Digital Elevation Models & Mining History

Digital elevation models are expressions of the surface of a planet as data. These, and the software to view or process the data, have become freely available over recent years. The software ranges from easily usable online viewers to PC-based tools requiring intermediate levels of IT skill. This all makes for an interesting and useful resource for people interested in mining history, whether as a casual interest or as a more focused amateur historian. This article seeks to provide an introduction to the topic and to illustrate what is possible. A shortened version is being published in the Peak District Mines Historical Society (PDMHS) members’ newsletter. If you find this article interesting, you may be interested in joining and participating in membership activities, but we also have public activities when pandemics allow.

I am avoiding technical detail, and not providing a “how to” guide; there is an abundance of information available on the web which should suffice, although a mining history focused “how to” guide may be created for PDMHS members if there is interest.

A Brief Introduction to Digital Elevation Models (and Lidar)

I am using “digital elevation model” (hereafter: DEM) as a generic term as well as two more specific terms: “digital terrain models” (DTM) and “digital surface models” (DSM). A DSMs give the surface with things on it – trees, buildings, walls, etc – whereas DTMs, in UK parlance, show the ground level (in the USA, a DTM is augmented with information about surface features such as rivers). Someone with an interest in forest canopies would normally be interested in a DSM, whereas we mining history people are much more likely to be interested in what is beneath the vegetation, hence in DTMs.

DEMs are usually created using the data from flying aircraft, with a considerable amount of sophisticated data processing to get from the raw data to usable DEM data. Radar has been used, although Lidar (an acronym for “light detection and ranging”) is the most likely source of DEMs which we will use. Lidar and radar use similar techniques: bouncing light or radio waves off objects and measuring the pulses which arrive at the sensor to “see” objects. Lidar equipment is extremely expensive. A cheaper alternative is to use photogrammetry, which entails taking numerous overlapping photographs from an aircraft, and using software to work out what the surface elevations must be to cause the observed changes between adjacent photographs. Photogrammetry can be achieved from professional grade unmanned aerial vehicles (aka “drones”). See, for example the 3D model of Magpie Mine, with online viewer, created by Peak Drone Imaging. Some specialist providers also fly Lidar drones.

What are DEMs Useful For?

Before listing what we can use DEMs for, it is fair to mention that even the higher-resolution images are no substitute for an experienced archaeologist on the ground. Such a person will see things, feel them under their feet, draw inferences from changes in vegetation, and observe traces of mineral, etc. Unfortunately such people are in short supply. Aerial photographs may also be more revealing than images created from DTMs, especially high resolution black and white images, but also when the character of vegetation is shown, although the best resolution of freely available aerial photographs is generally not good enough to compete with the latest Lidar-based DTMs.

Fieldwork (preparation) from your desk chair: before venturing out, 2m or higher resolution DTMs, can be a useful way of “seeing” what is there. This kind of prospecting can be really useful when the terrain is difficult. Alternatively, there may be places of interest on private property with no access.

Seeing though undergrowth is a major advantage of DTMs, making visible what would otherwise remain fairly well hidden.

Validating grid references and sketch maps; there are abundant imprecise locations from before the days of GPS, when sometimes an approximate 100m grid reference was the best that could be achieved. This also applies to British Geological Survey maps, which are often based on very old surveys and locate veins quite inaccurately.

Visualising the 3D character of the landscape is one nice application, and one which works even with 50m spatial resolution. This can be either quite subtle shading on a flat map, to given an impression of shadow, or a more virtual-reality-like presentation. Contour lines are really in this category too.

What Data is Available?

There are several good sources of freely-downloadable data (but with some licence terms):

Ordnance Survey OS Terrain 50 is a DTM with a spatial resolution of 50m. It is not available as GeoTiff (see below), but in alternative grid-based formats and as contours.

NASA mapped almost the entire land surface of planet Earth in the Shuttle Radar Tomography Mission (SRTM) at 1 arc second spatial resolution, which is approximately 30m. You can browse maps online and download the data using the USGS Earth Explorer, and there is a SRTM downloader plugin for GQIS (see below).

The Environment Agency has surveyed many parts of England over the last 20 years, at a range of spatial resolutions between 2m and 0.25m. This work was particularly driven by a desire to model and predict flooding, so the smallest resolution surveys closely follow draining networks near large urban/suburban areas such as Sheffield. Since 2016 they have been undertaking the National Lidar Programme, which will complete its survey of the whole of England at 1m spatial resolution in 2021/2 (Lidar surveys are taken during the winter months while vegatation is largely dormant). The vertical accuracy is claimed to be below 0.15m and both DTM and DSM datasets are available. This is all available under an Open Government Licence. This is outstanding!

While the OS and NASA datasets are useful for visualising the physical geography, a spatial resolution of 1m or less is essential for picking up mining historical features other than large scale open-casting. Consequently, most of the rest of this article will look at the Environment Agency data, in particular the 1m National Lidar Programme data as it is available in nice square blocks, whereas the 0.5m and 0.25m surveys are quite limited in coverage and tightly follow streams and rivers.

Visualising DEMs

The simplest way of representing a DEM is as a square grid of altitudes, and this is the most likely way in which data is provided. The grid spacing, known as the spatial resolution, typically varies from around 100m down to 0.5m, or sometimes as small as 0.25m. A grid of elevations is also quite easy for computer programmes to process. The most favoured current format is GeoTiff, which is an extension of the Tag Image File Format (Tiff) to include data about the geospatial location which the data relates to. The simplest way of rendering a DEM for viewing is to have the elevation values correspond with different brightness values, from black (lowest elevation) to white (highest elevation), so GeoTiff is a neat solution. A bit of desk research will turn up lots of alternative file formats. Fortunately, most geospatial software can handle several of them.

