Project Diary 2019-12-19: Greater Christchurch Maps Update

In the last few weeks our research has continued of Greater Christchurch maps, branching out into looking into railway housing in Heathcote and Lyttelton. The development of railway housing in large cities is naturally very different from small towns, as these two examples show. We plan also to take a closer look at railway housing in Addington, but not in any other part of Greater Christchurch except where it is within a station precinct.

The research has come to a pause for now due to the holiday season, but we intend to take a few weeks' break even when Archives reopens in the first week of January and probably catch up on some other areas of the maps project (i.e. some areas outside Greater Christchurch). There are only a few stations' worth of files to look through within the GC area, perhaps only at most two more months of research. The maps themselves haven't progressed much recently, but the break period will be used to get progress in that area as well.

Railway Housing in Greater Christchurch [1]: Introduction

As we all know, when NZR developed their network, staffing requirements especially in remote locations often required housing to be provided. There are a number of townships in NZ where the amount of railway staff housing was significant compared to the overall community, including several railway towns - Otira on the Midland Line being one of the best known examples and one of the few that is still substantially unaltered today.

In the course of the Greater Christchurch research several housing settlements have come to light around the city, some of which have long gone and others which still exist in some form, however in general Kiwirail does not have any houses of its own these days and only occasionally leases properties as needed such as at Otira. At a major station there would generally be a stationmaster's house which was separate from most other housing, the latter tending to be a row of houses close together, and generally without driveways or garages although a small shed was commonly seen on the majority of sections.

As we generally know, NZR housing became a major focus around 100 years ago to the extent that a housing factory was built at Frankton (the buildings still exist in some form) and prefabricated standarised designs were churned out for a number of years afterwards. Hence it becomes possible to identify ex-Railway houses by sight in some areas because of the characteristic appearance. Wherever possible, Railway housing settlements were often placed at the edges of railway yards.

We haven't specifically researched Railway housing in the Greater Christchurch area just yet but intend to do so as it is becoming a topic of interest and the map key does include a code for marking houses which has been undertaken on other maps we have developed, most notably the Otago Central line, so having stumbled across a number of files specifically related to housing in our research currently on Lyttelton, we now intend to research the housing more intensively in order to document it as part of the maps.

Our research some weeks ago on Christchurch Station, for example, documented that a house for the stationmaster may have been provided somewhere in the area in the early 20th century, but it was not too long before NZR started to pay their stationmasters allowances to live in their own houses elsewhere. For example, paying the telephone connection and monthly fees for the Christchurch stationmaster was undertaken for many years in recognition that they would be expected to be available on-call most of the time when not actually working.

Housing at Lyttelton appears to have been located all over the township rather than being clustered close to the station. So far we have uncovered that NZR owned houses as high up as College Road and Brenchley Road (these locations are near the northmost limits of housing development on the eastern side of the township). As our aerial photos at the moment don't go that high, we may have to extend them if possible in order to show the housing on maps.

Heathcote is an interesting place to investigate as well because a housing development was undertaken c. 1960 at Railway Terrace, between the upper section of Martindales Road, Port Hills Road and the Tunnel Road. At this stage we can only speculate that the intention was to provide for a number of staff housing needs in the general area rather than specifically for Heathcote, as it is evident on the map that there were approximately 20 houses constructed, and Heathcote has never been a major station that would have required a significant number of staff - by that stage probably only a stationmaster and/or signalman. Since NZR sold the housing, some of the sections have been redeveloped but there are still some of the staff houses in place. Further research has yet to be completed on this settlement and the other NZR housing that is known to exist around Heathcote.

Then-new housing development at Heathcote in 1961. Martindales Road has been extended to give access to the tunnel road formation during the construction of the Lyttelton Road Tunnel project.

View of some of the wider area in 1970, including a water reservoir on the hillside above that no longer exists today.

Same view as the first map but taken 2015, the settlement has changed much but some of the houses are essentially the same.

No survey of railway housing in Greater Christchurch would be complete without looking at the settlements around Addington Railway Workshops. At this stage we know little about this housing except that due to the extensive changes in the transport landscape that have taken place in Addington over the past 30 years, even some of the streets these houses were in have disappeared.

The first area that may have been a railway settlement at Addington is directly north of the Shops site in Lowe Street. Earlier aerial photos show this was almost all in housing, whereas today there is hardly anything left due to it becoming mostly commercial and light industrial in nature.

Addington in 1940. The area that may have been a railway housing settlement is bounded by Lowe Street, Tyne Street and what later became Blenheim Road (known as Alliance Street at the time). There already seems to have been some industrial development in Pope Street in the area opposite the saleyards that later became the NZR Publicity and Advertising premises.

 The same area of Addington in 2015, some 75 years later. Tower Junction occupies the Shops site, and the housing has almost completely disappeared from view. In fact there was just one of the old houses in a run down condition which appears to the left and slightly below the 0.5 km peg. But the latest imagery from Canterbury Maps (below) shows it too has gone recently.

The other main area of Railway housing we are aware of in Addington to date is directly to the west of the Shops. This was never huge - there may have been only about 20 houses there at its peak - but the streets themselves were removed completely when Tower Junction was developed.

