Spatiotemporal Posture¶
Not every checked-in source family is comparable in the same way through space and time.
That is not a documentation problem. It is a real property of the tracked evidence. Some source families contribute lake anchors, some contribute point inventories, some contribute density context, and only some contribute numeric BP windows that can support chronology-aware comparison.
The repository keeps that distinction explicit in the machine-readable registry:
data/source_spatiotemporal_posture_registry.json
Why This Registry Exists¶
Without one shared posture surface, readers have to infer too much from file names or map symbols alone.
This registry answers a narrower set of questions directly:
- is this family a lake anchor, a context point inventory, a density surface, or framing geometry?
- does the checked-in repository state carry numeric BP windows for this family?
- should this family affect chronology-aware ranking, broad context scoring, or only candidate-lake identity?
What It Says Right Now¶
- LandClim carries numeric BP windows in the normalized site-sequence layer and acts as supporting pollen context.
- Neotoma carries uneven BP site-span support and must not be treated as if the checked-in raw capture already includes chronology rows for every site.
- SEAD remains contextual archaeology inventory in the checked-in state and should not be promoted into same-period support unless explicit numeric intervals are present.
- RAÄ is rich Swedish archaeology context, but the repository-owned normalized surface is density-oriented rather than one exact local-distance site inventory.
- SVAR governs candidate-lake identity and location, not chronology.
- Boundary geometry constrains scope and framing only.
How To Use It¶
Use the registry before making or reading lake-ranking claims.
If a source family is marked as context only, do not inflate it into direct same-period support. If a family is marked as a lake anchor only, do not count it as environmental or archaeological evidence. If a family carries numeric intervals unevenly, keep the unevenness visible instead of smoothing it away in the report.