Mix and match

Here’s another obscure wrinkle similar to that of over-zooming which can produce an unwarranted degree of over-confidence in the accuracy of a vector chart, but for which the software may give no warning.

Let’s say we digitize a paper chart which has a scale of 1:100,000, and thus a chart compiler’s plotting accuracy of +/- 0.2 x 100,000 = 20,000 mm = 20 meters. Once the chart is vectorized and goes into an electronic database, the data can be applied to an electronic chart at any scale. It might turn up in a chart that is effectively displayed at 1:20,000, which, in its paper version, would have a plotting accuracy of +/- 4 meters, but now we have incorporated data that was only originally plotted to +/- 20 meters (and this does not take account of any survey errors: we are simply talking plotting errors). In other words the electronic chart will give us a false sense of security regarding its accuracy.

High end equipment will filter out such things, but once again when it comes to the low end equipment and electronic charting systems often found in the recreational market, there is no such guarantee (and, in fact, these electronic charts are often compiled from data that has come from paper charts at different scales). All modern paper charts have a source diagram that shows the user not only the age and origins of the survey data used for that chart, but also the scale of the various surveys; high end electronic charts are required to have the same data (known as meta data), but low end products may not have it. Even if present, it will not generally be displayed on a vector chart – the user has to go looking for it – as opposed to a paper chart, were it is clearly printed on the chart (with a raster chart, it will be on the chart, but quite likely not on the part that is visible on the screen – it is necessary to scroll around to find it).