The NZR housing settlement in Margaret Street, Levin Street and Bell Street, as seen in 1940. There were still some houses there in the 1980s although parts of the land by that stage were being used for Shops purposes such as carparking. Clarence Street at that time was the major through route of the area, with a level crossing over the Main South Line just west of the present passenger station.

The same area much altered in 2015. Whiteleigh Avenue has become the major roading route with a new level crossing and Clarence Street is now closed off and (as Troup Drive) ends at the entrance to the present Christchurch Railway Station (Addington).

Project Diary 2019-11-15: Greater Christchurch Maps Key Milestone

We are pleased to announce a key milestone has been reached in the Greater Christchurch maps we are producing as part of the NZ Rail Maps project. We have now completed the historic aerial photo mosaic projects for all of Greater Christchurch, that is, the following four sections of railway:

• Eastward: Main South Line from Christchurch to Lyttelton
• Westward: Main South Line from Christchurch to Rolleston
• Southward: Hornby Line from Hornby to Lincoln
• Northward: Main North Line from Addington to Rangiora

This means the remaining steps needed to produce the actual maps for Greater Christchurch are:

• Extract map tiles from the mosaic projects for GIS use 
• Complete basic historical research for every station and siding in the Greater Christchurch area
• Draw and label the maps for every location.

The extraction step is a slow one because of the use of multiple generations of historical aerial photography in the mosaic projects - the same reason why it has taken so long to produce those mosaics in the first place. In addition, once the resulting tiles have been imported into the GIS, it is common for mistakes to be detected at this stage, requiring more work to rectify.

The historical research stage is also slow because of the number of files that need to be consulted. At this stage we have been researching up to 50 files in a typical week over the past 3 months, which in total has been in actuality 500 files to date. We estimate it will take another 6 weeks of work to complete this but there may not be more than 200 actual files left to look at.

After obtaining the various diagrams that we have copied as part of the historical research, some of these will need to be imported into map mosaics where they are the only source of information for a particular historical feature. An example is the pumping station siding at 2.81 miles on the MSL that was the subject of a recent post. Since this did not appear on an aerial photo, the copied drawing was added into the existing aerial photo mosaic and another set of tiles (only 2 in this case) dated 1924 were produced for the GIS in order to be able to draw the maps that accompanied the blog post. There is likely to be more of this needed, so the mosaics will have to be added to with the diagrams that have been copied. However, unlike the aerial photos that have been used in the projects to date, most of these diagrams only cover small areas and therefore don't take very long to be added in. So that will not in actuality cause much additional delay, as most of the diagrams we have copied in actuality are just for labelling features that we can already see on aerial photos.

Then finally the maps get drawn and labelled. How much work is needed for each yard depends on how big the yard was. This in turn informs the complexity of the task. In the past we have gone to great lengths to document as much as possible historical changes visible on different generations of aerial photography. Our focus has changed with Retrolens because of being able to get the NZR yard and corridor surveys that are not available on our previous main source for the region, which was Canterbury Maps. This means we have changed focus into having really only one or two complete historical layers for a yard or section, which will be a yard survey if there was one, and a corridor survey if there was one that shows significant changes from the yard survey. There are a very small number of yards that have had two NZR surveys, and the corridor surveys are additional to these. In the case of a small number of yards there is no yard survey (Heathcote for example) and we are relying on other historical surveys that are available of the general area. For some stations these will be highway surveys if the yard was alongside State Highway 1 or another highway corridor.

Previously we would go to great lengths to try to document which parts of a yard in the modern era were based on historical features visible in the yard survey that might have been done 40 or 50 years ago. However as you can imagine this requires a lot of work, not the least because it is difficult to exactly align the historical aerial photos to modern ones, and because in the case of some yards (Lyttelton, Middleton and a few others that come to mind as examples) they have changed so much that there has been major track realignment. So right now the focus is on drawing every track in the historical layout as a "former" or "closed" track and then drawing the tracks again as "current" or "open" on the contemporary aerial background, and not bothering trying to work out which bits are common to both. Just that we will use the filtering capabilities built into the GIS to display the appropriate generation of track and structures etc with the aerial photo generation that is applicable.

As we have noted in previous posts the actual map drawing is the furtherest behind at the moment as we have focused most of our time in the last few months on the mosaics and research work. This has been changed this week with the completion of the mosaics and even though research is not completed, we are currently progressing as much map work as possible, with Lyttelton and Heathcote progressing well at present. There has just been a post about part of Heathcote and our next post will be about Lyttelton. However we still have to do historical research for Lyttelton so the maps posted in the next post will be necessarily limited until we can find out more historical information about the general layout of Lyttelton. On the other hand it is possible to push ahead with Heathcote to full completion because we have done most of the research on that yard although there is still a small amount, maybe one day's worth, to be carried out.

So expect to see a lot more posts about specific yards and locations in Greater Christchurch in the coming weeks. Once that is done, then the overall project work will shift back to the rest of NZ as detailed in other posts in the last few months with the total project schedule. We can then expect to see the volumes produced for all of the other maps for the whole country.

Project Diary 2019-11-02: Christchurch Maps

Our first blog post for November. 

