Gate result, conservation rules and sources

Result: FAIL (run 14 Nov 2026, scripts/analyse.py, numbers in data/analysis.json)

Addendum: series GBIF lacks (looked for 14 Nov 2026 after the fail, "aim higher")

What the page claims instead

Romania has no breeding griffons (sources disagree on when they stopped, see SOURCES). Neighbours grew. One tagged bird, of 18 in the dataset, flew into Romania three times in 2018–19 and crossed the Danube six times. Romanian releases are planned, not done.

Post-review notes (14 Nov 2026)

Conservation rules for the tracks

Never print or ship nest, feeding-station, roost, aviary or release-site coordinates.

Applied in scripts/analyse.py before any file reaches public/ (rule changed 14 Nov 2026 after two reviewers reconstructed roosts from the first version, which used 8 km):

5b. The 6 Danube crossing rings are placed on the river at the nearest point that is at least 20 km from every day-edge or perched fix and 40 km from every stationary-day fix (moved up to 30 km from the real crossing), and no two rings coincide.

Result: about 94% of the 2,571 Romanian fixes are not shipped.

Retest (14 Nov 2026, after the second version)

Three attack scripts written by the two reviewers were rerun on the shipped data: (a) midpoint of one evening's last and next morning's first shipped fix versus the evening's real last fix: 0 of 18 Romanian nights within 5 km, median about 22 km; (b) centre of the round hole the buffer leaves: 1 of 18 nights within 5 km, 0 within 3 km; (c) centroid of shipped points near the 27-28 June 2018 stop and near the Bulgarian home area: nothing shipped within 20 km, nearest shipped point about 74 km from the home area; no stationary-day fix is shipped. The 6 rings are placed by rule 5b, not measured against these scripts.

Sources (checked 14 Nov 2026)

Back