMWF's machine-learning global weather model, run operationally beside its physics-based system since 2025; weights published openly.
-
Geospatial representations and foundation models for downstream mapping tasks.
Agent-based, discrete-event and system-dynamics modeling; pedestrian and transport libraries.
-
Microsoft Research foundation model for the Earth system, fine-tuned for global weather, air pollution and ocean-wave forecasting; code and weights open under MIT.
-
Represent entities and relationships and connect data within twin applications.
-
Integrates infrastructure engineering, geographic, reality and sensor data.
-
Open-source Python library by James Adams that computes drought, fire and flood-potential indices such as SPI, SPEI, the Palmer family and KBDI from station or gridded precipitation and temperature.
Analyzes climate exposure, vulnerability and investment choices.
-
Open-source Android app from the non-profit Awana Digital for territory monitoring and mapping, co-designed with Indigenous communities; teams record observations on a map and share them peer to peer.
Open single-class fish detector built by Filippo Varini and a group of contributors, trained on seventeen open datasets pooled into one; finds fish in underwater images and video from any camera or habitat.
-
Models hazards and infrastructure dependencies to examine cascading impacts.
-
DAFNI supports infrastructure modeling; Wenwen Li’s research addresses GeoAI and geospatial integration.
-
Open-source pipeline from EPFL's ECEO lab that turns handheld underwater video of coral reefs into labelled 3D point clouds, orthomosaics and benthic cover statistics.
-
Feature of Deltares' Delft-FEWS forecasting platform, from version 2026.02, that pulls Google Flood Hub river forecasts and flood status through Google's Flood Forecasting API so forecasters can compare them with their own models and observations.
-
Models coastal, estuarine and river systems, including water flow and related processes.
-
Climate and extremes twins with supporting computing and data workflows.
-
-
Open object detector from the ConservationDronesAI group, built on RF-DETR, that finds people, mammals and vehicles in drone imagery for wildlife surveys and counts.
Simulates crop growth under specified weather, soil and management conditions.
-
Finds environmental features in satellite imagery through examples, search and review.
-
Simulates road vehicles, public transport and pedestrians on large networks.
-
Molecular atlas that links microbial genes and transcripts from wetland and coastal sediments to field measurements of methane, nitrous oxide and sulfur, as evidence for blue-carbon projects; contacts are at Ecosphere Blue, whose MethaNet product is built on it.
-
Explores combinations of climate policies and their modeled effects.
-
Maps, spatial analysis and geographic data for city and infrastructure applications.
-
Prototype web app by Hannah Nisonson of Esri's Applications Prototype Lab that links a 3D map of species observations in the San Bernardino National Forest with a 3D food-web graph, and shows what else is affected when a species is removed.
-
Wildfire risk platform from Opal AI Inc. of Marina del Rey, California, that combines imagery, lidar, GIS, weather and field data into risk rankings and mitigation recommendations for utilities, fire agencies and insurers.
-
EarthDaily's wildfire risk model, which combines satellite observations, daily weather and other geospatial data into ignition likelihood, fuel hazard and directional vulnerability layers, sold to insurers in the western United States.
-
Tropocen's browser-based GIS for urban heat adaptation, with a 3D model of a city district, PALM-4U microclimate simulation and a planning editor for comparing cooling measures.
-
Spatial agent-based models for urban mobility, evacuation, environmental and other applications.
-
Connects energy, water, land, economy and climate in regional and global scenarios.
-
Organizes spatial proposals, alternatives and stakeholder negotiation.
-
Experimental Gemini-based agent from Google Research that plans a multi-step spatial analysis, calls Earth AI models and geospatial data sources, and returns the result with its plan and sources shown.
-
Google's family of geospatial AI models and agents, covering satellite imagery, weather, floods and population, reached through Google Earth, Earth Engine, BigQuery, Maps Platform and Google Cloud.
-
Free public site from Google Research with riverine flood forecasts, flood maps and water trends for river basins in over 150 countries, updated daily; the same forecasts reach Search, Maps and Android and can be pulled by API.
-
Google Research's open-source hydrology framework, a Python package that replicates the LSTM river-flow models behind Flood Hub so forecasters can train or fine-tune them on their own watersheds.
-
Coordinate model execution or exchange model information.
-
-
Maps and values ecosystem services in terrestrial, freshwater, coastal and urban settings.
-
AI weather model from Schmitt, Jain and co-authors that retrains ECMWF's AIFS on satellite precipitation observations to forecast rain more accurately at the local scale.
