A method that produces a plausible result has not automatically demonstrated that it works for its intended users. Validation requires a defined question, suitable reference evidence, and an evaluation design that reflects how the method will be used. This perspective examines which observations should be compared, how they should be sampled, and which errors matter in practice. It also considers the difference between evidence used to develop a method and evidence reserved for evaluating it. The objective is a test that can reveal weaknesses as well as confirm useful performance.
This working guide considers how pointing and motion affect satellite imagery in banana plantation units. The immediate decision is to identify plantation units for closer review. Useful reference evidence includes plant age mixture, canopy continuity and dated field inspections. The case is an illustrative planning context, with assumptions to check against the actual sensor, landscape and project requirements.
Separate the different pointing requirements
A camera can point near its intended target while still producing an image with poor geometry or blur. Pointing accuracy describes how closely the spacecraft achieves a commanded orientation. Pointing knowledge describes how well that orientation is estimated. Stability concerns changes over a relevant interval, including the detector integration period. These quantities serve different purposes and should be specified separately with a clear reference frame and time basis.
Orientation estimates combine information from sensors such as gyroscopes, Sun sensors and star trackers, depending on the architecture. Actuators then change or maintain the attitude. Reaction wheels exchange angular momentum with the spacecraft, while magnetic torquers interact with Earth's magnetic field and may support momentum management. Their suitability depends on the orbit, operating modes and disturbance environment. A component's advertised accuracy does not automatically become the pointing performance of the integrated spacecraft.
Evaluate the complete image motion budget
Ground features move across a detector because the spacecraft travels along its orbit and may rotate during imaging. Exposure time, scan timing and any motion compensation must be consistent with that movement. Residual motion during integration spreads signal across pixels and reduces contrast. High frequency vibration can produce a different image effect from slow drift, so one average attitude error may conceal an important problem.
Build an image motion budget that includes control behavior, structural response, actuator disturbances and timing uncertainties. Consider the mechanical path between attitude sensors and the optical payload: a flexible interface can change their relative alignment. Thermal distortion can also alter boresight alignment over an orbit. Test realistic operating combinations, including wheel speeds and other mechanisms that may run during an exposure, and assess whether laboratory support equipment affects the measurement.
Geometric processing uses orbit information, attitude knowledge, timing and an Earth model to locate observations. Ground control and image matching can improve some errors, but processing cannot always restore information lost to blur. For repeat observations, test relative registration as well as absolute location. Report residual errors across the field and across relevant operating conditions. A useful requirement links a user consequence, such as registration between spectral bands, to a measurable spacecraft or payload behavior and a verification method. This keeps pointing discussions connected to the quality of the delivered imagery.
The setting: Banana plantation units
Banana production units can contain plants at different stages within a continuously managed system. Replanting, harvesting, leaf management and weather effects can change the canopy pattern at a local scale. A monitoring service should account for that mixture rather than expecting every unit to follow one simple seasonal curve. Define the management boundaries and record how stages are organised within them before interpreting a difference between neighbouring plantation areas.
In humid settings, clouds and shadows can reduce the number of usable optical observations. Retain those gaps explicitly and examine whether the information remains current enough for the intended inspection process. Plantation roads, drainage features and open replacement areas can influence a unit average. Where the proposed output is more detailed than the underlying image support, consider a broader summary or complementary field and drone observations instead of implying plant-level certainty.
Field teams should record stage mixture, canopy condition, recent management and visible events relevant to the sampled area. A vegetation anomaly may justify a visit, but it cannot identify a specific disease or establish production loss on its own. Evaluate models and inspection priorities against representative observations, including normal-looking units. Reports should distinguish the remotely observed pattern from the explanation established during follow-up, allowing plantation staff to use both sources of evidence appropriately.
Represent gaps without inventing observations
A monitoring sequence often contains periods without a usable observation. The reason may be cloud, shadow, missing coverage, a failed transfer, or a quality condition that excludes the data. Begin by representing these states explicitly. A missing observation is different from a low measurement, an unchanged surface, or a confirmed absence of the feature being studied. The distinction should survive both the analysis and the presentation of the result.
Define when a temporal gap is acceptable for the intended decision. A workflow may continue using the latest qualified observation, request another source, or suspend an alert until sufficient evidence is available. Whichever response is selected, show the age of the supporting observation. If interpolation is used for a suitable analytical purpose, preserve the distinction between an estimated value and a directly observed measurement, together with the assumptions needed to interpret it.
Review how gaps affect the coverage of the final evaluation. Locations with frequent missing data may differ systematically from those with regular observations. A performance estimate based only on clear, complete records can therefore describe a narrower population than the service is expected to support. Report usable coverage alongside analytical results and use recurring gaps to guide acquisition planning, reference collection, and the selection of complementary observation methods.
Define a representative reference sample
A reference sample should represent the situations in which a method will be used. Begin by defining the target population: locations, time periods, management conditions, surface types, or instrument states that belong within the intended application. A collection of convenient observations may be valuable for exploration, but it does not automatically support conclusions about that wider population. Make the scope of the sample visible before calculating a summary of performance.
Consider a stratified design when important conditions are unevenly represented. Relevant strata might include different landscape types, developmental stages, observation quality levels, or ranges of the variable being estimated. Choose sample locations within those groups using a documented rule, and retain the information needed to understand their selection. Where access restrictions affect the sample, record the excluded conditions so that the limits of the resulting evaluation remain clear.
Keep development and evaluation evidence separate where the purpose of the study requires an independent test. Nearby observations or repeated measurements can share information, so a random split of individual records may overstate independence. A split organised by location, period, or operational unit can provide a more demanding comparison. The appropriate design follows the intended deployment, the structure of the observations, and the errors that matter to the user.
Bring the evidence back to the decision
The validation record should explain the population represented by the test, the reference observations used, and the limitations of the comparison. Report meaningful errors alongside useful successes, and retain examples that challenge the method. Treat wider deployment as a new opportunity to evaluate transfer across relevant conditions. A well-designed validation programme gives users a defensible basis for deciding where the method is useful, where additional evidence is needed, and where another approach may be appropriate.
Sources & further reading
Explore the bhumisat spacecraft concept and our vision towards 2030, or return to the journal.