Unfortunately, just opening one of these GeoTiff files on your PC will usually just give you what looks like a plain white image, but with the right software we can get something looking a little surreal or medical.
(All images are linked to a larger version)
We can do a bit better by creating a pseudo-colour image, where the elevations are mapped to a colour. The colour scheme can be semi-naturalistic like a traditional atlas, a simple grading from one colour to another via white, a rainbow or a garish scheme designed to highlight a particular elevation range. One really effective technique is to generate “hill shading” by  simulating light and shadow from an imaginary sun. If you visit the OS Terrain 50 web page, there is an interactive map where you can adjust the azimuth to alter the hill shading. Hillshade works well when overlaying a base map or aerial photography.

Grey-scale representation of a DTM. Monsal Head is at the top and Lathkill Dale at the bottom

Wye to Lathkill Pseudocolour DTM

Wye to Lathkill Hillshade Overlaying OpenStreetMap

Hill shading turns out to be quite effective at picking out even quite slight surface features with the higher resolution DEM data. When hill shading is used to augment a map, steep slopes can appear very dark, so people often use multi-directional hill shading, with three or more imaginary suns all shining at the same time. This sounds very unrealistic, but the images are pleasantly usable.

Some other ways of representing a DEM are as contours, which are not so easy to work with “as data” but super if you are map-making, and as the surfaces of 3D objects, which would be useful for creating photo-realistic images and “virtual reality” environments. This article will not look at 3D visualisations.

Visualising DEMs on the Web

My top recommendation is to use the Environment Agency Geomatics Team’s Lidar Composite viewer, but there are other services to explore via their Open Data Products page. This is a very interesting resource which includes clear information about Lidar, the accuracy of the data, and the way raw data is turned into a DTM or DSM. At the time of writing, the latest data which is included is the “2020 Composite”, but this will change as the National Lidar Programme progresses (link to catalogue of all completed and planned surveys).

Environment Agency Lidar Composite Viewer, with key controls annotated: Map layer, Base map, Spatial Resolution, and Placename Search.

At this point, I urge you to go and explore! The most important controls are the spatial resolution selection and the map layer. Once the Map layer control is opened, click on the little eye symbols to switch different effects on and off, or on the three dots to change the opacity of the layer. Experiment with hill shading, pseudo-colour, overlaying hill shade onto base maps (you will have to decrease the opacity of the hill shade), and looking at slope and aspect. The area shown in the image, above, is Grin Low, above Poole’s Cavern in Buxton, providing a dramatic illustration of the legacy of lime burning which is only partially evident as one walks around the woodland. The Thatch Marsh and Burbage Colliery area, not far to the West of Grin Low shows the causeways and pit locations quite clearly, as well as the pack horse hollow-way leading from the pits towards the lime kilns where much of the coal was consumed. The area and its history is thoroughly described in “Coal Mining near Buxton: Thatch Marsh, Orchard Common and Goyt’s Moss” by John Barnatt, in Mining History 19-2 (not currently online) which has several maps drawn from proper fieldwork, which make for an instructive comparison with Lidar armchair exploration.

An alternative viewer is the Environment Agency Survey Open Data Index Catalogue. This gives access to more datasets, including DSMs (although only in the 2017 Composite, not the 2019 version), but I don’t find it quite as pleasant to use.

For the More Adventurous – QGIS

QGIS (formerly Quantum GIS) is open source software with professional-level geographical information system (GIS) capabilities. Software with such power comes with challenges, and I would only recommend people who are confident IT users even taking a look. I hope those who are confident will be able to repeat the following examples, and are maybe motivated to learn more. There are a lot of resources explaining how to use QGIS on the web; just be aware that the latest version is QGIS 3, and many resources refer to the previous release. If you do download QGIS, choose the latest “long term release” version. I am going to use some more technical terms in this section, and will not explain them all, expecting readers can do their own web searches.

Before doing anything else, I recommend adding OpenStreetMap as a base map layer, if only so you know where you are! This is achieved by adding an “XYZ Tiles layer” (see this guide). The URL to add is “{z}/{x}/{y}.png” (without the quote marks, but with those curley brackets). There are lots of XYZ servers available – try a google search – but the Bing Maps satellite imagery service is a good companion to the OpenStreetMap: “{q}.jpeg?g=1“.

The starting point for all which follows is QGIS with a map in view and two panels – headed “Browser” and “Layers” – on the left hand side. If these panels are not visible, use menu View > Panels.

Getting Data for QGIS

I am only going to consider using Environment Agency data, for which there are two ways of using the DEMs in QGIS: downloading GeoTiff files, and accessing a WMS map tile server. Once set up, the WMS approach is quicker and easier but you get less control over how the DEM is rendered. Consequently, I suggest using the Defra Survey Data Download service. The workflow for using this site is generally: draw your area of interest on the map, click “Get Available Tiles”, then choose which dataset you want. Ignore the shapefile upload option; the icons for drawing your area are just beneath the upload grey box. This catalogue covers all the published datasets, with downloads being 5km x 5km squares. At 1m spatial resolution, that amounts to 5000 x 5000 = 25 million data points, so even these tiles are quite large files.

