Create custom categorical maps from what you see on the globe—no code required.
If you already use Google Earth for your work or community projects, chances are you spend plenty of time doing visual inspection. You pan across the globe, zoom into a coastline or mountain valley, trace property boundaries, and inspect how neighborhoods, forests, or reservoirs are evolving over time. While the human eye is good at recognizing these spatial patterns, visual observations alone cannot easily be quantified, shared with non-technical stakeholders, or integrated into analyses to track changes over time.
Off-the-shelf global land cover maps can help, but they may be too generic for local decision-making. A global map might classify a region as “Forest” or “Agriculture,” when your project actually requires distinguishing native forest from commercial timber plantations, identifying specific crop types, or tracking specific invasive species. Until now, though, creating a custom categorical map may have required specialized expertise in satellite imagery analysis and machine learning code.
Today, we’re thrilled to dive into a new classification tool in Google Earth on the web, helping make the jump from visual inspection to custom categorical mapmaking much easier. You simply draw a boundary around your area of interest, define categorical classes, and label a few sample points. The tool trains a classifier on an underlying dataset of “Satellite Embeddings” (more on this below!) and generates a custom categorical map layer across your area of interest with zero coding required.
Creating your first classified map
Getting started takes just a couple of minutes, whether you follow our step-by-step documentation or jump right into Google Earth on the web. Open an existing map project (or create a new one) and navigate to Tools > Classify. You can then draw a boundary outlining your area of interest. When you're first experimenting with the tool, choose a focused region—such as a local watershed, neighborhood, or park boundary. This helps you quickly get a feel for how the classifier responds to your inputs before scaling to larger areas.
Once your boundary is drawn, select your classification year and establish your classes. You can define custom classes tailored to whatever questions you are exploring—such as distinguishing native forest from commercial timber farms, identifying specific crop varieties, or tracking invasive species. When designing your legend, aim for classes that are mutually exclusive (with no overlap between class definitions) and collectively exhaustive across your region. If your work involves analysis or reporting using standardized classification systems, you can choose “Use classification system” to select classes from global frameworks like the IUCN Global Ecosystem Typology or ESA WorldCover.
With your classes defined, you train the classifier by placing sample points across the embeddings in your boundary. You’ll need at least three sample points per class to establish a baseline. As soon as you place your initial points, the model computes and renders your custom classification layer across the landscape.
The technology under the hood
At the core of the tool is one of the world’s most cutting-edge geospatial technologies – the annual AlphaEarth Foundations Satellite Embedding dataset. The AlphaEarth Foundations model analyzes multiple types of satellite data like multi-spectral optical imagery and radar observations across an entire calendar year, compressing seasonal vegetation dynamics, moisture patterns, and surface textures into a unified representation called an “embedding” for every 10-meter patch of Earth. Datasets and models typically only accessible to enterprises and geospatial experts are now at the beck and call of professionals using Google Earth all over the world.
When you use the classify tool, Google Earth Engine is used behind the scenes to infer the class of each pixel in the selected site based entirely on those embeddings and the labeled sample points you provide. By combining global feature representations generated by AlphaEarth Foundations with Google Earth Engine's cloud compute and your ground knowledge, the tool turns what used to require complex coding into a code-free, interactive workflow.
How building classification maps in Google Earth works under the hood...
Practical tips to get the best results
When you zoom in and place a sample point inside a cell it is applied to that 10-meter pixel.
Understanding how the underlying embeddings interact with your visual basemap is essential for building dependable classifications. AlphaEarth Foundations embeddings capture land surface and environmental properties at a 10-meter resolution. This is coarser than what you see when you zoom in close on Google Earth’s high-resolution basemap imagery. When you zoom in and place a sample point inside one of the cells in the yellow grid, it is applied to that 10-meter pixel. This means that your classified map will capture aggregate properties, not fine-grained spatial details.
As your classification layer renders, you may notice some sections shaded in grey labeled “Inconclusive”. “Inconclusive” represents the tool’s way of communicating that the classifier has identified two or more classes as equally strong candidates for the pixel's likely class. “Inconclusive” zones act as an active learning guide, pinpointing where the model requires further guidance. You can resolve ambiguity by inspecting the satellite imagery in that zone, adding a clarifying sample point, or expanding your legend if a distinct land cover type was omitted.
To achieve high quality results, keep a few additional best practices in mind:
-
Capture the variation within each class. For example, if you have a single “water” class, place sample points in lakes, rivers, shallow coastal ocean, and deep ocean. This helps the classifier learn the full spectrum of each class.
-
Make sure your classes apply over the entire year. The annual satellite embeddings summarize an entire calendar year. You can use historical imagery to get a sense of what a sample point looks like over the year.
-
Maintain a balanced distribution of training sample points across your classes. If you provide thirty points for a forest class but only two points for a water body, the classifier can develop a mathematical bias toward predicting forest in ambiguous areas.
To support downstream quantitative analysis, the tool includes a built-in “Download sample points” feature that exports your labeled sample data as a GeoJSON file. This makes it seamless to bring sample data into tools like Google Earth Engine to re-build your map and create additional samples to perform a full map accuracy assessment.
Try classification in Earth today
Classification in Google Earth is available today as an experimental feature. To explore the feature, launch Google Earth on the web, open or create a project, and choose Tools > Classify. You can also consult our official documentation for comprehensive guidance on advanced settings, classification systems, and visualization tools.
Explore our other geospatial AI offerings
AlphaEarth Foundations is part of Google Earth AI, Google’s collection of geospatial models and datasets. Learn more today.
