Landsat MoreThanPrettyPicture 2013Apr11

8/10/2019 Landsat MoreThanPrettyPicture 2013Apr11

1/50

More than a Pretty Pic ture:

How Landsat Images Are Made

Malaspina Glacier,

Alaska


2/50

The wholeelectromagnetic (EM) spectrum consists

of the longest wavelengths (radio), shortest ones

(gamma rays), and everything in between.

People have grouped EM waves into these categories

in order to talk about them.


3/50

Visible light, the light we see with our eyes alone, is

a very small partof the whole spectrum of radiant

energy in the universe.


4/50

We measure radiant energy in wavelengths, from

crest to crest.

Wavelength (a)

is longer

than wavelength (b)


5/50

Colors have different wavelengths!

We see colors as different becausethey have

different wavelengths.

Red has the longestwavelengths of visible light, and

blue/purple has the shortestwavelengths of visible light.


6/50

NASA illustration by Robert Simmon. Astronaut photograph ISS013-E-8948

About 25 percent of the Suns energy is absorbed by the

atmosphere; about 50 percent is absorbed by the Earths

surface; and about 30 percent is reflected back to space.


7/50

In this picture,

IR is Infrared light

R is red light

G is green light

B is blue light

Consider a tree and its leaves.

Red, green, blue, and infrared light from the sun hit the

tree and its leaves.

Infrared and green light are reflected from the tree.

Red and blue light are absorbed by the tree.

Image Credit: Canada Centre for Remote Sensing


8/50

We seethe tree as green,

because wavelengths of light we call green

are reflected to our eyes by the tree.

Image Credit: Canada Centre for Remote Sensing


9/50

Wavelengths we see as green are about 525-550

nanometers (nm) in length. Wavelengths we see

as red are 630-800 nm in length.

The red petals of this poppy

flower reflect strongly at

wave-lengths of 700 nm.

Photo: Jeannette Allen


10/50

Every kind of surface reflects light differently, absorbing

and reflecting it weakly or strongly in different wavelengths.

Photos: Jeannette Allen


11/50

Butter reflects weakly in blue and strongly in yellow to red.

Tomato reflects weakly in blue and strongly in red.

redblue yellow

Every kind of surface has its own spectral signature,

somewhat like a fingerprint.


12/50

Notice that water and vegetation reflect somewhat similarlyin the visible wavelengths (about 0.4 to 0.7 nm) but are

almost always separable in the infrared.

This graph shows the spectral signatures

of vegetation and water.


13/50

(This graph

uses

micrometers

rather than

nanometers.)

More spectral signatures.Notice how different kinds of surfaces reflect strongly

or weakly at different wavelengths.


14/50

More spectral signatures


15/50

A farmer using remote sensing can tell which

sugar beet fields are healthy and which are not,

if she/he knows their spectral signatures.

If s/he were designing a sensor solely to measure the health of his sugar

beets, what wavelength range would he want the sensor to detect?


16/50

People measure the spectral signatures of

different surfaces on the ground. Then when they

look at the spectral signature of a surface in a

satellite image, they can tell what kind of surfacethe satellite was looking at.

Researcher with hand-held spectrometer


17/50

Lines (Rows)

Pixel

s

(Colu

mns)

Landsat scenes are made up of these

30-meter squares, or pixels.


18/50

Landsat 7 observes the Earth in 7 ranges (or

bands) of the electromagnetic spectrum.


19/50

Lets look at how it works for just 1 band of the 7. In this

illustration, each square represents a 30m x 30m piece of

land surface.


20/50

The Landsat instrument records the amount of reflected light

in each band for each 30 m pixel, on a scale of 0 to 255. A

numerical value of 0 represents no reflected light and a

numerical value of 255 represents maximum reflected light.

0: See the corres-

ponding black pixel in

gray-scale array.

255: See the

corresponding white

pixel in gray-scale

array.


21/50

7 bands of data looked at side by side

in shades of gray


22/50

Now how do we make color

images of all that grayscaledata

so we can work with it more easily?


23/50

We have to assignColorsto represent

Landsat bands(using computer software).


24/50

People can chose red, green, or blue to

represent any of the wavelength ranges they like.

One can make lots of color combinations. (This is Beverly, MA.)


25/50

5 6 74321

3,2,1

Visible Infrared

Green Data is shown as GreenBlue Data is shown as BlueRed Data is shown as Red


26/50

5 6 74321

4,3,2

Visible Infrared

NIR Data is shown as RedRed Data is shown as GreenGreen Data is shown as Blue


27/50

Visible wavelengths

Infrared wavelengths

Making images with different band combinations,

we see more than we could otherwise.

Same scene, different wavelengths


28/50

To view a short movie about how wavelength

ranges (bands) are combined to make an

image, go to:http://landsat.gsfc.nasa.gov/education/resources/L7_sanfran_onionskin.mov
http://landsat.gsfc.nasa.gov/education/resources/L7_sanfran_onionskin.movhttp://landsat.gsfc.nasa.gov/education/resources/L7_sanfran_onionskin.movhttp://landsat.gsfc.nasa.gov/education/resources/L7_sanfran_onionskin.mov


29/50


30/50

Some common band

combinations


31/50

True-Color Composite (3,2,1)True-color composite images approximate the range of vision for the

human eye, and hence these images appear to be close to what we

would expect to see in a normal photograph. True-color images tend to

be low in contrast and somewhat hazy in appearance. This is because

blue light is more susceptible than other bandwidths to scattering by theatmosphere. Broad-based analysis of underwater features and

landcover are representative applications for true-color composites.

