Friday 25 December 2015

NZ Rail Maps 2015 Review

It seems like a good time to review what progress has been made with this project in 2015 so here we are.

The community of the NZ Rail Maps project changed during the year with a shift to a secret Facebook group to work with the small community of interest that is helping me to develop the project. The Yahoo group was closed down. We did consider a public or closed FB group but the administrative workload is considerable when working with any type of group that has open membership. The NZ Rail Maps FB page was renamed to EnzedTransport and whilst I perhaps should have kept the page with its previous name, this was done because of the change to a FB group instead of a page. 

Due to the change of the type of group from public/closed to secret, a new public interface for the project was needed and therefore this blog was set up, and a significant amount of content from Enzed Transport's blog was migrated to it. However the posts that were from the other blog are still present there, but new content on this blog is not duplicated on the old blog. Articles from both this blog and Enzed Transport are automatically republished to the Facebook page. 

The need for a separate group and other separate structures is largely driven around the fact that this project is considered to be outside the mainstream of the railfan community in NZ and consequently there have been doors closed to me. Nevertheless the FB group does have significant people in it who were chosen due to the contributions they have made to the project and these people have been willing to go outside of their community to make their contributions, which are most appreciated.

Another major change was in the way data is structured in the project and where it is stored online. The data structure has been changed so that all of the Linz Data Service layers are incorporated into the online storage, whereas previously only the layers I had created were online. This means it is easy to make an archive that anyone else can access should that happen in the future. Because Microsoft has made yet another excuse for slashing the amount of OneDrive storage to a measly 5 GB, the online storage has been changed to Google Drive which supports 15 GB. At this stage I do not have a new bitly short URL for the new storage location, but some of the old ones will not work anymore, so older maps URLs at the bottom of each published map will be incorrect.

The main internal change in data structure was made possible because Qgis supports rule-based styles. What is a style you may ask? A style is simply the visual representation of the stored map data on a map. For example a railway station location in the system I use is shown as a number of different square or circular symbols. The exact symbols used are styles. When you create a shapefile to hold data then you have to have a way of structuring the data so that the correct style is displayed for each item in the file. The easiest way to do this when I first started the project was to have a separate data layer for each style, for example if there were six types of station there would be six different layers. Having a rule based style means I have an extra column of data in the table that stores the style code for each item. This means there are now far fewer data layers in each project but it has meant a lot of work to migrate the data from the old layers to new and in a number of projects this is still ongoing.

For the projects, the major works being done this year have revolved around the relatively cheaply available Archives New Zealand collection of contact prints of the NZ Aerial Mapping overhead coverage that was used by Lands and Survey to create their older map series. Having discovered this I have used it in several applications, most notably for the Otago Central Railway, especially the Cromwell Gorge. Although ANZ increased their prices at the middle of the year it is still a good resource and I am continuing with the rest of the line back towards Wingatui. In the fourth quarter of the year work had to stop due to other commitments but it will be resumed towards the end of January.

The idea of publishing volumes of maps online has also been canvassed and if this goes ahead then the Otago Central Railway volume (Volume 18) of the maps project will be the first such volume produced. Since the maps themselves will not be finished until sometime in 2016 at the earliest, producing the volume (via CreateSpace) could possibly happen towards the end of the year and would mark the official public launch of the project. The Cromwell Gorge subject in particular would likely be attended by publication in NZ Railfan magazine as part of that official launch. In saying that there will be an official launch, it is expected like most aspects of this project to be low-key and mainly an online event as our community is small and scattered. 

In recent weeks the main focus has been with the Canterbury-Westland and Nelson-Marlborough maps, mainly to get the Main North Line maps which were one of the oldest series published, up to date with the current styles used and bridge and km post numberings have been added from a new source, the Kiwirail ALCAM GIS maps. The CW project is still being migrated to the new layer structure mentioned above. During the year I also switched computers at home to have a computer with three screens available for the work and in the process reorganised the disk storage so that has also disrupted the migration but everything is coming together pretty well now.

In 2016 therefore work will continue in the foreground on the Otago Central project in particular and there will be some smaller scale work in other areas as time permits. 

Tuesday 17 November 2015

Project storage moving to Google

Since the beginning of this project I have used online storage for all of the project files and also published PDFs. Skydrive / OneDrive was used at the beginning because Microsoft initially gave out 25 GB of space way back about five years ago. They then cut this back to 7 GB for new accounts, but existing accounts were allowed to keep the 25 GB, so that was the skydrive I was using for the project. More recently Microsoft was giving 15 GB of storage plus 15 GB of camera roll space. Once again they have slashed the amount of free storage, now you will get no camera roll and only 5 GB of storage.

