Imagine a city on treads; if you've seen or read Mortal Engines, imagine that instead.

mortal engines city

beep beep losers here I come

They're constantly moving, and have decent populations. Don't worry about how it works, or anything like that, just know that they are major trade centers and need to be visited on a regular basis, similar to port cities like New York, London, Shanghai, etc.

Because of this, mapmakers would need to locate these cities on maps. While magic does exist, and it is the basis for the moving cities, I would prefer that these maps be static, regular, maps.

Late renaissance to early industrial technology.

Some more info:

  • I would rather not have these cities move on a regular route, and instead I would prefer them to move on the spur of the moment.
  • The cities are limited to an area about the size of Mexico, with a vaguely trapezoidal shape. This area is grassland with very little change in elevation.
  • The cities move at slow-traffic speeds, at about 10 mph.
  • Instant communications, akin to the early internet but with only text, exist; see this question
  • More info added on request.
  • That "magic communication" isn't universal; it's more like a private net, with only those who are privileged and wealthy being able to receive or send.

Not really sure what tech level to tag it as. Feel free to edit if you think it's needed.

  • 21
    $\begingroup$ "Because of this, mapmakers would need to locate these cities on maps" Would you though? Are you sure you aren't taking for granted that cities on the real world are always on maps because they never move? Doesn't seem that big a deal to be honest if they are being visited frequently. Just update a central map. Omit from other maps. $\endgroup$
    – DKNguyen
    Commented Mar 31, 2022 at 13:20
  • 2
    $\begingroup$ What level of technology do we have? Is there telephone, radio, fax or anything of the sort to allow for instant communication? If not, is there a printing press? $\endgroup$
    – KeizerHarm
    Commented Mar 31, 2022 at 13:22
  • 9
    $\begingroup$ Sailing ships never could use maps, because they'd always be in a different place the next day. This is why the Americas were never discovered until 1972, when Richard Feynman invented the backlit color LCD panel. True Story. $\endgroup$
    – John O
    Commented Mar 31, 2022 at 13:30
  • 2
    $\begingroup$ I don't know how meteorologists mapped cold fronts and warm fronts and rain and snow... These things move, right? How did they do it? If only we could have an idea... Maybe, just maybe, they had pre-printed maps with the fixed parts of the world and then they used something, maybe colored lines drawn with a pencil, or maybe colored pins, to represent the moving parts? $\endgroup$
    – AlexP
    Commented Mar 31, 2022 at 14:44
  • 4
    $\begingroup$ You write "they are major trade centers and need to be visited on a regular basis". How does that work? Do they come to big ports on a schedule, or visit other towns to trade just whenever, or are they so important that customers take wagons of trade goods and wander around looking for city tracks? $\endgroup$ Commented Mar 31, 2022 at 22:28

15 Answers 15


Updated daily, like a weather map

It won't be possible to use a static map to accurately depict a dynamic reality. Instead, you just need to update the map periodically. For a real-life example of this, look at weather maps - these show moving things like storm fronts overlaid on a static geography. For most people, there's not usually a need to track these things in real-time - knowing where the storm front is today, where it's headed, and where it's likely to be tomorrow are sufficient.

You just need some mapping authority to track each city's location, which can then be broadcast everywhere as the daily "city location forecast". An updated map could be published each day in the local paper or displayed in the town square.

I envision the development of a central authority to track city locations in order to promote commerce for these major trade hubs. After all, by virtue of being trade hubs, these cities depend on other people knowing where they are. This would cut down on communications, as each city just needs to transmit its location and receive a compiled list of coordinates of other cities, rather than dozens of cities each needing to transmit their coordinates to dozens of other cities and each compiling their own maps. This parallels how most weather maps are generated, with data from many remote stations aggregated and published by a central group.

  • 2
    $\begingroup$ A good and useful idea, albeit I don't think displaying it on a town square would be necessary since it doesn't concern the general public, usually, where all those cities are. Instead I propose the dynamic map is part of relevant services, like the post office and possibly high-end map makers, who can afford to pay a wizard to connect to the 'magic server' containing the info on city location. $\endgroup$
    – Mookuh
    Commented Apr 1, 2022 at 11:43
  • 2
    $\begingroup$ @Mookuh I had imagined this might be a free service provided a central national authority like the US National Weather Service (which interestingly is part of the Department of Commerce). Agree most people won't need to know specific locations on a daily basis, but businesses in City A won't want to have people in City B needing to pay just to find out where they are. Trade hubs will want as many people as possible to know their location, seems reasonable a government might bear that cost instead of individuals, but it depends on the level of centralization and cooperation in this society. $\endgroup$ Commented Apr 1, 2022 at 15:00

Probability Density

Probability density over two dimensions

You mentioned that the cities tend to keep within a "home range", if you will, of a defined size (up to the area of Mexico?) and defined shape (commonly trapezoidal?).

Even if the city doesn't follow a regular route, it stands to reason that there will be areas and locations within this range which it visits more commonly and/or at which it remains for longer. I would expect these to include points or areas in which they can harvest resources, make trades, replenish supplies (water? fuel?) and perform maintenance.

If these cities have existed for long enough that they've pretty much established their home range and where they want to be within it, then mapmakers should be able to gather the information about where they've been and extract a pattern of "probability density": a graphical representation of where the city is most likely to be found.

I would imagine two ways it could be represented on a static, paper map:

  1. As variably-shaded colour: similarly to how countries are represented on a political map, you could have differently-coloured shapes on a map that define the usual home ranges of the corresponding cities. Where the city spends more time/visits more frequently, the shade of the color is darker; in the parts of the range it rarely visits, the shade is lighter/more washed out.

  2. As sampled past locations: the map contains a statistically significant number of past locations of the city marked as single dots of the appropriate colour. The probability of the city being in a certain area is expressed by how densely the dots in that area are clustered. (If there is a particular, exact location that the city often occupies, the mapmaker can cluster multiple marks immediately around it instead of stacking them on top of each other, since this would better express the frequency to the reader.)

Using this information, a reader of the map can hopefully make some informed guesses about where the city is most likely to be, and then seek news or rumour to confirm one or the other (if no instant communication is available).

  • 3
    $\begingroup$ I can do you one better. I can join up the dots to show the city's route over the past ten years, rather than just the density. $\endgroup$
    – Daron
    Commented Mar 31, 2022 at 15:14
  • 2
    $\begingroup$ @Daron: Absolutely! As long as it doesn't make the map too busy, I suppose. $\endgroup$
    – Qami
    Commented Mar 31, 2022 at 15:15
  • 1
    $\begingroup$ Also, with enough knowledge of the layout and orography of the area within the range of each city, discriminating which places are likely (or unlikely) for the city to "stop by" should be relatively easy $\endgroup$
    – Josh Part
    Commented Mar 31, 2022 at 22:44
  • 2
    $\begingroup$ Much more realistic solution for low-tech worlds than the daily updates idea. Also probably results in very artistically-beautiful maps. +1 $\endgroup$
    – MaxD
    Commented Apr 1, 2022 at 23:30
  • 2
    $\begingroup$ I can't tell if you guys are map-makers, or pirates scouting a target and studying its path. And I kinda like it. $\endgroup$ Commented Apr 3, 2022 at 16:32

Instant Communication

enter image description here

There are maps of the landscape and there are moving cities. Fortunately your world has:

Instant communications, akin to the early internet but with only text

That is enough to send out coordinates. The moving cities frequently send out their coordinates to the other settlements, both static and moving. The city locations are broadcast in the morning and evening, after the weather forecast.

Unless it is a baddy city of course. Those guys do not want to be found. Then don't send out coordinates and they also disguise themselves by wearing a hat with bits of twig stuck to it.

That "magic communication" isn't universal; it's more like a private net, with only those who are privileged and wealthy being able to receive or send.

The coordinates are sent out by wealthy business owners. They want people to know where their city is, because then they can come here and spend money. So they send messages to the central post offices in nearby towns. The post office has a cork board with a local map. There are pins with pictures of the different cities. They use the coordinates to move the pins and update the city locations. People in that town walk into the post office and look at the map.

