Gradient Profiles for the UK

[splashoutradio]

Hopefully I am alright posting this here, feel free to remove if not...

As part of an ongoing project to use open and FOI data supplied by the rail industry in different ways, I've put together a (hopefully) comprehensive set of PDF's with gradient profiles of the Great British rail network.

They're only first editions, but just in case they're of any interest to anyone, they're available to browse at: www.railwaydata.co.uk/linefiles. Feedback always welcomed

 

[Llama]

ELR MVM seems to be missing some significant gradient information?

 

[splashoutradio]

Thanks Llama,
You're right - just had a look and it was missing quite a bit as far as accuracy. The data for some routes was more of a gradient summary than an accurate portrayal of what's actually there, so I've found a more detailed gradient dataset, and re-uploaded all the gradient profiles using this. Hopefully it'll look a little more complete now

 

[Cricketer8for9]

This is a great resource. One query. It may just be me opening each one on my phone, but in some cases a short section of LEVEL and the following gradient are “printed” together making the gradient itself hard to read.

 

[Senex]

One query from me too, if I may. Do you know the level of precision of your source? Sticking with ELR MVM, already mentioned, Miles Platting Bank is shewn simply as 1:59 rising steepening to 1:47. Yet we know that electrification has introduced a dip under the Cheetham Hill Road bridge at the east end of the station in order get sufficient clearance. There's a gradient-list on line on answer to an FOI request for the Newcastle to Carlisle line (https://www.whatdotheyknow.com/request/track_gradients) that seems to shew a much more complicated situation for that line. Or is it that as Miles Platting Bank was built as an incline it is much more consistent in its gradient?

 

[splashoutradio]

This is a great resource. One query. It may just be me opening each one on my phone, but in some cases a short section of LEVEL and the following gradient are “printed” together making the gradient itself hard to read.

This is a bit of an issue for routes where there are lots of gradient changes. Was thinking of including a table of gradient change locations, but didn't want to include too much info. Can include it if you think it'll be beneficial

One query from me too, if I may. Do you know the level of precision of your source? Sticking with ELR MVM, already mentioned, Miles Platting Bank is shewn simply as 1:59 rising steepening to 1:47. Yet we know that electrification has introduced a dip under the Cheetham Hill Road bridge at the east end of the station in order get sufficient clearance. There's a gradient-list on line on answer to an FOI request for the Newcastle to Carlisle line (https://www.whatdotheyknow.com/request/track_gradients) that seems to shew a much more complicated situation for that line. Or is it that as Miles Platting Bank was built as an incline it is much more consistent in its gradient?

The data used is from a Network Rail document called "Track Gradients by DU" (it's available from the attachments section here - https://safety.networkrail.co.uk/safety/on-track-plant-safety/rrv-safety-improvement-programme/) - it's by no means perfect and it does have its fair share of anomalies, but having crossed referenced it with the FOI request you mentioned, it does seem to contain the same data the only difference being, the FOI document joins all the ELRs together to make the complete route instead of it being split several separate documents. Happy to be corrected on this, if you've spotted something else though

 

[Senex]

This is a bit of an issue for routes where there are lots of gradient changes. Was thinking of including a table of gradient change locations, but didn't want to include too much info. Can include it if you think it'll be beneficial


The data used is from a Network Rail document called "Track Gradients by DU" (it's available from the attachments section here - https://safety.networkrail.co.uk/safety/on-track-plant-safety/rrv-safety-improvement-programme/) - it's by no means perfect and it does have its fair share of anomalies, but having crossed referenced it with the FOI request you mentioned, it does seem to contain the same data the only difference being, the FOI document joins all the ELRs together to make the complete route instead of it being split several separate documents. Happy to be corrected on this, if you've spotted something else though

Thanks for the info. What an interesting document! I wonder which department it was originally prepared for.

A further thought. The NR reference you give refers specifically to gradients steeper than 1:75, and the last column in the Excel file seems to bear that out, with gaps in the mileages, whereas the Newcastle-Carlisle table is continuous in mileages. (Or have I misread something? — always very possible!)

 

[ComUtoR]

This is a bit of an issue for routes where there are lots of gradient changes. Was thinking of including a table of gradient change locations, but didn't want to include too much info. Can include it if you think it'll be beneficial

When presented by each page and you can see the profile at the top I like that initial look.

My issues/questions/suggestions are :

I'd like a Key as it took me a while to figure out where the stations are. or where each reference point was.
I'd like to have each profile between two stations or customizable between two selectable points.

How difficult would it to be to have the gradient profile at a zoomable resource ? I understand that scale would be a problem but I'd really like to see the profiles through the station.

Too much information... No such thing On my local routes we have stations that change gradient through the platforms it would be great to see this. The more accurate it is, the better.

The changes I've suggested are because I'm looking at the document for a practical purpose as a Driver. Maps are nice but I want to be able to use it in some way and the more customisable/useful it is the more likely I am to use and share it. From an initial and general document. I liked it a lot.

Nicely done.

 

[Ploughman]

Will there be any info, at some point, in the future on the Preserved lines?

 

[Bevan Price]

Although not on line, gradients for all ex-North Eastern Railway lines (including North Yorkshire Moors) can be purchased from NER Association:
https://ner.org.uk/product-category/diagrams/
Scroll down page to "NER Gradient Sections". (and £4.50 represents very good value for the contents of that book.)

Gradients for all former L&YR lines appear in the histories of that railway by John Marshall, and also by Eric Mason.

Gradients for ex-GCR lines appear in George Dow's 3 volume history of that railway.

Gradients for ex-Midland Railway lines were published in the same series as the MR System Maps. You can probably find copies a lot cheaper than this copy offered for sale from America:
https://www.amazon.com/Midland-Railway-System-Maps-Gradient/dp/1899890289

 

[splashoutradio]

I'd like a Key as it took me a while to figure out where the stations are. or where each reference point was.

There are keys at the bottom of each file (or at least it should be ) - I'll have an experiment, and see if placing it somewhere a bit more prominent works better.

I'd like to have each profile between two stations or customizable between two selectable points.

Love this idea! The next part I was looking to do was a mileage search program - where you enter a start point and a destination, and then the system returns gradients, mileages, tunnels, that sort of thing. It's a fair way off yet though, and it'll require data of where the ELR's connect with each other, before work can really start.

How difficult would it to be to have the gradient profile at a zoomable resource ? I understand that scale would be a problem but I'd really like to see the profiles through the station.

A zoomable resource, yes, but unfortunately not to the level of detail you mention. NR data states the approximate location of a station and roughly where a gradient change takes place. As gradients are often smooth gradual changes and not at specific points NR have to do some averageing out to produce the data that is shown on diagrams and in signal boxes.

The original document idea was based very loosely on the industry's Five Mile Diagrams, but instead set over 4 miles per diagram. To an extent, this offsets some of the data inaccuracies and averagings.

Too much information... No such thing

As a trial, too see what people think, I've added the locations of recorded gradient changes to the Settle-Carlisle line - www.railwaydata.co.uk/linefiles/route/?ELR=SAC

Will there be any info, at some point, in the future on the Preserved lines?

Absolutely! If the information is available, why not?
I've got all the info for the East Lancs Railway and will add this before long, but should anyone have details for other lines (that they're allowed to disclose, that is ), I'd gladly add it to the site.

 

[kieron]

I have a few questions about some of the data in these files.

I'm having a little difficulty in seeing what WDB2 is showing. The only gradient indicator on it says that it is level from 13m 78ch. The first feature quoted is Hawarden Swing Bridge at 13m 24ch. The direction in which distances increase changes on the sectional appendix between Shotton and Hawarden Bridge stations, so I wonder if it could be that some distances are measured in the wrong direction.

It's also unclear to me what is being shown on WDB2 and what on WDB3. For instance, there's a bridge over a road just north of Hawarden Bridge station, with a barrow crossing running across the top of it. The crossing appears to be on WDB3 and the road bridge on WDB2.

WDB2 is all level, so that's not an issue, but I can't tell from these files how long the level section actually is.

On another note, I'm wondering if you could confirm where the structure data comes from, or how old it is? I know that WDB1 43 (a bridge over Killins Lane in Shotton) has had the name written on the "rail authority" signs on it changed from "Shotton Lane" to "Killens Lane" (sic) at some point in the last few years (I've seen both, but don't remember when). The WDB1 list uses the old name, so I'm wondering how old it is.

 

[splashoutradio]

The data for gradients, stations, bridges, tunnels and level crossings come from 5 separate NR datasets, either obtained via the "Safety Central" portal, NR's 'Information and Data' website or through FOI requests,

By and large the data is fairly accurate and gives a good overview, but as you've noticed, things don't always match up as they should - particularly with gradients.

There are some basic 'sense checks' and conversions that are carried out on the data we receive; the first is that where one gradient ends, the next (if there is one) starts immediately and as such it flows without any breaks. Distances are also received in varying formats (miles & chains, miles & yards and miles rounded to 2 dp.) For constancy, these are converted to miles to the nearest 10dp for plotting and to the nearest chain for reference. As such, there's no distance measuring done on our side, but it's not beyond the realms of possibility that the original data could be measured the wrong way round, and I'm sure such a huge dataset is liable to human error at times.

The bridge data we're currently using is dated 2016, but I shall request an updated version (as well as confirmation on the WDB2 situation) and recompile the files once received

 

[Senex]

Thanks for the information about the information and the sources—what a huge amount of effort has gone into this, for which much gratitude! And I for one really appreciate this month's addition of the text information about the points of gradient-change.

How reliable is the NR information? Should one always be aware that error might have crept in? I was looking at SPC2 out of curiosity, since the Midland main line is one I know well. The climb to Sharnbrook effectively begins at milepost 56. You have, from NR, 1:100 from 55m75ch, 1:119 from 56m19ch, 1:130 from 56m44ch, and then 1:200 from 56m72ch to the summit level at 59m44ch. The descent starts at 59m54ch and drops at 1:120 for three miles to 62m46ch.

Whilst the official Midland Railway diagram of 1902 agrees almost exactly for the section from the summit almost into Wellingborough, it is diferent on the ascent side, shewing a three-mile incline at 1:119 and with significant differences south of that. Unless my memory deceives me, the 1:119 ties in with the lineside markers. And NR's lengthy climb at 1:200 would not gain enough height to allow for the subsequent steep descent!