Extent, concentration, forecast, chart: four different questions
Most disagreements about whether a route is open come from comparing products that were never meant to answer the same question.
Four kinds of ice information reach a bridge, and they are routinely treated as interchangeable. They are not.
Extent is the total area where ice concentration is at least 15 per cent. It is a climate-scale metric, published daily, ideal for comparing this August with the last forty. It says nothing about what is inside that area.
Concentration charts from national ice services describe how much of a given area is covered, usually with stage of development and floe size alongside. This is the level at which route decisions are properly made. The charts are analyst products: an ice service combines satellite passes, ship reports and aircraft observation into an assessment, with a date and an issuing authority attached.
Forecasts project that picture forward. Their skill falls off with lead time and varies by region and season — a five-day drift forecast in a stable pressure pattern is a different animal from the same forecast during a rapid synoptic change.
Operational imagery, principally radar, shows the ice as it was at the moment of the pass. It is the only product at a resolution useful for choosing a track through a specific field, and it is a snapshot: ice drifts, and a twelve-hour-old image of a lead is a historical document.
Operational perspective
The practical rule is to match the product to the decision. Whether to plan a voyage at all: extent and seasonal outlook. Which route and which window: ice service charts and forecasts. Which track, this watch: operational imagery, and eyes.
Errors follow from mismatching. A voyage plan built on extent anomalies will describe a season that looks open and put the ship into a field the plan never anticipated. A route argument settled by pointing at a satellite image from three days ago is settling it with data that has since drifted, possibly tens of nautical miles. The most consequential version is planning to a forecast at a lead time where the forecast has no useful skill, and treating the resulting track as though it were surveyed.
Two habits are worth carrying. First, every ice product should be quoted with its age and its source — an assessment that does not state when the information was valid cannot be evaluated by the person acting on it. Second, the plan should say what happens when the picture turns out to be wrong, because at some point in an ice-affected passage it will be. A passage plan without a stated abort point and a place to wait is not finished, however precise its waypoints look.