Applied navigation

Map reading test scored on four skills separately

Four skills, three items each: the grid bearing between two features, the six-figure reference of a feature, a distance converted through a stated scale, and a height or a climb read off contour lines. Eleven of the twelve items take a typed number or a typed reference, so there is nothing to eliminate and nothing to guess between, and each skill is marked against a tolerance printed before you start. The map is generated for your run from a height field, which means a second attempt is genuinely a second map rather than the same one reshuffled.

100% free
No signup
12 items, 4 skills
Typed answers, not multiple choice
90 seconds an item
New map every run

Four skills, three items each: taking a bearing, giving a six-figure grid reference, converting a scale, and reading contours. The map is 4 km across with one-kilometer grid squares and contours every 10 m, and a new one is generated for every run. 90 seconds per item.

This is the applied corner of spatial ability, and the audience for it is specific: people going for a Land Navigation module, a mountain leader award, an orienteering season or a surveying ticket, plus everybody who has stood at a path junction with a map that suddenly disagrees with the ground. What it asks is what those settings ask — a number, in the units the discipline uses, not a choice from four.

What is asked and how it is marked

Map: 4 km square, one-kilometer grid, contours every 10 m, generated per run
Items: 12 — three bearings, three grid references, three scale conversions, three contour readings
Window: 90 seconds each; an unanswered item is recorded as unanswered, not wrong
Bearing: grid bearing from grid north, clockwise, marked within 5 degrees
Grid reference: six figures, eastings then northings, marked exactly
Distances: marked within 10 percent, and the pure scale conversion within 3
Contours: marked within half a contour interval, which is 5 m here

The worked items show how the answer is arrived at, step by step, which is worth taking if you have never given a grid reference before — eastings first, then northings, and the whole of the rest follows from that one convention.

How to take the map reading test

One map, four skills, a tolerance stated for each.

Run the two worked items first if grid references are new to you
They are untimed and each ends with the answer worked through in full — eastings before northings, tenths of a square, and where grid north actually points. That convention is the only piece of prior knowledge the scored run assumes; everything else on the page is arithmetic you can do from what is drawn. If you already give references for a living, skip straight past them.
Answer in the units the question names
A bearing is degrees clockwise from grid north, a reference is six figures, a distance is meters unless the question says centimeters. Type the number and nothing else — a stray unit is stripped, but a bearing given as a compass point cannot be marked. Ninety seconds is the window per item, which is generous for a bearing and tight for a contour question you have not thought about, so read the whole question before starting the arithmetic.
Read the four scores as four scores
The result gives bearings, grid references, scale and contours separately, each with the tolerance it was marked against and the median time it took you. Underneath, every question is listed with your answer, the true answer and the working, so a wrong item is something you can learn from rather than something you are told about. Three items per skill is a thin basis for a verdict, which is why an interval is printed beside the weakest skill instead of a label.

Technical specifications

The map	4 km square with one-kilometer grid squares, contours every 10 m, six named features and four lettered points, generated from a sum of three Gaussian hills evaluated on a 65 by 65 lattice
Contours	Traced as the level sets of that height field by marching squares, with every fifth line drawn heavier. They are computed from the relief rather than drawn beside it, so the height at a lettered point is exact rather than approximately what somebody intended
Items	12 scored — three bearings, three grid references, three scale conversions, three contour readings, shuffled — plus two worked items with the method shown and no clock on them
Tolerances	5 degrees on a bearing, exact on a six-figure reference, 10 percent on a measured distance, 3 percent on a pure scale conversion, and half a contour interval on a height
Guessing	11 of the 12 items take a typed answer, and the one shortlisted item offers four lettered points. No correction for guessing is applied or needed — arriving at a correct six-figure reference by chance is a one-in-a-million event
Window	90 seconds per item. An item you do not answer is recorded as unanswered rather than wrong, and the two are reported separately because they are different failures
Scales used	1:25,000 for the map you are looking at, where one centimeter is 250 m, and 1:50,000 in the conversion items, where the same route draws half as long
What voids an item	Sending this tab behind another window. Ninety seconds is the only pressure the page applies, and a clock that kept running inside a different application would be applying it to the wrong thing

Frequently asked questions

Who is this test actually for?

Anyone who has to produce a number from a map rather than an impression of one — orienteers, hill walkers working towards a leadership award, candidates for military and emergency-service selection, surveying and forestry students, and search teams who pass grid references over a radio. The four skills here are the four those settings ask for, in the same units, which is why the answers are typed. If you only ever need to know roughly where you are, this is a harder test than you need.

Why are the answers typed instead of multiple choice?

Because a four-option map question hands a quarter of a mark to somebody who cannot read a map at all, and because nothing in the field is multiple choice. Typing also lets each skill carry its own tolerance rather than a shared right-or-wrong: five degrees on a bearing reflects what a protractor on paper actually delivers, while a grid reference is either the right hundred-meter square or it is not. A shortlist could not express that difference.

Is this a real map?

