Tutorial: Getting Started with COG Data

Divya Sakpal
Divya Sakpal
  • Updated

Introduction

Cloud-native geospatial data formats are designed to be accessed and processed directly from remote cloud storage without the need to download entire datasets, essentially allowing you to “read what you need”. One example is the Cloud Optimized GeoTIFF (COG), a type of GeoTIFF designed to be read quickly over the Internet by loading only the needed parts of the file.

This tutorial demonstrates how to inspect a COG dataset using FME's Data Inspector. It provides a step-by-step guide to opening a remote COG file and configuring the appropriate reader parameters, setting the spatial extent and resolution. For more information about the reader, please see the documentation.

 

Step-by-Step Instructions

Let’s take a closer look at how to inspect a COG in FME Data Inspector. The COG format has been supported since version 2023.0. However, it is recommended that you use the latest version of FME for all cloud-native formats.

1. Open FME Data Inspector and Open a Dataset

In FME Data Inspector, click on the Open button, which will add the dataset to a new view directly. You could also use the Add button, which will add the dataset to the list at the bottom. From there, you can open the dataset in a new view.

 

Continue to Step 2 before clicking OK. 

 

Copy and paste the path in the Dataset field. Do NOT open in a browser - The file is a 31GB COG dataset. If you accidentally click the link, cancel the download.

 

2. Set the Parameters

Click the “Parameters..” button to set additional parameters. To view the entire COG dataset:

  • Use Search Envelope: Unchecked
  • Pyramid Level to Read: 10

Then click OK. 

Using pyramid level 10, which is the lowest resolution of 16 km x 16 km, will ensure the lowest resolution for this large dataset and render the image faster.

 

To view a certain area of the DEM at the highest resolution of 30m, reduce the extent of the raster by enabling Use Search Envelope.

  • Use Search Envelope: Checked
    • Minimum X: -123
    • Minimum Y: 49
    • Maximum X: -122
    • Maximum Y: 49.5
  • Search Envelope Coordinate System: LL84
  • Clip to Search Envelope: Checked
  • Pyramid Level to Read: 1

Then click OK. 

Level 1 is a 30 m resolution.

 

3. Explore the Data

Once either the full dataset or a subset is added to the view, it can be inspected using the background map, and details for each pixel will appear in the Feature Information window.

Below is a screenshot of the full dataset. This is the DEM of Canada without the search envelope and at the lowest-level resolution of 16km x 16km.

 

And below is the screenshot of the subset of the same DEM. Pyramid Level = 1 displays the highest resolution of 30m x 30m.

 

Summary

In this tutorial, we used FME to read and inspect a Cloud Optimized GeoTIFF (COG) file hosted in a public AWS S3 bucket. Using the COG Reader in FME Data Inspector, we demonstrated how to open the dataset either fully or by defining a search envelope to focus on a specific area.

We explored how setting different pyramid levels affects the resolution of the data being viewed, with level 10 used for full overviews at low resolution, and level 1 enabling high-resolution (30m) inspection over a defined extent. We also set the appropriate coordinate systems for both the dataset and the search envelope.

This exercise demonstrates how FME supports efficient inspection of large cloud-native raster datasets and emphasizes the importance of parameter tuning for scalable data exploration.

 

Articles

Converting a STAC Asset to Cloud Optimized GeoTIFF (COG)

 

 

Was this article helpful?

We're sorry to hear that.

Please tell us why.

As of January 14th, 2026, comments on knowledge base articles have been closed. To make sure questions don’t get missed and to enable more community support, we’ve moved discussions to the FME Community. If you have a question or a comment about this article, please create a new post or create a support ticket.