Spectral bands

Optical sensors capture light across the electromagnetic spectrum, with different wavelengths suited for different insights. A spectral band is a specific range of wavelengths on the spectrum. To select the right data for a given use case, it’s important to understand the different spectral bands.

Optical imagery in the visible region typically spans the wavelengths from 400 to 800 nm.

Optical imagery in the infrared region typically spans the wavelengths of 800 to 14,500 nm.

Multispectral bands capture data in distinct narrow wavelength ranges, whereas the panchromatic band captures a broad range of wavelengths.

In panchromatic products, the light is captured from across the visual region of the electromagnetic spectrum. It spans wavelengths from around 400 to 800 nm. The resulting panchromatic product captures a wide range of colors in the form of grayscale imagery. This spectral range typically provides higher spatial resolution when compared to other multispectral bands and is suited for applications that require spatial accuracy.

Panchromatic products are also used for pansharpening. It’s a process that combines higher-resolution panchromatic imagery with lower-resolution multispectral imagery to create a single multispectral image with an enhanced spatial resolution. You can use the UP42 platform to perform pansharpening on optical imagery.

Panchromatic products are used for the following use cases:

• Urban and infrastructure monitoring
• Cartography and mapping
• Pansharpening

The red, green, and blue spectral bands broadly cover the visual region of the electromagnetic spectrum. The blue wavelengths span from around 400 to 450 nm, green wavelengths span from around 500 to 600 nm, and red wavelengths span from around 600 to 700 nm.

RGB bands are used for the following use cases:

• Landcover classification and change detection
• Urban and infrastructure monitoring
• Vegetation and crop monitoring

RGB bands are used to create true-color and false-color imagery:

• In true-color imagery, the colors in the image closely resemble the natural colors of the objects as seen by the human eye.
• In false-color imagery, the colors assigned to the different bands aren’t representative of the natural colors of the objects. Because other multispectral bands fall outside the visible spectrum, the choice of bands highlights specific features that aren’t distinguishable in true-color imagery.

False-color displays can be different, depending on which combination of bands is chosen. The most common visualization combines red, green, and NIR bands. It displays the red band in green, the green band in blue, and the NIR band in red.

True-color imagery

False-color imagery

The yellow spectral band is located between the green and red regions of the electromagnetic spectrum. It spans wavelengths from around 580 to 620 nm. This spectral band is sensitive to organic matter, minerals, and chlorophyll content in vegetation. It can also detect suspended particles and algal blooms in water bodies.

The yellow band is used for the following use cases:

• Vegetation analysis
• Soil and mineral identification
• Waterbody monitoring

The deep blue spectral band is on the edge of the visible region and captures blues and violets. Coastal aerosol bands also come under this spectral band. It spans wavelengths from around 400 to 450 nm. This spectral range can penetrate water, making it useful to determine the depth, quality, and clearness of water.

The deep blue band is used for the following use cases:

• Waterbody monitoring
• Cloud detection
• Bathymetry

The red edge spectral band is located between the red and NIR regions of the electromagnetic spectrum. Coastal aerosol bands also come under this spectral band. It spans wavelengths from around 700 to 800 nm. This spectral range displays higher reflectance when compared to other multispectral bands. It’s sensitive to changes in chlorophyll content and vegetation health.

The red edge band is used for the following use cases:

• Vegetative health monitoring
• Biodiversity monitoring
• Ecological analysis

The near-infrared (NIR) spectral band captures radiation just beyond visible red light on the electromagnetic spectrum. Some water vapor bands also come under this spectral band. It spans wavelengths from around 800 to 1,000 nm. This spectral region is sensitive to water and vegetative surface reflectance. Vegetation strongly reflects light in this spectral region, which changes in reflectance indicating health, biomass, and stress.

The NIR band is used for the following use cases:

• Water and water vapor detection
• Vegetative health monitoring
• Spectral indexes (NDVI, EVI, and NBR)

This spectral band is typically combined or bundled with RGB and other bands.

The short-wave infrared (SWIR) spectral band spans wavelengths from 1,000 to 2,500 nm. This spectral range provides information about material composition and surface characteristics. It can penetrate atmospheric effects like haze and moisture, and is useful for studying the Earth’s surface.

The SWIR band is used for the following use cases:

• Infrastructure monitoring
• Soil composition analysis
• Mineral identification
• NBR index

The thermal infrared (TIR) or long-wave infrared spectral band spans wavelengths from around 8,000 to 14,500 nm. This spectral range captures the thermal radiation emitted by objects and surfaces.

The TIR band is used for the following use cases:

• Industrial and infrastructure monitoring
• Land surface temperature monitoring
• Geothermal exploration
• Wildlife monitoring

^*Only in EROS-C3 PAN & MS.

Why is there an extra band in my imagery?

Some raster products are delivered with an additional alpha band that acts as a transparency mask.

• An introduction to optical satellite imagery
• How pansharpening improves satellite imagery
• Pansharpening: An initial view into our advanced analytic capabilities
• Taking Earth’s temperature: Land surface temperature monitoring (LSTM) explained
• Track forest fires with satellite imagery: The normalized burn ratio and other techniques