5 Tips for Interpreting Satellite Images

 5 Tips for Interpreting Satellite Images

Satellite images are like maps: they are full of useful and interesting information, provided you have a key. They can show us how much a city has changed, how well our crops are growing, where a fire is burning, or when a storm is coming. For interpreting satellite images to unlock the rich information, you need to:

  1. Understand the image scale
  2. Look for patterns, shapes, and textures
  3. Define the colours (including shadows)
  4. Find north
  5. Consider what you know about the area

These tips come from the Earth Observatory’s writers and visualizers, who use them to interpret images daily. They will help you get oriented enough to pull valuable information out of satellite images.

Based on an article courtesy of NASA.

1 – Understand the image scale

One of the first things people want to do when they look at a satellite image is identify the places that are familiar to them: their home, school, or place of business; a favorite park or tourist attraction; or a natural feature like a lake, river, or mountain ridge. Some images from military or commercial satellites are detailed enough to show many of these things. Such satellites zoom in on small areas to collect fine details down to the scale of individual houses or cars. In the process, they usually sacrifice the big picture.

High-resolution satellite view of the Boulder water treatment plant.
Full-resolution Landsat image of the Platte River and Boulder, Colorado.
Image of the same area at a different scale (source: DigitalGlobe and NASA)

Images from the commercial WorldView-2 satellite (top) can show street by street details of the September 2013 flood in Boulder, Colorado, while the scientific Landsat 8 satellite (lower) can be zoomed in to give a city size scale.

The level of detail depends on the satellite’s spatial resolution. Like digital photographs, satellite images are made up of little dots called pixels. The width of each pixel is the satellite’s spatial resolution.

Commercial satellites have a spatial resolution down to 50 centimetres per pixel. The most detailed NASA images show 10 meters in each pixel. Geostationary weather satellites, which observe a whole hemisphere at a time, are much less detailed, seeing one to four kilometres in a pixel.

Full MODIS scene of Boulder, Colorado and the central United States.
the MODIS satellite provides a very wide view of an area (image: MODIS)

Depending on the image resolution, a city may fill an entire satellite image with grids of streets or it may be a mere dot on a landscape. Before you begin to interpret an image, it helps to know what the scale is. Does the image cover 1 kilometre or 100? What level of detail is shown? Images published on the Earth Observatory include a scale.

You can learn different things at each scale. For example, when tracking a flood, a detailed, high-resolution view will show which homes and businesses are surrounded by water. The wider landscape view shows which parts of the county or metropolitan area are flooded and perhaps where the water is coming from. A broader view would show the entire region—the flooded river system or the mountain ranges and valleys that control the flow. A hemispheric view would show the movement of weather systems connected to the floods.

2 – Look for patterns, shapes, and textures

If you have ever spent an afternoon identifying animals and other shapes in the clouds, you’ll know that humans are very good at finding patterns. This skill is useful in interpreting satellite imagery because distinctive patterns can be matched to external maps to identify key features.

Bodies of water—rivers, lakes, and oceans—are often the simplest features to identify because they tend to have unique shapes and they show up on maps.

Other obvious patterns come from the way people use the land. Farms usually have geometric shapes—circles or rectangles—that stand out against the more random patterns seen in nature. When people cut down a forest, the clearing is often square or has a series of herring-bone lines that form along roads. A straight line anywhere in an image is almost certainly human-made, and may be a road, a canal, or some kind of boundary made visible by land use.

Satellite image of Reese Michigan and the surrounding fields.
(NASA Earth Observatory image by Jesse Allen and Robert Simmon, using ALI data from the NASA EO-1 team

Straight lines and geometric shapes in this image of Reese, Michigan, are a result of human land use. Roads cut diagonally across the squares that define farm fields.

Geology shapes the landscape in ways that are often easier to see in a satellite image. Volcanoes and craters are circular, and mountain ranges tend to run in long, sometimes wavy lines. Geologic features create visible textures. Canyons are squiggly lines framed by shadows. Mountains look like wrinkles or bumps.

These features can also affect clouds by influencing the flow of air in the atmosphere. Mountains force air up, where it cools and forms clouds. Islands create turbulence that results in swirling vortices or wakes in the clouds. When you see a line of clouds or vortices, they provide a clue about the topography of the land below.

South American ocean, forest, mountains, and plains.
NASA image courtesy Jeff Schmaltz LANCE/EOSDIS MODIS Rapid Response Team, GSFC

Central Chile and Argentina offer a wide range of geographic features, including snow-covered mountains, canyons, and volcanoes.

