Skip to content
Lab Overview
This lab introduced the use of ArcGIS Pro to analyze multiband UAS imagery and explore how different spectral bands can be used to identify features on the ground. Using preburn and postburn datasets from the Doaks Burn site, the lab focused on understanding metadata, individual spectral bands, multiband composites, and vegetation indices such as NDVI.
The main goal of the lab was to build familiarity with ArcGIS Pro tools while learning how spectral information from UAS sensors can support analysis of vegetation health, burn severity, and land surface conditions.
Data Preparation and Metadata Review
Before working in ArcGIS Pro, the provided dataset was copied into a structured local folder to maintain organization. Metadata from the flight was reviewed to understand how the data was collected. This included information about the aircraft, sensor type, altitude, flight time, and sensor orientation.
Reviewing metadata is important because it provides context for the imagery and helps analysts understand resolution, accuracy, and limitations of the data before beginning any analysis.
Viewing Individual Spectral Bands
Once the ArcGIS Pro project was opened, individual spectral bands from the Altum sensor were viewed one at a time. Each band showed different surface characteristics based on its peak reflectance wavelength.
The blue, green, red, red edge, and near infrared bands all highlighted different features across the burn site. Vegetation, trails, and field boundaries became more or less visible depending on the band being displayed. Viewing each band separately helped demonstrate how no single band tells the full story, and why multiband analysis is valuable.
Thermal Imagery Comparison
Thermal imagery was also examined using long wave infrared data. Comparing the standard LWIR layer with the postburn LWIR layer clearly showed heat retained in recently burned fields. Areas that had been burned appeared much brighter immediately after the burn, while surrounding vegetation appeared darker.
This comparison demonstrated how thermal data differs from near infrared imagery and how it can be used to analyze surface temperature rather than reflectance.
Exploring Raster Properties
Raster properties for multiple layers were reviewed to better understand the data. This included examining pixel size, radiometric resolution, and coordinate systems. Differences in cell size between multispectral and thermal layers highlighted why resolution varies depending on sensor type.
Understanding raster properties is critical when performing analysis, as these values directly affect accuracy, scale, and how datasets can be compared.
Multiband Composites and False Color Imagery
The lab then focused on multiband composite imagery. Preburn and postburn composites were added to the map and viewed in both true color and false color infrared. Assigning the near infrared band to the red channel made vegetation stand out clearly, allowing healthy and unhealthy areas to be easily distinguished.
The swipe tool was used to compare preburn and postburn imagery, making changes caused by the burn clearly visible.
Figure 1: Preburn Scan False IR
Figure 2: Postburn Scan False IR
NDVI Analysis
Normalized Difference Vegetation Index layers were examined for both preburn and postburn conditions. NDVI made it easy to identify healthy vegetation, burned areas, and bare ground using color differences. Comparing NDVI to composite imagery reinforced how indices can simplify complex multiband data into an easy to interpret product.
Figure 3: My Personal False IR Band
Key Takeaway
This lab demonstrated how ArcGIS Pro can be used to analyze multiband UAS imagery for environmental and vegetation analysis. Viewing individual bands, composites, thermal data, and NDVI showed how each tool provides different insights into the same landscape. Overall, the lab reinforced the importance of understanding spectral data and metadata when using geospatial software for UAS analysis.
Want to print your doc?
This is not the way.
This is not the way.
Try clicking the ··· in the right corner or using a keyboard shortcut (
CtrlP
) instead.