You can reach the whole town with a single message so it's an inexpensive form of advertising.

Every large enough town will have a receiver. People from smaller towns have to wait for the postman to arrive from the larger town. The postman brings the coordinates with him.

  • $\begingroup$ Sorry, I should add it; that "magic communication" isn't universal; it's more like a private net, with only those who are privileged and wealthy being able to receive or send. $\endgroup$ Commented Mar 31, 2022 at 13:54
  • 2
    $\begingroup$ @Firedestroyer The coordinates are sent out by wealthy business owners. They want people to know where their city is, because then they can come here and spend money. So they send messages to the central post offices in nearby towns. People walk into the post office and look at the chart. You can reach the whole town with a single message so it's a relatively cheap for of advertising. $\endgroup$
    – Daron
    Commented Mar 31, 2022 at 13:58
  • 3
    $\begingroup$ @Daron this is very much what happened with a lot of early governmental services. Sending mail to the East Indies was very expensive and yet vast amounts of "trivial" archive information were sent just so that businesses could function. $\endgroup$
    – Borgh
    Commented Mar 31, 2022 at 14:21
  • 1
    $\begingroup$ +1 Mobile Shanghai's traders have a valid interest in publicizing their current and predicted location information if they wish to remain relevant (and wealthy). The local merchant will go to a convenient market to trade if the lucrative market is too hard to find. $\endgroup$
    – user535733
    Commented Mar 31, 2022 at 14:59
  • 3
    $\begingroup$ "How was I supposed to know it was Hoboken coming up behind us? It was wearing a dozen square kilometers of trenchcoat and fedora! It could have been any seventy-meter tall person with tank treads! " $\endgroup$
    – notovny
    Commented Mar 31, 2022 at 22:29

Frame Challenge: No Map Required

In a steppe environment, the smoke from a steam-punk style city will be visible for MILES. An object is visible at a distance of $1.22\times\sqrt{eyeheight_{feet}}\space\mathrm{miles}$ or $3.57\times\sqrt{eyeheight_{m}}\space\mathrm{km}$.

The top of a 20 story building is visible about 17 miles away. If the dust from the treads and the smoke from heating and lighting and industry is visible until it's 5000 feet in the air, then it should be visible from nearly 90 miles away.

The described range of these cities is roughly a square of 900 miles on a side, which means at 10mph, the cities can cross the entire range in about 4 days.

Since the city is visible from so far away, travelers just go to any convenient point near the center of the known range and wait. Within a week or two, they will see the city.

If the stationary cities around the edges of the steppe are keeping track of roughly where each of the mobile cities are, that would cut down on waiting time significantly. There may be a cottage industry collecting and distributing this information.

Catching up to the city once you've found it... using renaissance tech... that's an exercise left to the reader.

  • 2
    $\begingroup$ Yes you can wander for 2 weeks until you see a smokestack and then go towards it. But this doesn't make it easier to locate the particular city where your sister lives. $\endgroup$
    – Daron
    Commented Apr 1, 2022 at 10:15
  • 1
    $\begingroup$ Sounds very slow too. We cannot have market day once per week if it takes weeks and weeks to find the market. And what happens to their farms when the farmers go to the shops for a month? Someone still needs to milk the cows and turn hay and plant potatoes and repair the fence and trim hedges and scare off the local roughains and do weeding and make a scarecrow and. . . $\endgroup$
    – Daron
    Commented Apr 1, 2022 at 10:17
  • $\begingroup$ Plus what it you buy something large like a barrel of pickled anchovies for the Winter? Then I need to transport it all the way back home. If we traded locally the anchovy seller might drop it to my house; or I could get a lift off Anghony up the road. "Hey Anghony, are you taking your wagon to market tomorrow? You know the local market where we all go on the same time each week. Do you mind if I chuck a barrel of anchovies onto the back please?" $\endgroup$
    – Daron
    Commented Apr 1, 2022 at 10:25
  • 1
    $\begingroup$ It seems I am happy to handwave stuff about the fantastical elements of the world like the mobile cities, but less happy to handwave stuff about the mundane stuff like how non-mobile-city farmers get to market. $\endgroup$
    – Daron
    Commented Apr 1, 2022 at 16:26
  • 2
    $\begingroup$ I had the same idea as I read the question, but you can improve it by enhancing the smoke to be more visible. Also, you could colour-code it to make the cities distinctive! You could even use strong lamps during the nights and focus them to colourful beams in the smoke. $\endgroup$
    – EdvinW
    Commented Apr 3, 2022 at 17:37