Occasionally, shadows can make it hard to tell the difference between mountains and canyons. This optical illusion is called relief inversion. It happens because most of us expect an image to be lit from the top left corner. When the sunlight comes from another angle (especially from the lower edge), the shadows fall in ways we don’t expect and our brains turn valleys into mountains to compensate. The problem is usually resolved by rotating the image so the light appears to come from the top of the image.

3 – The Meaning of Colours

The colours in an image will depend on what kind of light the satellite instrument measured. True-colour images use visible light—red, green and blue wavelengths—so the colours are similar to what a person would see from space. False-colour images incorporate infrared light and may take on unexpected colours. In a true colour image, common features appear as follows:

Sediment plume from the Zambezi River.
Zambezi River Delta – NASA Earth Observatory images by Robert Simmon, using Landsat 8 data from the USGS Earth Explorer

In above image you can clearly see how sediment colours the sea near the mouth of the Zambezi River. The water grows darker offshore as the sediment disperses.

Water

Water absorbs light, so it is usually black or dark blue. Sediment reflects light and colors the water. When suspended sand or mud is dense, the water looks brown. As the sediment disperses, the water’s color changes to green and then blue. Shallow waters with sandy bottoms can lead to a similar effect.

Sunlight reflecting off the surface of the water makes the water look grey, silver, or white. This phenomenon, known as sunglint, can highlight wave features or oil slicks, but it also masks the presence of sediment or phytoplankton.

Sunglint and island wakes around the Canary Islands.
NASA image courtesy Jeff Schmaltz LANCE/EOSDIS MODIS Rapid Response Team, GSFC

Sunglint makes it possible to see current patterns on the ocean’s surface around the Canary Islands.

Frozen water—snow and ice—is white, grey, and sometimes slightly blue. Dirt or glacial debris can give snow and ice a tan colour.

Plants

Plants come in different shades of green, and those differences show up in the true-color view from space. Grasslands tend to be pale green, while forests are very dark green. Land used for agriculture is often much brighter in tone than natural vegetation.

In some locations (high and mid latitudes), plant color depends on the season. Spring vegetation tends to be paler than dense summer vegetation. Fall vegetation can be red, orange, yellow, and tan; leafless and withered winter vegetation is brown. For these reasons, it is helpful to know when the image was collected.

4 satellite images showing seasonal changes in temperate climates.
NASA images courtesy Jeff Schmaltz LANCE/EOSDIS MODIS Rapid Response Team, GSFC

The forests covering the Great Smoky Mountains of the Southeastern United States change colors from brown to green to orange to brown as the seasons progress.

In the oceans, floating plants—phytoplankton—can colour the water in a wide variety of blues and greens. Submerged vegetation like kelp forests can provided a shadowy black or brown hue to coastal water.

Bare ground

Bare or very lightly vegetated ground is usually some shade of brown or tan. The colour depends on the mineral content of the soil. In some deserts such as the Australian Outback and the southwestern United States, exposed earth is red or pink because it contains iron oxides like hematite (Greek for blood-like). When the ground is white or very pale tan, especially in dried lakebeds, it is because of salt-, silicon-, or calcium-based minerals. Volcanic debris is brown, grey, or black. Newly burned land is also dark brown or black, but the burn scar fades to brown before disappearing over time.

Cities

Densely built areas are typically silver or gray from the concentration of concrete and other building materials. Some cities have a more brown or red tone depending on the materials used for rooftops.

Satellite image of Warsaw, Poland.
Satellite image of Warsaw, Poland. Image courtesy NASA/USGS Landsat

The contrast between Warsaw’s modern and historic neighbourhoods is easily visible by satellite. The new Stadion Narodowy is brilliant white. Śródmieście (Inner City) was rebuilt after World War II and most areas appear beige or grey. But some neighbourhoods rebuilt with older-style buildings, such as the red tile and green copper roofs of Stare Miasto (Old Town).

Atmosphere

Clouds are white and grey, and they tend to have texture just as they do when viewed from the ground. They also cast dark shadows on the ground that mirror the shape of the cloud. Some high, thin clouds are detectable only by the shadow they cast.

Smoke is often smoother than clouds and ranges in color from brown to gray. Smoke from oil fires is black. Haze is usually featureless and pale grey or a dingy white. Dense haze is opaque, but you can see through thinner haze. The colour of smoke or haze usually reflects the amount of moisture and chemical pollutants, but it’s not always possible to tell the difference between haze and fog in a visual interpretation of a satellite image. White haze may be natural fog, but it may also be pollution.

himalaya_amo_2013305
The Himalayas (Image adapted from MODIS Worldview)

Clouds, fog, haze and snow are sometimes difficult to distinguish in satellite imagery, as in this MODIS image of the Himalaya Mountains.

