Blog

None of which matters very much because most GPS units do a pretty good job, most of the time, and provided we understand and accept that errors or inaccuracies can occur, and don’t put absolute faith in what they tell us, then they are an incredibly useful tool.

If you want a little more explanation, please do read on. 

Horizontal error

A route (or track) is recorded by your GPS unit as a series of data points. Each an individual record of a digital exercise of triangulation between your unit and any available number of satellites. The greater the number of satellites, the more accurate that exercise of triangulation is likely to be (at least five satellites are required for an acceptable level of accuracy in most scenarios). This is the first variable to consider. Satellites are moving in orbit, so a combination of your uninterrupted view of the sky and the length of time you are at a given location, combine to increase the likelihood of an accurate position record, or perhaps more accurately I should say, the degree of accuracy of that record. When we are walking, we are moving continuously and so the degree of accuracy of the unit’s measurement will inevitably vary between data points. The designers of your GPS unit are of course aware of the accuracy issues or varying levels of success in triangulation with satellites, so they design clever algorithms embedded in your device that ask a series of questions about the likely coherence and consistency of recorded data points. These work pretty well, but algorithms make assumptions so small variation can creep in.

Uninterrupted view of the sky is the issue we tend to be most aware of. If you are deep in tree cover, among tall buildings or in deep gullies or canyons, then the ability of the unit to establish reliable communication with sufficient satellites is compromised.

Then there is the issue of standing still. Wait, ‘that’s a good thing surely’ I hear you say, it will increase accuracy by enabling a more accurate satellite fix to be obtained. In theory yes, but in practice no. That is because the data sampling frequency is temporal, so it records every so many seconds in time, not by how much distance you move from the previous recorded point. As a result, a static GPS unit (let’s say, every time you pause at a junction to consider the view or your direction of travel) will take multiple triangulation measurements. Because the accuracy of those measurements will vary, many GPS units will show that pause as a squiggly line that jumps back and forth as though you are in fact moving around continually in a relatively small area. You know you are not, but the GPS unit cannot know that. In practice for the user trying to record and share a ‘sensible’ route line this can be frustrating and result in the time-consuming need to manually delete dozens, if not hundreds, of data points manually. That doesn’t happen in my unit/phone/watch you might be thinking, but I can assure you it does. The reason you may not see it in your recorded routes is because once again, clever people that make GPS units now design algorithms that seek to detect and correct these ‘squiggles’ in a route line without you having to manually intervene. While in most cases these do a pretty good job, there can be small inaccuracies creeping in because an algorithm has to make assumptions and cannot know your exact movements.

Then we have the sampling period. To create a ‘route’ the unit records a series of data points and then creates straight lines to join up the dots, so the frequency of recording impacts how close the route line created is to the line you actually walked. So, let’s have the most frequent recording possible and lots of data points to get the best line right? Wrong. Great in theory, but you end up with a massive data file and in practice many software platforms that you might upload that to, for example to view it, edit it, or share it, will reduce the number of waypoints in order to make handling the file more resource efficient, and to make you less likely to think that their software is unbearably slow! The effect that this reduction in data points has we tend to refer to as ‘smoothing’ and it’s not dissimilar to having recorded less data points in the first place. 

A route line is expressed as a straight line between corresponding data points, so the result of reduced data points tends to look rather like cutting across corners. Of course, in a relatively straight line a reduction in data points has a very limited effect on the overall distance or impression of the route, but the sharper a turn the route makes the more likely it is that reduced data points will be expressed as a rounding of corners. The sharper corners the route takes, the greater the overall discrepancy will be between the original recorded distance and the ‘new’ displayed route in some other software package or GPS unit. In my experience this discrepancy can show up even between units and software from the same brand. 

Of course, it is essential to be familiar with the set up menu of your chosen device too, as the frequency of data point recording is usually a user setting, so if you are setting out to compare two units, be sure to check that they are set up comparably.

Vertical error

So far, we have talked about route recording in two dimensions, the equivalent of working on a flat map as it were. But we all appreciate that the ground is rarely ever truly flat and in many cases our route will take us over undulations, hills and even mountains. So how is this measured and how do discrepancies happen in the vertical dimension. It’s useful to think about a mountain or hill as though it were effectively a triangle when looked at from the side. We all understand that when we walk up one side and down the other, we are walking two sides of a triangle and that the ‘straight-line’ distance along those two sides is considerably greater than the distance across the short side of the triangle (effectively a straight line through the mountain from the base on one side to the base the other side). The difference (or extra distance) between the direct measurement across and the longer measurement up and back down, we want to measure as ascent and descent, ensuring that our recorded distance measurement reflects the two-dimensional line between two points that will match the same as that measured on a flat map.  How does the GPS unit do that?

