Introduction

Port congestion (ships waiting to dock, cargo piling up) is a leading economic indicator. Congestion predicts supply chain delays, higher shipping costs, and supply constraints. Satellite imagery enables counting ships queued at major ports, providing real-time visibility into port congestion. For traders, congestion data predicts commodity prices, shipping stocks, and manufacturing activity. This guide covers satellite-based port monitoring for trading signals.

Identifying Ships in Satellite Imagery

Modern satellites (like Maxar's) have 30cm resolution enabling ship detection. Analysts or automated ship detection algorithms count vessels anchored or moving near ports. Historical satellite archives enable tracking congestion trends over time.

Automated detection: train object detection models (YOLO, Faster R-CNN) on labeled satellite imagery to identify ships. Models must handle various ship sizes, orientations, and water conditions (calm water makes ships clear, rough water blurs detection). Accuracy: good models achieve 85-95% ship detection rate.

Queue Length Metrics

Simple metric: count ships within X miles of port anchoring zone. Higher count indicates congestion. More sophisticated: estimate wait time by modeling port capacity and arrival rate. If port can handle 5 ships/day and 50 ships are waiting, expected wait is ~10 days.

Temporal tracking: compare current queue length to historical average. Anomalously high queue lengths indicate congestion. Trending: is congestion increasing or clearing? These trends are tradeable signals.

Trading Applications

Congestion at container ports predicts shipping costs (more ships waiting = higher costs for new arrivals). Shipping stocks (FedEx, shipping companies) benefit from congestion (higher rates). Importers/exporters suffer (higher costs, delays).

Commodity implications: congestion at ore ports predicts ore delivery delays, potentially supporting ore prices. Congestion at grain ports predicts grain shipment delays, supporting grain prices (limited supply reaching market).

Data Collection and Processing

Satellite providers (Maxar, Planet Labs) offer port monitoring services. Curated datasets provide daily or weekly port queue counts for major ports (Shanghai, Rotterdam, Singapore, Los Angeles). Alternatively, order satellite imagery and perform manual/automated ship counting.

Validation: Comparing to Official Data

Port authorities publish official vessel statistics. Compare satellite-based ship counts to official numbers for validation. Discrepancies reveal: 1) satellite detection errors, 2) differences in measurement definitions (anchored vs arriving vs docked), 3) actual port congestion not captured in official stats.

Practical Challenges

Challenge 1: Cloud Cover. Rainy weather obscures satellite views. Tropical ports have low satellite visibility. Recent developments (SAR radar, which works through clouds) help but are more expensive and harder to interpret.

Challenge 2: Interpretation Difficulty. Not all ships near ports are waiting for berth. Some are anchored for maintenance, refueling, or crew changes. Distinguishing waiting from other reasons requires analysis of ship behavior (movement patterns, port schedules).

Challenge 3: Information Already Priced. Shipping companies and large traders monitor port conditions directly. Satellite data confirms what's already known. Advantage exists only if your satellite analysis is faster or more accurate than other sources.

Advanced Analysis: Ship Type and Cargo Class

Different ship types (container, bulk carrier, tanker) serve different markets. Container ship congestion affects import/export timelines. Bulk carrier congestion affects commodity markets. Detailed ship classification enables targeted trading signals.

Integrating with Other Supply Chain Signals

Combine port congestion data with other signals: AIS ship tracking (where are ships going?), BoL data (what cargo is being shipped?), port rate data (dock fees, congestion surcharges). Triangulation across sources produces robust signals.

Conclusion

Satellite imagery enables quantifying port congestion (ship queue lengths) in real time, providing supply chain visibility. Congestion predicts shipping costs, commodity prices, and manufacturing delays. Automated ship detection from satellite imagery works well (85-95% accuracy) but requires careful validation and interpretation. Most valuable when combined with other supply chain and economic indicators. For traders focused on supply chain disruptions and shipping/commodity markets, satellite-based port monitoring provides meaningful edge.