Map production this week has been limited to a complex fix for a problem discovered in the Heathcote-Ferrymead maps for 1985. Trying to align two different rail lines in four places simultaneously can be very complex. Cutting the Main South Line out through the back of Ferrymead is a partial solution. The rest of the fix was to cut the aerial photo covering Ferrymead in half and align the two halves separately. Due to the amount of work needed, no action has been taken with other generations, although most of them satisfactory as they are. Most of the intention in having all the generations is not to document Heathcote, but more to do with the early history of Ferrymead.

Research this week has taken in Addington Station, including all of the private sidings, Templeton Station, and Papanui Station.

Project DIary 2019-10-26: Christchurch Maps

It's been another busy week of gathering research information and a lot of work pulling out map tiles. As noted in previous posts, we spend a lot of time each week, currently around 6 hours, researching all the stations in the Greater Christchurch area. This will involve viewing around 50 files a week and copying mainly track diagrams from them in order to be able to label everything on the maps properly. We have spent most of the time actually just gathering the information and very little lately has been put into maps because the process of creating the historic mosaics and getting map tiles out of them is relatively time intensive. 

So at the moment things appear to progress slowly while the main focus is in these phases of gathering information and investing a lot of work into creating the historical map background images to show the various locations around the Greater Christchurch railway network. This week's research had a look at Linwood Station which was a passenger station on the MSL between Christchurch and Lyttelton. It was located between Linwood Loco Depot and the Wilsons Road crossing and it served exclusively suburban passenger traffic. Like the other suburban stations on the "Port Line", Linwood closed in the early 1970s when the commuter trains to Lyttelton ceased operating, and there is practically no trace of it to be found today. However, NZ Railways Corporation still owns the entire Linwood site that it occupied until the mid 2010s and which is now leased to a number of entities. 

We also took a look at the early history of Linwood Loco, but aren't specifically interested in the entire history of that site, and haven't yet examined all of the records relating specifically to Linwood Loco. We also took a look at Opawa Station, which was the next stop up the line from Linwood and was also passenger specific. Ashley is another small station near Rangiora on the Main North Line and it was between those two stations that a couple of minor realignments of the MNL took place in the 1960s and 1970s relating to the replacement of two bridges, one of which is a major structure crossing the Ashley River. Ashley Station closed in the early 1960s partly as a consequence of the construction of the Ashley Bridge deviation but a siding is still in place for the nearby Canterbury Timber Products factory although disused for some years. Addington Station and also the Way and Works "Plant Zone" and the Signals Depot have been researched and we are currently looking at the numerous private sidings around Addington, including the NZ International Exhibition of 1906 siding which was laid from Riccarton Station across Hagley Park, and the lengthy Fletchers Siding also known as "The Alley" which served many industries and ran from Addington to nearly Riccarton Road.

Map tile production this week has taken in Heathcote and Ferrymead, so that we now have full sets of tiles covering 1940, 1950, 1961, 1970, 1973, 1976, 1985, 1995, 2000, and we also have the Linz aerial coverage for the 2000s, 2015 and some additional Canterbury Maps coverage for updating parts of Heathcote and Ferrymead to 2019. Normally for a smaller station like Heathcote we wouldn't cover such a large number of generations, however the interest for the widest range of coverage is more geared to Ferrymead, which was in significant development from the mid 1960s to the mid 1980s. Now having these full sets of maps will enable work to go ahead particularly with a high priority for Ferrymead; given that we already recently got full map sets for Lyttelton but have done very little with them so far, our actual resources (time) for map drawing itself are a bit scarce at the moment.

A partially completed map of the Ferrymead Railway Moorhouse Village site in relation to the Main South Line in 2019.

The Ferrymead site in 1961, when it was still a dairy farm.

Project Diary 2019-10-18: Christchurch Maps

Progress on the maps this week has continued to focus on the Main North Line section of Greater Christchurch. This inches on towards completion with, we estimate, just a week more work to finish it. Quite a bit of work was needed just to add a 1955 aerial image showing the old route between Chaneys and Kaiapoi (deviated in 1958), and the similar but less lengthy realignment between Rangiora and Ashley (1961) is also going to be included. We also expect to show the first couple of km of the Eyreton Branch having found an image dated 1963, one year before the purported closure date. As of this moment, all of Belfast from 1950 to 1985 has been put in, and what is actually left is some of the imagery for Papanui, Styx, Kaiapoi, Eyreton Junction, Southbrook and Ashley. Flaxton isn't currently being mapped, but if I can find anything about it in Archives I may add it.

The extraction of the map tiles for any part of G.C. has proceeded quite slowly. The updates for Lyttelton have been extracted, paving the way to move on to Heathcote / Ferrymead. Once the maps themselves are completed then we can focus our full resources onto extraction. This is quite an involved process as it is not uncommon

Only one day of research has been possible this week, focusing mainly on Addington Station. So far we have avoided looking into Addington Workshops, planning to leave that for another time, but will try and find out if there is any coverage of Plant Zone.

In the wake of the local government elections with the result not showing much potential for the desirable level of change we sought, we are considering what level of campaigning to continue with. Christchurch Transport Blog has more details.

