Dataset Modeling (Sub‑components)
Dataset modeling approach (next step)
Section titled “Dataset modeling approach (next step)”We currently have two datasets decomposed and mapped in one large diagram. It’s useful but too complex to maintain. Our next step is to split the model by sub‑components and build each module independently—starting with entities / objects of interest.
Why this approach
Section titled “Why this approach”- Reduces complexity and makes reviews tractable
- Lets different contributors focus on their specialty
- Produces smaller modules that can be composed later
Sub‑components to model
Section titled “Sub‑components to model”1) Entities / Objects of Interest (start here)
Section titled “1) Entities / Objects of Interest (start here)”- Salmon life stages (juvenile, smolt, adult)
- Demographic units (population, CU, stock)
- Sampling units (site, station, survey)
- Physical specimens and collections
Outputs: entity taxonomy, labels, and mappings to SOSA + I‑ADOPT ObjectOfInterest.
2) Variables
Section titled “2) Variables”- Define “what is being observed” as a composed variable
- Example: “juvenile sockeye fork length at ocean entry”
Outputs: variable templates, decomposition into I‑ADOPT components, links to methods.
3) Properties
Section titled “3) Properties”- The measurable characteristic (length, weight, condition factor)
- Standardized naming + QUDT units
Outputs: property vocabulary + constraints.
4) Events & Activities
Section titled “4) Events & Activities”- Sampling events, surveys, capture/recapture
- Observation events in SOSA
Outputs: event classes + relationships to entities and methods.
5) Methods & Protocols
Section titled “5) Methods & Protocols”- Measurement methods (otoliths, scales, gear type)
- Protocol identifiers and provenance links
Outputs: methods vocabulary + provenance templates.
Suggested order of work
Section titled “Suggested order of work”- Entities (objects of interest)
- Properties (what is measured)
- Variables (compose object + property + constraints)
- Events (observation/sampling context)
- Methods (procedures and provenance)
We’ll then re‑compose the modules into a single integrated model.
Simple alignment diagram (SOSA + I‑ADOPT)
Section titled “Simple alignment diagram (SOSA + I‑ADOPT)”flowchart LR OOI[I‑ADOPT Object of Interest] PROP[I‑ADOPT Property] VAR[I‑ADOPT Variable] EVENT[SOSA: Event / Sampling] OBS[SOSA: Observation] MEAS[SOSA: Result / Measurement] METHOD[SOSA: Procedure / Method] VAR -->|has Object of Interest| OOI VAR -->|has Property| PROP OBS -->|observedProperty| VAR OBS -->|hasFeatureOfInterest| OOI OBS -->|hasResult| MEAS OBS -->|madeBySensor / Procedure| METHOD EVENT -->|hosts| OBS METHOD -->|used in| OBS