Species Distribution Modeling with Google Earth Engine
- Start Date:
- July 7, 2026
- End Date:
- July 14, 2026
- Description:
- Species distribution models (SDMs), sometimes referred to as ecological niche or habitat suitability models, are a commonly used analytical technique to link known species locations with environmental predictor variables and assess patterns of species occurrence and habitat suitability. SDMs have been successfully used to model the distributions of various species, from butterflies to baleen whales, with a goal of using associations with habitat variables — such as elevation, vegetation greenness, or distance to human settlements — to map the potential distribution to a species across a landscape or seascape. To implement SDMs, we will fit models using Google Earth Engine (GEE), a cloud-based spatial analysis platform that can simplify the process of accessing and analyzing huge quantities of remotely sensed data. This course begins by introducing key SDM concepts for participants working in ecology, conservation, or wildlife biology. The course then provides a basic introduction to Google Earth Engine (GEE) and JavaScript coding, before moving on to spatial data manipulation and example workflows for species distribution mapping in GEE. Participants need no prior experience working with GEE, but we assume participants have a basic understanding of GIS data and concepts, such as rasters (including stacks or multi-band images), vectors, projections, and spatial resolution.
- Targeted Audience:
- NGOs and technicians, GIS officers, and academics, Early career scientists and postgraduate students.
- Expected Outcomes:
- By the end of this training attendees will be able to: Conduct basic operations and functions in the Google Earth Engine (GEE) web interface using JavaScript code. Identify core concepts and applications of species distribution modeling (SDM). Assess key SDM workflow decisions to ensure occurrence point independence, to select background points, and to choose ecologically relevant predictors. Access and manage analysis-ready spatial data from the GEE catalog as predictor variables. Fit species distribution models using machine learning models (e.g., Random Forest) in GEE using a reproducible code-based workflow. Critically evaluate the appropriateness of the model design and interpret predicted distribution and habitat suitability results.
- Host:
- NASA Applied Remote Sensing Training Program (ARSET), University College Cork, Smithsonian Institute
- Organizer:
- NASA Applied Remote Sensing Training Program (ARSET), University College Cork, Smithsonian Institute
- Format/Training Type:
- Webinar
- Language:
- English
- Attendance:
- Open
- Application Deadline:
- July 14, 2026
- Qualifications:
Part 2 of Visualizing Land Cover and Land Use Change with NASA Satellite Imagery (https://www.earthdata.nasa.gov/learn/trainings/visualizing-land-cover-land-use-change-nasa-satellite-imagery), General knowledge of GIS, Some coding experience.
- Contact Name:
- Brock Blevins
- Contact Email:
- brock.blevins@nasa.gov
- Link:
- https://www.earthdata.nasa.gov/learn/trainings/species-distribution-modeling-google-earth-engine
- Share this event: