In our experience of practical navigational assessments conducted at sea, there appears to be a widespread misunderstanding in the marine industry between the meaning of position fixing and position verification. There has been some discussion on this subject in the Nautical Institute section of “Linked-in”, but this is not necessarily seen by the wider maritime audience.
This confusion probably derives from the way the term ‘position fixing’ is applied to navigating on paper charts and its subsequent uncontrolled evolution into the ECDIS environment. Whenever new technology is introduced, those implementing and using the system tend to bend old terminology – as best they can – to fit the new system: as with all evolutionary processes, divergent interpretations inevitably occur, and it takes time before a unified interpretation is reached by the industry. Academic organisations may take the lead with well-conceived usage, but their fine distinctions and subtle nuances are not always explained with reasoned validation, nor are they necessarily widely publicised or appreciated by the wider maritime community. In this case, the confusion has unfortunately been perpetuated by extensive careless and interchangeable use of these terms by marine professionals at all levels within the industry, including those charged with making ship inspections and audits.
In the days of paper charts, the system was simple: a position fix comprised at least three (mutually independent) position lines, the third and any subsequent position lines being the verification of the position.
With the coming of ECDIS, it is inherent that the position fixing function must now be ‘real-time’ and can only practically be done by automated electronic means such as GNSS. The traditional concept of “position fixing frequency” or interval is therefore not relevant to navigation by ECDIS, as the position is continually fixed, and recorded in the ECDIS history.
By contrast, the position checking – verification – role very much remains with the watchkeeping officer; but it now refers to the periodic checking of the GNSS position by other independent means, e.g. from terrestrial or celestial observations. The question then arises of what frequency this verification should to be conducted?
This is more complex than the traditional concept of position fix frequency on paper charts, which is based solely on proximity to hazard. Furthermore, position verification frequency criteria on ECDIS needs to take the following into account:-
- Provision and functionality of the Radar Overlay on the ECDIS
- Functionality and performance of track monitoring facilities on Radar (i.e. parallel indexing)
- Accessibility and functionality of the manual position plotting facilities on ECDIS
- Skill sets and fluency of the watchkeeping officer in terrestrial fixing and track monitoring techniques
- Capability of the ECDIS equipment to automatically detect and alert the watchkeeping officer to potential degradation of position quality from input sensors.
The first two items are self-explanatory. Radar overlay (the superposition of the entire radar screen image onto the ENC map, allowing real time direct positional comparison of the two pictures) will show instantaneously whether the electronically generated position is correct, or the extent of any error. It also reveals any radar alignment error. If the radar overlay shows no significant error in the GNSS position, the operator can simply drop a single-push (GNSS) “Manual Fix” to record the fact that he has verified the position at that time, and this will be recorded in the ECDIS history log. This becomes a position verification, and it is important to realise that this is just as valid as (and indeed probably superior to) a fix obtained from an ECDIS manual terrestrial plot.
Position verification from the radar overlay is sometimes called “the thousand-point fix” – and not without justification. The importance of having radar overlay available cannot be overstated; notwithstanding, we still see vessels where the facility is not provided or available – often simply not connected.
The proper use of Parallel Indexes and clearing ranges (on radar) remains an excellent and simple real-time check on the GNSS position. Although this is only a single position line, it lies in the most important positional axis – the ship’s fore and aft line – and directly gives us cross track error, and the information necessary to correct it. Setting up the parallel index for the next course gives all the information needed to turn onto that course at the correct time and with precision, and this process can be repeated, to enable effective safe-water navigation in the event of electronic position fixing system failure. It should be remembered that a vessel can theoretically be navigated in safe water with precision from port to port, using nothing but a succession of parallel indexes, and without further reference to the navigational chart.
Unfortunately, the facilities provided in most ECDIS systems for plotting terrestrial position lines are often awkward to use, time consuming and counter-intuitive. It would appear that they were given a very low priority at the initial concept and software design stage, and this has not improved with time. The seamless integration of electronic position fixing and electronic charts in ECDIS inevitably implies a high degree of reliance on its electronic position fixing sources and there appears to be little concession by system architects to traditional methodologies or the constant cross-checking which was so necessary to resolve the doubt and uncertainties inherent in traditional position fixing.
In times gone by, Deck Officers were highly practiced and experienced in taking and plotting terrestrial positions, as this was the only way to fix the ship’s position on the coast. An officer might plot as many as 50-60 individual bearings and/or distances in the course of a single coastal watch. In the process, officers inevitably became both fast and accurate. Equipment had become refined and improved, culminating in the introduction of the open-horizon (sightless) prismatic azimuth ring. A multiple range and/or distance fix would therefore typically be taken in a few seconds and plotted in a few more – the whole operation typically taking less than a minute. This was only possible because of the fluency of the officers with prismatic azimuth rings and their manual plotting skills. Even with a required fix position frequency of 5 minutes, it was still possible to maintain an adequate lookout and attend to other bridge watch duties.
In the last 30 years, we have witnessed the prismatic azimuth rings mostly being replaced with fore and back-sight devices, of much older and inferior design. This precludes even the most experienced officer taking multiple bearings at speed and undoubtedly explains the modern officer’s tendency to use radar bearings in preference, despite their acknowledged poorer accuracy. These elements, combined with the awkwardness of the manual fixing facilities provided in ECDIS means that a fix which hitherto could have been plotted with precision on paper within a minute would now take around five minutes or more to plot on ECDIS, and is probably much less accurate.
From the above it should be clear that it is neither necessary nor is it desirable to require officers to plot frequent manual fixes on ECDIS. The equipment has (or should have) the capability of real-time instantaneous position verification of the electronic position by the use of the radar overlay, and radar parallel indexing provides additional real-time validation. The manual position plot is both poorer quality and can be dangerously time-consuming.
