Landsat 5 Bands: Complete Guide to Wavelengths, Uses & Applications
Launched in 1984 and still returning usable data long after its design life expired, Landsat 5 became one of the most productive Earth-observing satellites ever flown. Its longevity stretched a three-year mission into a nearly 29-year record of the planet's surface, and that record survives today because of a simple design decision: the Landsat 5 bands were built to capture distinct slices of the electromagnetic spectrum, each one revealing something different about the land below.
Whether you are mining archival imagery for a land-cover change study, assembling a three-decade vegetation index time series, or simply learning the fundamentals of remote sensing, knowing what each band measures and how it behaves is the foundation everything else rests on. This guide walks through the Thematic Mapper's seven bands, their resolutions, their quirks, and the band combinations that made this sensor a workhorse for an entire generation of scientists.
What the Landsat 5 Thematic Mapper Bands Actually Measure
Landsat 5 carried two imaging instruments: the Multispectral Scanner (MSS) and the Thematic Mapper (TM). When researchers say "Landsat 5 bands," they almost always mean the TM, a seven-band sensor that represented a substantial leap over the four-band MSS that flew on the earlier Landsat 1 through 3 missions. The TM added narrower, better-placed spectral windows, improved spatial detail, and, for the first time on a Landsat platform, a thermal channel dedicated to measuring emitted heat rather than reflected sunlight.
Each band occupies a defined wavelength interval, and the choice of those intervals was not arbitrary. They were selected to line up with the physical behavior of vegetation, water, soil, and rock. Chlorophyll absorbs red light and reflects near-infrared energy, so a sensor needs both a red band and a near-infrared band to detect plant vigor. Water absorbs almost everything beyond the visible range, which makes infrared bands excellent for mapping shorelines and moisture. Clay minerals and iron oxides have diagnostic absorption features in the shortwave infrared, which is why two SWIR bands were included rather than one.
The Thematic Mapper Instrument and Its Design
The TM was a whisk-broom scanner orbiting at roughly 705 kilometers in a sun-synchronous polar orbit, crossing the equator mid-morning and repeating its ground track every 16 days. It collected a 185-kilometer swath with a scan mirror that swept side to side, building an image line by line. That mechanical scanning design is a key detail: it explains why TM data is stored as 8-bit digital numbers rather than the 12-bit or 16-bit ranges used by modern pushbroom sensors like the Operational Land Imager on Landsat 8.
Its radiometric and geometric calibration were maintained across the mission with onboard lamps, solar diffusers, and a network of ground reference targets, which is precisely why Landsat 5 imagery remains scientifically trustworthy decades later. The satellite itself outlived its planned three-year life by an enormous margin. Routine TM acquisitions continued until November 2011, and the spacecraft was formally decommissioned in June 2013 after gyroscope failures and a degraded transmitter finally ended operations.
Spatial and Radiometric Resolution Explained
Six of the seven TM bands were delivered at a 30-meter pixel size, which was a meaningful improvement over the 60- to 80-meter MSS pixels. The thermal band, TM Band 6, was collected at 120 meters because a thermal detector needs a larger instantaneous field of view to gather enough photons for a reliable measurement. In most standard Level-1 and Level-2 products distributed by the USGS, Band 6 is resampled to 30 meters so that all seven layers stack cleanly, but the true native resolution remains 120 meters and the data is noticeably softer than the reflective bands.
On the radiometric side, TM data is quantized to 8 bits, giving 256 possible digital number values per band. That is limited compared to Landsat 8's 16-bit OLI, and it means bright surfaces such as snow, playa, and bare gypsum can saturate, especially in the shortwave infrared. When you convert digital numbers to top-of-atmosphere reflectance or surface reflectance using the published calibration coefficients, you are stretching a fairly narrow measurement range, so careful attention to saturation and haze is worthwhile in any rigorous analysis.
A Band-by-Band Breakdown of the Landsat 5 Bands
The seven reflective and thermal channels of the TM cover a continuous range from visible blue to thermal infrared. Together they support everything from true-color visualization to quantitative indices for vegetation health, burn severity, and surface temperature. The table below summarizes the core specs before we dig into each group.
It is worth noting up front that Landsat 5 had no panchromatic band. Pan-sharpening, the technique of fusing a high-resolution grayscale image with lower-resolution color data, became standard practice with Landsat 7's ETM+ and its 15-meter pan band. With Landsat 5 you work at 30 meters and accept it, or you fuse the data with imagery from another sensor, such as SPOT or a modern commercial satellite.
| Band | Name | Wavelength (micrometers) | Resolution |
|---|---|---|---|
| Band 1 | Visible Blue | 0.45 - 0.52 | 30 m |
| Band 2 | Visible Green | 0.52 - 0.60 | 30 m |
| Band 3 | Visible Red | 0.63 - 0.69 | 30 m |
| Band 4 | Near Infrared | 0.76 - 0.90 | 30 m |
| Band 5 | Shortwave Infrared 1 | 1.55 - 1.75 | 30 m |
| Band 6 | Thermal Infrared | 10.40 - 12.50 | 120 m |
| Band 7 | Shortwave Infrared 2 | 2.08 - 2.35 | 30 m |
Visible Bands: TM Bands 1, 2, and 3
Band 1 spans the blue portion of the spectrum and is the most strongly scattered by the atmosphere. That sensitivity makes it useful for coastal water mapping, turbidity and bathymetry work in shallow seas, and distinguishing bare soil from vegetation, but it is also the noisiest of the three visible bands. In vegetated or hazy scenes, many analysts either exclude it or apply a dark object subtraction or other atmospheric correction before using it. Band 1 also saturates easily over bright, highly reflective targets.