Open-source streaming 3D reconstruction model from the Robbyant team at Ant Group; turns a video or image sequence into camera poses, depth and point clouds in one feed-forward pass.
-
Large-scale agent-based transport modeling, including travel demand and replanning.
-
Open-source family of wildlife detection models and tools from Microsoft's AI for Good Lab; MegaDetector finds animals, people and vehicles in camera-trap images, and PyTorch-Wildlife is the framework that runs it and its siblings for aerial and acoustic data.
-
Ecosphere Blue's planned assessment service for blue-carbon projects, rating project performance with methane-risk projections drawn from microbial genomic evidence in mangrove sediment.
-
Hosted mapping and visual positioning service from MultiSet AI that turns one drone flight over a site into a localisation map which drones, phones, headsets and robots then query for their position.
-
-
Catalogue from Niantic Spatial of 100 real-world environments captured as simulation-ready USDZ scenes, for training and testing robot policies in Isaac Sim and other OpenUSD simulators.
-
NVIDIA's family of world foundation models and tools for physical AI; generates and reasons over video of the physical world for robot training, simulation and synthetic sensor data.
-
NVIDIA's open-source robotics simulator built on Omniverse and OpenUSD; used to train, test and validate robots in physically based virtual environments and to generate synthetic training data.
-
World model from Odyssey that generates an interactive 3D environment from a prompt and predicts how it changes as a person or agent moves and acts in it.
-
-
Universal Scene Description, the open 3D scene format and software platform created by Pixar and now stewarded by the Alliance for OpenUSD; the scene layer under NVIDIA Omniverse and many simulation pipelines.
-
Open-source detector from the Microsoft AI for Good Lab that finds wildlife in aerial and drone survey imagery; ships the OWL model family and extends MegaDetector from camera traps to overhead views.
-
Open-source data analysis and visualization application from Kitware, used to look at the output of simulations from laptop to supercomputer scale.
-
Part of Google Earth AI; given a user's local outcome data such as disease rates or crop yields, it picks relevant population and satellite signals and trains models for spatial prediction, downscaling and forecasting.
-
Google Research geospatial foundation model that turns aggregated search trends, place busyness, maps, weather and air quality into an embedding vector per postal code or county, for predicting local health, economic and environmental indicators.
-
Models power systems and energy transitions; PyPSA-Eur also represents transport, buildings and industry.
-
-
Pilot by The Nature Conservancy Argentina and the startup ReForest Latam that seeds native trees from drones on degraded farmland in Argentina's Gran Chaco.
-
Modeled travel activity and scenario tools for transport and regional planning.
-
Open-source, solar-powered edge device from Microsoft's AI for Good Lab that runs camera-trap and bioacoustic AI models on site and sends the results out by satellite or GSM, for biodiversity monitoring in remote places.
-
Real-time underwater SLAM from the University of Michigan field robotics group that fuses learned stereo depth with IMU, pressure and Doppler velocity log measurements; developed and tested on shipwreck surveys.
-
Models wildfire behavior and exposure for fire agencies and electric utilities.
-
-
Explores linked urban scenarios using a simulation engine and browser interfaces.
-
Models location choice, development and urban growth, with transport-model integration.
-
The Amazon Conservation Team and the AARIMO network use Vantor satellite imagery to find and track illegal gold-mining dredges along the Puré River in the Colombian Amazon and pass the findings to enforcement agencies.
-
Feed-forward 3D vision model from Meta AI and Oxford's Visual Geometry Group that predicts camera poses, depth and point clouds from video; used as a reconstruction backend in DeepReefMap for coral reef mapping.
-
VU.CITY supplies 3D planning context; CLARA models urban environmental conditions.
-
Caribbean project testing whether dense restored stands of elkhorn coral reduce wave energy and sediment movement along coasts, with sites in Jamaica, Barbados and Antigua & Barbuda; funded by the Caribbean Biodiversity Fund.
-
Weather prediction and supporting components, with different distribution and developer options.
-
Wildflow's platform for coral reefs, where divers upload footage and get georeferenced 3D models, orthomosaics and bathymetry for tracking a reef over time, with the stated aim of a digital twin of any reef.
-
Learns spatial representations for scene reconstruction, generation and robotics simulation.
-
-
Robotic surface vehicle from Woods Hole Oceanographic Institution, built by Peter Traykovski and adapted for Anne Cohen's reef work, that photographs shallow coral reefs along GPS-tagged tracks so the same colonies can be checked again after each heat wave.