Project Diary 2019-10-12: Christchurch Maps

Local government election results are in and the unfortunate conclusion is we will not see much that improves Public Transport over the next three years in Greater Christchurch. More about this in Christchurch Transport Blog in the next few days.

Meanwhile Greater Christchurch maps are the main focus this week again and in recognition of the time this is taking, a project development schedule post went out earlier this week allowing for several months more work for the rest of the year and possibly the earlier part of next year on completing the GC maps.

The work is continuing on the last section of the Greater Christchurch map tile mosaics which is the section from Riccarton to Ashley. The tile work has been pushed ahead in the last few days. The research is also going ahead quite well with 50 files a week being worked through from Archives New Zealand's local office. However major areas are still to be researched including the stations of Lyttelton, Heathcote, Woolston, Opawa, Linwood, Addington*, Papanui, Belfast, Kaiapoi, Flaxton, Rangiora, Ashley, Middleton, Sockburn, Islington, Templeton. (* Addington has partly been completed, we have not yet looked at either of the Workshops or Plant Zone however and will have to prioritise these). The research has to do a lot of catching up as we only decided to start researching Greater Christchurch stations towards the end of August.

The tiles for most of Lyttelton have now been extracted but a small change put into the mosaic means re-extracting a small number. In the coming week we can expect to see tiles extracted for Heathcote and Ferrymead.

Project Diary 2019-10-03: Christchurch Maps

Well here's another update, and promises, promises, promises.... the last week has slowed things down a bit due to having to deal with other issues, but we are back up to speed now. Plus there have been more than a few issues with Gimp that caused crashes, meaning having to go back and redo stuff and try to find out what wouldn't crash.

The reason there are issues with Gimp and needing a lot of hardware resources (a PC with 32 GB of RAM and a 200 GB SSD swap partition to use for the tile cache), is our preference to use the highest resolution imagery we can. For base imagery that is a resolution of 0.075 metres per pixel (i.e. 7.5 centimetres) and everything snowballs from there. Take for example Lyttelton. We could just do the main railyard but have chosen to cover all the areas where tracks went to, and that means a canvas of 12 tiles wide and 4 tiles high. That translates into a size in pixels of 57600 by 28800, or 1.6 billion pixels total. The highest resolution imagery from Retrolens (usually around 1:4300 scale but occasionally 1:3400) fits in well with that base resolution. By the time we've added in a few eras of historical imagery (from 1940 to 2000 generally) then the result, in this case, is a file size of 21 GiB. It's only since Gimp 2.10 came out that it's been possible for it to handle files greater than 4 GiB in size, which is a real blessing because otherwise multiple files would be needed which would be very inefficient and time consuming. 

We started out some of these Gimp projects covering multiple stations but then as more layers were added (we use a rule of thumb that no project will have more than 100 layers) some of these projects grew too big and had to be split into multiple files. That happened in the case of Lyttelton, which used to have Heathcote along for the ride, which is what was happening earlier this week when everything kept crashing because the system or Gimp hit some sort of limit. Heathcote is a big area to cover because it takes in Ferrymead, so the result was a lot of area to cover and eventually the need to split the files, which we can now recognise because the current version of the Heathcote file is 30 GiB and that is the upper limit of file size we're prepared to work with, from experience. With a file that size it takes ages to load into Gimp, the system will easily use half the total tile cache just to bring it up with no edits, and it can take three hours to save (and Gimp can still crash partway through the save if the file is too big and lose everything). So you can see we are working with some real limits.

The resource issues for doing these highest quality images is the key issue that affects the kind of coverage we do of stations around the country. Christchurch is a particular exception that is going to be the highest quality for all of Greater Christchurch, across multiple generations, and backed up by research as well. That simply can't be justified across NZ, and only major yards will get done to that level. Most sites will be done with just the NZR survey, an example being Bluff that was recently draw up, it was done to high quality but only one generation of historical images. With corridor surveys we can cover the smaller stations as well, but if we can't get coverage of them when they were open, there won't be any. Before we dropped the Midland Line coverage to focus on the Greater Christchurch stuff, we had covered just about all the stations to Kotuku with corridor surveys plus the major stations and some research as well. The same will go through for the MNL and MSL eventually as far as we can get to.

Anyway back onto Greater Christchurch, we only have the MNL from Riccarton through to Ashley to complete. The tiles for the rest will start to get spat out as quickly as we can from here, because map drawing is way behind and really needs to be pushed along as quickly as possible now. So all that MSL stuff will be pushed out as map tiles and then be drawn alongside finishing that stretch of MNL as quickly as possible.

Project Diary 2019-08-31

In today's update, we can report we have finally completed the historical map mosaics for the section of the Main South Line from Woolston to Waltham, after splitting the Gimp project into two parts to deal with persistent crashes caused by project file size. The last section to be added this morning was the historical coverage of the section for the year 2000, which is the most recent aerial photography available from Retrolens.

The historical eras covered in these mosaics of the Main South line are 1940s, 1950s, 1960s, 1970s, 1980s, 1990s and 2000s, with Linz aerial photography available after 2000. NZR official station surveys cover from Opawa to Waltham in 1970, Woolston in 1974, and Woolston to Waltham in 1981. NZR corridor survey is available from 1984, but only for Waltham; for some reason, an entire run (B) is missing from this survey. The rest of the coverage is general for Christchurch and of a lower quality and only the NZR surveys in general will be able to fill in the historical data for the section.