Once downloaded, you can just drag-and-drop the file into QGIS. If you do this, you will see that QGIS chooses a grey scale for each based on black = the lowest elevation and white = the highest elevation. This makes things look blocky if you load several 5km square tiles since each has a different min/max elevation. This can be fixed quite easily, but doesn’t matter for some uses (see below). I generally work with the OpenStreetMap and satellite imagery as the bottom layers and put DEM layers above (and the notes below assume this).

Playing with Hill Shading, Pseudo Colouring, and Overlaying Maps

For this section, we will work within a single 5km tile. I chose the SK16NE DTM from the 1m National Lidar Programme (aka DTM_1565) to work through what follows, which has Magpie Mine almost at the centre. The Magpie Mine is easily accessible and has been thoroughly surveyed and described, so it is a good place to see what Lidar data can (and cannot) reveal. Choose a place you know!

All of the following is achieved by double-clicking on the DTM layer in the “Layers” panel then choosing the “Symbology” option. When you open this up, it will not look exactly as below, because the default “Render type” is the rather boring “Singleband gray”. Change this to “Hillshade”. Also change “Zoomed in” to “Cubic”; if you don’t do this, the appearance when zoomed in looks a little bit like linen (this is an easy thing to forget). You must either “Apply” the changes or “OK” to close the properties window for the settings to take effect.

To begin with, “Multidirectional” will be un-ticked. In that state you can play around with the altitude and azimuth (compass direction) of the imaginary sun for which the shading is created. It will quickly become clear that different features are revealed for different angles, but that if you are interested in showing several features, there isn’t a good answer. This is where multidirectional hillshade comes in; it combines the effect of several simulated suns (3 for QGIS). Since the hillshade algorithm uses gradient and aspect to determine the shade of grey, it will work if you have several 5km tiles loaded.

To combine the hillshade with either the OpenStreetMap, change the “Blending mode” from “Normal” to “Multiply”. This merges the DTM and the map images so that the map appears to be draped over a 3D surface. I find this the best way to interpret features in relation to landscape features. A similar effect can be obtained with the Bing aerial photography layer, although if there are buildings and trees in the area of interest, it may be better to use a DSM instead. Combining the DTM and aerial photographs can often be very revealing since the photographs can show variations in vegetation or surface material in topographically indifferent ground, whereas the DTM reveals aspects which are indifferent in ground cover.

An alternative to hillshading is to set the “Render type” to “Singleband pseudocolor” (the TIFF images are “single band” because there is a single value for each pixel, which corresponds with the height, whereas colour photographs usually have three bands with separate red, green, and blue values per pixel). The magic button is “Classify“! There are lots of colour schemes to experiment with – change “Color ramp” – but most are horrible.

The Min/Max values can be useful to limit the colour range to the elevation range in a particular area of interest. A 10m range can nicely bring out spoil heaps and hollows. These values can also be used to force several 5km tiles to have the same colour scale, so to appear seamless.

Most of the other options will not be useful for general experimentation, but the brightness/contrast/saturation slides can be useful for preparing images for print.

Some tricks:

  • Right-click on the DTM layer and choose “Duplicate Layer”. Now set one of the two layers to be hillshade and one to be pseudocolour. This can really bring out relief.
  • Make your own multidirectional hillshade by duplicating the DTM layer 2 or more times and setting the azimuth separately on each layer. This can help to reveal features which are not showing nicely with the QGIS fixed azimuth angles.

Here is a composite image of Magpie Mine using the DTM mentioned above (so the buildings have been magically transported away). The spoil heap near the 1869 engine house and the reservoir are easily seen, as are the main veins outside the heavily re-modelled central area, but can you make out the four gin circles and the crushing circle? Gin/crushing circles are much easier to see on the ground than from a DTM when the terrain is quite flat. How about the covered flue from the square chimney to the long engine house, the slime ponds and dressing area, and the straight tramroad from Dirty Redsoil into the centre of the main site? In this case, adding the pseudocolour actually makes it harder to make out the more subtle features where the changes in slope and aspect are more significant than changes in elevation.

Composite of Magpie Mine comprising OpenStreetMap base map overlain by a multidirectional hillshade and a single band pseudocolor constrained to the altitude range of 315m to 325m. (Click to open larger image)

QGIS can also generate contour lines from DTMs; change the “Render type” again. The contour options are fairly self-explanatory and work well down to as low as 1m interval with the National Lidar Programme data. The “index contour” can be used to make every 5th, 10th, etc contour be styled differently. The “input downscaling” setting default of 4 usually works well; this smooths out the contour lines to make them less jittery as the limits of the spatial resolution (and to a lesser degree the vertical accuracy) come into play. Larger values make for smoother contours.

I find there isn’t a single magic setting which works for all cases; different sites and different exploratory questions indicate different settings, but there is enough variety and power in what QGIS provides that there is usually a satisfactory approach, within the limits of the data.

