Unlock raster analytics & visualizations - now in your lakehouse!

We’re thrilled to announce a major milestone in spatial data analysis for the modern data lakehouse: Raster support in CARTO is here!

With this release, CARTO becomes the first platform to support all of your raster needs - from ETL to analysis to visualization - directly inside your cloud data lakehouse. Until now, working with raster data in cloud environments has often meant relying on multiple external tools or systems. This fragmented tech stack has created inefficiencies, increased costs, and posed data governance risks.

With this announcement, raster analysis and visualization can now happen natively within your cloud data lakehouse, providing unparalleled scalability, flexibility, and efficiency. Keep reading to learn more and find out how to get started!

Above: global forest cover

Raster data is critical in various industries where spatial insights are essential, such as:

• Telecommunications: mapping network coverage using raster-based datasets.
• Insurance: calculating risk exposure using weather and physical data.
• Climate analysis: identifying ESG risks such as deforestation.

The problem? It’s big. This is particularly the case if you’re working with high-resolution, large-extent or multi-band data. For data of this size, cloud-based processing would seem like a natural fit. However, raster data has - until recently - been incompatible with cloud data lakehouses, as well as the wider data tech stack, requiring specialist GIS and earth observation tools to transform it into insights.  

All of this has meant that users have had to rely on multiple different platforms for their raster needs, placing it in a silo not just from other spatial data, but from their wider data ecosystem as a whole.

The sheer size of raster data has meant that this approach could be a slow and inflexible process - not to mention the data governance issues that stem from a fragmented tech stack, with half of your spatial data “living” outside the cloud.

This is a thing of the past.

With the release of Raster support in CARTO, you can now unlock the power of raster data in the cloud, making it just as accessible and efficient to analyse as vector data. CARTO operates entirely inside your cloud data lakehouse, eliminating the risks of a fragmented tech stack, and enabling users to analyze and visualize raster data in one place.

Why does this matter?

• Time and cost savings: avoid data duplication and conversion, and save storage costs by working directly in the cloud, at an enormous scale.
• Minimize Risk: eliminate issues caused by managing multiple dataset versions.
• Industry-First Innovation: CARTO is the first cloud-native GIS platform with full raster support, paving the way for new open data standards like Parquet raster in the cloud.

Above: global Photovoltaic Power Potential

Openness is at the heart of our raster data approach. Just as CARTO is a leader in supporting GeoParquet for vector data, we’re exploring innovative formats to bring similar advancements to raster data. Our vision includes:

• Leveraging the strengths of Parquet, due to it acting as a foundation for many open-source frameworks and commercially available SaaS and PaaS solutions, such as Delta Lake and Apache Iceberg formats.
• Integrating this with wider open formats like Cloud Optimized GeoTIFFs (COGs), SpatioTemporal Asset Catalogs (STAC), and the exploration of array-based storage technologies like Zarr and Xarray.

You can learn more about our vision for raster on the cloud here.

Above: using raster-based wind data to assess hazards at US airports

Let’s check this out in action! CARTO provides its users with an end-to-end environment for working with raster data - from processing and analysis to large-scale visualizations. Don’t just take our word for it - try it out yourself with this quick walkthrough! If you don’t have access to a CARTO account you can sign up for a free 14-day trial here!

In this tutorial, we’ll walk through how to import raster data to the cloud, run some simple analysis and visualize this in an interactive dashboard.

You’ll need a raster dataset - we’ll be using Global Shipping Traffic Density which you can download from the World Bank here - but you can use any raster. We’ll be using this to determine which global ports have the highest density of shipping activity to help determine supply chain and investment risk.

Not sure where to find the data you need? We have a great list of open data sources in our free ebook Modernizing the spatial analysis stack, or you can skip the import step and use one of the rasters in your CARTO Data Warehouse.

So, you have your raster table - the first step is to get it into our cloud environment. There are a couple of steps we need to take first to ensure a seamless experience once in the cloud. We recommend using GDAL to run these pre-processing steps:

1. Set a No Data value
2. Reproject the raster to EPSG:4326 (WGS 84)
3. Convert the table to a Cloud Optimized geotiff

You can find a full guide to these processes in our Documentation for Snowflake and Google Cloud’s BigQuery, with Databricks support is coming soon.

