Skip to main content

Reference data

Reference data is the set of shared lookup lists that the rest of Erde points at: the Matrices, Analytes, Methods, Measurement Units, Site Types, and dozens of other classifications that fill the dropdowns on your forms. Centralizing these values means a Sample's matrix or a Lab Analysis's method always come from one curated list — so records stay comparable and analysis stays clean. Administrators and Data Managers maintain reference data in one place: the top-level Reference Data area.

Why Erde centralizes lookups​

If every record carried free-typed text for its matrix or method, "Groundwater", "groundwater", and "GW" would all be different values, and screening or Crosstab results would fragment. Instead, each domain field references a single reference-data row by ID. You manage the list once; every picker, import, and report draws from it.

Reference data is grouped by domain in the catalog. Each Reference data type classifies one kind of record:

Reference-data typeClassifies / supplies
MatricesThe sample medium — Groundwater, Soil, Soil Gas, Surface Water, and the like
AnalytesSubstances the lab measures (PFAS, VOCs, metals); carries a CAS (Chemical Abstracts Service) number and aliases
Analyte GroupsNamed collections of analytes
Analysis Methods / Preparation Methods / Leachate MethodsThe lab Methods behind a result
Measurement UnitsUnits of measure, with conversion grouping
Detection Limit TypesKinds of detection or reporting limit
QualifiersLab-result qualifier codes (J, U, R, and the like) and what each conveys — its disposition and how it propagates when results combine
Lab Instrumentations / Method OrganizationsLab instrument and method-source lookups
Site Types / Location Types / Site Area TypesClassifications for Sites and Locations
Site ProgramsRegulatory programs a Site can be registered under
Well Types / Drilling Methods / Casing Materials / Screen Types / Annular Materials / Surface Completion Types / AquifersWell construction and hydrogeology
Borehole Types / Coring Methods / Decommission Methods / LithologiesField-artifact and geology lookups
Coordinate Systems / Vertical DatumsSpatial reference systems and elevation datums
Project Types / Organization TypesProject and organization classifications
Sample MethodsHow a sample was collected
ParametersField and water-quality parameters for monitoring
Document CategoriesThe optional filing buckets for free-standing Site and Project Documents — QAPP, Work Plan, Permit, and the like; each carries a Site/Project/All type
Equipment TypesField-equipment classifications

The common shape​

Every classification-style reference-data type shares the same four-plus-one fields. Types with extra domain data — Matrices carry a Matrix Type, Coordinate Systems carry an Srid, Analytes carry a CAS number, Qualifiers carry disposition flags and propagation ranks — add columns on top of this base, but the core is always the same.

FieldTypeRequiredDescription
NametextYesThe descriptive label; max 200 characters
CodetextNoShort identifier; max 50 characters
DescriptiontextNoOptional notes; max 500 characters
Activeyes/no—Controls whether the value is selectable in pickers; defaults to yes
Pinnedyes/no—Sorts frequently-used values to the top; defaults to no

Name is required and unique; the Code is unique when set. The Code is optional for most classification lookups, but some types — the lab Methods, Measurement Units, Analyte Groups, Matrices, and Detection Limit Types — require one, because for those the Code is the operational identifier labs and EDDs key on. Both Name and Code are matched case-insensitively, so GW and gw collide. Adding a duplicate Name or Code returns a clear conflict rather than a silent overwrite. Standard codes accept letters, digits, and ., _, -; Measurement Unit codes additionally allow / and %. See Codes and names for the Code-versus-Name convention across Erde.

How it fits together​

Reference data sits beside the data model: domain records hold a reference to a lookup row, never a copy of its text. Each reference-data type is an aggregate root, so it carries its own audit trail and comments.

Figure: Domain records reference reference-data rows rather than storing their own text.

Two consequences follow from that relationship:

  • A value in use cannot be deleted. When a Site references a Site Type, or a Sample references a Matrix, deleting that lookup is blocked with a conflict telling you the value cannot be deleted because it is still in use. Clear its Active checkbox instead to retire it from pickers without breaking existing records.
  • Activity and audit follow each row. Because each value is an aggregate root, every create, edit, and delete is recorded, the same as for a Site or Project.

Standard values vs. custom values​

Erde ships a curated set of standard values for each type — common matrices, units, well types, and so on. They are not inserted automatically. A Data Manager loads them on demand with Load Defaults, which previews only the defaults not already present and adds the ones you select. Seeding is additive: it skips any default whose Name or Code already exists and never overwrites or deletes.

You can also add custom values at any time through each type's Add form. There is no schema difference between a seeded value and one you add yourself — a standard value behaves exactly like a custom one once it is in the list.

Who can manage it​

ActionAny signed-in userData ManagerAdministrator
View reference data✓✓✓
Add / edit / delete / seed—✓✓

Editing reference data requires the Data Steward policy, held by the Administrator and Data Manager roles. Without it you see the catalog read-only, with the banner "Reference values are managed by Data Managers — you have read-only access." For the full role matrix, see permissions and scoping.