Creating Terrain Profiles

Since the DEM encodes height on a grid, it is possible to construct profiles. The standard QGIS plugin “Profile tool” is the easiest way of doing this and there are a few more advanced plugins available. The tool is quite self-explanatory, so here is an example without a “how to”. The area around Wass’ Level (also known as Moorhigh New Level) on Maury Rake is in an accessible grassland location just South of the old Midland Railway bed through Millers Dale. A visit to the site allows the spoil heaps and settling ponds to be easily seen along with the presumed collapsed level at around 245m elevation, at the edge of the woodland. The Lidar image reveals some interesting features up-slope which cannot be seen from the open area but the elevation helpfully reveals enough to motivate a bit of struggling through the undergrowth. Could the feature at 255m be evidence for the excavation from the surface of a winze down from Wass’ Level to the now-lost Upper Level (destroyed when the railway was build just above it)? The paper entitled “The Maury and Burfoot Mines, Taddington and Brushfield, Derbyshire” by John Barnatt & Chris Heathcote in Mining History 15-3 describes the site and its history.

Aside: the vertical line near the right is just an artefact of having used two DEM tiles which have been separately hill-shaded; the hillshade algorithm can’t work out what the slope/aspect is at the edge of a tile. This can be avoided by combining the tiles into a single layer in QGIS, which should be done for publication but I’ve left it in here as there is educational value in showing and explaining the issue.

The Area Around Wass’ Level (SK149730) on Maury Rake, Millers Dale.
(Click for larger image)

Some Oddments…

Once you have got to grips with the basics, there are numerous plugins and processing algorithms to experiment with. The Visibility Analysis plugin uses the DTM to determine those parts of a map which show what can and cannot be seen from a vantage point. There are algorithms which will calculate surface slope and aspect, or roughness (Processing Toolbox > Raster Terrain Analysis); these values can then me mapped to colours using the same technique as above. Mapping slope to pseudocolour does a better job of revealing gin circles than hillshade when the terrain is quite level.

Using a WMS (Web Map Service) is generally more convenient than adding 5km square Tiff tiles in QGIS but the available services provide a combined hillshaded and pseudocoloured image and only a DTM is currently available (things might change). The separate services for 2m and 1m DTM (etc) are:


The following URL may be added as a WMS to discover the extent of coverage without the DEM: . NB this will not work in a web browser, but the same source can be accessed using an online service via web browser.

The elevation data is available (as opposed to a hillshaded and pseudocoloured image), but only using the ArcGIS Image Server protocol. In QGIS this requires the use of the “ArcGIS ImageServer Connector” plugin, which works but the image redraws rather slowly. The URL to use is of the form:

We can hope that, once the National Lidar Programme has concluded, the WMS provision will improve, along with the rather confusing array of data download, WMS, and online map viewers. Quite a few things changed during the drafting of this article.


OpenStreetMap base maps have been produced from the OSM raster map tile service. This is © OpenStreetMap contributors and used under the terms of the Open Database Licence:

Digital terrain models used for creating profiles and DEM visualisations are from Defra, used under the Open Government Licence v3.0:

Searching Ordnance Survey Names for Evidence of Mining History

This article describes some informal/experimental work in which various sources of data were processed to find traces of mining history in the names of places.

A few of the results from the place name search, displayed over an Open Street Map

The Geographic Area

My area of interest is broadly-speaking the Peak District and adjacent areas. For practical purposes, I have defined both a detailed boundary and a rectangular bounding box. The latter takes in all of the former, which is defined using Parish (and similar town/urban) boundaries. Since the boundary of Sheffield stretches into outlying areas which are of interest, this way of defining the boundary ends up including the city and Eastern areas. The same is not the case for Greater Manchester. Noteworthy cases where areas outside the Peak District are included are: the Churnet Valley and Alderley Edge.

The bounding box is (E-min, N-min, E-max, N-max): (381355, 340577, 445079, 402069)

Coordinates will either use fully-numerical eastings and northings to 1m, or use 100km letters (SK covers most of the area) and a variable number of digits.

Where source data is chunked according to 100km or 20km tiles, the small excursion of the boundary north of 400000 is neglected.

Data Sources and Pre-processing

All data sources are ultimately from the Ordnance Survey, but with some qualifications:

  • OS Open Map Local* for NGR squares SJ and SK, was used, limited to shapefile data files “NamedPlace” and “Road”.
  • OS Open Names* is provided in 20km tiles; CSV files with the following names were used: SJ 84, 86, and 88; SK 04, 06, 08, 24, 26, 28.
  • OS 1:50,000 Scale Gazetteer is no longer available from OS but I had a copy from 2009 on disk. This contains names which are not present in currently-available OS downloads. It was processed to extract entries occurring in the same area as used for Open Names (the data includes the tile designator).
  • The Visions of Britain “GB 1900 Gazetteer” (the abridged version) was initially limited to the area extent then an attempt was made to remove entries which are descriptive of features (e.g. “Old Lead Mines”) rather than being names. This is a somewhat subjective exercise. Proper names for lead mines are quite common in this gazetteer; these were left in, although the main intent of the activity is to find place names which had arisen from mining activity, rather than finding named mines.

* – obtained from the OS Open Data download area.

A final spatial filter was applied in all cases to limit the input data to the “detailed area”.

Search Terms

The range of possible name-parts of interest is split up, largely according to variation-spellings of the same root, but with some “misc” categories which contain various thematically-related words. The categories/terms are given below, with the category name in bold.