Essentially the primary method for GPS elevation calculations is the same as for horizontal positioning, it is the triangulation of your position with two or more satellites. So, it would be easy to assume that the error risks would be the same, but that would be wrong. Satellites themselves orbit on a horizontal plane, and we are now asking them to calculate change on a vertical plane, something inherently harder to do accurately. GPS triangulation alone is generally considered to be accurate to within 50m – that’s a pretty wide margin of error! If you then consider the effect of reduced satellite coverage in trees, or narrow valleys, canyons and so on, the likelihood of error is even higher. 

For this reason, GPS providers don’t tend to rely entirely on triangulation to establish elevation, or subsequent changes in elevation. They supplement the recorded measurements with known spot heights taken from topographic surveys using orbiting radar or lidar. These costly and complex methods produce a much more accurate picture of the elevation of specific points on the earth (where they can see all the way to the earths surface), and these are combined into a digital elevation model (DEM). So, wherever your GPS unit can, it will ‘correct’ any triangulation data for elevation with pre-established DEM data that is known to be more accurate. Great right? Except where a correction is made there is a fictional gain or loss relative to the previous triangulated recorded elevation. So, while the spot height may be more accurate, the accuracy of the elevation change between points may well decrease as a result.

Finally, many units will come equipped with the ability to measure barometric (air pressure) change and will combine this with other data to factor into calculations. If the weather is stable and does not change, then barometric pressure change between altitude points can be quite an accurate way of measuring elevation change. However, this too can be subject to error as the unit is typically cross-checking between available DEM data and triangulation data simultaneously and adjusting where it finds discrepancy. In the process it is always trying to establish whether you have actually changed elevation, or whether perhaps the barometric pressure has changed.

Weather-related error

There are a lot of myths around the impact of weather on the accuracy of normal consumer-based GPS equipment. Variations in humidity, temperature, and air pressure cause slight, but measurable, delays in the GPS signal’s travel time. Since GPS calculates distance based on signal travel time, these delays introduce very minor positioning errors. 

It is not uncommon to hear the suggestion that rain, fog, and clouds reduce GPS accuracy. However, as these are vapor (effectively a gas), signals pass through relatively ok. Only when rain becomes quite heavy can it affect the signal.

Where weather does cause problems, it is usually because of what is known as multipath interference (or refraction). In adverse weather, signals are more likely to reflect off surfaces like wet ground, buildings, or trees before reaching the receiver. This “multipath interference” causes timing discrepancies and reduces accuracy. I’ve most commonly experienced this when recording routes in extended sections of wet forests.

Equipment-related error

Provided our GPS devices are working correctly and are not faulty, then we tend to discount our equipment as being the source of potential inaccuracy. Afterall, accurate recording is what the device was designed for wasn’t it? Not always!

When we are using a dedicated GPS unit whose sole purpose is position and route measurement then you would expect it to be designed to be accurate. However, like anything else, there is a trade-off between the cost and quality of the parts and overall design of the unit. Cheaper units simply won’t have the same quality of components, and this can have an, admittedly small, effect on accuracy. This seemingly small level of inaccuracy increases when you consider devices that are not designed solely as GPS units. The most common example of this is the now ubiquitous use of mobile phone handsets as GPS measuring devices. To create a device that takes on the slim form factor that we desire in a mobile phone requires compromises in the design of some components.

In a mobile phone the GPS antenna is smaller than in most dedicated units and is internal not external. Then there is the issue of interference from other antennae in the unit for different purposes. Most mobile phones will include a Bluetooth antenna, a multi-band wifi antenna and one or more cellular phone antennae. Each of these has the potential to interfere with your GPS antenna to some degree, and consequently each are usually positioned as far apart as possible in the (already small) phone casing. These compromises of course increase further in even smaller units such as when compressing GPS equipment into your wristwatch.

Older and/or cheaper mobile phones will also only have a single-band GPS antenna, typically on the L1 frequency. While all satellites broadcast on L1, this frequency is more prone to interference, jamming and atmospheric issues. The trade-off is down to cost, power consumption and device design constraints. A significant issue is also the reliance on what is known as ‘assisted-GPS’. These single band antennae rely on assistance from cellular network signals to get an initial ‘fix’ and then add accuracy through satellite triangulation. This can work ok, provided you have a cellular signal. Once you lose signal from a cellular network, then the GPS antenna can take much longer to get even a basic ‘fix’, and if you are moving the accuracy is consistently impaired.

Newer high-end mobile phones began to use multi-band GNSS receivers from around 2019. These systems receive both L1 and L5 bands simultaneously and this greatly improves accuracy in most scenarios. Lower end phones continue to use single band receivers however, so if you intend to use your phone for navigation then it is worth checking which system your phone is on.

Another ‘myth’ you may hear around GPS equipment accuracy is that the government deliberately limit signal accuracy to civilian devices for security reasons. This used to be partially true but is no longer the case. In the past the US government did deliberately limit GPS accuracy to around 100m (330ft) except to military devices. This policy was abandoned in May 2000, making civilian devices as accurate as military ones. There are no other known governmental restrictions on GPS accuracy elsewhere in the world (although this would likely change in the event of a major war). 

Bearing in mind all the factors described above that indicate a potential difference in the level of accuracy between GPS devices, in general it is accepted that consumer grade GPS units have the potential to be accurate to somewhere between 5m (16ft) and 15m (49ft) in ideal conditions. If you are using a GPS to seek out, or record, a specific waypoint or feature then that level of inaccuracy is generally perfectly acceptable because from that point your target is usually close enough to be visible. However, if you bear in mind that for the purposes of recording a route, each data point has the potential to be inaccurate by up to 15m (49ft) even in ideal conditions, then it quickly becomes apparent that over a long distance the inaccuracy can increase to quite significant levels. There are commercial grade GPS units that are inherently more accurate, but none of us would want to carry one, or bear the cost of buying one for that matter.

Why none of this matters!

Clearly there is a lot that can go wrong with what otherwise seems like ‘simply measuring a route’ using an apparently accurate modern GPS unit. In just my high-level summary of the issues I count eleven different broad sources of potential error leading to inaccuracy. Consequently, it is not uncommon to follow a GPX track that someone else recorded and find that your own unit records the route differently, but it is important to put this into context. 

The differences that can occur are usually relatively minor. They will occur most on routes with a lot of ascent and descent, and those with a lot of sharp turns in the route line. The greater the distance of the recorded route, the greater the likelihood of a level of inaccuracy – even in otherwise ideal conditions. 

If we understand and accept that it is almost impossible to ensure complete accuracy in recorded GPS routes, then we can use these tools in a much more informed way. Modern units that contain advanced algorithms do an incredibly good job most of the time. Knowing why a route may not look quite right, or completely match the ground, lets us use them without blindly following them, and encourages us to question when something seems amiss. Ultimately this makes us better navigators, capable of using the range of tools and observations at our disposal to make informed decisions. This is why it doesn’t matter – provided that we are aware of it.

However, if you were viewing tent ‘reviews’ on YouTube with a view to purchasing a tent capable of surviving a night in the mountains, then you might easily find yourself spending more money than necessary by being led to believe that you need a tent that will stand up to gale force winds on an exposed summit. Many of my early tents were definitely somewhat less than capable of such a feat, and yet I’ve enjoyed many a happy mountain night under nylon, even in bad weather. 

Site selection in these reviews seems to be much neglected topic in favour of kit comparison in the competition to find (and buy) the ‘best’ tent for wild camping in the worst conditions possible. For most people it just isn’t necessary. However, in claiming to have something to offer in terms of advice, I should also admit that while I have learned over the years from the company of many older and wiser (at the time) than me, I have also learned through experience (which is the more stylish way of saying, I’ve made a ton of foolish errors and had some cold, wet and uncomfortable nights too!) Indeed, I would encourage learning from mistakes as the ‘premier’ way to learn. Not only do the lessons tend to be remembered more effectively, but in the process, you can accumulate a significant amount of bragging rights for late night tales with like-minded folk, a favourite pastime in Scottish bothies.

There is another good reason to highlight this issue however – not everyone can afford the top kit, and no one should be denied the opportunity to experience the outdoors in all the glorious weather challenges that are typical in the UK, just because the prices of the best outdoor kit are now in the same league as Rolex watches and designer handbags (they are often seen in the same bars too, thanks to Gorpcore!)  

Wild camping offers a rare sense of freedom and connection with nature, but you don’t have to sell the kids to be able to afford the experience. Where you choose to pitch your tent can make the difference between a great night outdoors in any adequate tent, vs a sleepless night in a very expensive tent – and possibly a damaged tent (that I hope you didn’t put on the credit card with the intent of returning it after the review).

Like with many things, lots of the factors to consider when choosing a great tent spot (and one that shouldn’t expose your tent to the full force of a mountain storm), are not obvious until you know them. So, in the hope that its useful to someone, I’ve set out my top ten things to consider for a good wild camp spot.