On the other hand Google has recently upped free storage from 7 GB to 15 GB. Google storage has always gone up. Microsoft has had a history that you can see of allocating generous amounts of storage and then changing their minds later and slashing it to the bone. Google still give free photo storage for photos of 1600 pixels or less which happens to be the size that I have scaled photos for the web for years.

So I have set up a new Google drive and it will be used for the project files to be synced to from the computer I use to edit the maps. At the moment the published files (PDFs) will stay where they are as there may not be enough space for all the files to be in one place and there is no problem that they are causing with space where they are at present.

Sunday 20 September 2015

Clyde Today

The site of the second Clyde station as it appears today (closed 25 years ago). The engine shed is now part of a rural fire service premises, the turntable still sits in the ground, while the goods shed and loading bank are in a site used by rail trail operators. Public access to the rail trail terminus was enhanced recently by opening a pedestrian subway under the main highway. Work continues to map the pre-closure layouts of tracks when the line was in operation in this yard.

Tuesday 1 September 2015

Current Progress: Otago Central and Clyde

Currently I am looking at some more aerials but these will not enable new detail to be added to the map as they are going to be too low in resolution due to the change of Archives New Zealand's pricing structure. All they will be used for is to pinpoint where I can get higher resolution scans done at a later stage.

I am still looking at other scans I can get done for the Clyde area itself that will help to fill in some of the details. There will now be three maps for the 2nd Clyde station area: 1977, 1981 and current. This is due to the fact there were significant changes in the area during those periods. which cover construction, operation and post-closure rail trail (i.e. what exists today).

So while there have been no new aerials, we are still working on the Cromwell Gorge and Clyde but it will be some months before any additional areas are mapped with aerial photography.

Saturday 22 August 2015

Next Aerial Batch

A request has been sent to Archives New Zealand for a new batch of aerials. These will cover from Galloway down to Wedderburn, with a couple of Clyde as well. The ballast pit at Chatto Creek should also be included. The specific photos requested are:
  1. R18344219: Survey 71, Run C, Photo 4
  2. R18344220: Survey 71, Run D, Photo 3
  3. R18347187: Survey 1745, Run X, Photo 11
  4. R18347185: Survey 1745, Run V, Photo 10
  5. R18347185: Survey 1745, Run V, Photo 11
  6. R18347184: Survey 1745, Run U. Photo 13
  7. R18347184: Survey 1745, Run U. Photo 14
  8. R18353409: Survey 3858, Run C, Photo 17
  9. R18353410: Survey 3858, Run D, Photo 16
  10. R18353415: Survey 3858, Run I, Photo 2
  11. R18353415: Survey 3858, Run I, Photo 3
  12. R18353409: Survey 3858, Run C, Photo 18
  13. R18344440: Survey 112, Run H, Photo 7
  14. R18344439: Survey 112, Run G, Photo 6
  15. R18353407: Survey 3858, Run A, Photo 24
  16. R18353408: Survey 3858, Run B, Photo 24
  17. R18347178: Survey 1745, Run O, Photo 4
  18. R18347178: Survey 1745, Run O, Photo 5
  19. R18359828: Survey 8319, Run B, Photo 3
  20. R18359829: Survey 8319, Run C, Photo 2
Although my preferred era has been the 1960s, aerial photography from around that period is hard to locate in Central Otago which apparently doesn't get photographed as often as some areas do. The reason there was so much around Cromwell-Clyde would be due to the hydro development which was a long standing plan for the area. Some of these aerials are therefore from the 1970s and others are smaller scales that will be unusable at 300 dpi, they are only being used in order to pinpoint the location for rescanning at hi res. Unfortunately the last batch of scans confirmed that these 300 dpi images are nearly unusable,

I expect probably only one more batch will be needed to get down to Middlemarch and from there I will consider whether to get a batch to cover the Taieri Gorge stations. There will not be a great deal of map making until I can get those hi res scans so I expect things will appear to go slower until the end of year holiday period.

It is possible I may get some hi res scans of the 1992 Cromwell Gorge set done at some point for further reference but this is not such a priority as the rest of the line and enough time has been spent looking at this area already.

Saturday 15 August 2015

Galloway, Chatto Creek, Omakau, Lauder, Auripo

So as we move on down the line, then I am about to order the next batch of scans for the line. How usable these are as I have mentioned, is difficult to know.

