Field Sampling Plan¶
Sampling plan — instantiates Representative Sampling Design
Spreads observation effort across sites, seasons, and conditions on a spatial-temporal design so field data speak for a whole environment, not just its most accessible corners.
Field Sampling Plan decides where and when to observe across a physical environment — sites, times, seasons, habitat gradients — so that the easy-to-reach and the fair-weather don't quietly stand in for the whole. Its distinctive axis is space and time: the population is an environment, the selection is a spatial-temporal design, and the enemy is accessibility bias — the roadside site, the summer-only visit, the remote reach dropped for logistics. It defines the environmental extent, distributes effort across it, and checks that the achieved sites actually span the physical and seasonal gradient rather than clustering where the truck can park.
Example¶
A watershed council wants to characterize stream health across a 500-square-kilometer basin (scenario illustrative). The temptation is to sample the three bridges with parking, in summer, and call it the basin. The plan instead names the target as all wadeable stream reaches, basin-wide, across the year. Selection uses a spatially balanced random design that places sites across the whole stream network — including headwaters reachable only on foot — with repeat visits scheduled in both wet and dry seasons. The coverage check overlays the achieved sites on a GIS map of the network, broken out by elevation, stream order, and surrounding land use, to expose zones that are under-sampled.
The outcome is a stream-health index that represents the basin, carrying an explicit note of which high-elevation reaches remain thin — not a verdict of "the streams are healthy" inferred from three convenient bridges in July.
How it works¶
- Map the environmental extent. Define the full spatial and temporal universe the claim is about.
- Select across space and time. A spatially balanced or systematic-random design places sites across the network and schedules visits across seasons and conditions.
- Overlay achieved on intended. Compare the sites actually sampled against the extent map to find spatial and temporal gaps.
- Revisit to fill. Add effort where the gradient is thin, or narrow the claim to the covered extent.
Tuning parameters¶
- Spatial density — more sites sharpen spatial resolution but raise field cost sharply in remote terrain.
- Temporal replication — more seasons and repeat visits capture real variation but multiply effort.
- Access weighting — how much to spend fighting accessibility bias (a helicopter to remote reaches) versus staying on budget.
- Habitat-stratum depth — finer habitat classes improve coverage of the gradient at the cost of design complexity.
When it helps, and when it misleads¶
Its strength is converting "where we could park" into "where the environment actually is," so a field index reflects the basin rather than its trailheads.
Its failure mode is subtler than access alone: imperfect detection. Even a well-placed site can miss a species or condition that is present but cryptic[1], so an absence in the data is not an absence in the field, and a single visit can undercount a seasonal feature. Accessibility bias also creeps back whenever remote sites are quietly dropped for weather or logistics and the coverage map is not updated to show it. The classic misuse is generalizing a roadside survey to a wilderness interior. The guarding discipline is to build in repeat visits or detection adjustment and to keep the coverage map honest about every site that was planned but skipped.
How it implements the components¶
target_population— the environmental extent: all reaches, basin-wide, across seasons.selection_method— the spatially balanced, systematic-random placement of sites and timing of visits.coverage_gap_check— overlaying achieved sites on the extent map to expose spatial and temporal gaps.
It does not maintain a response_pattern_monitor for participant dropout or a weighting_or_adjustment_plan — those handle people and belong to Representative Survey Protocol and Stratified Sample; a field plan samples places and times, which cannot decline to answer.
Related¶
- Instantiates: Representative Sampling Design — the spatial-temporal instance, guarding field evidence against accessibility bias.
- Sibling mechanisms: Representative Survey Protocol · Stratified Sample · Audit Sample · User Research Panel · Quality Inspection Sample · Public Consultation Panel · Benchmark Dataset
Editorial Notes¶
Form Classification¶
Form family: Representation, Specification & Plan
Rationale: Field Sampling Plan operates as a non-executable information artifact that externalizes static or prospective structure because it spreads observation effort across sites, seasons, and conditions on a spatial-temporal design so field data speak for a whole environment, not just its most accessible corners.
Independent corroboration: The frozen evidence defines Field Sampling Plan as 'Spreads observation effort across sites, seasons, and conditions on a spatial-temporal design so field data speak for a whole environment, not just its most accessible corners', so its operative form is Representation, Specification & Plan.
Nearest alternative: Protocol, Workflow & Routine — The artifact prospectively specifies where and when observation should occur; sampling execution is handled by a separate procedure.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Statistics & Experimental Design
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Sampling designs for representative inference across space, time, and strata belong to statistics and survey methodology.
Related originating lineages:
- Biology & Ecology — Ecological field surveys materially developed habitat- and season-balanced sampling designs.
- Environmental Science & Climate Studies — Ecological and environmental monitoring materially developed spatial-temporal field sampling practice.
Review resolution: Both reviewers agree that statistics_experimental_design is primary. I retain environmental_climate, biology_ecology only as formative origin lineage(s), without treating every later application as an origin. cross_disciplinary_synthesis is appropriate because the exact artifact combines contributions from multiple professional lineages. Reach is multi_domain as a separate applicability judgment: it does not widen or narrow the recorded provenance. Encyclopedia synthesis is false because the artifact is already established enough that encyclopedia-specific synthesis is not required. The secondary differences are reconciled with no unresolved primary-provenance ambiguity.
Review outcome: Reconciled after independent review; high confidence.
References¶
[1] MacKenzie, D. I., Nichols, J. D., Lachman, G. B., Droege, S., Royle, J. A., & Langtimm, C. A. "Estimating Site Occupancy Rates When Detection Probabilities Are Less Than One". Ecology 83(8), 2248–2255 (2002). Shows that imperfect detection can make a present species go unobserved at a sampled site, so nondetection is not evidence of absence. registry ↩