It is important to be aware that wild camping is not legal in most places but is generally tolerated if you leave no trace. Ideally you would pitch somewhere discreet just before dark and be gone as soon as it is light. That way no one should ever know you were there. 

Always check the weather forecast regularly too. Knowing what to expect can be especially helpful in considering some of the points below.

1. Avoid Summits, Ridges and Tops

While tempting for the views, summits, ridges and open tops are particularly exposed to strong winds. When horizontal wind meets a slope, it accelerates as the air is forced into a narrower channel. This is the ‘ridge lift’ used by paragliders to take off. Don’t expose your tent to the risk of high winds if you want it to last. Drop below the ridge or top to find calmer, more stable conditions, on the lee side of the slope.

2. Prioritise Wind Shelter

Protection from the wind can also be found using the nature of the terrain such as shallow bowls, depressions, rock outcrops, or other such land features. However, be aware that wind turbulence can form behind large or steep crags, so these are best avoided. Further down a slope a stand of trees can slow and dissipate the wind, a dry-stone wall or even a sheep fold wall can be useful too. Pitch with the narrowest end of your tent (usually the rear) to the wind.

3. Check Drainage Carefully

Mountain weather changes fast and you should always assume overnight rain could happen. Avoid gullies and meltwater channels, that can turn into streams overnight. Will any water pool or drain in that shallow depression? Slightly raised, compacted ground drains far better. Watch out for a reddish coloured coarse grass – it tends to grow where water pools.

4. Avoid Cold-Air Sinks

Cold air sinks at night in mountains when the ground cools rapidly, particularly under clear, calm skies, chilling the air above it through conduction. This cold, dense air becomes heavier than the warmer air further up the slope, causing it to flow downhill (known as a katabatic wind) and pool in valleys, often creating a temperature inversion where the valley floor is colder than the mountain slopes above it.  Camping a little higher up the slope can mean warmer temperatures and less condensation by morning.

5. Minimise Condensation from Water

You often hear that you should camp away from water, but this is usually to protect water sources. There is another good reason to do so. Condensation formation in your tent will be worse around moist air from water sources. Being even 50m away from a stream or lake can make a difference.

6. Minimise Condensation from Clear Skies

Tent reviewers would have you believe that vents are the only thing preventing condensation. However, even these will not prevent condensation forming in your tent under clear, cold skies. Finding a spot under the cover of trees, shrubs or even tucked close against a hedgerow will drastically reduce condensation. But of course, always check above to ensure there is no risk of branches, or anything else falling onto you.

7. Assess Rock and Slope Stability

Particularly important during freeze–thaw conditions. Look for loose scree, fractured rock, or signs of recent rockfall. Even small stones skipping down a slope can become a significant hazard.

8. Consider Snow and Seasonal Hazards

In winter mountains, ensure you are well away from avalanche paths, cornices, and steep snow-loaded slopes. Check recent weather and snow conditions and understand avalanche risk. Snow itself can insulate from colder winds, so digging down into snow to pitch your tent lower can protect from wind and cold. Just ensure that you also dig a channel from your tent downhill a small distance so that any cold air sinks away from your tent.

9. Consider Sun Orientation

A south or east-facing site can offer early morning warmth and quicker drying in cold conditions, however this is a lower priority than avoiding a pitch on the windward side of a slope. 

10. Consider Retreat Options

Even with all the precautions above, weather can change and a night in a tent in the mountains can become uncomfortable. If you have the choice, try to pick a site that enables you a relatively safe descent if you decide that may be wiser than staying put. Bear in mind that you will be descending in the dark by headlamp (of course you have a headlamp), in bad weather, so avoiding crags, steep drops and loose scree is preferable.

Commencing from Nufenen Pass between the Swiss cantons of Valais and Ticino, it is a short walk to the Italian border where the view unfolds south across the Lepontine then Pennine Alps. Leaving the glaciers and moraine of Switzerland behind, the route descends through a series of long, flower-rich valleys punctuated by steep, densely vegetated climbs over high passes. Stone roofed huts (baita del pastore) enable shepherds and herders to stay in the highest alpine pastures and move with their cattle and flocks during the summer. Dairy products are produced on site, using age-old methods and can often be bought directly.

Exploring the GTA is a far more than a mountain journey. With considerable ascent and descent, the path takes the walker over high passes daily, typically descending to small settlements. It is a glimpse into lives and practices that have changed little in hundreds of years. Without the tourism or income that enables visitor infrastructure in so much of the alps, the GTA requires a little more independence and effort, and is consequently a wilder and more personally rewarding journey. 