Dust ranges in colour, depending on its source. It is most often slightly tan, but like soil, can be white, red, dark brown, and even black due to different mineral content.

Volcanic plumes also vary in appearance, depending on the type of eruption. Plumes of steam and gas are white. Ash plumes are brown. Resuspended volcanic ash is also brown.

Colours in Context

Looking at a satellite image, you see everything between the satellite and the ground (clouds, dust, haze, land) in a single, flat plane. This means that a white patch might be a cloud, but it could also be snow or a salt flat or sunglint. The combination of context, shape, and texture will help you tell the difference.

For example, shadows cast by clouds or mountains can be easy to mistake for other dark surface features like water, forest, or burned land. Looking at other images of the same area taken at another time can help eliminate confusion. Most of the time, context will help you see the source of the shadow—a cloud or mountain—by comparing the shape of the shadow to other features in the image.

4 – Find North

When you get lost, the simplest way to figure out where you are is to find a familiar landmark and orient yourself with respect to it. The same technique applies to satellite images. If you know where north is, you can figure out if that mountain range is running north to south or east to west, or if a city is on the east side of the river or the west. These details can help you match the features to a map. On the Earth Observatory, most images are oriented so that north is up. All images include a north arrow.

5 – Apply Other Knowledge

Perhaps the most powerful tool for interpreting a satellite image is knowledge of the place. If you know that a wildfire burned through a forest last year, it’s easy to figure out that the dark brown patch of forest is probably a burn scar, not a volcanic flow or shadow.

Satellite image of the Rim Fire burn scar, Yosemite National Park.
Wildfires in Yosemite National Park, California – NASA Earth Observatory images by Robert Simmon, using Landsat 8 data from the USGS Earth Explorer

Land burned by Yosemite’s Rim Fire is grey brown in comparison to the unburned brown and green landscape around it. See this linked map that helps differentiate between burned land and non-burned land.

Having local knowledge also allows you to connect satellite mapping to what’s happening in everyday life, from social studies, economics, and history (for example, population growth, transport, food production); to geology (volcanic activity, tectonics); to biology and ecology (plant growth and ecosystems); to politics and culture (land and water use); to chemistry (atmospheric pollution); and to health (pollution, habitat for disease carriers).

For example, land ownership and land use policy is contrasted in the pair of images below. In Poland, small parcels of privately owned land surround the Niepolomice Forest. The government has managed the forest as a unit since the thirteenth century. While the canopy isn’t a solid, unbroken green, the forest is largely intact. The lower image shows a checkerboard combination of private and public land near Washington’s Okanogan-Wenatchee National Forest. The U.S. Forest Service manages the forest under a mixed use policy that preserves some forest, while opening other sections to logging. Lighter green areas indicate that logging has occurred on federal, state, or private land. Parcels of private land are much larger in this part of the western United States than in Poland.

Niepolomice Forest, Poland.
Land use and conservation policies define the forest area in Poland (Image: USGS)
Okanogan-Wenatchee National Forest, United States.
Land use in Washington State, USA (image: USGS)

Land use and conservation policies define the forest area in both Poland (top) and the U.S. state of Washington (lower).

Use a map

If you lack knowledge of the area shown, a reference map or atlas can be extremely valuable for interpreting satellite images. A map gives names to the features you can see in the image, and that gives you the ability to look for additional information. Several online mapping services even provide a satellite view with features labelled. Historic maps, such as those found at the Library of Congress or in the David Rumsey Map Collection, can help you identify changes and may even help you understand why those changes occurred.

Whether you are looking at Earth for science, history, or something else, also consider the Earth Observatory as a key resource. The site hosts a rich, deep archive of more than 12,000 interpreted satellite images covering a wide range of topics and locations. The archive includes images of natural events as well as more diverse featured images.

Data Sources

This article was written by NASA, using examples of satellite data provided by NASA Earth Observation satellites, most notably the Landsat and MODIS programmes. These same tips apply of course to other sources of satellite data, like Europe’s Copernicus programme.

We help you interpret satellite data

At Groundstation.Space we are happy to help anyone new to remote sensing to get the most of it. We organise events, masterclasses and other activities to help you use space data for your organisation. Find more info here.

Remco Timmermans

Related post