# ZooDB holoplankton — per-taxon export provenance

**Source portal:** ZooDB (SIO Ocean Informatics / Mark Ohman Lab / SIO Pelagic
Invertebrate Collection) — <https://oceaninformatics.ucsd.edu/zoodb/>
**Access:** public login (name/email), `save.php` CSV export endpoint.
**Downloaded:** 2026-06-26 by CalCOFI data integration (issue CalCOFI/workflows#31).

## Query parameters (held constant across all files)

| param | value | meaning |
|---|---|---|
| `beginYear` / `endYear` | 1949 / 2026 | full temporal range |
| `month[]` | 1–12 | all months |
| `timeType` | `all` | day + night |
| `locType` | `all` | all locations |
| `GS[]` | `.*` | all genera/species within the higher taxon |
| `PS[]` | `.*` | all phases/stages combined |
| `pooled` | `combined` | returns BOTH unpooled tows and pooled region-aggregates (see `Source` column) |
| `calcType` | `individual` | one value per sample (no temporal mean) |
| `corrected` | *(off)* | RAW counts, net correction factor NOT applied |
| `questionableID` | *(off)* | excludes questionable identifications |
| `calcUnit` | `1000m3` / `m2` | one file each per taxon |
| `HT[]` | one higher-taxon per file | 33 taxa + `.*` (all-taxa total) |

Files: `{taxon_slug}__{1000m3|m2}.csv` (33 taxa × 2 units) plus
`_all_taxa__{unit}.csv` (community total, QA reference). See `_manifest.csv`
for per-file HTTP code, byte size, and unpooled/pooled row counts.

## Semantics (important for ingest)

- One row = the taxon was **analyzed** for that sample. `Abundance = 0` means
  analyzed-but-absent; a **missing** taxon×sample means **not analyzed**
  (coverage differs by taxon/cruise — do NOT impute zeros).
- `Biomass` (mg C / m²) is **unit-independent** — identical in the `1000m3`
  and `m2` files; only `Abundance` changes with `calcUnit`.
- `Source` = `Unpooled` (single net tow, has Line/Station/lat/lon/datetime) or
  `Pooled` (region aggregate of N tows per cruise; Line/Station/position = N/A,
  `Tows` = number pooled).

## Snapshot note

The live ZooDB is flagged "WORK IN PROGRESS: not yet complete." This snapshot's
all-taxa unpooled count (225 tows) differs from an earlier 2026-06-19 manual
export (276 tows) — plausibly euphausiids (a separate database, ingested via
`ingest_cce-lter_euphausiids`) were split out of the holoplankton "all
taxonomies" aggregate. Treat this directory as a single self-consistent snapshot.