1. Introduction
  2. Background on Conventional Methods to Retrieve Landsat Imagery
  3. Esri Landsat Explorer: A New Web-based Explorer that Democratizes Scientific Analysis Using Satellite Imagery for Everyone
  4. Creating a Timelapse of the Tweedsmuir Wildfire Using Landsat imagery
  5. Recap
  6. References

Introduction

Photography freezes a moment in time, preserving memories and providing visual documentation of events. While personal photography centers around capturing significant events in our lives, image capture for research includes the systematic recording of objects and phenomena, enabling retrospective analysis and generating valuable insights on changes and developments over time.

Image Capture for Research

For studies spanning extended periods that range from days to decades, time-series analysis is an invaluable tool for tracking and identifying changes. One powerful technique is timelapse, which involves compiling a series of images captured at regular intervals and played back at normal or accelerated speeds. It condenses lengthy processes into easily digestible video clips, illustrating changes and trends over time. Central to the creation of timelapse videos is the use of time-aware imagery that contains temporal metadata, including the specific time and date of capture, ensuring each frame can be accurately placed within the larger sequence. There are many methods for image capture, a source of images that I would like to discuss is the use of remote sensing to obtain information about an object without direct physical contact with it. Specifically, the use of special sensors to capture the reflectance of light as a form of electromagnetic radiation to inform us about the biophysical characteristics of our planet.

Landsat – Spaceborne Satellite Imagery

Enter the world of satellite imagery. Among the many satellite missions carried out in human history, Landsat is the longest running continuous spaceborne earth imaging and observation program in history that offers over 50 years of imagery from 1972 to the present (USGS, 2022). Since Landsat imagery’s availability to the public through the U.S. Geological Survey (USGS), an American agency for Earth, water and biological sciences, it has become a critical data source for observation and analysis of Earth processes and changes over time. In this article, I share my experience using a newly launched Esri Landsat Explorer to create timelapse visualization of the 2018 Tweedsmuir wildfire event occurred in British Columbia, Canada.

Background on Conventional Methods to Retrieve Landsat Imagery

USGS EarthExplorer

There are numerous ways to gain access to Landsat satellite imagery. A traditional method is by downloading the multispectral images from the USGS EarthExplorer website. From there, bundled images are distributed on a per band basis (i.e. blue, green, red, near-infrared, etc.) and need to be combined or stacked for meaningful analysis such as calculation of vegetation and moisture indices. Spectral composition and map symbology also need to be customized to enable identification and visualization of specific land cover features. Using this method, time needs to be spent on large data downloads and processing before insights could be extracted.

Landsat Imagery Made Available on the Cloud in 2020

As of early 2020, USGS began storing the entire Landsat Collection 2 operational archive onto the Amazon Web Services (AWS) cloud platform. The transition to cloud was a major milestone that eliminated or reduced the need to download and maintain a local copy of the imagery. It also enables the use of third-party applications such as those offered by cloud providers including AWS, Google Cloud and Microsoft Azure to run custom scripts on the cloud for advanced analysis or application development.

Esri Landsat Explorer: A New Web-based Explorer that Democratizes Scientific Analysis Using Satellite Imagery for Everyone

In early 2024, Esri released the Esri Landsat Explorer web-based application to make it incredibly easy for anyone with access to the internet and a web browser to explore Earth’s changes over time and perform instant analysis for a given location without the need of advanced remote sensing skills (Waterman, 2024). Esri Landsat Explorer allows users to select Landsat scenes captured by Landsat missions 4 to 9, covering years from 1982 up to the present.

Figure 1: Esri Landsat Explorer user interface

The following are the key features of Esri Landsat Explorer illustrated in Figure 1:

For more tips on how to use the Esri Landsat Explorer, you may view the quickstart guide written by the ArcGIS Living Atlas Imagery team.

Creating a Timelapse of the Tweedsmuir Wildfire Using Landsat Imagery

To demonstrate the power of satellite imagery using the web-based application, I decided to investigate the before and after scenes of one of the most severe wildfires in Canada in recent years – The Tweedsmuir Complex Wildfire that occurred in 2018. In 2017 and 2018, approximately half of the forests in Tweedsmuir Provincial Park were lost to severe wildfires (BC Parks, 2022). To navigate to the location where wildfires occurred, I started by entering “Tweedsmuir Provincial Park” into the location search bar. It quickly zoomed into the my location of interest as shown in Figure 2.