You track their positions in a separate almanac

An almanac is a book that publishes forecasted information (tides, weather, planting seasons, etc.) using tabular data associated with a calendar. However, unlike the weather, these rolling cities WANT to be accurate so that they can trade their goods. So, the person driving the city has a vested interest in also keeping a copy of the Moving Cities Almanac to make sure that he is staying on schedule and being where merchants expect him to be at any given date.

So yes, you have a map, but your city is not marked out on the map anywhere, rather it is kept track of in this separate book. Each year, the moving city sends the almanac publisher its schedule for the following year to be printed, distributed, and sold to merchants and anyone else who wants to be able to find these cities. When you consider Code Monkey's answer about how far away these cities can be seen from the smoke cloud, it becomes obvious that fairly large map grids would be sufficient. So using the following map and almanac, you could go anywhere in C5 on the morning of Feb.4th and keep an eye on the Northern horizon, and it wont take long to spot the giant smoke stack coming your way.

But to get even more accurate, you can add Points of Interest as places these cities may stop specifically to do trade; so, if you know the city will be parked just outside of Valonoce from the 9th to the 13th, then this can give a very specific point to meet up with the moving city; so, even if the grid sections are 50 miles wide each, knowing it is going to a specific town or trading post means no time wasted looking around for it at all.

Also, just because these cities can move at 10mph does not mean they are always moving in straight lines over ideal terrain. For starters, they need to stop to actually load/unload cargo and conduct maintenance and they need to also be very carful because of their size to go around bad terrain and avoid farmlands etc that they would destroy if they drive through; so, while two PoIs may only be 30mi appart, it might take the better part of a day to get there if it needs to go slower and more indirectly to get through the marshlands or whatever that separate the 2 points; so, while it can move 240 miles in a day, in practice, it might cover much smaller distances making them easier to track down than you might think.

enter image description here

For context, the first printed almanacs started to appear in the mid 15th century and became very popular up until the internet phased them out; so, they are very appropriate for your tech level, and don't require any magic at all to use.

This makes them accessible to your common merchant and farmer which is FAR better for business than only being able to trade with the magical elite. When every farmer in a community can know when the moving city will show up, then they can co-ordinate their planting, harvesting, etc. to line up with the arrival of the moving city which will maximize the available trade goods for when your moving city does shows up.

Minor Frame Challenge

I would rather not have these cities move on a regular route, and instead I would prefer them to move on the spur of the moment.

This is a terrible idea. Even today with the internet and instant communication, this is not how bulk freight is done (IE: we can but we don't). All bulk supply lines happen on a schedule that has to be planned months and sometimes years in advance or they do not work. They need to line up with planting, fertilizing, and harvesting seasons to make sure the city is where it needs to be to meet supply/demand. If you don't arrive at a certain town in time for the harvest, then the food rots in its silos and people who needed to buy that food die of starvation and the people who needed to sell it suffer massive economic hardships from loosing their whole income for that year. If you don't arrive at another town in time for the first cold front, then they won't have the wood they need to warm their homes and people die from hypothermia. Every time a bulk freighter deviates from their route for even a few days, it will result in shortages, missed trade opportunities, and very palatable human suffering.

On demand goods have historically only been handled by small trade vessels for a good reason; so, if you're going to build a big moving city like this, your only real concern will be in bulk goods and that means keeping to a schedule. If you want a moving city to try to capitalize on spur the moment opportunities, then it should have smaller trucks it can send out to do these on-demand trades, and then those trucks can meet back up with the city later by following the almanac.


Static Map

If you want a static map, simply make one without the cities on it. The easiest way to deal with it.

