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 testing whether a remote sensing model transfers in shaded cocoa agroforestry. The immediate decision is to identify sites for canopy assessment. Useful reference evidence includes cocoa canopy and overstory observations collected separately. The case is an illustrative planning context, with assumptions to check against the actual sensor, landscape and project requirements.
The setting: Shaded cocoa agroforestry
A shaded cocoa farm contains multiple vegetation layers whose contributions cannot always be separated in a satellite image. Large shade trees may dominate the upper canopy, while the cocoa plants of commercial interest occupy a lower layer. The first task is to define what the observation is intended to represent. Monitoring tree-cover change, evaluating shade structure and estimating a cocoa parameter are related objectives, but they should not be treated as interchangeable outputs.
Build a farm register that captures boundaries, management units and known differences in shade arrangement. Compare imagery at suitable dates and exclude contaminated observations, while recognising that stable upper-canopy greenness can conceal changes below it. Canopy openings, recent pruning, intercropping and storm damage may alter the signal for different reasons. Contextual information helps turn these patterns into questions for field investigation instead of unsupported explanations about cocoa health.
A validation visit should record cocoa condition and overstory characteristics separately. Photograph or describe the sampled area consistently, and retain its location and date so that it can be related to the observation footprint. Sampling only accessible farm edges risks overlooking different conditions in the interior. Reports should explain whether they describe whole-farm vegetation or a validated cocoa-specific measurement, and identify the additional evidence needed before a management recommendation can be made.
Identify what changes between domains
A model trained on one region or season learns relationships present in its development data. Applying it elsewhere assumes those relationships remain useful. New crop varieties, forest types, soil colours, management practices, illumination, or sensor characteristics can violate that assumption. Define the intended transfer explicitly: a new district, another year, another sensor, or a different class distribution. Each change creates a different test. A model that transfers across nearby fields has not thereby demonstrated reliable performance across climates or observation systems.
Keep the measurement pipeline consistent before attributing differences to the model. Reflectance scaling, band order, cloud masking, spatial resolution, composite windows, and missing value handling can all change predictions. Sensors with similarly named bands may have different spectral responses. Harmonization can reduce some differences but does not establish equivalence for every target. Examine predictor distributions, missing data rates, and examples from the new setting. Flag inputs outside the conditions represented during training so that extrapolation remains visible.
Test the destination, then adapt carefully
Design evaluation splits around the intended deployment. Hold out geographic regions to test spatial transfer and later years to test temporal transfer, keeping neighboring samples or repeated observations of the same site together when needed. Report performance for meaningful destination groups rather than only a pooled score. Class prevalence affects some metrics, so interpret changes alongside the underlying confusion matrix or residual distributions. A random pixel split can look excellent while providing little evidence about performance in an unseen region.
If local labels are available, reserve a destination assessment set before fine tuning or recalibration. Use the remaining observations to compare straightforward baselines with more complex adaptation methods. Check whether improvement comes from genuine transfer or simply from adding representative local data. Retain an option to withhold predictions when input quality or novelty makes them unreliable. Document the training domain, destination evidence, preprocessing versions, and conditions where use is unsupported. Continue monitoring once new observations arrive, since seasonal and environmental shifts can emerge after an initial evaluation. Transferability is a measured property of a particular model, workflow, target, and destination; it should not be inferred from architecture name or strong performance on the original test set.
Keep an auditable review record
A review record should explain how a result moved from an analytical output to an accepted interpretation or operational decision. Preserve the product identifier, the evidence considered, the reviewer role, and the outcome of the review. A short, consistent record is often more useful than an extensive free-text note that cannot be compared between cases. The objective is to make important decisions traceable without turning routine review into unnecessary administrative work.
Separate observations from conclusions within the record. Describe the evidence visible to the reviewer, the alternative explanations considered where relevant, and the reason for accepting, qualifying, or rejecting the result. If a correction is required, identify what should change and how the revised product relates to the earlier version. Keep unresolved questions visible so that they can be addressed by another observation or by the person responsible for the next action.
Use the accumulated record to improve the workflow. Recurring review reasons can reveal unclear requirements, missing contextual information, unstable processing steps, or an evaluation gap. Examine whether similar cases receive similar decisions across reviewers and over time. Preserve the relationship between feedback and subsequent revisions so that improvements can be assessed against the evidence that motivated them. An auditable record supports learning, accountability, and consistent delivery while allowing the technical process to evolve.
Locate the uncertainty that matters to the decision
An uncertainty budget is a practical description of where a result could become unreliable. Begin with the observation chain: acquisition, calibration, geographic matching, reference measurements, feature calculation, modelling, and delivery. Identify the assumptions at each stage and ask which could materially change the interpretation. The purpose is not to assign a precise number to every limitation. It is to locate the sources that deserve attention for the intended decision.
Separate limitations that can be measured from those that currently require a qualitative description. Repeated measurements may help characterise variability, while comparisons with independent references can reveal systematic differences. Some uncertainties interact, so combining individual error estimates without examining their relationships can produce a misleading impression of precision. Keep the method of assessment visible and avoid treating an unexamined assumption as though it were a negligible contribution.
Use the budget to prioritise further work. The largest visible source of variation is not always the one whose reduction would most improve the decision. Consider which additional observation, test, or contextual record would change confidence in an actionable way. Revisit the budget when the application, observation source, or deployment conditions change. This turns uncertainty analysis into a guide for improving the workflow rather than a qualification added only after the result has been delivered.
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.


