The Coast That Keeps Moving

There is a sandbank at the mouth of the Wester that has been legally inside two municipalities since 1994, and inside neither of them for eleven weeks in 2011. Nobody moved a boundary. The bank moved, and the boundary was written in a way that made the bank the boundary.

This is not a curiosity. It is what happens when a line drawn for one purpose is later asked to do another job. The 1958 chart that first fixed the estuary's edge was made for navigation: its job was to tell a pilot where a keel would touch. The line it drew was the low-water mark on an average spring tide, because that is the line that matters if you are in a boat. Forty years later the same line was used to decide who levies the harbour tax, and the tax followed the mud.

Four coastlines, one estuary

Ask four people where the coast is and you will get four honest answers. The hydrographer says: the lowest astronomical tide, because that is the worst case for a hull. The geomorphologist says: the seaward edge of the salt marsh, because that is where the system changes behaviour. The municipal surveyor says: the line in the 1994 annex, because that is what the ordinance cites. The person who lives on the dyke says: where the water was last February, and points.

Each of those definitions produces a different length. Measured at low astronomical tide the estuary's north shore runs to nineteen kilometres. Measured at the marsh edge, eleven. The difference is not measurement error. It is two different questions being asked of one landscape, and neither answer is wrong.

A coastline is not a fact you can measure more carefully. It is a decision you can make more explicitly.

What makes the Wester unusual is only that the disagreement has consequences you can read in a ledger. Between 1994 and 2011, the sandbank sat above the ordinance line and paid mooring fees to Nieuwdam. In the winter of 2011 a storm surge cut a channel across its neck and the bank, briefly, stopped being attached to anything. The ordinance had no clause for that. The fees went unpaid, and no one asked for them.

What the surveys actually record

We pulled six surveys of the same three kilometres, from 1958, 1971, 1986, 1994, 2008 and 2023, and redrew them to a single vertical datum. That last step is the tedious one. Two of the older sheets are referenced to a local tide gauge whose zero was itself re-levelled in 1979, which means the sheets are internally consistent and mutually incompatible. The correction is thirty-one centimetres. On a mudflat with a gradient of one in nine hundred, thirty-one centimetres is two hundred and eighty metres of horizontal disagreement.

Once the sheets are on the same datum, the picture is calmer than the argument. The bank has migrated east-north-east at an average of nine metres a year for sixty-five years, with two step changes — 1976 and 2011 — that both follow a named storm. The average is almost useless; the steps are the story. A landform that moves in jumps punishes any map that is revised on a schedule, because the schedule and the jumps do not agree.

This is the argument for surveying to an event rather than to a calendar. It is also, in practice, impossible: survey budgets are annual, and storms are not. The compromise most agencies land on is a fast, coarse post-event capture and a slow, precise scheduled one, and then the problem becomes reconciling two datasets of different quality that disagree about the same week.

The clause that would have helped

The 1994 annex could have written the boundary as a fixed line in coordinates, with the note that it approximates the low-water mark as surveyed in 1993. That is the boring option, and it is the right one. A fixed line is wrong in a knowable way: you can measure how far it has drifted from the thing it was meant to represent, and you can decide, in public, when to move it.

A line defined as "the low-water mark" is wrong in an unknowable way. It is always exactly correct and never locatable. Every party can point at a different sheet and be right. The 2011 gap in the ledger is not an administrative failure; it is the predictable output of a definition that outsourced its own position to the weather.

Since 2019 the province has run both: a fixed cadastral line for anything with money attached, and a monitored morphological line, updated after any surge above a threshold, for anything ecological. The two are published on the same sheet, in different weights, with the date of each in the margin. It is not elegant. It is the first version of this map that does not require a footnote to be honest.

Coda: the bank has a name now

In 2022 the bank was formally named — Wester Hoek — which is the administrative equivalent of admitting it is going to be around for a while. Naming a landform is a quiet claim that it is stable enough to be referred to. On the evidence of six surveys, that claim is worth about forty years. The next chart will tell us whether the name outlives the sand.