Night Landfall with a GPS/ECDIS Position Error
M/V Nordkapp Trader, a 24,000 dwt general cargo ship drawing 7.2 m even keel, is 14 NM off the Sunnmøre coast at 0300 local time, bound for Ålesund. Visibility is good, wind NW 15 kt, sea slight. The ECDIS is in route-monitoring mode. The planned track passes between a 6.5 m shoal to starboard and a steep-to islet to port; the safety contour is set to 10 m and the cross-track limit to 0.08 NM. The GPS position puts the ship 0.06 NM to port of the planned track. However, two visual bearings taken by the second officer to a lighthouse and a lighted radio mast, plotted on the ECDIS, place the ship about 0.15 NM to starboard of the GPS position, and the radar range to a charted headland is 0.12 NM less than the ECDIS indicates. No differential GPS correction is in use and HDOP is 1.6. It is one hour after high water and the predicted height is 1.9 m.
1. As the OOW, what do you do in the next few minutes, and how do you establish which position source is correct?
Read the worked answer
Immediate action. Reduce to manoeuvring speed (about 6 kt) and be ready to stop; the ship is 14 NM off a coast and the next waypoint is close to a 6.5 m shoal.
Establish the truth by independent cross-checks:
- Plot the two visual bearings on the paper chart and compare with the plotted ECDIS fix. Two bearings that agree with each other, plus a radar range that agrees with the bearings, are strong evidence that the GPS position is wrong.
- Check the ECDIS: datum setting (WGS 84 against a local chart datum), the GPS antenna offset entered, the radar overlay alignment, and whether the route was built on another datum.
- Check the GPS itself: HDOP 1.6 is usable, but without differential correction a 280 m error is entirely possible; suspect multipath or antenna shading by the superstructure. Compare a second receiver if fitted.
- Call the master, note the fix in the log and on the chart.
Until the sources agree, navigate by visual bearings, radar ranges and parallel indexing, not by the ECDIS position.
2. Assuming the visual fix is right, assess the effective cross-track situation and the adequacy of the ECDIS safety settings, and state what you would change.
Read the worked answer
The displayed cross-track error is misleading. GPS shows 0.06 NM to port of track; if the visual fix is right, the ship is about 0.09 NM to starboard of track (0.15 NM starboard of GPS, less the 0.06 NM offset). More importantly, the position uncertainty is at least ±0.15 NM, nearly twice the 0.08 NM XTD limit. An XTD alarm set inside the position error gives no protection at all.
What to change:
- XTD limit: base it on the navigable width available and on the accuracy of the position source - the distance to the nearest danger minus the expected position error. Use the ECDIS accuracy indication and the chart CATZOC.
- Safety contour and safety depth: with draft 7.2 m and a 20% UKC policy the minimum depth is 7.2 + 1.44 = 8.64 m. A 10 m safety contour is sensibly conservative, but set the safety depth near 9 m so that depths between contour and safety depth are highlighted, and confirm the 6.5 m shoal is actually enclosed by the contour at the chart's CATZOC.
- Move the route waypoint to the deep-water centre of the channel, add a no-go area over the shoal, and set parallel index lines on the islet and the headland.
3. Using the tidal information and the company's 20% UKC policy, decide whether the ship may pass the shoal on this route and tide, and give your plan.
Read the worked answer
Depth and UKC. Charted depth over the shoal 6.5 m + tide 1.9 m = 8.4 m. With draft 7.2 m, UKC would be 1.2 m. The 20% policy requires 0.20 × 7.2 = 1.44 m, so the ship is already 0.24 m short before allowing for squat (roughly 0.2–0.4 m at 8 kt in shallow water), swell or chart accuracy. The tide is one hour after HW and falling, so it will get worse, not better.
Decision. Do not take the planned track over or close to the shoal on this tide. Options, in order:
- Use the deeper channel to the west (charted 14 m), accepting extra distance and time.
- Hold off or anchor clear of the approach and wait for the next high water (about 11–12 hours); a long wait is still better than forcing 1.2 m UKC at night.
- If the pilot or port authority offers a surveyed deeper route with a stated tide, take it at reduced speed with continuous visual and radar fixing.
Call the master, record the calculation and the decision, and inform the port and VTS of the delay.