Exploring Digital Spectral Band Concepts through Satellite Imagery

 Exploring Digital Spectral Band Concepts through Satellite Imagery

    We will be using an online application through an ESRI tutorial to view and work with Satellite Data. It's important to think of how UAS data is similar yet different in how it can be and is applied. For example, We will look at some temporal aspects of the data. It's important to think of how UAS data can be analyzed in the same way. Also, we will be looking at different bands of satellite imagery. Thinking of UAS data, we can evaluate what we can and cannot do with UAS data coming off a normal "color" image.

Working with Satellite Imagery:

The tutorial first takes you to the Sundarbans mangrove region. You can see a clear distinction between the heavily forested Sundarbans region and the highly urbanized city of Kolkata to the northwest (Figure 7).

Figure 7: Sundarbans Mangrove Region

To distinguish the mangrove vegetation more clearly, we'll use the Color Infrared band combination. It combines the bands Near Infrared, Red, and Green. The Near Infrared band shows a very clear difference between vegetation and non-vegetation features. As you can see below (Figure 8) typical Forest, Urban, and Water features have very different values.

Figure 8: Shows Color Infrared Band Combination

Then in the toolbar, you select the "Color Infrared" button. The mangrove forest now appears bright red, This is because the healthy, dense vegetation has been changed to red. The bodies of water going through the mangrove are high in sediments and appear blue. Residential areas, such as the city of Kolkata now appear dark and grey. Areas with agriculture appear as a lighter shade of red, signifying some vegetation, but much less dense than in the mangrove.

Figure 9: Color Infrared Of Mangrove Forest

To start off, it's important to understand what color infrared is.

• Color infrared is a combination of the bands 5, 4, and 3 or near-infrared, red, and green

This helps the thick and healthy vegetation stand out because it creates a contrast in colors. The near-infrared band distinguishes very clearly between vegetation and non-vegetation features. This is important because the forest is broken up by several rivers and waterways. Many of its small islands are accessible only by boat, which hinders ground observation and intensifies the need for satellite imagery to monitor the forest.

The next section of the tutorial takes you to the Takla Makan Desert (Figure 10). It's primarily made up of sand dunes, its surface appears smooth and uniform from above. While the desert appears dry, we will be manipulating the color band index to find areas of moisture. 

Figure 10: Takla Makan Desert

Unlike Color Infrared, which is a combination of three different spectral bands, the Moisture Index is a more sophisticated use of spectral bands. The Moisture Index shows areas of high moisture in blue tones and low moisture areas in orange tones. Most of the Takla Makan Desert appears low moisture. However, the Moisture Index reveals several places that aren't as dry as they seemed by the natural eye (Figure 11). 

Figure 11: Takla Makan Desert With Moisture Index

Disregard the dark blue areas because that is cloud coverage, but you can see the light blue color surrounding the center of the desert, which is an area of rich moisture. The Moisture Index shows that this area has a fair amount of moisture overall which is a great tool for monitoring drought levels and measuring desertification trends. If the NIR band value is higher than the SWIR 1 band value, the result of the formula will be a higher value. Otherwise, it will be a lower value, denoting a low moisture level. Lush Grass and Water features will typically result in a positive value, and Desert features in a negative value (Figure 12).

Figure 12: Shows Moisture Index

ArcGIS also allows you to see changes over time. The Time Line slider shows all available imagery from as far back in time as Landsat satellites captured imagery of a given area. It also allows you to filter the imagery based on cloud cover or seasonality. A great example of this is to look at Las Vegas, Nevada. The oldest imagery of this area goes as far back as June 19, 1991 (Figure 13). While looking at this image, you can see the beginning developments of Las Vegas. Now if we jump forward to the latest imagery which was collected on August 25, 2021, and you can see how much the area has grown and expanded (Figure 14).

Figure 13: Las Vegas In 1991

Figure 14: Las Vegas In 2021

Major Takeaways:

1. Satellite imagery, UAS imagery, and manned aircraft imagery have a lot in common. They are all used to collect some form of data that can be used to tell us things about the world around us that you can't see with the naked eye. The only difference they have is the scale to which they collect this data. Satellite is used for a much larger scale while UAS imagery is used on a smaller scale, and manned aircraft imagery falls in the middle. 
2. There are a few advantages and disadvantages that satellite imagery has over UAS. One advantage that satellites have over UAS is that they can cover much more land than UAS can. They have a much longer range so that they can have a wider view. A disadvantage of this is that the resolution isn't as good. Since UAS are much closer to the ground, they are able to collect better quality data.

3. UAS falls more into the category of satellite imagery for a couple reasons. The first is that they are unmanned, this almost defines the whole niche. They also perform the same sort of task that manned aircraft data collection, just without the man.
4. The use of spectral bands and index manipulation has many real-world applications. being able to change the imagery to clearly see areas of high or low vegetation, moisture, and many other features can help geologists, or anyone better identify changes in landmass. There are so many possibilities that it would be impossible for us to name them all.