Mapping a species' occurrences — GBIF + OBIS¶

Two of the biodiversity-cluster backends record where a species has been observed: GBIF (the global biodiversity record — museums, citizen science, monitoring) and OBIS (the marine record). This showcase pulls both for the loggerhead sea turtle (Caretta caretta) — which is logged both on land (nesting beaches, in GBIF) and at sea (in OBIS) — and maps them together.

Both backends are vector and anonymous: each download() returns a points FeatureCollection in EPSG:4326, so they drop straight onto one map.

Install. This notebook needs both occurrence SDKs: pip install earthlens[gbif,obis]. No credentials are required — both searches are anonymous.

Setup¶

Import matplotlib for the map and the unified EarthLens entry point.

In [ ]:

import matplotlib.pyplot as plt

from earthlens import EarthLens

import matplotlib.pyplot as plt from earthlens import EarthLens

Area and window¶

The wider Caribbean / north-west Atlantic — a loggerhead stronghold — over a five-year window. A bounded bbox keeps both queries fast.

In [ ]:

bbox_lat = [10.0, 35.0]
bbox_lon = [-90.0, -60.0]
start, end = "2015-01-01", "2020-12-31"
species = "Caretta caretta"  # loggerhead sea turtle

bbox_lat = [10.0, 35.0] bbox_lon = [-90.0, -60.0] start, end = "2015-01-01", "2020-12-31" species = "Caretta caretta" # loggerhead sea turtle

GBIF occurrences¶

variables=["taxon:Caretta caretta"] resolves the binomial to a GBIF backbone taxonKey live, then searches occurrences in the bbox.

In [ ]:

gbif = EarthLens(
    data_source="gbif",
    variables=[f"taxon:{species}"],
    start=start,
    end=end,
    lat_lim=bbox_lat,
    lon_lim=bbox_lon,
    path="out/showcase",
    max_records=1500,
).download()

print(f"GBIF: {len(gbif)} occurrences")

gbif = EarthLens( data_source="gbif", variables=[f"taxon:{species}"], start=start, end=end, lat_lim=bbox_lat, lon_lim=bbox_lon, path="out/showcase", max_records=1500, ).download() print(f"GBIF: {len(gbif)} occurrences")

OBIS occurrences¶

The marine record for the same species and bbox.

In [ ]:

obis = EarthLens(
    data_source="obis",
    variables=[f"species:{species}"],
    start=start,
    end=end,
    lat_lim=bbox_lat,
    lon_lim=bbox_lon,
    path="out/showcase",
    size=1500,
).download()

print(f"OBIS: {len(obis)} occurrences")

obis = EarthLens( data_source="obis", variables=[f"species:{species}"], start=start, end=end, lat_lim=bbox_lat, lon_lim=bbox_lon, path="out/showcase", size=1500, ).download() print(f"OBIS: {len(obis)} occurrences")

Map both sources together¶

GBIF in green, OBIS in blue — the land/coast records and the at-sea records form a fuller picture of the species' range than either source alone.

In [ ]:

ax = gbif.plot(markersize=6, color="darkgreen", label="GBIF")
obis.plot(ax=ax, markersize=6, color="navy", label="OBIS")
ax.set_title("Caretta caretta occurrences — GBIF + OBIS, NW Atlantic")
ax.set_xlabel("longitude")
ax.set_ylabel("latitude")
ax.legend()
plt.show()

ax = gbif.plot(markersize=6, color="darkgreen", label="GBIF") obis.plot(ax=ax, markersize=6, color="navy", label="OBIS") ax.set_title("Caretta caretta occurrences — GBIF + OBIS, NW Atlantic") ax.set_xlabel("longitude") ax.set_ylabel("latitude") ax.legend() plt.show()

Licensing¶

Both sources mix permissive and restrictive licenses; each backend raises a LicenseWarning when a result carries a non-commercial (CC-BY-NC) record. Honour the per-record obligations before redistributing, and attribute GBIF and OBIS as their data policies require.