  • coal: words starting with “coal”, “cole”, or “collier”. Historically, charcoal was also referred to this way.
  • pit: names ending, or having a word ending, in “pit” or “pits”.
  • lead: words beginning “lead”, “led”, “lyd”, “lidgat” or “lidyat”.
  • mine: words starting “mine”.
  • cost: words starting “coars”, “cost”, or “coast”. These could be due to Old English names indicating a mining trial (see PDMHS Bulletin 7-6 pp 339-341).
  • mining misc: a word beginning with one of “rake” or “raike”, “delf” or “delph”, “gin” or “engine”, or “ochre”.
  • bole: a word containing either “bole” or “brenn”. The latter catches “brenner”, the operator of a bole.
  • smelting misc: terms other than “bole”, which signify smelting: “” (where . is any letter), “cupola”, “pig”, or “slag”.
  • misc: catches names with words containing “jagger”, “belland”, “forge”, or beginning with “copper”, “bloomery”, “furnace”, or “furnes”.

While it is clear that these search terms will include obviously-erroneous names, these are not excluded from the results; given the status of this work as “for interest and as stimulus”, this feels appropriate.

Search results were further processed to try to remove unwanted multiplicity arising from either the same name appearing in several sources, and from linear features such as roads having multiple entries. This was done by checking for identical names within 500m. This sometimes fails, as different sources may differ in capitalisation or make composite words such as “Bolehill”.


Three map exports are available:

Three maps were created due to difficulties combining all the options using the Leaflet technology which underpins the web maps. For maps which do not immediately show the names, simply place the mouse cursor over a point.

Some Observations

Some of the search terms will give less-obvious false “hits”, and so all should be taken with some caution; place name specialists rely on written records from far earlier than those used here. Some example false friends are: “coal” might originate from “cold”, although we might reason that local geology makes the former more likely; “lydgate” is often though to derive from Old English “hlid-geat”, a swing gate. I will consider the case of “lyd” in a later post.

Almost all of the hits for “mine” are named mines, with a few exceptions such as “Miner’s Cottage”, “Miners Hill”, or “Miners’ Standard PH”.

Licences and Acknowledgements

This work includes data from:

The maps were created in QGIS using the QGIS2Web plugin and the Leaflet web mapping open source toolkit (see links at the bottom of the map screens).

The maps use data from Open Street Map and the National Library of Scotland, under an open licence for non-commercial use.

This blog post and the derived data created by me are licensed under Creative Commons BY-SA.


Adding British Geological Survey (BGS) WMS Layers to

This took a little while to do; there doesn’t seem to be anywhere explaining how to add WMS layers and controls in simple terms… so here it is for anyone else. Pieced together from various places, the leaflet documentation, and some guesswork, here is how to add WMS layers for BGS solid and superficial deposit geology and linear features to a view. The layer opacity has been set to 0.5 to allow the base map to be seen.

You will need to access the developers’ console (F12 on Firefox).

It will look like this:

Enter the following commands where the “>>” is:

var solid = L.tileLayer.wms("", {layers: 'BGS.50k.Bedrock', format: 'image/png', version: '1.3.0'});
var drift = L.tileLayer.wms("", {layers: 'BGS.50k.Superficial.deposits', format: 'image/png', version: '1.3.0'});
var linear = L.tileLayer.wms("", {layers: 'BGS.50k.Linear.features', format: 'image/png', version: '1.3.0'});
var layer_control = {"Solid": solid, "Drift": drift, "Linear Features": linear};
L.control.layers(null, layer_control).addTo(;

I find I have to leave the console now.

One thing to note is that the BGS WMS will not return an image if you are very “zoomed out”. Zoom in until the scale bar shows 500m or 1km.


Other BGS WMS layers to add:

var artificial = L.tileLayer.wms("", {layers: 'BGS.50k.Artificial.ground', format: 'image/png', version: '1.3.0'});
var movement = L.tileLayer.wms("", {layers: 'BGS.50k.Mass.movement', format: 'image/png', version: '1.3.0'});

Chinley Churn Underground Quarrying (Mines)

From a visit with Beth Knight earlier this year, here are the locations of some underground workings on Chinley Churn. This post is not a recommendation to visit; the workings contain dangerous loose rock and areas liable to collapse.

Point Name WGS84 Longitude WGS84 Latitude OS Grid Ref. 10m Altitude
1 -1.94595 53.34832 SK 0369 8349 407.5 m
2 -1.94561 53.34882 SK 0371 8355 416.6 m
3 -1.94563 53.34927 SK 0371 8360 415.2 m
4 -1.94546 53.34964 SK 0372 8364 423.6 m
5 -1.94625 53.35133 SK 0367 8382 422.1 m

Photographs showing the entrances, in the same order as the points above:

Chinley Churn Entrance 1


Chinley Churn Entrance 2


Chinley Churn Entrance 3


Chinley Churn Entrance 4


Chinley Churn Entrance 5

Orchard Common and Drystone Edge Geological Field Trip Notes

The following are field notes from a visit to the valley just to the W of Drystone Edge to inspect the Upper Namurian, especially the Ringinglow coal outcrop and associated seatearth and marine shales (contorted beds). The locations were logged using a Garmin GPS60, and may be downloaded in GPX format. The points are named in the format “ORC %N”, and referred to below.

ORC 001 –  SK 02262 68251 – Roadside parking.

Proceed N up the road to Readyleech Green. Just past the house, the road swings NW, following a fault with downthrow on the SW side. To the right is the Upper Namurian Chatsworth Grit (CG), while the higher ground to the W of the road (Knotbury Common) is Westphalian (Coal Measures) Woodhead Hill Rock (WH). The right of way track around the S side of Knotbury Common approximately follows the lower boundary of the WH.