Once your raster has been optimized for the cloud, you simply need to navigate to the Data Explorer tab of your CARTO Workspace, select Import Data and follow the steps to import the raster - just like you would do when uploading any other file to your lakehouse via CARTO.

Please note that for files larger than 1024MB, you will need to use the CARTO raster loader - you can find instructions for this in your lakehouse's Analytics Toolbox Documentation.

With the upload process complete, you should now be able to see your raster data under your connection in the Data Explorer - and now the fun can start!

Now, let’s use this data to establish which ports around the world have the highest shipping density. We can do this in a few easy steps with CARTO Workflows, our low-code tool for automating multi-step analysis. We'll be doing that by building the below workflow:

1. In the CARTO Workspace, head to the Workflows tab. Create a new workflow through whichever connection you uploaded your shipping density data to. Our end-result workflow will end up looking something like this:
2. In the Connections tab, navigate to where you saved your raster data, and simply drag it onto the canvas.
3. Next, we need to bring our port location data into the workflow; in this case we’ve accessed this via an Import from URL component - which you can find in the  Components tab. This component is sourcing data from this Google Sheet. Note your Workflow needs to be immediately run after using this specific component.
4. We want to create a search area of 1,000 meters around each port, which we’ll use to calculate both the average and maximum shipping density. Connect the Import from URL component to an ST Buffer component with a distance of 1000 meters.
5. Our shipping density raster is - well - massive! It covers the whole world with a resolution of 500 meters. Because of this, we’ll need to break the ports table into two subsets to run the next section of the analysis. Connect your buffer to a Simple Filter component, with the condition that the rank is equal or less than 25. Run your workflow here.
6. Now, time to calculate shipping density! Drag an Intersect and Aggregate Raster component onto the canvas. Connect the shipping density raster to the top input, and the top (positive) output of the filter to the bottom input. Here, you can choose the input bands and aggregation types - we’ll use band 1 (shipping density) and both an average and maximum aggregation type.

7. Repeat this with a second Intersect and Aggregate Raster, this time using the bottom (negative) output of the filter for the bottom input. This will repeat the aggregation process for the second half of the ports table.
8. Run the workflow! You should see that each of your Intersect and Aggregate components now have a band_1_max and band_1_avg field. Now let’s bring them together - connect both to a Union All component to re-merge the two halves of our analysis, then finally connect this to a Save as Table component… now we’re ready to make a map!

You can learn more about the raster functionality that CARTO makes available in your lakehouse through the Raster module of your Analytics Toolbox.

Visualizing raster data in CARTO is as easy as any other spatial layer - just with a few small enhancements! You can choose from a range of raster-specific styling options, including:

• Color Range for continuous data like elevation.
• Unique for categorical data such as land-use.
• RGB for true-color or false-color satellite imagery.

CARTO Builder automatically adapts to your raster's metadata and ensures seamless visualization, even excluding "No Data" pixels for cleaner results. As with vector layers,  you also have full control over opacity, zoom level visibility, layer blending and interactions.

Here’s how we’ve used these powerful capabilities for our example! In addition to creating a stunning map with the massive ship density raster, we’ve overlaid this with the results of our analysis from earlier. Additionally, we've added a series of interactive widgets and pop-ups. This allows users to easily pinpoint the ports - such as Hamburg, Rotterdam and Ningbo-Zhoushan - which have the highest shipping density, helping them to prioritise actions to mitigate supply chain and investment risk.

Why stop there? Enable our new CARTO AI Agents in your map to help users get the most out of your analysis!

Find more tutorials to level-up your geo game in the CARTO Academy.

By enabling raster in the cloud, we’re not only empowering organizations to work smarter and faster but also leading the industry toward new data standards like Parquet raster.

Unlike traditional platforms, Raster in CARTO offers:

• A truly cloud-native approach: Direct integration with cloud data lakehouses. No caching, no copying, no compromises.
• Seamless workflows: Analyze and visualize raster data within the same platform.
• Scalable and secure insights: Built for modern data demands and governance standards.

Ready to take your spatial analysis to the next level? With CARTO’s cloud-native approach, your raster data can drive scalable, actionable insights like never before - start a free 14-day trial to see for yourself!

Want to get started?

Start free trial