Staying close to the French border, the route moves south through the Alpe Veglia and Alpe Devero protected areas without crossing a major road for four stages. Beyond, in the Natural Park of Alta Valle Antrona, a series of turquoise lakes are nestled between steeply forested slopes. At Alagna Valsesia, the GTA touches on the Tour of Monte Rosa. The nine peaks of the Monte Rosa massif can be viewed from several places on route, but a day off in Alagna enables a cable car ride to Passo dei Salati where a ridge-top restaurant enables you to dine among these giants. Only after crossing the south-eastern corner of the Aosta valley at Oropa, does the route descend toward Quincinetto, the end of the northern section of the GTA and the first of only two major centres of population you will pass through. 

Now heading west into the Graian Alps, the route drops into the Valle Soana and meanders along the riverside before climbing momentarily into the Gran Paradiso National Park and joining the Senteiro Italia along the Valle Orco to the great dam at Lago di Ceresole. From here a steep climb through fragrant pines levels briefly for the Piano del Morti (plain of the dead), before ascending through great slabs of crumbling rock to the dramatic Colle della Crocetta (pass of the cross). Continuing south, the trail crosses a series of 2500m passes, dropping at each stage into the heads of the valleys for rest and accommodation.  

Close to the mid-point of the route, Susa, considered the capital of the Cottian Alps, pre-dates Roman rule and is the second major centre of population visited by the trail. Inducted into the Roman Empire in the 1^st century BC the city hosts a fine Roman amphitheatre and the Arch of Augustus dating to 8BC. Further south the trail passes within a few hundred metres of the upper battlements of the great fortress of Fenestrelle. Stretching some 3km and over 600m in height above the Val Chisone, this structure is sometimes referred to as the ‘Great Wall of Europe’.

The remote crossing of the Colle dell’Albergian precedes a descent to Balsiglia, a ghostly hanging valley and site of one of the defining battles between the Duchy of Savoy and Waldensians, persecuted for their religious division from the Catholic Church. This densely wooded area is sparsely populated and rich in nature, and undulates between ridge and valley, passing isolated farms before rising again at the Colle della Gianna. From here, a magical descent to Pian del Re sees you step across the spring at the source of the mighty Po river, before climbing to one of the mountain highlights of the route, the desolate and pyramidal Monviso, rising from a moonscape of rock and scree. 

The southern section of the GTA commences at Castello in Valle Varaita and continues south across the head of Valle Maira, crossing the expansive Gardetta plateau and past the soaring pale form of Rocca La Meja, a rock climbers Mecca. The trail takes advantage of many former military tracks and passes bunkers at the Colle della Lombarda. A brief dip into Valle Valscura precedes another great climb to the Colle del Chiapous passing the ‘queen of the maritime alps’ and mountaineers’ favourite, Monte Argentera. As the route turns south for the final time it crosses from Piedmont into Liguria. Here the environment and flora become clearly more Mediterranean, water is sparse and climbs become shorter, descents longer, but the trail hugs the remaining high ground, until within touching distance of the sea at the Roman coastal settlement of Ventimiglia.

Anyone taking on this trek is likely to face a range of challenges that at times can include overgrown or obscured paths, cattle across the trail, fierce patou sheepdogs guarding flocks and language difficulties. If you view your trek as a holiday and want everything made easy and accessible for you, then look elsewhere. This is just not that kind of trek. It is far more rewarding. This is an adventure in the truest sense of the word. The experience will push you. It will expand your comfort zone and require you to engage with real local people in a way that you don’t in most areas of the alps. At some point you will probably stumble into a remote settlement, in the dark, drenched by a mountain storm and knock on a farmhouse door seeking shelter. An experience more akin to travellers of the past than of the modern age.

However, having overcome your anxiety about barging in on a family having dinner, about not being able to communicate or about the pool of muddy water you are leaving on their kitchen floor, you will experience the kind of hospitality that is equally rare in modern travel. You will be welcomed with smiles all round. Your inadequate grasp of the language and your desperate attempts at mime will generate much amusement, and you will be welcomed like old friends. Soon you will be warm and dry, be shown to a chair and fed the kind of food that makes you wonder why supermarket food taste so bland. With a warming glass of red wine that is equally astounding, you’ll find that basic conversation about your journey and your life is possible with a few gestures and a phrasebook. Overall, you’ll discover the happiness and hospitality that mountain people are famous for, despite their simple lifestyle and lack of modern amenities.