Here are a couple of thoughts:

1. Most National Hydrographic Offices, and hence the government behind them, make some sort of assurance or even guarantee about the quality of their charts.

2. Electronic (ENC) charts are generally considered to be an improvement over paper charts. It is easier to be more accurate in using them and updating them for example.

Putting these two assertions together you might be tempted to think that the governments can now make and even stronger statement as to the reliability of their chart data. However scratch under the surface a little bit and this notion can come unstuck.

There are some fundamental differences between paper and vector charts; here I am just going to focus on display options. Paper charts don’t really have any, vector charts have lots. With a traditional paper chart the choice as to exactly what goes onto the chart and how it is displayed is determined by a cartographer (chart compiler). So when the chart is published it can come with as assurance about the accuracy of the data and (this is the important point so pay attention) exactly what the chart looks like. Two skippers in completely different vessels with totally different equipment will be looking at exactly the same image.

With ENC charts, the data will be just as accurate but control over the image has been diluted. If we were to compare several different types of chart viewer using the same data, showing the same area at the same scale then chances are that they would show a different image. Sure they would all be similar but sometimes the devil really is in the detail. In fact (I can already hear the pedants) if you were to try and match the display settings on each of these units them you would still find that there were some differences between the images.

The starting point of this problem is that there is too much data to display. There are some 180 different classes of symbol which are arranged into groups. 20 of these groups must always be displayed and 90 are optional. Of the optional groups 51 are normally visible, switched on and the remaining 39 are usually switched off. Ok – so if you are worried about switching something off that you should really be displaying then you might consider simply switching everything on. Here is what it looks like:

On the left is the paper chart and on the right that nasty mess is all the ENC data. These two charts are intended to tell you the same thing so clearly this is sub-optimal. There is too much clutter on the vector display – something will have to go. In fact clutter is possibly the biggest problem with ENC data. So, you need to switch some stuff off – but what? Actually doesn’t it strike you as odd that right out of the box this chart is virtually unusable and so things have to be switched off? When are these things ever going to be switched on and why? How are you going to decide what should be on or off? This is the crunch; there is no easy way. I guess you could read a manual, learn about S52 viewing groups, brush up on cartography and give it a stab but that is actually quite a tall order. It is quite a lot to expect of your average mariner before they can use and electronic chart. It is also a moving target. You may well get the display looking just so but as soon as you start changing scales it can all go a bit pear shaped and if you use charts from a different producer then all bets are off. With paper charts you need to learn what they mean, with ENC you need to first decide what they should look like.

In all probability what you are really going to do is to take some reasonable default values. Fiddle with anything that you can understand and stop once you have a half decent display. This is a quite rational approach and this is one reason why all the ENC displays mentioned above are going to look different. There are many other reasons and these can rapidly get very technical but I hope you are getting the gist of this now. So the chart data may all be reliably accurate in line with the assurances of the Hydrographic Office but that is not going to help if the thing you just hit was not being displayed on the chart.

With Nuno we have taken the fairly pragmatic approach that the skipper is more likely to be interested in the chart as a navigation aid rather than a computer game. To this end we have attempted, to the best of our ten years’ experience in messing with ENC data, to make the chart display just work. There is a lot of clever stuff involving dynamic positioning of symbols, subtle re-scaling, jiggling labels around, changing fonts, adjusting for display scale, merging multiple cells and so on. The end result is a half decent chart display with very little messing around. It is all on the chart, well apart from one switch which we have termed anchoring mode. This enables additional data to be displayed in a way that would be appropriate if you were looking for somewhere to tie up. So this switches sounding, bottom type and some other stuff on. This is a feature in the new version of Nuno and it is coming very soon now.

Today’s blog is the first of a short series from Andy Watkin. When not writing code Andy is often to be found poring over charts. Lately he has been thinking about soundings…

Soundings

In the beginning, someone stuck an oar over the side of their boat and found that the water was deep enough at that single point for their rowing boat. Now we can measure the depth of a whole area simply by flying a small (suitably equipped) airplane over it and display all these depths on a screen.

As part of the development of the chart display for Nuno™ Navigator, I looked at the way soundings have evolved.

A single line of soundings indicates measurements made during a voyage – rather than a dedicated survey where the ship passed back and forth over the area. This was common in Captain Cook’s time and is still obvious on some current charts.

As each sounding was measured manually with a lead and line there are only a few and they fit easily onto the chart. A lead was a substantial weight: www.nmm.ac.uk. Great skill was required to throw it forward so that by the time it hit the bottom the line was vertical (thus allowing for the forward movement of the ship). At the depths above, they probably had to heave to each time.

The depths were marked on the line using bits of leather, bunting, etc, as shown in www.navyandmarine.org. Of course, on old sailing ships, the line had to be readable in the dark as well as in daylight – so each of the mark had a unique feel. Taking soundings from the deck of a heaving ship in the dark in a gale (with little idea where you really were) must have been a terrible job.

There is a whole language around the use of the lead and line, complicated by not marking all the depths.

· “By the mark ten” is 10 fathoms (60 feet).

· “By the deep eleven” is one fathom deeper than the 10 fathom mark. A fathom is of course approximately the distance between outstretched finger tips, which was presumably used to estimate the intervening lengths.

The lead line was not just used for measuring the depth. When the only positioning devices were a sextant and a sharp pair of eyes, finding out what the sea bed was made of could be a vital clue to your position. A hollow in the end of the lead was filled with tallow (which is sticky) and would bring up a bottom sample if you were lucky. This is why charts are marked with “mud”, “shingle”, “sand” etc.

Modern Soundings

Rather than measuring one depth at a time with a piece of string, a ship can use a side scan sonar to build an image of the sea bed for a significant width either side of the ship, all the while steaming along. The sensor can be built into the hull of the ship or towed in a ‘fish’ – a floating torpedo shaped housing. A fish can be controlled (it has underwater wings) to stay at a constant depth providing the ship’s speed is maintained constant. An advantage of using a fish is that, being entirely under water, it is not subject the pitch, roll and yaw experienced by a ship.

The sonar images can be remarkably detailed. www.abc.se contains some good examples.

The most amazing technique though is LiDAR - the word is a composite of light and radar. Two lasers are shone downwards and swept from side to side from an aircraft. One is a frequency (color) designed to be reflected from the sea surface. The other is a frequency designed to penetrate the sea and be reflected by the sea bed. Comparing the two, with a lot of computer processing allows the depth over the entire width of the scan to be determined – and the aircraft is moving forward at a rate so huge areas can be covered by repeating the process quickly enough. The depth that can be measured is limited by the sea opacity and does best in shallow areas; which is ok as that is where we generally care about most.