ORC 002 – SK 02198 69014 –  Shaft Hummock. This is in the shales between the Chatsworth Grit (CG) and Rough Rock (RR). The coal sits just above the CG.

Section through Drystone Edge and Orchard Common
Section through Drystone Edge and Orchard Common looking North

At this point you are on Open Access Land. Descend to the stream, which is easy enough if you follow the fence approx Southwards on its stream-side.

ORC 003 – SK 02170 68842 – Good clean exposure just above the stream bed of  CG followed by seatearth then coal. There may be a detectable marine band here but I didn’t look. The seatearth shows interesting patterning. There is also some evidence of fossilised vegetation.

ORC 004 – SK 02330 69112 – Contorted Bed.

ORC 005 – SK 02319 69450 – Quarry in Rough Rock. Look at the texture, grain size, bedding etc to compare with CG on Drystone Edge (ORC 009). The top of the RR is the top of the Namurian; a band of marine shale with distinctive fossils marks the bottom of the Westphalian (Coal Measures).

ORC 006 – SK 02598 69608 – Contorted Beds.

ORC 007 – SK 02680 69725 – Some nice pieces of loose coal (apparently not in situ, but 5-10cm longest dimension).

ORC 008 – SK 02781 69373 – CG boulders showing Karstic features.

ORC 009 – SK 02698 69152 – Scarp exposure of CG on Drystone Edge. Look NW towards the RR quarry visited earlier and consider the section (above), which follows the same line.

A 5km x 5km square of the solid geology with a lower left corner at SK 000 650 may be obtained using the British Geological Survey map server using the link,365000,405000,370000&WIDTH=500&HEIGHT=500

These are personal notes, shared for whoever may find them useful. I am not a qualified geologist; more details may be found in the Geologist Association Guide No. 26, by F. Wolverson Cope. I used this book to visit the site, although the account above contains additional sites, and more precise location for some of Cope’s sites.

Look up: upper carboniferous cyclothems.

Using Mapyx Quo to Create Virtual Field-trips with Google Earth

This post describes an approach I have adopted to create virtual field trips using Mapyx Quo as a starting point to produce interactive annotated virtual field-trips. The initial impetus was to look at the geology around Millers Dale, specifically at the outcrops of the Millers Dale Lava Beds. Details of this trip will follow in the next post.

Quo is reasonably good but the KML exports it produces loose some of the folder-like structure that can be used inside Quo. It alsi has the very annoying feature of inserting timings into all “tracks”, no matter whether theses are logged from a GPS or inserted using a mouse or tablet inside Quo. Tracks are the best way to define areas and linear features for visualisation in Google Earth but timings and associated waypoints just make for clutter. Furthermore, the colours set for tracks and waypoints in Quo is lost on export.

My solution was to write some software (for MS Windows) to:

  1. tidy up the KML export, removing unwanted aspects and putting back some folder structure
  2. generate a javascript configuration file to drive a web page that uses the Google Earth Plugin (API)

The resulting KML may be used with the desktop Google Earth or from a web page (with #2 as the intended case).

The source code is available on GitHub, with an MIT licence. At some point I will produce an installer (sooner if someone asks). The source code should be consulted for specifics that are omitted from this post; there are sufficient comments for non-readers of C#.Net.

There are a number of assumptions about how Quo will be used, and how the software (Quo2GE) will be used, and this post is really an aide memoir for myself.

Organising Waypoints and Tracks in Quo

In Quo, create three groups (or two groups and make the first two be layers within the same group), and as content as follows:

  1. A group for points of interest.
    • Within this, create waypoint sets. For each set, specify a different “Point Name Pattern” consisting of a number of alpha-numeric characters followed by “-%N”. The characters comprise a prefix that will be used in Quo2GE to group the waypoints (Quo destroys its own group/layer/waypoint-set structure on export to KML). Note that the “-” sign MUST be present.
    • Any routes will be ignored by Quo2GE.
    • Quo2GE assumes three types of waypoint sets, which can be assigned ad hoc: sites to visit, landmarks, and places that make good viewpoints. These are treated slightly differently by Quo2GE, see below, with implications: for landmarks, add the name of the landmark to the “Note” field for each waypoint; for viewpoints, a place to look at from the viewpoint may be specified by adding to the “Note” field a “@” character followed by the name of any other point of interest, followed by a space character if it is not at the end of the “Note”.
  2. A group containing the walking/driving routes and maybe waypoints at key turns.
    • The waypoints will be ignored by Quo2GE but may be useful if a GPS is used in the field.
    • Several routes may be added, for example to break an over-long route into more walkable units or to separate walking and driving routes. Distinguish the routes by prefix, changing the default “Track 1” etc to something more meaningful. Note: for routes, use a space character (rather than “-” as before) to separate the prefix from the index number. Use different prefixes to allow different groups of 1 or more tracks to be separated in Google Earth. Otherwise use the same prefix.
  3. A group containing geological/geographical features designated using the Quo “track” tool.
    • Use layers to organise the features by type, e.g. a layer for geological faults.
    • Within each layer name the tracks with a prefix followed by a space and then an index number. e.g. “Fault 1”, “Fault 2” etc. The prefix is used by Quo2GE to group the features and to allow different colours to be used for each group.
    • Waypoints will be ignored by Quo2GE.
Example of Quo structure.
Example of Quo structure.

Export each of these as a separate KML file (right-click on the group or layer and “Export to…” (file), selecting Google Earth (kml) as the file type).