To back up the map data, this week we also have commenced three half-days of research at Christchurch Archives at the rate of one half-day per week for the Christchurch Station, which includes Waltham. We are prepared to carry out the more intensive scale of research just for Christchurch, and do not intend to spend any significant amount of time on the other stations, each of which would have less than one half-day spent on them. In each half-day session we would typically expect to view and copy content from around 40 files, and the time scale of one half-day per week is governed by limits on the rate at which Archives can produce and issue the files for our research. The main purpose of this research is to obtain additional map information about the locations and track layouts of sidings and time limitations mean we are not able to take in general historical research of the areas concerned, even though much of the information about Christchurch in the 1980s and 1990s is very interesting, being an era we lived through and had extensive knowledge of from a personal perspective.

At the moment our focus is going more into producing all of the mosaics as quickly as possible for Greater Christchurch, and less into drawing actual maps, and that is why we are publishing project diary updates lately rather than map coverage of the areas. This is driven by the needs of Christchurch Transport Blog for map tiles for project work, being a higher priority than NZ Rail Maps at present. Once all of the Greater Christchurch map tiles have been completed, then NZ Rail Maps can go back into map development around the country according to its existing goals. This of course also means the original project schedule for NZ Rail Maps that we outlined for 2019 will not be kept. We estimate at the moment it may take for the rest of 2019 to be able to complete the Christchurch Transport Blog map tiles, and therefore will have to rearrange our goals. 

This will mean the overall schedule for completing the bulk of the NZ Rail Maps project work for all of New Zealand will be altered. Against that, having increased resources has improved our efficiency considerably this week and will speed up the mosaic production.

Tile extraction for Woolston to Waltham will start early next week, so expect to see some actual historical mappables displayed in the next update. Mosaic production is shifting to the Middleton to Hornby section of the Main South Line, the Lyttelton to Heathcote section of the Main South Line (including Ferrymead) and the Bryndwr to Ashley section of the Main North Line which as far as possible will be worked on more or less simultaneously, as is now easier to achieve with the resourcing improvements.

Project Diary 2019-08-29

This week work is still continuing on the historical map tiles for Waltham, Linwood, Opawa and Woolston. We have gained some additional resources this week to enable us to speed up the tile production work, so we are also simultaneously working on tile production for Middleton to Hornby section of the Main South Line, and the Bryndwr to Ashley section of the Main North Line. So the production of these tiles should be able to be sped up noticeably, as part of the reason for it being slow is the inherent slowness Gimp has when working with very large mosaic projects. Some of these cover a canvas surface having a working area of 3.5 billion pixels (3,500,000,000) and around 100 layers in total. The Woolston-Waltham project is now completed and tile extraction will be commencing and could be completed tomorrow but will probably have to be deferred for a couple of days due to other commitments, and we have had numerous Gimp crashes, so it has been slowed down a lot from what we expected last week.

The issue with a city like Christchurch or any main centre is you are looking at an entire section of a corridor, rather than a number of individual stations along it. Because there are sidings and stuff all the way along that corridor that you want to document, and that is certainly the case for the Main South Line from Woolston to Islington in particular, and on the Hornby Industrial Line as far as Prebbleton. It is not quite the same on the Main North Line which has historically much less industrial development. So that dictates Gimp projects that cover large areas. We could split these into multiple smaller projects but then there is too much problems with overlapping at the edges of these projects because those edges all have to line up when the tiles are put together in Qgis, and the historical aerial photos boundaries don't neatly end on a tile boundary. So a smaller number of larger projects is definitely the way to go, but we have had a few instances of having to split some of them up when they got too large. For example Addington to Ashley started off as one project, but then Addington and Riccarton got combined into another project with Christchurch and it ended up too large to put the rest of the MNL suburban area into it, so Bryndwr to Ashley is what it has ended up as. Same with Lyttelton to Waltham originally being one project that now has Lyttelton and Heathcote in one file and Woolston to Waltham in another. And so on. 

The extra resources mean the time Gimp wastes on saving large projects, which can actually take several hours to complete, can be used to work on another project at the same time. So in fact today we really were working on three at the same time at one stage because the first one was being saved. After completing the progress on the second one, it too was saved, and while that was happening, we worked on the third. Another time saving measure is to use Cubic interpolation for unified transforms instead of LoHalo. My hunch is that Cubic is a lot faster than LoHalo and makes very little difference to the quality of practically any layer we throw at it.

Project Diary 2019-08-22

Since last week's diary report we have continued working on the maps for the Christchurch area. Putting together all the aerial photos can take a while to complete but we now have complete historical aerial photo tiles for Christchurch, Addington and Riccarton, and have started working on ones for Waltham, Linwood, Opawa and Woolston. These ones should be ready in a couple of days. Then after that we still have, as far as Greater Christchurch goes, to look at Lyttelton, Heathcote, Middleton, Sockburn, Hornby, Islington, Weedons and Rolleston on the MSL; Prebbleton and Lincoln on the Hornby Line; and Bryndwr, Papanui, Styx, Belfast, Chaneys, Stewarts Gully, Kainga, Kaiapoi, Flaxton, Southbrook and Rangiora on the MNL, to complete all of the stuff for GC.