Assuming that the movement of the cities doesn't affect the terrain aversely to the point it becomes beyond recognition on a regular basis, a geographic map can be created with everything that isn't a moving city on it.

Displaying the cities

As for displaying the cities? Simplified wooden versions of cities can be created with little to no effort and placed on the map. Then using colour coding or letters they can be given unique identifiers.

The routes can be displayed using graphite pencils on the map, or a plastic overlay, on which arrows can be drawn to indicate the direction of movement.


Now, as long as cities use a pre-arranged coordinate system, like latitude & longitude it is easy to simply find and paste those coordinates. With help of a communication network, only the city-identifier, coordinates, and direction need to be sent to generally get a good idea of where the city is and will/can be in a few hours. Just send/request updates a few times a day.

Secure Communication

As for how the data is distributed amongst the cities, there is little need for anyone that doesn't control where THE city is going to know where the others are. So they simply don't know, or only get 'older' data (like the maps from the previous years).

Now, this brings me to an interesting point:

Map trustworthiness

People can lie. Cities' controllers will lie to each other for their personal gain. Now I would display the known routes (like those checked by scouts) with a different colour than the coordinates gotten from first contact from an unknown city.

  • Green for certain (seen by scouts. Checked the tracks)
  • Yellow for allies (Very trustworthy, except when they suddenly are not)
  • Orange for estimates(Take it with a pinch of salt)
  • Red for unknowns(Enemies won't tell you where they are)
  • Etc.

You could go very far in this with city A telling ally city B that they checked and saw tracks from their mutual enemy, city C, around those and those coordinates. This would likely warrant at least yellow level trustworthiness for a hostile city.


Now back to the problem, how would mapmakers deal with moving cities? well depends on who is asking. The public gets old but reliable/checked data with routes drawn on them and separate maps for the cities themselves. The mayor gets a 'live-updated' map with sculptures, drawn-on older routes, estimated paths in erasable pencil, colour-coded and all, and probably a whole 'war room' to go with it.


NOTE: if you put your thought-based internet in, with instantaneous communication, there's no issue.. No need to draw any cities on a fixed map using ink. You can connect to the appropriate source and inquire where the coordinates are, e.g. in section BN-39 on the map. Then, put a pawn or token on your map. Or a magic pointer beam of light, whatever.. I'll assume you can't.. no life info..

Your city would be like an oil tanker, not change course quickly

Suppose there wouldn't be life coordinates on your magic internet..

You have a Napoleon-time civilization that built moving cities. These cities would bring a fantastic military advantage at the time.. they are on large wheels, moving 10km/h, they could outpace an enemy pursuing it on the ground.

Military: city can't change course quickly, so extrapolate and intercept

enter image description here

Say, Napoleon needs to attack a city like that. I suppose the best way to do that is anticipate its path. These cities are like oil tankers, they don't turn easily. They set out a course and could move in one direction for days. So all you'd need to have is a spy with doves to send you a message from inside the city, using a dove, or a fast courier. When you have your news in 24 hours, Napoleon could intercept the city.

In peace time, folks wanting to reach the city for trading could use a similar method. Travel to a place where the city will be, in a few days. All the time of the world, to take your sample and direction. Ask people where it was and where it went a few days ago, chance is you'll find it easily.


You state that your cities travel SLOW at 10 mph. While that may seem slow to us, used as we are to cars, trains and planes, it is very FAST given your tech-level.

Your traders will be moving their goods around at the speed of horse (or oxen) and wagon, which will be more like 3 mph.

Your traders can never catch up to a moving city.
The only way to trade is to wait for the cities to visit at a known (and fixed) meeting/trading location.
Or a choke point in the landscape (like a ford in a river or a mountain pass) were all cities must travel through when they are nearby.

So your maps don't need to display the cities themselves at all. They depict the meeting places and other places were you can expect a city to show up sooner or later.