Quo2GE should work if any of the above groups/layers are omitted.

Using Quo2GE


Use “Set Output Folder” to locate the KML files exported from Quo. The same folder will be used for the output of Quo2GE.

Load each of the exported KML files using the appropriate tab. The list on the left (with check-boxes) should enumerate all of the prefixes used for waypoint or track names in Quo. Use the checkboxes to de-select any that you do not wish to see in Google Earth.

For each prefix, the colour to be used in Google Earth may be changed, as may the title (which is initially set to the prefix). The title will be shown in the web page to allow groups of waypoints or tracks to be shown or hidden. They also appear as folder names in Google Earth Desktop.

For points of interest, there are three types that can be assigned to each prefix-set. The points are handled differently in Quo2GE and in the web page:

  1. Points of type “Site” are intended for places a person would visit. In the virtual tour, users can jump to have an aerial view of any of these. The points are shown in Google Earth with the waypoint name, e.g. WPT-002.
  2. Points of type “Landmark” are intended as key landmarks (!). They are shown in Google Earth with the waypoint Note (e.g. “Hammerton Hill”).
  3. Points of type “Viewpoint” are intended as places a person would view their surrounding from (in real life and in the virtual tour). In the virtual tour, users can choose to go to these places; the Google Earth view changes to a near-ground-level view looking north, or towards any point specified in the Quo waypoint Note using “@”.

Set the Main Title and click “Export”.

Using the Output

Several files will be created in the chosen output folder. Those starting “clean” are tidied-up versions of the KML files exported from Quo but without any organisation into folders, application of styles according to Quo2GE. The file allSets.kml has been produced from the cleaned KML and does contain the folders and styling. Use this with  Google Earth desktop.

Use on the web requires use of allSets.kml, gm.html and config.js. hg.html reads config.js to locate the kml file to be used and to set the controls that appear in the web page to allow the view to be manipulated.


  • edit config.js so that the kmlHref variable contains the full URL to the ultimate location of the allSets.kml file.NOTE: the kml must be on a web server for the Google Earth plugin to be able to use it, although you can run gm.html from your local machine.
  • add text, change styling etc for gm.html according to your taste.
  • copy all three files to the web space.
  • enter the URL for the gm.html file in your web browser…. it should all appear!

As an alternative, which may be useful if you have several virtual field trips is to place the gm.html file in one folder and to create sub-folders for each trip, which contain the kml and js files. In this case, tell gm.html the name of the sub-folder (e.g. “trip1”) by adding it to the end of the URL, after a “?” (e.g.

LED Caving Lamp – Adapting an “Oldham” Caplamp

Some incomplete notes from a recent conversion of 3 old cap lamps to run off white LEDs extracted from a cheap (yet astonishingly bright) headtorch and powered by Lithium-based cells salvaged from an old laptop battery. One of the 3 caplamps was from a 1991 Apex 2 and the others were ex-NCB pit-lamps of unknown age.

Raw Materials

  • some pieces of brass (see below)
  • a 5W, 270 lumen headtorch powered by 3xAAA cells. Cost less than £5 incl postage from Hong Kong via ebay. I bought a few cheap torches looking for suitable de-mountable modules (see the photos below). A simple on-off type was chosen rather than multi-mode (bright, dim, flashing…)
  • A pair of 18650 3.7V Lithium cells with solder tags or salvaged from a laptop battery
  • A cheap “travel charger” for these cells. I bought a pair of cells with a charger for about £5 via ebay and use the cells in a hand-held zoomable torch that serves as a caving backup and dark-corner-probe.
  • A potting box and potting resin (Maplin does suitable)
  • Heat-sink compound (see Maplin’s again)
  • a few solder tags, a couple of M3 machine screws
  • 4x #40-40 stainless steel cap-screws at least 3/4″ long (this is an American standard thread. Try radio control model suppliers)
  • An old caving lamp
  • A through-headset charging part. I had one of these “lying about” but they may be hard to find these days. If so, adapt the design with charging connectors on the battery some-how.

Headset- Components to Make and Fitting

Remove all the components from the headset except the  negative wire (fixed to the body) and the switch moving part and its spring-contact. I actually removed all parts to overhaul everything.

See the dimensioned drawings and photo of the finished result below (both are available in larger size by “clicking” the image). The dimensions on the plate are approximate; be prepared to make some adjustments and in particular note that the photo shows a small cutout to clear the on/off switch. I tapped the holes to secure the LED modules but you could just drill to clear an M3 screw and use a nut on the rear.

Cap Lamp Parts to Make
Cap Lamp Assembled

The smaller part on the drawings is a pillar that will be located at the top of the lamp where the pilot lap is. This needs rounded corners to fit. For the Apex 2 caplamp, this fits neatly without any work on the caplamp body. For my other two, there is a lump to machine away. I used a slot-drill in a pillar-drill but a router-cutter would probably work. Beware that this process exposes some brass. This brass is actually part of the (negative) circuit embedded in the body and it must not be allowed to contact the pillar. The pillar should, however, have a good contact with the body in order to conduct heat away from the LED. I simply milled a recess over it to avoid contact.

Heat-sink compound was applied between the LED module and the plate and both top and bottom of the “pillar”.

The #4-40 screws will have to be made the correct length: 2 at 3/4″ and 2 at slightly less than 5/8″ (both dimensions include the head of the cap-screw – sorry I failed to measure thread-length.) . These screws fasten the brass plate to the cap lamp body. “Clear” in the diagram means that the screw will easily pass through; these are plain drilled holes.

