Skip to content

GeoNet data via eolas

GeoNet is the joint geological hazard monitoring programme run by GNS Science, the Earthquake Commission (Toka Tū Ake EQC), and Toitū Te Whenua LINZ. It operates the seismic + GNSS + volcano monitoring networks. eolas serves 3 datasets from GeoNet — the canonical reference for recent quakes and current volcanic alert levels.

If you're doing seismic-risk modelling, volcanic-hazard work, or building real-time earthquake-aware applications — GeoNet is the source.

For historical major earthquakes + fault locations, use the council-published earthquake-hazard layers (e.g. ECan / Canterbury's ecan_earthquake_faults_2024).

What's in the catalogue

Dataset Description
geonet_quakes_recent Current rolling window of NZ earthquakes with Modified Mercalli Intensity ≥3 — magnitude, depth, location, intensity, origin time.
geonet_strong_motion_sensors Reference data for strong-motion sensor stations across NZ (network code, station, location).
geonet_volcanic_alert_levels Current alert level (Green / Yellow / Orange / Red), activity description, and hazards for each monitored NZ volcano.

Refresh schedule

Weekly. For real-time earthquake feeds, use GeoNet's own WFS/JSON APIs — those update within seconds of an event. Our weekly snapshot is suited to research / dashboard use cases, not emergency-response.

meta = client.info("geonet_quakes_recent")
meta["last_refreshed_at"]

License

All GeoNet data is published under CC-BY 4.0. Commercial use is fine; attribution required.

Recommended attribution: "Source: GeoNet (GNS Science / Toka Tū Ake EQC / LINZ), served via eolas (eolas.fyi). CC-BY 4.0."

Common patterns

Recent quake magnitude distribution

import matplotlib.pyplot as plt

q = client.geonet("geonet_quakes_recent")
q["magnitude"].hist(bins=30)
plt.title("Recent NZ earthquake magnitudes (M ≥ 3)")
plt.xlabel("Magnitude (ML)")
plt.show()

library(ggplot2)
q <- eolas_get_geonet("geonet_quakes_recent")
ggplot(q, aes(magnitude)) + geom_histogram(bins = 30) +
  labs(title = "Recent NZ earthquakes", x = "Magnitude (ML)")

Volcanic alert dashboard

v = client.geonet("geonet_volcanic_alert_levels")
print(v[["volcano", "alert_level", "activity_description"]])

Strong-motion network map

import geopandas as gpd
sensors = client.geonet("geonet_strong_motion_sensors", as_sf=True)
print(f"Strong-motion sensor stations: {len(sensors)}")
sensors.plot(figsize=(8, 10), markersize=5)

Source-specific notes

  • Recent ≠ historical: geonet_quakes_recent is a rolling window (typically last ~30 days). For historical quake catalogues, use GeoNet's data archives.
  • Intensity vs magnitude: Modified Mercalli (MMI) measures perceived shaking at a location; magnitude measures the energy released at the source. A M6 quake 100km offshore might produce MMI 4 onshore; a M4 directly underneath could produce MMI 5.
  • Volcanic alert levels: 5-level scale (0=no unrest through 5=major eruption). Each volcano has its own characteristic activity baseline. Whakaari/White Island has been at Level 2 since the 2019 eruption; Tongariro/Ruapehu fluctuate between 0-2 routinely.
  • Use the live API for real-time apps: GeoNet publishes earthquake events within seconds via api.geonet.org.nz. Our weekly refresh isn't appropriate for emergency-response use cases.
  • Coverage: GeoNet's seismic network is densest in populated areas + active volcanic regions. Strong-motion sensors are biased toward critical infrastructure + dam sites.

Where to find more