Band 2, the green band, captures the reflectance peak of healthy vegetation and is the primary input to several water indices, including the Normalized Difference Water Index that pairs green with near infrared. Band 3, the red band, is where chlorophyll absorption is strongest, making it the second half of the classic NDVI equation when paired with Band 4. Stacking bands 3, 2, and 1 in red, green, and blue channels produces a natural-color composite that looks roughly like what a human eye would see from orbit.
Infrared Bands: TM Bands 4, 5, 6, and 7
Band 4, the near-infrared channel from 0.76 to 0.90 micrometers, is arguably the single most important band on the sensor for vegetation science. Healthy plant canopies reflect strongly here because leaf cell structure scatters near-infrared energy, while water absorbs it almost completely. The boundary between high NIR reflectance over land and near-zero reflectance over water is so pronounced that Band 4 alone can produce a clean land-water mask with simple thresholding.
Band 5, the first shortwave infrared channel, is exceptional for measuring vegetation and soil moisture content, since water absorbs strongly at these wavelengths. Band 7, the second shortwave infrared band, is sensitive to clay and carbonate minerals and is the workhorse for geological mapping. Combined with Band 4, Band 7 powers the Normalized Burn Ratio used to delineate fire scars. Band 6 is the thermal channel, recording emitted radiation at 120-meter resolution and enabling land surface temperature retrieval, urban heat island analysis, and geothermal and volcanic monitoring when the calibration coefficients are applied correctly.
Practical Applications and Band Combinations
Raw bands on their own are informative, but the real analytical power of Landsat 5 comes from combining them. Ratios cancel out much of the illumination and atmospheric variation between scenes, which is what makes multi-decade time series possible despite changes in sun angle, sensor drift, and atmospheric conditions on any given acquisition date. Choosing the right combination depends entirely on what you are trying to see.
Applications that relied heavily on this sensor include global forest cover mapping, the long-running monitoring of Amazonian deforestation, rangeland productivity assessments, drought tracking, coastal change analysis, and agricultural crop inventories. The 29-year archive means a single study area can often be compared across four decades of consistent measurements, something few other civilian sensors can match.
Essential Band Combinations for Common Tasks
For a natural-color view, assign Band 3 to red, Band 2 to green, and Band 1 to blue. To highlight vegetation, the classic false-color infrared composite uses Band 4 in red, Band 3 in green, and Band 2 in blue, rendering healthy plants in vivid red. A widely used agriculture composite places Band 5 in red, Band 4 in green, and Band 3 in blue, revealing moisture and crop condition differences that are invisible to the eye. Urban and built-up analysis often benefits from swapping the first channel for Band 7, since concrete and asphalt reflect strongly in shortwave infrared while vegetation does not.
Beyond three-band composites, index math extends the sensor's reach considerably. NDVI uses Bands 3 and 4 to quantify greenness. The Normalized Burn Ratio uses Bands 4 and 7 to map fire severity and post-fire recovery. The Modified Normalized Difference Water Index uses Bands 2 and 5 to isolate surface water even in built-up areas. The Tasseled Cap transformation, developed specifically for Landsat data, converts the six reflective bands into brightness, greenness, and wetness components that track land-cover change with a clarity that raw bands often obscure.
Accessing and Processing Landsat 5 Data Today
Landsat 5 TM data is freely available from the USGS EarthExplorer portal, the Earth Resources Observation and Science Center archive, and cloud platforms such as Google Earth Engine and the AWS Open Data program. Most users should start with Collection 2 Level-2 surface reflectance products, which come atmospherically corrected and ready for index calculation, saving hours of preprocessing. Level-1 top-of-atmosphere products remain useful when you need full control over the correction workflow.
A few practical cautions apply when working with this archive. Because the data is 8-bit, saturation in bright targets is more common than with Landsat 8 or 9. Scenes vary in quality, so filtering by cloud cover metadata and visually inspecting candidate dates remains good practice. Older scenes may carry different calibration versions, and mixing pre- and post-2011 imagery without accounting for those differences can introduce spurious trends. Once those details are handled, the payoff is substantial: a clean, radiometrically consistent record stretching back to the mid-1980s.
The Landsat 5 archive is not simply a historical curiosity. It is the backbone of nearly every long-term land change study covering the late twentieth century, and it remains a free, well-documented training ground for anyone learning to work with multispectral data. Pull a scene over your own region, stack the bands, and see what four decades of pixels can tell you.