The wiring is mostly obvious from the photograph. The negative wire from the LED is attached to one of the switch contacts using a solder-tag. The other switch contact is not used and was not refitted. Don’t trust the pictures – work it out!

Small padding washers are used for the LED module hold-down. These could be old inner tube, although I used some gasket paper. The idea is just to reduce the risk of damage.

The reflector will need to be sawn off as in the photograph. A Dremel tool (etc) is quite handy for this. Find a way of marking off the cut height or mount the Dremel at a fixed height over a flat surface to get this cut neat, even and to avoid cutting off too much. The LED module I used had quite good combination of central beam and spill-out and the end result with the reflector as shown is quite satisfactory.

Battery Pack

This is just a pair of 18650 cells in parallel cast in potting compound. A piece of rubber was placed beneath the cells to avoid direct contact with the box anywhere. Make sure the hole for the cable is tight otherwise compound will leak out. I would have used a grommit but bought the wrong size. An outer protective layer of inner tube and gaffer tape finishes the job off. This will he helmet mounted and only be used for light caving so is not “bomb proof”; I do not expect to bash my head against many long awkward passages.


This basically involves mounting the “travel charger” on the side of a box, having soldered mains supply to its circuit board and soldered some take-off wires for the head-set contacts. It is probably best to avoid charging via headset at the same time as separate cells.

Finished Product

NB: the Apex 2 headset was bodged to allow for through-headset charging.

GPS/GPX Waypoints for PDMHS Newsletters 122-133

The download below is a GPX file suitable for GPS and digitak mapping software. The locations are of sites mentioned in the Peak District Mines and Historical Society newsletter editions 122-133. These are not sites I have GPS-located but have been transcribed from grid references in the newsletter. These are sometimes 6 figure OS Grid refs and sometimes 10 figure. They may be wrong! The comments are generally very brief and the intent is that reference is made to the appropriate newsletter, using these “waypoints” as a spatial index. The waypoint names are of the form “PD-N122-03” which means the reference is in Newsletter 122 and is the third one (generally in the “Observations and Discoveries” section).

(At present Newsletters are not on the PDMHS website but many of the old journals are. Join PDMHS!)

GPX download: pdmhs-newsletters-122-133-for-web

Using the British Geological Survey WMS Service with Mapyx Quo

Major kudos to BGS for making their 1:50,000 geological maps available online including via a WMS Service.

I’ve been struggling with scanning BGS maps and importing them into Mapyx Quo for a while. Its an arduous process and the scanned maps contain the roads, placenames etc as well as the OS topographic map I have in Quo so its not as pleasing to use as it could be.

I was thus motivated to write a little Windows .Net programme to access the BGS WMS and automate the import into Quo. The process is not completely automated (partially because Quo uses a non-XML file for storing the user-loaded maps) but is still quite simple.

The programme is not as polished as it could be – its good enough for me – and has only been written with the BGS service in mind (although it could well work with other services).

I am making the Windows installer (I use XP still) available (no warranty blah blah) but please do not distribute it (refer people to this web address instead): download installer (about 350k)

Usage is simple:

  • In Quo set the coordinates to WGS 84 decimal degrees
  • Right-click and copy the location of interest to the clipboard
  • Paste it into WMS2Quo (my programme)
  • Choose which layers you want. e.g Bedrock and Superficial (bedrock is the default)
  • Click “Fetch”. This uses the geonames service to find the name of a nearby location. This will be used in the name of saved files. You can edit/change this in a text-box.
  • (the segment of the geological map should appear)
  • Save the map and a Quo calibration file
  • In Quo “Explorer”, select the “Loaded Maps” tab and use the document-with-green arrow icon to import the map image. This will cause the image AND saved calibration file to be read. I usually now change the transparency to 80%. You should see the geological map in Quo.
  • You can “query” the map one layer at a time to find the kind of rock (etc) at a given point by clicking the mouse on the image. This information is remembered and can be saved out as a GPX file containing “waypoints” for each location. Import this into Quo and set it to show the waypoint “note” and you will get geological labels showing.

There is a “settings” button which can be used to alter the save location, image size etc. Take care and NB that the BGS WMS server will sometimes return a blank image if your image size/map size combination are out of its range. A known bug also means you need to restart the programme if you change the save location. I recommend you change the save location as the first thing you do. Also, watch out for your firewall blocking web access; if you get an error on “Fetch” this is the first place to investigate.

I would like to acknowledge Paul Dixon ( as I adapted code of his (GPL Open Source) for the coordinate transformations that are used. This is JavaScript whereas I used C#.Net. I’ve uploaded this for use, adaption or what you will under the same licence: DLL, Source code.

If you would like the C# source code for the “WMS2Quo” app please contact me. Similarly, I’d like to hear of any bugs (you know what I mean by “like”).

GPS Waypoints for Bradwell Area Caves and Mines

The following GPX file uses data from HNH/PeakDistrictCaving, specifically their index to the Bradwell Catchment (PDF).

I created this after a trip around Bradwell Dale, mostly looking at the geology, and compared with a set of GPS waypoints and the tracklog. Based on this, I am sceptical of the accuracy of the Grid Ref for Bradwell Parish Cave. I would place this at SK17255 80605.

Bradwell Catchment GPX file.