090°Open data
Open datasets, fully documented — searchable here, and readable by any LLM.
2 datasets
North America container gateway throughput (agent-curated)
Derived coast x month panel over the gateway throughput data: west (LA/Oakland/Seattle-Tacoma), east (NY-NJ/Charleston/Savannah), gulf (Houston), canada-pacific (Vancouver), canada-atlantic (Montreal). Each row gives the number of reporting ports, coast total TEU, the coast's share of covered traffic, 12-month YoY and 3-month momentum on the coast sum, and the 0-100 gateway_momentum_index on coast-level terms. A port missing for a month contributes nothing (never filled). Deterministic in code from the curated port panel; monthly cadence.
Agent-curated monthly TEU panel for 9 North American container gateways (7 US: Los Angeles, Oakland, Oakland, Seattle/Tacoma, New York & New Jersey, Charleston, Savannah, Houston; 2 Canadian: Vancouver, Montreal), one row per port x month. Columns: loaded-import / loaded-export / empty TEU splits where the port publishes them (honest nulls otherwise), split shares, 12-month YoY and 3-month momentum on total TEU, and a documented 0-100 gateway_momentum_index = round(100*(0.6*clip01(teu_yoy_pct/30) + 0.4*clip01((teu_mom3_pct+25)/50))) with neutral 0.5 fill for a missing term (null when both are missing). Per-row source_url + verification_level (official_stats_page / official_press_release / secondary_citing_authority) + preliminary_flag record the provenance. Join on (year_month, port_code / coast / state_code) to time purchase orders, promo calendars, and freight-cost forecasts against import momentum. Caveats: ports revise the latest 1-2 months; split detail varies by port; Halifax is quarterly.
Once today's free allowance is used up, AI features can run on your own provider account.
Kept in this browser tab only (cleared when you close it) and sent with each AI request. Our servers use it for that request and never store or log it.