Research has shifted to the GC area as well. After a break from the Midland Line work is starting with Christchurch Station which will take about 3 weeks to complete with over 100 files to view. The plan is to look at every station within the GC area in order to be able to fully detail the maps being drawn of the area.

Project Diary 2019-08-12

This week we have switched back to the Christchurch area maps due to the impeding local government elections. We are currently working with several candidates and preparing material in order to assist some of their campaigns, in association with Christchurch Transport Blog.

Because of this the Midland Line work has temporarily halted, although research at the local branch of Archives New Zealand has continued. We estimate approximately two more days of research at Archives will be sufficient to complete the Midland Line maps, but are unsure exactly when these maps will actually be ready.

We are taking the time to complete more of the historical maps of Christchurch as well as some of the other stuff we are doing so I have pulled up a number of Gimp projects and revised them as necessary to keep them within reasonable file sizes. For the Main North Line it is now split into one project that covers Addington and Riccarton along with Christchurch (from MSL) and a second one covering from Bryndwr to Ashley, previously it was all one project from Addington to Ashley. 

For the Main South Line we have Lyttelton to Heathcote including Ferrymead in a project, then from Woolston to Waltham, then from Middleton to Hornby, and then from Islington to Rolleston. The reason Christchurch was put in with Addington going north to Riccarton was because we had Addington in two projects, one for the MNL and one for the MSL, which was an unwarranted duplication.The one we haven't started on really is Hornby to Lincoln, and it's likely we will take Hornby out of the MSL Gimp projects at this point and put it into the one for the Hornby Industrial Line.

It's timely to comment briefly about maps of local government boundaries. Linz Data Services does not have this type of map in the list of layers that can be downloaded through their site on Koordinates. However, Statistics New Zealand, another Government department, does have them through its Datafinder site, http://datafinder.stats.govt.nz also hosted on Koordinates. As a result we have taken the opportunity to download six layers together: Maori Constituency 2019, Constituency 2019, Community Board 2019, Regional Council 2019, Territorial Authority 2019 and Ward 2019. This information is needed in order to discover the boundaries of each of the wards for the Regional Council elections and in so doing, be able to assist certain candidates in the wards of the transport related issues in their constituency.

Project Diary 2019-06-02

In the past week there have been two main areas of work in this project. The first area is the Christchurch City maps referred to in a previous diary entry. It has been decided to combine existing maps of the city created for non-rail purposes with the existing rail maps data drawn for the city, which means there will be a combined Volume 10 and Volume 11 Qgis project that covers only Christchurch. To make things easier this project will share the existing Volume 10 and 11 layers with the rest of those volumes existing projects and will not have separate layers except for the aerial photos that are not of rail corridors or yards. At the same time the existing Volume 10 and Volume 11 projects will have all aerial photography removed that is inside the boundaries of the city.

The other area is the NZ Rail Maps website on Wordpress.com. This currently is accessed by the address http://nzrailmaps.wordpress.com . It will generally be known that the project also uses the domain name nzrailmaps.nz which currently redirects to a Trainweb site for historical reasons. Given the length of time that the Trainweb site has not been used for the project, it is likely the redirection will be changed soon to point directly to the Wordpress site. However there is no intention for the present to change the site URL to nzrailmaps.nz as this requires a paid Wordpress hosting plan. The other change has been driven by the decision that the Wordpress site has enough free space to host the map volume PDF files, which means they can be removed off Scribd thus reducing the number of external sites used to host project content. However the free version of Wordpress is not conducive to hosting the map tiles which are best to remain as Google Photos albums for the time being. Generally we are well behind in publishing completed project content and this needs to be put together and put on the site as more of a priority in future.

It is hoped that the computer used to do the map mosaic tiles will get a bigger SSD soon to enable it to deal with very large project files more easily, or to be able to have two 100-layer projects open at the same time.

Main North Line [0E]: Volume 10 Progress Report [4]

As we have reported in the last few posts in this series, we have been busy collecting aerial images for the Main North Line corridor from Christchurch to Picton and this work is now practically complete. Although a few extra downloads from Linz have been needed, most of the work has been completed and 150 GB of disk space was recovered by removing all of the download files that had been previously collected for the MNL corridor, which was the main object of this exercise. The majority of the work has been the effort needed to locate as many Retrolens aerial images of old stations as possible. 

Apart from the corridor surveys from NZR from the 1970s and a small number of dedicated station surveys, we have been fortunate to be able to draw on corridor surveys of State Highway 1, where it has been located close enough to the rail corridor, and since the earlier series of these date from 1961, it has been possible to capture many stations at an earlier time when they may have been open, and have closed by the time the NZR surveys were carried out.

Our knowledge of the Main North Line in Marlborough has been added to by discovering a previously undocumented series of realignments that were made on the former Picton Section prior to 1947/48 which in many cases is when the earliest historical aerial photos are available for the area. One of these is at Hog Swamp Creek just south of Seddon and is seen below.