As the vessel moves into confined waters, officers should therefore be concentrating more on position verification by the former real-time methods and spending less time on manual position plots. Unfortunately, some Company bridge procedures – apparently based on the traditional paper chart position fixing criteria – actually require the opposite. This is contrary to several of the essential concepts of ECDIS – that of simplifying navigation, reducing the probability of error and reducing the workload on the watchkeeping officer: more importantly, it is potentially dangerous, particularly where it distracts the navigating officer from other essential watchkeeping duties.
But what of the vessel that is not fitted with radar overlay? Here, the only real time method of position verification available is parallel indexing (plus the intelligent use of leading lines in ports and harbour areas, at the planning stage). In this case, the manual position plot becomes more important, and in the absence of radar overlay has more value. However, the shortcomings of the process are still obviously just as relevant, and so the emphasis shifts to making the best of it, in terms of:
- Improving speed of application of plots, and
- Minimising the scope and consequence of operator errors
From traditional position line error theory, the fix taken when passing exactly abeam of a salient point is hard to beat. Rate of change of distance is nil, and so it is easy to take time and care to obtain the measurement exactly from the radar. Crucially, it also directly gives cross track distance, in harmony with the parallel index. As the closest point of approach, it is also generally the most accurate distance we will ever obtain from that point. Equally, while the rate of change of bearing is at its maximum, the distance is at its closest, making the resulting position the least sensitive to any bearing error, the consequences of which will thus equate to little more than a small error in the fix time. The abeam fix can, in practice, be predicted and set up on the radar (and ECDIS) shortly before it happens. It is then little more than noting the time when it occurs, and marking the ECDIS accordingly. Coincidentally, the most accurate terrestrial manual verification also happens to be the easiest and quickest to plot – and least prone to error.
There is also a tendency for the more significant navigational hazards to lie off salient points, and for them to be provided with visually and radar conspicuous navigation aids to aid positive identification. This, together with the intelligent use of parallel indexes, means that we are often verifying position at the most appropriate moment to guarantee the planned clearance.
The above suggests that rather than using arbitrary time intervals to specify the frequency of manual plotting, we should plan to use it in a way which is geared to passing salient points or other hazards safely.
Precisely the same principles will apply in the event of a GNSS shut-down or interference; however, in this case, the above techniques are “promoted” to become the fix rather than the verification method. In this case, we will need to increase the frequency of manual terrestrial plotting, and this may well require additional manpower on the bridge to cover the task. Dead reckoning (based on gyro course and log speed) will replace the GNSS feed, and a task of the watchkeeping officer will then be to periodically adjust the DR position to the position obtained by use of the radar overlay or the manual plot.
This is why it is so important that officers maintain and practice their skill with using the manual position plot facility on ECDIS. It could even be argued that the required frequency to exercise manual plot verification of the GNSS fix on ECDIS is more related to the need to maintain the officers’ fluency than to the ship’s navigational situation, as meeting the former will be more than adequate to cover the latter.
Only when we have a clear idea of the distinction between position fixing and position verification on ECDIS can we develop sensible guidance on the required frequency for position verification, and this should distinguish between the important real-time methods such as radar overlay / parallel indexing and the historic terrestrial manual plotting. It is the former which needs to increase in frequency as the vessel approaches hazard, not the latter. Only in the case of a vessel not fitted with operational radar overlay will the vessel need to increase the frequency of manual plotting verification, and this should be carefully targeted to optimise the balance between the value of the information obtained to the watchkeeping officer and time taken to obtain it: arbitrary time intervals – and especially short intervals – are almost certainly not going meet that balance.
Since the introduction of ECDIS, we have seen ship operators have divided approximately equally into two camps: there are those have reduced the required frequency of manual (terrestrial) ECDIS position plots to once an hour or even once a watch on the coast; this is combined with continuous track monitoring by parallel index and frequent position verification by the use of radar overlay. We consider this good practice and are not aware of any negative port state/vetting observations being raised where this system is in operation, and properly documented.
Other operators have retained manual terrestrial position plotting criteria on ECDIS similar or identical to those specified for fixing on paper charts; in some cases, this has included the requirement to plot manual terrestrial fixes at 5-minute intervals, in confined waters. In some cases, this a no more than a fallacious implementation of an assumed requirement of external auditing bodies. In most of these cases, the distinction between position fixing and verification has not been made, and in some cases, no reference has been made to requirements for the use of radar overlay to verify the ships position. It is these latter cases which gives us cause for concern, as these watchkeeping officers are spending an inordinate amount of time on such “position fixing” – generally of negligible value – to the detriment of their other watchkeeping duties, and contrary to the whole ethos of ECDIS.
In conclusion, we therefore consider that industry best practice should be to use real-time and frequent position verification, such as radar parallel indexing, transit headings and radar overlay to ensure that the primary electronic system (GNSS feed) remains correct.
In order to meet external audit requirements, it is necessary for officers to record their verifications. The easiest way to do this is to check for satisfactory alignment between the radar overlay and the ECDIS chart and record this by manually recording or ‘dropping’ the current GNSS position, and labelling it accordingly. Passing abeam of salient point whilst on the parallel index can be similarly marked by ‘dropping’ the GNSS fix and labelling it in the same way. It should be understood that this use of radar overlay and radar parallel indexing and a coincident GNSS position dropped from them meets the requirements for checking and monitoring the position as set out in industry publications. Of course, if the positions are not considered sufficiently coincident, it will be necessary to carry out manual plots by other means, and record these as such.