The multispectral Landsat layer that you added to the map in the previous lesson contains all the images captured by the Landsat satellites since the program began. As such, the layer can be time-enabled and filtered to select out and display only the images you need for your animation. In this lesson, you'll enable the time capabilities of the Multispectral Landsat imagery layer. Then, you'll configure a time animation to show historical images at an appropriate speed. You'll filter the dataset to use only specific, relevant, and high-quality images in your animation. It will also be important to select images from approximately the same time of year (within a three-month period) to accommodate seasonal changes in moisture and vegetation health. Finally, you'll emphasize vegetation in the imagery by changing its spectral bands to color infrared. Doing so will make the distinction between types of land use more apparent to your audience. Then you'll identify a region of Samut Songkhram that has experienced significant land-use change for your web app.
Enable time animation
First, you'll enable the time function for the layer. Turning on this functionality allows you to search through 40 years of satellite imagery for your study area.
- In the Contents pane, point to the Multispectral Landsat layer. Click the More Options button and choose Enable Time Animation.
The time animation toolbar appears at the bottom of the map window. The map also changes to show only imagery within the first date range on the timeline: July 24, 1972, to July 24, 1977. Your map viewer turns a color, such as white or blue, because of an error in capturing the first image of this area.
The Landsat program began in 1972, and early Landsat imagery can often be of poor quality or, in this case, show only blank tiles. Later, you'll filter this and other problematic images out of the dataset so only high-quality images are shown. First, you'll look at all of the images to see which ones might be useful for your presentation.
- On the time animation toolbar, click the Play/Pause button.
The timeline plays, cycling through all available images in chronological order. The speed at which the animation progresses is not ideal for observing changes, so you'll change the default settings of the time animation tool to lower the playback speed.
Some images may not load before the timeline continues to the next image. You may need to play the animation more than once to see all of the images.
- When you are finished watching the time animation, click the Play/Pause button again to pause the animation.
Configure the time animation tool
In this section, you will change the default settings in the time animation toolbar to play the animation at an appropriate speed, and to display images by decades to start building your presentation.
- On the time animation toolbar, click the Configure button.
The Time Settings window appears. This window allows you to change the playback speed of the timeline so that more or less time will pass before the timeline jumps to the next image.
- Adjust the playback speed to the slowest setting and click OK.
- Play the time slider again.
While the time animation plays more slowly, sometimes the imagery changes even when the time slider does not. The time slider is currently divided into four-year intervals, and it's possible for more than one image to have been taken during these intervals. You'll lower the display interval to show only one image per tick of the timeline. Doing so will cause the time animation to take a long time to span the entire chronology of Landsat imagery, so you'll also narrow the time encompassed by the timeline to show only imagery between 1972 and 1980. Later, you'll look at imagery from the other decades. This way, you can look at the imagery in more manageable 10-year time periods and also choose the best image from each decade to include in your presentation.
- Click the Configure button. In the Time Settings window, click Show advanced options.
The advanced settings give you the option of customizing the time span and display intervals of the animation.
- For Time Span, change the start time to 7/24/1972 12:00 AM and the end time to 1/1/1980 12:00 AM.
- For Time Display, change the display interval to 1 year.
- Click OK. Play the time animation again.
You can also use the Previous and Next buttons to the right of the timeline to switch between the imagery.
Most of the years between 1972 and 1980 have no imagery or imagery of too low quality to use. In 1978, however, the Landsat program captured a usable image of your study area:
You'll use this image in your presentation to show land use during the 1970s. Later, you'll filter the Landsat dataset to include only this layer, as well as a few others. To do that, you'll need to know the layer's acquisition date, or the date the image was captured by the satellite.
- Click the image in the map window to open a pop-up for the image.
The acquisition date for this image was December 15, 1978. You'll use this date in the next lesson when you filter the dataset. First, you'll repeat the process for the 1980s, 1990s, and 2000s to find an image for each decade to use in your presentation. You will also do a search from 2010 until present day to find the most recent high-quality image.
- On a piece of paper (or in a Word document), write the following information:
Remember to select images from the same time of year (within a three-month period). Seasonal differences in rainfall can affect the appearance of vegetation health.
- Click the Configure button again to open the Time Settings window, and click Show advanced options.
- Change the end time to 1/1/1990 12:00 AM and the start time to 1/1/1980 12:00 AM.
- Optionally, you can adjust the playback speed to your preference.
- Click OK. Play the time animation again. Remember to play the animation more than once for the images to appear properly.
After playing the animation once or twice, using the Previous and Next buttons can make it easier to assess the image quality and cloud cover.
The Landsat program did not capture an image for your study area during the 1980s. In your presentation, this decade will appear blank. Fortunately, there are good-quality images in the next three decades.