Figure 2: Navigate to the Tweedsmuir Provincial Park in British Columbia, Canada

Animating Landsat Scenes

Next, I turned on the “Animate” mode so that I could add chronological Landsat scenes to create a timelapse. Knowing that the Tweedsmuir fire occurred within August 2018, I selected scenes from before, within August 2018, as well as those in the summer months from the years that follow with the least cloud coverage to create my animation. The result enabled me to see the devastation wildfire did to the forests in the impacted regions. Figure 3 illustrates the contrast across scenes from July 2018, August 2018 and June 2019. Given shortwave infrared’s ability to penetrate through smoke and haze, I used the related renderer that is readily available to reveal the active wildfire as seen in the August 22, 2018 scene. We can clearly see the leading edges of the fire and smoke in the August 2018 image when the fire was active. The severity of wildfire event was reportedly due to numerous forest fires that occurred simultaneously and merged into a larger fire [3]. This was evidently reflected in the imagery. The June 2019 imagery showed a large area of vegetation was burnt away – see the brownish patch on the left-hand-side of the image.

July 29, 2018 – Before wildfire
August 22, 2018 – During wildfire
June 30, 2019 – After wildfire
Figure 3: Before and after scenes of 2018 Tweedsmuir Complex Fire located in the Tweedsmuir Provincial Park in British Columbia, Canada

Figure 4 below shows the timelapse of the Tweedsmuir site before and after the occurrence of the 2018 wildfire. The timelapse comprised of six scenes extracted between July 26, 2018 and August 25, 2025 in summer months. We can see that the regreening took a matter of years for the large area that was engulfed by fire.

Figure 4: Timelapse of Tweedsmuir Complex Fire site before and after the wildfire from July 2018 to August 2025

You can view the timelapse that I created directly on the Esri Landsat Explorer here. The animation plays immediately in a loop.

Analyzing Temporal Profile of Vegetation Index in a Location Impacted by Wildfire

In remote sensing studies, spectral indices calculated by using the reflectance of specific combinations of spectral bands from satellite imagery are often used to analyze changes in land cover over time. To understand the impact of wildfire on healthy vegetation in the impacted regions, a temporal profile created by plotting the vegetation index known as the normalized difference vegetation index (NDVI) over time could reveal the temporal changes of vegetation at a given location. By turning on the “Analyze – Temporal Profile” mode from the Esri Landsat Explorer, you have the option to instantly create a temporal profile by year of the vegetation index. I inspected the temporal vegetation profiles of both a location impacted by the 2018 Tweedsmuir wildfire and one that did not. See Figure 5 and 6 for the results.

Figure 5: NDVI temporal profile of a location impacted by the 2018 Tweedsmuir Complex Fire in British Columbia, Canada

From Figure 5, we can see a sharp decline in the vegetation index at and after the 2018 mark for the location impacted by the wildfire. This signified a drastic impact to healthy vegetation in the area. The vegetation index does not climb back up to pre-2018 level until nearly 4 years later in 2022.

Figure 6: NDVI temporal profile of a location not impacted by the 2018 Tweedsmuir Complex Fire in British Columbia, Canada

In contrast, we can see that the vegetation index was maintained at a healthy level before and after 2018 as shown in Figure 6. This signified the location was not disrupted by wildfire.

Recap

Today, you learned about:

  • Features available on the Esri Landsat Explorer web application that enable users to dynamically explore biophysical changes on our planet through point and click operations
  • Creation of land cover timelapse at a user selected location using Landsat scenes selected by date
  • Analysis of vegetation health at a user selected location using temporal profile of NDVI

Aside from the Esri Landsat Explorer, Esri also offers similar user interfaces for users to explore imagery from the Sentinel-1 and Sentinel-2 missions. I encourage you to experiment with these tools yourself to make discoveries and revisit planet events that interest you.

References

BC Parks. (2022, August 12). Reconstructing historic fire activity and whitebark pine distributions in Tweedsmuir Provincial Park North. BC Parks News. https://engage.gov.bc.ca/bcparksblog/2022/08/12/reconstructing-historic-fire-activity-and-whitebark-pine-distributions-in-tweedsmuir-provincial-park-north/

USGS. (2022, July 23). Fifty Years of Landsat: Observing Earth to Look Forward. USGS Communications and Publishing. https://www.usgs.gov/news/featured-story/fifty-years-landsat-observing-earth-look-forward

Waterman, R. (2024, February 22). Landsat Explorer – Quickstart Guide. ArcGIS Blog. https://www.esri.com/arcgis-blog/products/arcgis-living-atlas/imagery/landsat-explorer-earth-science-and-observation-for-all