  • 1
    $\begingroup$ Magic is nigh-universal in this world; traders are also equipped with things that allow them to move quite fast. Thanks for the input though. $\endgroup$ Commented Apr 1, 2022 at 12:17
  • $\begingroup$ Have to -1 duplicate answer, worked out the same idea last night, see answer above $\endgroup$
    – Goodies
    Commented Apr 1, 2022 at 13:40
  • $\begingroup$ @Firedestroyer The only thing you said about magic is that it is involved in moving the cities. I assumed a low-magic environment because of that. But that doesn't really void my answer. As long as you keep chasing after the cities you're always a step behind. $\endgroup$
    – Tonny
    Commented Apr 1, 2022 at 14:32
  • $\begingroup$ @Goodies Not quite the same in my opinion. $\endgroup$
    – Tonny
    Commented Apr 1, 2022 at 14:33

Don't map them. Map their routes between port cities, as well as the port cities. Label the route, not the city. The data would be outdated as soon as you published the map anyway.

Or just update their position daily and only publish online :-)


Maps with Expiration Dates

Using your communications network, cartographers would regularly update their maps with known and reasonably certain locations to a particular degree of accuracy – the map's best by date. A city known to be at coordinates x, y on publication day would have concentric circles around it each representing the maximum distance it could travel (factoring terrain) within a given period from that date.

By collating multiple editions, someone could track the progress of a given city and reasonably conclude the general direction of it, so even if the map were several months out of date they would have a good idea where they could find it.


The city route may not be regular, but it still can be predicted.

The moving cities have their very own reason to move, it is something like harvesting.

Cities deplete whatever the resource they harvest around themselves (be it wealthy customers, unwashed masses of pesantry, particular product made or mined, whatever), so they move on, avoiding places where other moving cities have been recently.

Few different types of cities may compete on few different combinations of possible resources, to add some complexity.

The land has different productivity and different pattern of productivity depending on the season.

And to sum up, a good map maker can predict the city leader's decisions pretty well by knowing the land productivity behavior and the needs of the particular cities.

If cities are numerous, a comitee of map makers, each knowing a particular group of these may be needed in order to predict the whole swarm.

The map may show predicted tracks of these objects with points with dates on them.

The regular maps will be good for a while - say, a year or so.

A map good for 2 or 3 years will be more expensive and less accurate.


As a mapmaker, I've have dealt with similar situations and have seen them dealt with by others. The short answer is that your mapmakers don't map the cities.

When an object isn't fixed or trackable, you don't map it as though it was. Wildfires, sandbars, and herds of animals can all be located and mapped in real time today, but historically, it just wasn't practical. Your cities act as though they are animals rather than wildfires or sandbars, moving frequently and somewhat irregularly, but with intent. Like animals, there are going to be patterns. So what the mapmakers map is the pattern.

In Colorado (chosen because it's got excellent digital maps for big game hunters at https://ndismaps.nrel.colostate.edu/index.html?app=HuntingAtlas), biologists map the seasonal ranges of the animals, the areas herds migrate through, and the overall ranges. Unlike a herd's current location, these things are determined by climate, terrain, and the number of animals, so they change very little year to year and a decade old map will still be fairly accurate.

If I was mapping a movable city, I'd ask what matters to a moving city's survival. Presuming the land is peaceful enough that staying hidden is not a priority, cities require things like food, trade, and communication with allies.

Food is never randomly placed. The moving cities presumably have farmers or herders they're allied with to supply food. So every year around harvest time, the city will be making a route around its farmlands. A more herding oriented city might simply follow where the animals can find feed in any given season. There could certainly be war if a herding city were to encroach on a farming city's allied farmers land.

In order to trade with other cities, there must be large marketplaces where multiple cities can gather. These would be at least somewhat regularly scheduled, and the fairgrounds would be markable on a map. The Grand Fair in the south might be the third week of May, or it might be in spring a few weeks after the first thaw, but it must be regular enough that merchants can plan to attend without going bankrupt. The best quality maps might show that Adams City and Brandon City almost always attend the Grand Fair, while Carver City is more often found at the North Fair, held across the country at a similar time.

Meetings with allies might be planned regularly year to year (think major international summits today) or they might be planned on an ad hoc basis between cities. If the second, the mapmakers would mark the areas that could support meetings of cities, but wouldn't provide much else. If there were summits, then the mapmaker could detail what cities and when they should be expected.