No, and that is deliberate. The relief is three smooth hills summed on a lattice, the contours are that function's level sets, and the features sit on the hundred-meter lattice so that every one of them has an exact six-figure reference. A real extract would bring copyright, a fixed set of answers that circulate, and features whose true bearing is whatever the draughtsman drew. A generated map gives exact answers, a fresh set every run, and no license to worry about.

Why is grid north not the same as magnetic north here?

It is not the same thing anywhere, but this test asks only for grid bearings, so the difference never enters the arithmetic. Grid north runs up the vertical grid lines; magnetic north wanders by a few degrees depending on where and when you are standing, and converting between them is a step you take with a compass in your hand and a current declination figure from the map margin. That step is real and important in the field and it is not measurable from a browser, so nothing here pretends to test it.

My distances are always slightly out. What am I doing wrong?

Usually one of two things: measuring along the route instead of straight, or converting with the wrong number of zeros. The straight-line items are the hypotenuse between two features, so if you have added the east-west and north-south legs you will be about 15 to 40 percent high depending on the angle. The scale items are the other common slip — one centimeter at 1:25,000 is 250 m, not 25 m and not 2.5 km, and the 10 percent tolerance on measured distances is wide enough to absorb a shaky ruler but not a decimal place.

How do I tell which slope is steepest from contours alone?

By spacing, and by nothing else. Contours are drawn at a fixed height interval, so lines close together mean the same height change is being crossed in less horizontal distance — that is the definition of steeper. Reading the shading, the color or the size of the hill is a habit worth unlearning, because none of those is on the map. The steepest-point item here is the one shortlisted question on the page, and it is shortlisted precisely because comparing four gradients is a judgment rather than a number.

Does doing well here mean I can navigate?

It means you can do the arithmetic navigation is built on, which is a necessary part and not the whole. What a screen cannot ask for is the rest of it: relating what is drawn to what is in front of you in poor light, keeping a pace count, deciding to aim off a stream junction rather than at it, and staying oriented while walking. Those are learned on ground with a compass. This page measures the part that can be measured exactly, and it is honest about which part that is.

What a map asks for, and why it splits into four skills

The four things on this page are related and they are not the same operation. A grid reference is a convention: eastings before northings, tenths of a square, no interpretation involved beyond reading a lattice correctly. A bearing is geometry — an angle between a line and grid north, which most people can estimate to within about fifteen degrees by eye and to within five with an instrument. A scale conversion is pure arithmetic and the commonest place to lose a factor of ten. Contours are the only one of the four that asks you to see a surface: a set of nested closed curves has to become a hill, and the same lines read as a hollow if the spot heights are ignored. People who are fluent in three of these and slow in the fourth are the normal case, which is why the report keeps them apart.

The tolerances are as much a part of the measurement as the questions. Five degrees on a bearing is roughly what a protractor and a steady hand deliver on paper, so demanding the exact degree would score the mouse rather than the reader. Ten percent on a measured distance is the width of a pencil line at 1:25,000 carried through to the ground. A grid reference gets no tolerance at all, because a six-figure reference names a hundred-meter square and the next square along is a different place — that is the entire point of the convention, and softening it would teach the wrong habit. Tolerances are printed before the run starts rather than revealed with the score.

This is the applied end of a cluster whose other members are deliberately abstract. The transformation underneath the bearing questions — working out a direction from a viewpoint you are not standing at — is measured on its own, with no map and no arithmetic, on the spatial orientation test, and it is the ability most likely to explain a run where the arithmetic was right and the bearings were still wrong. Reading a contoured map as a surface is closer to what the spatial reasoning test calls a cutting item, since both ask what a shape looks like along a plane through it. Two sensory checks are worth a minute before blaming any of that: fine contour lines are a low-contrast target, so the contrast sensitivity test will say whether they are actually reaching you, and the stereopsis test covers the depth cue that a flat map removes on purpose and real terrain gives back.

A reaction time here is the interval between the frame that painted the stimulus and the timestamp the browser attached to your key, both read from the same monotonic clock. What neither can see is the display pipeline behind it, so on a 60 Hz screen roughly 16 ms of every figure below is the machine rather than you. That is the timing floor: two numbers closer together than that are the same number, and this page reports no precision it cannot support.

A frame is invisible against a ninety-second window, and no score here depends on reaction time. The per-item durations in the results table exist to show which skill cost you the most thinking, and that is all they are precise enough to show.

This is a measurement exercise, not a clinical assessment. It reports what you did on this page against a stated reference and nothing more — it cannot establish a navigational or visuospatial difficulty. Only a qualified professional, working with more than a browser, can make that judgment.

Nor is it a qualification. Awards and selection boards that assess navigation do it on ground, with a compass, under supervision, and no browser result substitutes for one.

Where the map is drawn and the answers marked

Every number on this page is worked out by JavaScript running in the tab you are reading it in. Your answers, your reaction times and your score are never uploaded, logged or kept — which is also why the test carries on working after you disconnect from the network, and why nothing here can be held back behind an email address.

The height field, the contours and the twelve answers are all computed in this tab, and the map is inline SVG rather than an image requested from anywhere. Nothing you type is transmitted, no location of yours is involved at any point, and the only place a result can go is your own clipboard, if you press the copy button.