Coal trading — global logistics and thermal/metallurgical flows

Coal is the world's most politically awkward commodity — declining in the West, growing in Asia, increasingly routed around sanctions and climate policy. The physical trade persists at ~1.3 billion tonnes per year, moving on the same bulk carrier fleet as iron ore and grain.

NTHMAP tracks that fleet.

The global coal trade

Major flows:

• Australia → Japan / Korea / China — thermal + metallurgical
• Indonesia → India / China / SE Asia — thermal
• Russia → China / India — post-2022 shift from Europe
• Colombia → Europe / Asia — shrinking
• South Africa → India — thermal
• US → Europe — residual exports from Appalachian producers

Each of these routes uses Panamax and Capesize bulkers which NTHMAP tracks under the Bulk Carrier type.

Newcastle benchmark

Newcastle, Australia is the largest coal export port in the world by throughput. The Newcastle forward curve is the global thermal coal benchmark. Watching Newcastle loadings in real time gives you a direct signal into the physical market.

# Bulk carriers at Newcastle
nthmap vessels list --bbox 151.7,-33.0,152.0,-32.8 --types "Bulk Carrier"

# Waiting at anchorage (queue)
nthmap vessels list --bbox 151.7,-33.0,152.0,-32.8 --types "Bulk Carrier" --max-speed 1

When the anchorage fills up, it's either labor action, weather, or a rail loading issue at the mine. Each has different trade implications.

Russian coal rerouting

Post-2022, Russian thermal coal that previously went to Europe was redirected to China and India via:

• Vostochny / Vanino (Russian Far East ports)
• Black Sea ports
• Indirectly via Turkey and UAE

NTHMAP tracks all of these:

# Vostochny area (Russian Far East)
nthmap vessels list --bbox 132.6,42.6,133.0,42.9 --types "Bulk Carrier" --min-load 80

# Vanino
nthmap vessels list --bbox 140.0,48.9,140.3,49.2 --types "Bulk Carrier" --min-load 80

Aggregated by destination, you see the rerouted flow in real time.

Indonesian coal exports

Indonesia is the largest thermal coal exporter by volume. Most moves through:

• East Kalimantan — Bontang, Samarinda, Banjarmasin
• South Kalimantan — Banjarmasin terminal
• Sumatra — Tarahan

# East Kalimantan coal corridor
nthmap vessels list --bbox 116,-3,119,2 --types "Bulk Carrier" --min-load 75

Sanctions considerations

Russian thermal coal is subject to the G7 / EU ban. Tracking the physical shift to non-sanctioned buyers is a compliance concern for some traders and a market signal for others.

NTHMAP doesn't do sanctions screening, but it shows you the physical pattern. A bulker that left Vostochny and arrived in Mumbai is a factual observation. Whether the trade is sanctioned is a separate legal question.

Simple coal trade monitor

import nthmap
from collections import defaultdict

client = nthmap.Client(api_key="ntm_live_...")

EXPORT_PORTS = {
    "Newcastle (AU)":   [151.7, -33.0, 152.0, -32.8],
    "Hay Point (AU)":   [149.2, -21.3, 149.3, -21.2],
    "Dalrymple Bay":    [149.3, -21.3, 149.4, -21.2],
    "Balikpapan (ID)":  [116.8, -1.3, 117.0, -1.1],
    "Richards Bay (ZA)":[32.0, -28.8, 32.1, -28.7],
    "Vostochny (RU)":   [132.6, 42.6, 132.9, 42.9],
    "Vanino (RU)":      [140.0, 48.9, 140.3, 49.2],
}

for name, bbox in EXPORT_PORTS.items():
    loaded = client.vessels(
        bbox=bbox, types=["Bulk Carrier"], min_load=70
    )
    total_mt = sum(v["est_cargo_mt"] or 0 for v in loaded)
    print(f"{name:22} | {len(loaded):3} loaded bulkers | {total_mt:,.0f} MT")

Run daily, chart the series, compare to the Newcastle forward curve.

Metallurgical vs thermal

NTHMAP doesn't distinguish coking coal from thermal coal at the vessel level — both move on bulk carriers. Distinguishing them requires destination logic: vessels heading to steel mill ports (Incheon, Kashima, Vizag, Paradip) are likely carrying coking coal; vessels heading to power plant regions are more likely thermal.

Getting started

Pick one coal corridor — Newcastle → Japan is a good start — and build a daily monitor. You'll develop intuition for the trade's rhythm and start catching disruptions before they hit the forward curve.

See the CLI docs.