We assume the key reason was to eliminate and bypass the bridge there, probably when it became due for replacement, as the curve was only slightly reduced in radius.

The next three photos show a series of realignments between Tuamarina and Para.

We don't have these mapped at present and it will be some time before we expect to draw these onto maps as probably mosaics will have to be made to enable them to be traced readily.

As one would expected some other useful discoveries have also been made along the way. 

Just north of Wharanui is the Waima River (previously known as the Ure River). Wharanui was the southernmost extent of the Picton Section, to which the line was opened in 1915. The section between Wharanui and Ward took four years to complete, which included 15 km of track, the No.3 tunnel (currently No.21 on the MNL and known colloquially as the Tar Barrel), and the bridge. This was the last single deck combined bridge on the Main North Line. To our best knowledge there were in fact only two single deck combined bridges on the whole of the MNL, the other one being at Parnassus and that was replaced about 1940. The road bridge at Clarence was originally intended to be a combined bridge, but in fact when the railway eventually reached Clarence in the early 1940s it got its own bridge. The double deck combined bridge at Seddon became rail only in the early 2000s when the new state highway bridge was built, but the redundant lower deck is expected to be reinstated for a cycleway.

Waima River combined bridge in 1969.

By 1972, work was starting to replace the bridge. It looks like the new railway bridge was built first.

By 1975 both new bridges had been built alongside the original. The highway originally crossed the railway on a one lane overbridge north of the river. This was replaced by a new overbridge south of the river, which took several more years to be built.

Another major river crossed in Marlborough is the Wairau between Spring Creek and Tuamarina. We understand there may have been some erroneous suggestions this was a combined bridge but that is not the case.

1937 aerial photo of the bridge. The highway at that time was some distance downstream and did not cross the river alongside the railway, as it does today.

This 1977 aerial shows work was underway to build a new curved railway bridge alongside the existing straight structure. Anecdotally we have heard the replacement was needed due to scouring of piers in the original but have not confirmed this. 

To conclude this post here is a 1986 view of the Awatere River double deck bridge just north of Seddon. There were a number of occasions that the road deck was closed for repairs (the bridge was built c.1902 and the deck was wooden) and generally, a temporary road was put through the riverbed as there was no convenient detour. The route of the most recent temporary highway on the upstream side of the bridge can be seen fairly clearly here (it was different from other occasions) and sealed highway in the riverbed, albeit on the downstream side, was still visible until the new highway bridge was built, after which it seems to have been dug up.

Main North Line [0D]: Volume 10 Progress Report [3]

We have decided to push ahead with the MNL corridor LDS images and setting up Gimp mosaic projects as a top priority because of the disk space issue. As mentioned before the downloads for this particular corridor are presently using up lots of disk that we want to recover.

As it turns out we already have most of the base imagery downloaded and selected so most of what has been worked on today has been ascertaining how much Retrolens coverage there is of the Main North Line stations. 

NZR had basically two types of aerial surveys done for them: station specific surveys, which are easy to spot because the run number is always A, meaning the survey/run/photo number sequence will usually be XXXX-A-1 through to usually no more than 5, another giveway is the earlier ones always have a scale of 1:4300. The other type of survey done is for a whole route. Generally these were just the main lines and were done in the 1970s. A long corridor like the MNL or MSL will be broken into several surveys that must have been a lot of fun to fly as these long main lines are far from straight so the plane would have made many passes to capture all the different runs that make up each survey.

North of Christchurch there are relatively few station surveys as they are generally only for the major stations such as Belfast, Rangiora, Amberley and Waipara. However for many of the other stations, the route surveys done in 1976/77 often are showing the many small stations that were still open at that time (surprising great number as many are long gone these days). We intend to incorporate these into mosaics later on but for the present we are only setting up mosaic projects rather than actually making mosaics. Deviations and realignments also provide reasons for mosaic making.

The only completed set of mosaics north of Rangiora to date are the ones for Amberley, Waipara, Hundalee, and former Tunnel 4 near Oaro.  As was mentioned before, we did Picton under a different system some time back. These will be redone under the current system. A couple of stations on the Waiau Branch have also been mosaiced and the rest that we have coverage for can also be done in the future.

Between Parnassus and Wharanui the base coverage we have at the moment is the Kaikoura earthquake coverage shot just after the 2016 Rotherham quake. We are currently considering whether it is suitable to use as a background for the maps. At the moment that is what we are using, but if we decide it isn't suitable, there will be a lot of work to download the previous base images and they are not as sharp as the EQ stuff, which is at 0.2 metre pixel size.

Main North Line [0C]: Volume 10 Progress Report [2]

Of late we have been working on some of the rail corridors around Christchurch by amalgamating Gimp projects covering single station sites into projects that cover a whole section of corridor. This includes both the MSL and the MNL. The first one for the MNL covers from Addington to Ashley, which except for the area covering Flaxton station, which appears to be a tiny flag stop between Eyreton and Southbrook, is continuously in 0.075 metre imagery from several different layer groups.

Since we have already downloaded imagery for nearly the whole MNL, we are pushing ahead to get the entire MNL corridor organised in my computer. It is important to do this because we have so many downloaded aerial layers that are consuming massive amounts of disk space on the computers and need this space to store the mosaics that are going to be created for the project. As an example we have currently just over 200 GB of downloads for the MNL and could expect that further mosaics for this corridor might use 100 GB, giving a net gain of perhaps 100 GB. 

We need to do this work for every of the 12 volumes of the maps project but few of them have as much aerial imagery downloaded as for the MNL which has probably got a lot of duplications, but also it is a long corridor and therefore will have a lot of basemaps.

After Ashley, the little station on the north bank of the Ashley River 33 km north of Christchurch, the next section of mosaics would be at 0.3 metres and take us from Sefton through to Balcairn. We skip Amberley and Waipara, which are in 0.125 metre coverage from Hurunui District Council and will be in their own project, and carry on with stations north of Waipara and I haven't really looked at what there is good coverage of. We are not actually drawing maps but we are laying groundwork to be able to draw the base maps at a future time and the mosaics after that. Since there weren't that many major stations on the MNL we are not yet sure what aerial coverage will exist. Addington-Ashley covers really the old suburban passenger network north of Christchurch. Amberley and Waipara have been worked on recently as a separate effort, and north of there we started looking at a few places.

Some may recall there was an earlier aerial mapping effort that produced images of Picton, amongst other areas. That is indeed the case. It was carried out using similar principles but the technique was to overlay the historical aerial images into Google Earth and then trace over them, and then import the traces into Qgis overlaying them on current Linz basemaps for the produced output. This effort has been overturned because of licensing concerns with Google's maps system which we definitely want to avoid. We published a series of articles on Picton in October 2017. We did our first map project in Gimp, which was for Alexandra on the Otago Central Line, in March 2018. Since then we have fully embraced this combination of Linz basemaps with Retrolens overlays produced in Gimp and imported directly to Qgis for the tracing and it can produce historical map outputs as well. This means Picton will be redone in Gimp for a new set of maps sometime.

This need of completing the base maps of a corridor being urgent has caused work to be suspended on the MSL maps for the time being but we expect to be drawing some of them again this weekend even as we continue to finish off the MNL corridor, the only reason we haven't been doing any this week is because of other work priorities.

Main North Line [14]: Rangiora

Rangiora is the major intermediate station in the former tablet section from Addington (Christchurch) to Waipara (junction of the Waiau Branch). It is also one of the largest towns in rural North Canterbury. It is 30 km from Addington, almost exactly halfway between there and Waipara. 

The Great Northern Railway from Christchurch reached Southbrook, 3 km to the south, in 1872, and was then in the broad gauge of 5 feet 3 inches (c. 1600 mm). It was opened to Ashley (c. 3 km north of Rangiora on the north side of the Ashley River) in 1875, then to Balcairn in the same year, and to Amberley in 1876. This was the maximum northern extent of broad gauge in Canterbury, and it was converted to the NZ standard gauge of 3 feet 6 inches (1067 mm) in 1878.

Rangiora was from 1874 to 1959 the junction for the branch line to Oxford, which from 1884 to 1930 had an extension to Sheffield (a station on the Midland Line) via the Waimakariri Gorge bridge, which is still in use today as a road structure. The Oxford branch line also joined at Bennetts Junction with a line from Eyreton (Kaiapoi) that was first opened in 1878 and progressively closed between 1931 and 1965. The working of these branch lines required a locomotive depot at Rangiora and the turntable used to turn the steam engines remained long after the end of branch services into the 1970s at the north end of the station yard, while an engine shed was closed some time earlier.

Rangiora's passenger station has been at its present site for decades. A branch train platform behind it was still in place in the mid 1980s, but seems to have been demolished about 30 years ago. The importance of freight traffic at Rangiora can be seen in the extent of the sidings, nearly all removed including several private spurs on the west side. Rangiora's main importance today is as a passenger halt for the Coastal Pacific service, and for crossing freights.

The aerial photos are in sequence from 1942, 1975 and 2014.

Main North Line [13A]: Glasnevin

Glasnevin was the station on the Main North Line that was located between Greneys Road and Waipara. Apart from the platform and building or shelter, there was a separate siding for the Amberley Lime Co and and for the ballast pit just north of the station. Just north of the pit, the highway used to cross over the railway line and go down to an old wooden single lane bridge across the Waipara River, upstream from the railway bridge. This was replaced with the present two lane concrete bridge about 1971/72. Like Greneys Road, Glasnevin was closed to the public in 1966. The date of the lime siding being closed is unknown to me at present, and when the ballast pit ceased to be used or the track, which is clearly visible on the aerial photo, was lifted, is also unknown.

The lime siding consisted of a facility for receiving lime loads from vehicles, which dumped into a pair of hoppers, from which it appears conveyors took the lime into a building or shelter under or in which rail wagons were loaded. The truck dumping facility with its two hoppers, made of concrete, is the part which can still be seen beside the railway track to this very day, the rail loading structure having been removed. The ballast pit remains clearly visible as it has not been filled in to any significant extent.

Lime loading facility remains seen in 2008 from the Coastal Pacific.

Glasnevin ballast pit which for many years was bare gravel but has in more recent times been dressed with soil and sown in pasture.

Here are the maps.