Near Infrared Composite (4,3,2)

Adding a near infrared (NIR) band and dropping the visible blue bandcreates a near infrared composite image. Vegetation in the NIR band is

highly reflective due to chlorophyll, and an NIR composite vividly

shows vegetation in various shades of red. Water appears dark, almost

black, due to the absorption of energy in the visible red and NIR bands.

Shortwave Infrared Composite (7,4,3 or 7,4,2)A shortwave infrared composite image is one that contains at least one

shortwave infrared (SWIR) band. Reflectance in the SWIR region is due

primarily to moisture content. SWIR bands are especially suited for

camouflage detection, change detection, disturbed soils, soil type, and

vegetation stress.

Mount Rainier


32/50


33/50


34/50

Another infrared composite


35/50

Another infrared composite


36/50

A farmer using remote sensing can tell which

sugar beet fields are healthy and which are not,

if she/he knows their spectral signatures.

If s/he were designing a sensor solely to measure the health of his sugar

beets, what wavelength range would he want the sensor to detect?


37/50

Color Composites in Landsat TM Data

True-Color Composite (3,2,1)

approximates range of vision for human eye

images appear as we would expect to see in

photograph. low in contrast, somewhat hazy - blue light more

susceptible to scattering by atmosphere.

Broad-based analysis of underwater features and

landcover are applications for true-colorcomposites.


38/50

Band 3 (Visible red) = red

Band 2 (Visible green) = green

Band 1 (Visible blue-green) = blue


39/50


Near Infrared Composite (4,3,2)

Add near infrared (NIR) band and drop visibleblue band.

Vegetation in the NIR band is highly reflective NIR composite vividly shows vegetation invarious shades of red.

Water appears dark, almost black, due to theabsorption of energy in the visible red and NIRbands.


40/50


41/50


Shortwave Infrared Composite (7,4,3 or7,4,2)

contains at least one shortwave infrared(SWIR) band.

Reflectance in the SWIR region is dueprimarily to moisture content.

SWIR bands are especially suited for changedetection, disturbed soils, soil type, andvegetation stress.

B d 7 (SWIR) d


42/50

Band 7 (SWIR) = red

Band 4 (NIR) = green

Band 3 (red) = blue

Or...

Band 7 (SWIR) = red

Band 4 (NIR) = green

Band 2 (green) = blue

vegetation- shades

of green.

Bare soils

and


43/50


44/50

Station Fire - August 30, 2009 - 140,000 acres burned in Angeles National Forest - Mt.

Wilson Observatory threatened. Blue and white haze is smoke, pinkish areas have been

burned, active fires appear bright red.

Landsat-7ETM+

Stripin

g


45/50

Landsat 5 image - Esperanza Fire in San Bernardino National Forest, Oct. 26, 2006. The

fire caused a huge smoke plume; reaching ~3.0 miles high. Large image is a natural-

color composite, small inset is false-color composite. The active fire fronts in the false-

color image appear bright yellow.

T f l l L d t 5 i (1984 d 2009) ]d k l id f it t t d


46/50

Two false-color Landsat 5 images (1984 and 2009). ]dark purple grid of city streets and

green of irrigated vegetation.

Images were created using reflected light from the shortwave infrared, near-infrared and

green portions of the electromagnetic spectrum (Landsat 5 TM bands 7,4,2)


47/50

Venice - Landsat 7 - 2001. pan-sharpened false-color composite of ETM+

bands 7, 5, and 3. Venice can be found on Landsat WRS-2Path 192 Row

28.

Cli t Ch
http://landsat.gsfc.nasa.gov/about/wrs.htmlhttp://landsat.gsfc.nasa.gov/about/wrs.htmlhttp://landsat.gsfc.nasa.gov/about/wrs.htmlhttp://landsat.gsfc.nasa.gov/about/wrs.html


48/50

Climate Change - Lake Chad borders Chad, Niger, Nigeria, andCameroon, as well as on a natural border, the Sahel: a grassland which divides

the Sahara Desert to the north and the more humid savannah to the south.

Because Lake Chad is a shallow lake, depths of 16 - 26 feet, its surface area

fluctuates markedly with changes in climate.


49/50

The newest Landsat satellite, Landsat 8, measuresthe amounts of reflected light on a finer scale than

Landsat 7s scale of 0-255.

Landsat 8 measures light on a scale of 0-1024.

With Landsat 8 we can see many more shades of

the light we are measuring, and we can study

greater nuance in our scenes of interest.

A little more to know:


50/50

Landsat 8 also uses more bands of light

than Landsat 7, and the band numbers

are a little different.

To learn the differences between Landsat 7

and Landsat 8 spectral bands, go to this URL:

http://landsat.usgs.gov/L8_band_combos.php

Documents

Landsat MoreThanPrettyPicture 2013Apr11