Skip to content

DOC data via eolas

The Department of Conservation (DOC) manages New Zealand's ~8.8 million hectares of public conservation land — about a third of the country — plus the network of huts, campsites, and tracks that sit on it. eolas serves 10 datasets from DOC: the canonical layers behind every "find a hut", "where can I camp", and "what's protected" workflow.

If you're building outdoor / tramping / tourism tools, doing conservation-policy analysis, or producing recreation reports — DOC is the source.

What's in the catalogue

Recreation infrastructure

Dataset Description
doc_huts All ~1,400 DOC huts (backcountry, serviced, basic, historic). Point geometry + facilities + bookable flag.
doc_campsites All ~300 DOC-managed campsites (basic / standard / scenic / serviced). Point geometry + facilities + booking system.
doc_tracks All ~3,200 walking, tramping, and back-country tracks. Polyline geometry + length, difficulty, status.
doc_walking_experiences_locations Curated "experience" point locations — the marketing-quality subset of tracks/huts featured on DOC's visitor-facing site.
doc_walking_experiences_routes Polyline routes for the same curated experiences. Pair with _locations for the full set.

Protected areas

Dataset Description
doc_public_conservation_land ~11,000 parcels of public conservation land — national parks, conservation parks, ecological areas, reserves.
doc_marine_reserves All gazetted marine reserves under the Marine Reserves Act 1971, plus indicative areas.
doc_marine_mammal_sanctuaries Boundaries of gazetted marine mammal sanctuaries (e.g. Banks Peninsula, Te Pewhairangi/Bay of Islands).

Operational alerts (current closures + warnings)

Dataset Description
operational_alerts_campsites Current alerts on campsites — temporary closures, facility outages, access advisories.
operational_alerts_huts Current alerts on huts — closures, water-supply outages, track damage.

Refresh schedule

Weekly, Wednesday morning NZ time.

The recreation-infrastructure layers (huts, campsites, tracks) update from DOC's authoritative system as new bookings go live or assets are added. The operational-alerts feeds refresh weekly but the source is updated daily — for absolutely up-to-the-minute closures, also check the DOC alerts page before trips.

meta = client.info("doc_huts")
meta["last_refreshed_at"]
meta["source_last_modified_at"]

License

All DOC data is published under CC-BY 4.0 via the DOC Open Data Portal. Commercial use is fine; attribution required.

Recommended attribution: "Source: Department of Conservation, served via eolas (eolas.fyi). CC-BY 4.0."

Common patterns

All huts in Fiordland

import geopandas as gpd

huts = client.doc("doc_huts", as_sf=True)
# Filter to Fiordland National Park (most DOC layers tag region/conservancy)
fiordland = huts[huts["region"] == "Southland"]
fiordland.plot(figsize=(8, 10), markersize=10)

library(sf)
library(ggplot2)

huts <- eolas_get_doc("doc_huts", as_sf = TRUE)
fiordland <- huts[huts$region == "Southland", ]

ggplot(fiordland) + geom_sf(size = 0.6) + theme_void() +
  labs(title = "Southland DOC huts")

Active closures for trip planning

hut_alerts = client.doc("operational_alerts_huts")
camp_alerts = client.doc("operational_alerts_campsites")

# Active alerts in the Tongariro region this week
import pandas as pd
all_alerts = pd.concat([hut_alerts, camp_alerts])
tongariro = all_alerts[all_alerts["region"].str.contains("Tongariro", case=False, na=False)]
print(tongariro[["asset_name", "alert_type", "summary"]].to_string())

Track network length by region

tracks = client.doc("doc_tracks", as_sf=True)
# `tracks_meters` exists; if not, compute from geometry
tracks["length_km"] = tracks.to_crs("EPSG:2193").length / 1000
by_region = tracks.groupby("region")["length_km"].sum().sort_values(ascending=False)
print(by_region.head(10))

Marine protection overlay

import geopandas as gpd

reserves = client.doc("doc_marine_reserves", as_sf=True)
mammals = client.doc("doc_marine_mammal_sanctuaries", as_sf=True)

# Total area protected (km²)
print("Marine reserves:", reserves.to_crs("EPSG:2193").area.sum() / 1e6, "km²")
print("Marine mammal sanctuaries:", mammals.to_crs("EPSG:2193").area.sum() / 1e6, "km²")

Source-specific notes

  • Walking experiences vs tracks: doc_walking_experiences_* is a curated subset of doc_tracks (the marketing-quality "Great Walks" + featured tramps). For a complete network, use doc_tracks; for the consumer-friendly set with photos and route narratives on DOC's site, use the experiences pair.
  • PCL is a parcel layer: doc_public_conservation_land is ~11k discrete parcels, not 13 national-park polygons. To analyse "national park X total area", aggregate by the legal_status or park_name field.
  • Booking flag ≠ everything bookable: is_bookable on huts/campsites is DOC's booking-system flag. Some seasonal sites take payment via honesty box (no online booking) and are flagged false despite being usable.
  • Geometry: all DOC datasets expose geometry_wkt; use as_sf=True (Python) or as_sf = TRUE (R) to get a GeoDataFrame / sf object.
  • Marine reserves overlap with LINZ: DOC publishes the legal boundaries (doc_marine_reserves); for the underlying coastline/seabed/EEZ, use LINZ. For a "where can I fish?" analysis you typically want both.
  • Crown vs DOC land: not all Crown-owned land is DOC-managed — pastoral leases, defence land, river/lake beds are Crown but outside doc_public_conservation_land.

Where to find more