- Repeat steps 10 through 13 to view the imagery for the 1990s and 2000s and the most recent high-quality images. When you find an image for each time period, remember to click within the study area and record the date in your table. Use the following time settings for each decade:
- Start time: 1/1/1990 12:00 AM
- End time: 1/1/2000 12:00 AM
- Start time: 1/1/2000 12:00 AM
- End time: 1/1/2010 12:00 AM
- Most recent
- Start time: 1/1/2010 12:00 AM
- End time: <Today's date>
After reviewing animations for each time period, you should have images for the following dates:
- Save your map.
Filter the highest-quality images
Next, you'll filter the multispectral layer to display only the best image from each decade.
- In the Contents pane, point to the Multispectral Landsat layer and click the Filter button.
The Filter: Multispectral Landsat window opens. This window allows you to set a filter on the data to only display parts of that data. You can build filters based on attribute values. You'll build a filter based on the Acquisition Date attribute.
- At the top of the window, click the Edit tab.
By default, the Filter function provides two expression boxes to create the filter. You'll use four expression boxes, one for each of the images you chose.
- Click the Add another expression button twice.
You now have four expressions you can use.
- For each of the filter expressions, change the first drop-down menu to Acquisition Date. Confirm that the second drop-down menu is set to is on.
- Change the third drop-down menu in each filter expression to one of the four dates of the images you chose for your presentation: 12/15/1978, 10/24/1994, 12/14/2004, and <Today's date>.
- Click Apply Filter.
The filter is applied, but the images are not shown because the time animation toolbar is still configured to only show images from a small range of dates.
- On the time animation toolbar, click the Configure button and open the advanced options.
- For Time Span, change the start time and end time to show the entire range of available dates.
You can quickly set the start time and end time by dragging the slider handles to the respective ends of the timeline.
- For Time Display, change the display interval to 10 years.
- Click OK.
- Click the Play/Pause button to view the images.
Only the images for the four dates you chose when you created the filter appear. No image appears between 1982 and 1992, because you added no image from this time period to the filter.
Enhance the imagery with infrared
Even with only four images, it's difficult to tell how much land use has changed over time. When viewing imagery in natural color, as you are now, differences between vegetation, land, and water are not always clear. It can be especially difficult to distinguish land-use differences in earlier images, when satellites had less advanced sensors. To better see how land use has changed over time, you'll change the spectral band combinations of the imagery.
Each Landsat satellite captures red, green, and blue wavelengths of visible light to display imagery the way a human eye sees it. However, they also have sensors that capture information not normally visible. The data from these sensors can be combined into different combinations of spectral bands to emphasize different features. For example, infrared sensors show healthy vegetation as bright red and deep water as black. You'll use infrared to better show land-use change.
- Point to the Multispectral Landsat layer, click the More Options button, and choose Image Display.
The Image Display pane appears, allowing you to change the band combination of the imagery.
- For Renderer, choose the Color Infrared with DRA band combination.
- For Image Enhancement, change the Stretch Type to Standard Deviation.
Using the stretch functions will apply different contrast settings to help accentuate clusters of values. The standard deviation setting helps to brighten the red values so that vegetation and water are more distinct.
- Click Apply and click Close.
The image, for July 24, 1972, now shows healthy vegetation as brighter red, and water as darker blue to black.
- Play the time animation again.
As expected, the greatest amount of vegetation loss occurs between 1978 and 1994. From 1994 to 2004, it appears that there was a dramatic decrease in land-use change, which suggests that the 1991 regulations on the aquaculture industry may have been effective in thwarting the spread of shrimp farms. After 2004, some areas even experienced a rebound, transitioning from water back to vegetation. These areas are important to your organization and donors in supporting the idea that education and conservation can result in restoration.
- Zoom to the coastal areas of Samut Songkhram. Play the animation along different sections of the coast. Take note of where you see the most reduction and rebound in healthy vegetation.
The area of Samut Songkhram that has seen the most significant amount of change is in the southernmost portion of the province.
- Zoom in to this area.
In 1978, this area appears as a solid distribution of bright red, or healthy vegetation. Playing the animation through to present day, you'll observe the mass conversion from healthy vegetation to water. Much of the change occurs between the 1978 and 1994 images, which is the result of the uncontrolled growth in the shrimp farming industry. After 1994, the decimation of healthy coastal vegetation appears to have slowed, with some areas even rebounding back from water to vegetation. This is the area that you will use for your web app because it shows potential for successful restoration efforts.
- Save your map.
In this lesson, you enabled time animation for your data. You looked at images of Samut Songkhram and chose the best image per decade. You then filtered the Landsat data to show only those images. Lastly, you changed the band combination of the imagery to emphasize vegetation and make the differences in land use clearer. Your web map now contains only relevant data for your presentation. In the next lesson, you'll create a web app from your map in order to present your findings to potential donors to your nonprofit organization.