Of course, if the cities are warring frequently, then all of this becomes much less predictable because a predictable city would become dead as fast as a starving city. The principals of mapping would work the same, but it would be far less reliable.

The cities' leaders would have a maproom where they tracked the last known location of cities via pins placed into maps, but the accuracy would only be as good as their communication and intelligence sources. Placing the pins would be the responisibity of traders or generals, not mapmakers.


TL;DR: Tracking crews (aerial and ground), line of sight triangulation, radio signal triangulation, publishing of future travel routes, publishing of past position data (assumed no satellites).

I'm assuming that there isn't satellite surveillance (someone could probably generate an ASCII representation using a couple of satellites and triangulation).

I would think that certain moving cities whole industries would be dependent on tracking, for example shifts would go on excursions and set up temporary scout sites to visually triangulate positions by relaying, in relation to compass direction, what angle they can visually spot the city, and roughly how large it looks in their scopes and at what zoom level. In that way a minimum of two scout teams can set up a monitoring period of a city nearby, assuming they are close enough. A single crew would be able to determine direction relative to themselves but only roughly guess the distance based on the zoom level and known dimensions of observed cities. That city could then charge a subscription access to live updated maps, provided that the equipment is built to do so.

A few crews can go far and wide, assuming they are equipped for high endurance, and relocate to keep within a certain signal range of their home city; the crews would be the feelers of individual cities, too. This can also be implemented by having each city have their own crews and a "detection net".

Cities could also publish their scheduled itineraries, lists of where they have planned in advanced to go to, plus direction and speed. A city is generally going to be very slow to change course.

A few more major cities could archive broadcasted locations and provide probability maps (as another answer had already said) based on hourly, daily, weekly, monthly, yearly, etc. data points and can filter from today, 5 days, 1week, 1 month, 1 quarter, 1 year, etc. data sets.

There can be an open source database of city sightings, where citizens can submit tips on where they had last seen a city. This method is dubious for usability and probably would be used for search and rescue of a downed/stranded city.

Because electronic communication is already possible, cities can broadcast a ping signal like radar containing the angles they had measured nearby pings to be. Cities can collect this information to mathematically triangulate where other cities are. This can also be coupled with the aforementioned crews (the crews plus the city itself can measure angles based on location and triangulate for themselves, assuming a great enough distance is between the crews and the city itself).

Overall I would think that the cities would have at least a static point on the ground they are travelling on agreed upon to "ground" their relative position data.

An aerial lookout could be a thing. A towed blimp with a lookout crew could do most if not all of the aforementioned tracking methods.

Most of this kind of tracking would require a lot of cooperation between friendly cities.

I would assume there would at least be a few static cities that drill the land or otherwise require staying in place... there aren't many sources of energy aside from solar water heating, biomass farming, or solar panels (yes, they existed before computers, invented in 1883).


A map is a record of where a city was, at the time the map was made. For most cities, that's a static location. In this world, any map has a time-to-live and anything older than a week means the city could be 1680 miles away if driven 10 mph day and night.

So a map becomes more of a "newspaper" or a "telephone book" that has decreasing validity as time moves on. Since you know where the city was then if the city leaves a trail that can be identified by a tracker when it is crossed, then a profession of "city finder" can exist. Much like a "native tracker" or similar.

These tracks are going to be deep grooves in soil, or non-existent in rocky areas. How long it takes a track to be erased by erosion or similar could be a story point - a sudden rainstorm and the barely-visible track has vanished and the city turned sometime in the night.

I suggest that a city rolls along at a much slower pace than you've quoted. 10 miles in 24 hours, or far slower on bad terrain would make this more practical. 5 miles in 12 hours then 12 hours to "sleep" may be a more plausible speed, allowing runners to "catch up" and anyone in the way to move herds sideways out of the city's path.
Also, a week's travel is only 35 miles and it is plausible the city's smoke stacks/signal tower can still be seen from that distance.

Aside, driving in the dark may be considered a "last resort" because of the dangers of the unseen terrain.


Use a static map with a reusable transparent sheet overlaying it, updating the locations. I suppose the sheet can be made of glass and wiped clean for updates.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .