The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Compare an Earth-observation mission on three separate axes: the sensor’s band-specific spatial resolution, the width of ground imaged in one pass (swath), and the interval between opportunities to observe the same location. Then qualify every figure by satellite versus constellation, nominal capability versus usable imagery, and optical, thermal, or panchromatic mode.
Sentinel-2 and Landsat 8/9 illustrate why there is no universal winner. Sentinel-2 MSI combines 10 m, 20 m, and 60 m bands, a nominal 290 km swath, and a five-day revisit for its two-satellite constellation. Landsat 8 provides 30 m multispectral, 15 m panchromatic, and 100 m thermal data across a 185 km swath, repeating every 16 days per spacecraft; Landsat 8 and 9 together provide an approximately eight-day interval. Clouds, latitude, acquisition planning, spectral needs, and historical continuity determine which system is more useful for a particular job.
As an Amazon Associate I earn from qualifying purchases.
The three measurements answer different questions
Spatial resolution: what size of ground sample is represented by a pixel?
Spatial resolution describes the nominal ground sampling scale. A 10 m pixel represents a much smaller ground cell than a 30 m pixel, so smaller metre values generally make it easier to distinguish fine patterns. That does not mean every 10 m product is automatically better: the answer also depends on the spectral band, signal quality, viewing geometry, processing, and the feature being mapped.
Always attach a resolution number to its instrument and band. A mission’s headline figure can describe only its finest bands. Sentinel-2’s MultiSpectral Instrument (MSI) has 13 bands distributed across three resolutions: four bands at 10 m, six at 20 m, and three at 60 m. Calling all Sentinel-2 imagery “10 m” hides that difference. Landsat 8’s Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) likewise use different sampling scales: 30 m multispectral bands, a 15 m panchromatic band, and 100 m thermal bands.
#1 Best Overall
Coverage: how wide a strip is imaged during a pass?
Swath is the cross-track width of ground observed during one orbital pass. A wider swath can cover a broader strip in that pass, all else being equal. It is not the total area a mission can map during a month, and it is not a guarantee that every kilometre inside the strip will produce cloud-free, usable data.
Sentinel-2’s cited swath is 290 km; Landsat 8’s is 185 km. Those figures are nominal instrument geometries. Overlap between adjacent paths changes with latitude, and acquisition plans, edge effects, clouds, haze, shadows, and quality screening reduce the portion of a swath that is useful for a specific analysis.
Revisit: how often can the mission observe the same place?
Revisit or repeat time is the interval between scheduled opportunities to image a location under the mission’s orbit and acquisition rules. State whether the number applies to one spacecraft or to a constellation. Sentinel-2 technical documentation gives a 10-day cycle for one satellite and a five-day combined revisit for the constellation. Landsat 8 repeats every 16 days as an individual satellite; Landsat 8 and Landsat 9 are phased to provide an approximately eight-day observation interval at a location.
A scheduled opportunity is not the same as a usable image. Optical sensors can be blocked by clouds, and the specifications cited here do not provide a universal cloud-free probability. Treat revisit as potential cadence unless you have a location-specific analysis of cloud, daylight, viewing, and quality conditions.
Side-by-side comparison of the official examples
| System | Sensor and resolution examples | Nominal swath | Nominal repeat or revisit | How to interpret it |
|---|---|---|---|---|
| Sentinel-2 MSI, one satellite | Four bands at 10 m, six at 20 m, three at 60 m | 290 km | 10 days in technical documentation | Use the band-level value rather than the 10 m headline; this is a single-spacecraft cycle. |
| Sentinel-2 MSI constellation | Same 13-band MSI distribution | 290 km per satellite | 5 days for the combined constellation | More frequent opportunities result from two spacecraft, not from changing the sensor’s pixel size. |
| Landsat 8 OLI/TIRS | 30 m multispectral, 15 m panchromatic, 100 m thermal | 185 km | 16 days per satellite | Optical, panchromatic, and thermal measurements have different nominal resolutions. |
| Landsat 8/9 constellation | Landsat-family OLI/TIRS characteristics on each spacecraft | 185 km per satellite | 8 days at a location | Pairing the satellites improves temporal opportunity while retaining a long-running land-imaging record. |
The figures come from European Space Agency and Copernicus Sentinel-2 specifications, U.S. Geological Survey Landsat documentation, and NASA mission pages. They are technical mission specifications, not a ranking of every government or commercial constellation.
Rank #2
How to read resolution without being misled
Separate spatial, spectral, and thermal detail
Spatial resolution answers how finely the ground is sampled. Spectral resolution concerns the number and breadth of wavelength bands, which determines whether a sensor can distinguish vegetation, water, minerals, aerosols, or other materials using the relevant portion of the spectrum. A panchromatic band is usually broad and can have finer spatial sampling than multispectral bands, while a thermal band measures emitted heat at a coarser scale in the Landsat 8 example.
For Sentinel-2, report the actual MSI band group needed by the analysis: 10 m visible and near-infrared bands, 20 m bands, or 60 m atmospheric and calibration-oriented bands. For Landsat 8, identify whether the result uses a 30 m OLI multispectral band, a 15 m panchromatic product, or 100 m TIRS thermal data. Do not merge these into one undifferentiated “satellite resolution.”
Free tools Windows power users keep installed
One-click scans. No signup required.
Use the smallest relevant feature, not the smallest advertised pixel
A feature that occupies only a fraction of a pixel may be mixed with its surroundings, even when the nominal pixel size appears suitable. Conversely, a broad regional pattern can be mapped effectively with a coarser band. Ask what boundary, patch, field, road corridor, water body, or temperature pattern must be distinguished, then select the band and product that preserve the needed signal.
Do not treat panchromatic detail as multispectral detail
Landsat 8’s 15 m panchromatic measurement is not equivalent to having every multispectral band at 15 m. If an analysis needs a specific color or near-infrared band, its native 30 m sampling remains the relevant specification unless a documented processing method creates a derived product. A sharpened image can improve visual appearance, but it does not create new independent spectral measurements.
How swath width translates into practical coverage
Swath is a strip, not a guarantee of complete mapping
Imagine the satellite tracing a ground track while its sensor records a strip 290 km or 185 km wide. That width helps determine how quickly adjacent areas can be reached, but the mission still needs suitable acquisition geometry and a clear enough observation. Edge portions may have different viewing angles or quality characteristics, and a planned scene can be rejected or masked after acquisition.
Latitude changes path overlap
USGS notes that Landsat swath overlap varies from about 7% at the equator to approximately 85% at extreme latitudes. Consequently, a 16-day repeat figure and a 185 km swath do not imply identical geographic coverage behavior everywhere. Higher-latitude path convergence can create more frequent opportunities for some locations, while equatorial areas have less overlap. Apply the same geographic qualification to any comparison based on nominal swath.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteDo not convert width directly into usable area
Multiplying swath width by an assumed distance or number of days would produce a theoretical geometric estimate, not a guaranteed clear-image area. Clouds, haze, shadows, snow, off-nadir limits, acquisition priorities, and quality masks determine what can actually be analyzed. If your project requires a percentage of cloud-free observations, obtain a location- and season-specific assessment rather than inferring it from swath alone.
How to interpret revisit and repeat time
One spacecraft versus a constellation
Sentinel-2 demonstrates the distinction clearly: one satellite’s technical cycle is 10 days, while the two-satellite constellation’s combined revisit is five days. Landsat 8’s individual repeat is 16 days, while the paired Landsat 8/9 system observes a location at an eight-day interval. Label the number in every table, chart, and project requirement.
Nominal opportunity versus usable observation
Revisit is an orbital opportunity, not a promise of an interpretable image. A scheduled optical acquisition may be cloudy, hazy, dark, snow-covered, or otherwise unsuitable. The sources for these specifications do not quantify cloud-free cadence, so do not present five, eight, or 16 days as the guaranteed interval between clear images.
Orbit geometry and acquisition policy matter
Even when a spacecraft passes near a target, imaging can depend on pointing constraints, daylight, instrument operations, scene priorities, and processing availability. For emergency monitoring, ask the data provider how acquisitions are scheduled and delivered. For long-term mapping, define an acceptable interval and quality threshold, then evaluate whether the nominal repeat creates enough usable observations for that threshold.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Choose the system by the job
Fine visible or near-infrared mapping
If the task benefits from 10 m visible and near-infrared sampling and frequent constellation opportunities, Sentinel-2’s four 10 m bands and five-day combined revisit are relevant advantages. Confirm that the specific spectral bands required by the algorithm are among the 10 m group; other MSI bands are 20 m or 60 m.
Broad-area, consistent land monitoring
Landsat 8/9 offers a 185 km swath per satellite, an eight-day paired interval, and a long-running Landsat-family record. That combination can be valuable when consistent historical analysis matters more than the finest nominal pixel size. Use the 30 m multispectral, 15 m panchromatic, or 100 m thermal specification according to the measurement, not according to the mission name alone.
Thermal questions
When the question involves surface temperature or heat patterns, Landsat 8’s 100 m thermal resolution is the directly relevant figure. Sentinel-2’s 10 m headline resolution should not be used as a substitute for a thermal specification. A finer optical pixel does not answer a thermal question unless the mission carries an appropriate thermal sensor.
Rapid change detection
For changes that may occur within days, start with constellation-level revisit and then estimate the expected clear-observation cadence for the location and season. Sentinel-2’s five-day combined opportunity may suit a faster optical monitoring schedule than a single Landsat 8 spacecraft’s 16-day repeat, but clouds can erase that apparent advantage during a particular period.
Historical continuity and comparability
If the project spans many years, sensor lineage, calibration, band definitions, processing standards, and archive depth can outweigh a difference in nominal resolution. Landsat 8/9’s paired observation interval should be considered alongside the continuity of the Landsat record. A newer or finer pixel is not automatically preferable if it breaks the time series or changes the measured phenomenon.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.A repeatable comparison workflow
- Define the feature and decision. Write down the smallest feature, boundary, or change that must be detected, and whether the output is a map, a time series, a temperature measurement, or a visual reference.
- List required wavelengths and sensor type. Specify visible, near-infrared, shortwave-infrared, thermal, or panchromatic needs before looking at headline pixel sizes.
- Record band-level resolution. For each candidate, name the instrument and the exact band group: for example, Sentinel-2 MSI 10 m, 20 m, or 60 m; Landsat 8 OLI 30 m or 15 m panchromatic; Landsat 8 TIRS 100 m thermal.
- Record swath separately. Enter 290 km for Sentinel-2 or 185 km for Landsat 8 in the coverage column, and label the number as nominal swath width.
- Separate spacecraft and constellation cadence. Write both values where relevant: Sentinel-2’s 10-day single-satellite cycle and five-day constellation revisit; Landsat 8’s 16-day repeat and the eight-day Landsat 8/9 interval.
- Check geographic effects. Examine latitude, path overlap, target size, viewing geometry, and whether the location sits near a scene edge. USGS’s 7% equatorial to about 85% extreme-latitude overlap range shows why a global average is unsafe.
- Estimate usable cadence. Add cloud, haze, daylight, snow, quality-mask, and delivery constraints. Keep this estimate distinct from the nominal orbital number.
- Check continuity and processing. Confirm that the archive, calibration, band definitions, and product processing support the length and consistency of the analysis.
- State the trade-off in the final recommendation. Explain which requirement drove the choice instead of declaring a universal highest-resolution or fastest satellite.
Common comparison errors and their fixes
- Error: “Sentinel-2 is 10 m.” Fix: say that four MSI bands are 10 m, six are 20 m, and three are 60 m.
- Error: “Landsat is 15 m.” Fix: identify the 15 m panchromatic band; Landsat 8 multispectral bands are 30 m and thermal bands are 100 m.
- Error: treating swath as revisit. Fix: swath is cross-track width in one pass; revisit is the time between observation opportunities.
- Error: mixing a single satellite with a constellation. Fix: label Sentinel-2’s five-day constellation revisit separately from its 10-day single-satellite cycle, and Landsat 8’s 16-day repeat separately from the Landsat 8/9 eight-day interval.
- Error: promising a clear image on every repeat. Fix: call the figure a nominal opportunity and discuss clouds and quality screening.
- Error: assuming global coverage behaves uniformly. Fix: account for latitude-dependent overlap, including the Landsat range from about 7% at the equator to about 85% at extreme latitudes.
- Error: choosing by pixel size alone. Fix: include spectral bands, thermal capability, processing, archive continuity, area, and required cadence in the decision.
How to present a defensible satellite comparison
Use a table with separate columns for instrument, band or mode, spatial resolution, nominal swath, single-spacecraft repeat, constellation revisit, and practical limitations. Put the qualifier in the same cell as the number. For example, write “Sentinel-2 MSI: 5 days, combined constellation” rather than “5 days” alone, and “Landsat 8 TIRS: 100 m thermal” rather than “Landsat 8: 15 m.”
Then add a short use-case statement: identify the feature size, spectral requirement, area, monitoring interval, cloud tolerance, and historical period. This makes the comparison auditable and prevents a nominal specification from being mistaken for a guarantee.
What these specifications do not establish
The cited mission pages establish the listed sensor resolutions, swath widths, and nominal repeat or revisit figures. They do not establish current commercial imagery prices, licensing terms, tasking availability, exact cloud-free cadence for a particular location, or every product-level processing difference. Those questions require a separate provider, archive, or study-specific check.
Recommended Free Tools
They also do not support a universal ranking of all Earth-observation constellations. Sentinel-2 and Landsat 8/9 are official examples that expose the comparison method: tie resolution to a band, treat swath as nominal geometric coverage, and distinguish one spacecraft’s repeat from a constellation’s combined revisit.
Bottom line: compare the requirement, not the headline
Choose Sentinel-2 when its 10 m bands and five-day constellation opportunity fit the spectral and temporal needs, while remembering that the same mission also has 20 m and 60 m bands. Choose Landsat 8/9 when the 30 m multispectral, 15 m panchromatic, or 100 m thermal measurements, 185 km swath, eight-day paired interval, and long-running record fit the project. In either case, report the band, sensor, spacecraft count, latitude, and cloud-related qualification with every number. That is the difference between a technically correct comparison and a misleading resolution, coverage, or revisit claim.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