The current map of Galloway, showing aerial photo footprints. Since Survey 1452 only covers Alexandra to Clyde, I will have to find a new survey that is suitable, at around the same time, to work up the Manuherikia Valley. Galloway was only a small station so there won't be much detail to copy.

Chatto Creek was also a small station with not much to see. However, it also had a ballast pit, the bridge for which can be seen to the west of Bridge 76. It should be possible to pick the details of this up on aerial photography, as I understand the ballast pit was still in operation until the 1950s or 1960s.

Omakau was a more substantial station, which at one time had its own engine depot and turning triangle for locomotives. It will be useful to pick up this detail to add to the map.

Just east of Omakau we have the phantom curves. I don't think these warrant any further investigation; it seems clear enough that the surveyed corridor was never used by the railway at all.

Lauder was small enough that I don't expect to pick up much there but it will be useful for completeness.

I hope to be able to find further aerial proof of this deviation near Auripo. The nearby Auripo station is another small halt.

Friday 14 August 2015

Otago Central Railmaps

Things are going a bit slower at the moment, for a variety of reasons. I'm busier in term time with work and study commitments.

However there is also a change due to Archives New Zealand's change in their pricing. I have budgeted to spend not more than $50 per month on aerial photos. Previously this would get me 40 scans at 600 dpi but because of their price increases, they have now chopped this to 20 scans, and worse still, have chopped the resolution of those scans to 300 dpi. If the image is a very good quality then you can still see enough detail at that resolution to be useful. If it is less in quality then the 300 dpi scan is useless.

Because of that, what I get out of each batch will vary a great deal and sometimes it will be very little. The last batch had one hi-res scan of the 1980-1990 Clyde yard, which produced a useful outcome although incomplete. The 300 dpi scan from 1979 of the same general area was good enough to add some additional detail but not really much at all. The rest of the scans have been of little use, including the ones of Alexandra. They serve only to pinpoint areas I will have to pay more for later to get scanned at a higher resolution e.g. 1200 dpi. The photos of the Cromwell Gorge taken just before the dam filling started in 1992 were of little use.

Archives New Zealand have advised they can do high resolution scans on the basis that the first scan will cost $50 and each subsequent scan in the same batch will cost $10. So I will have to batch up a pile of images for hi res scan and get them all done in one go. For obvious reasons that will happen as soon as I have a big enough batch - say $200 worth. And that itself will have to wait until I have identified enough low-res areas of interest. So for now I will be heading down the line to find all the other stations and yards and sidings etc which are of interest and will just have to see what turns up in each batch, which may not always be a lot. In the meantime there is still a bit of tidying up of what I have already to do. 

So while the maps may appear to go slowly, I expect that I will probably put in a big batch of hi res scans towards the end of this year and then be able to make best use of having the holidays available to push through a lot of progress all at once.

Sunday 2 August 2015

Archives New Zealand's new charges in relation to aerial photos

On the 1st July 2015 Archives New Zealand introduced a new scale of charges for remote research services. The key points of these charges as they relate to aerial photographs include:
  • The base fee of $25 per half-hour has increased to $50
  • The scan resolution that is defined as "low resolution" has been decreased from 600 dpi to 300 dpi.
Since I have determined that 600 dpi is the minimum resolution suitable for the aerial photos I am using (around 1:10000 scale), I am now faced with either paying a lot more money to Archives, or making do with the lower resolution. The indication from resampling the images I already have at 300 dpi is that I will struggle to pick out some of the detail that is visible at 600. It indicates that 600-1200 dpi is the optimum resolution for these around 1:10000 aerials but it also depends on the print size. It is pleasing to know that the option for scanning these contact prints has worked out from a technical point of view, compared to scanning from the original film, which Opus charges an enormous scale of fees for.

The fees for high resolution copying start at $30 although there is a volume copying rate of $10 that may apply to subsequent scans. The $50 fee is also payable as a minimum charge. When this fee is applied to low resolution scans it includes 20 "free". I am currently seeking clarification from Archives as to the application of the scale of charges for high resolution scanning since I may inevitably resign myself to obtaining a limited number of high res scans at the higher pricing. If the charging scale works out that there is significant advantage in getting a number done together then they will be batched. The use of the high resolution scans is going to be limited to Volume 17 otherwise it just ends up costing too much. So everywhere else will be low res except where there is free or cheaper high res coverage available.

The chainage chart referred to in a previous post turns out to have 24 pages, therefore it would only cost about $50 to have it scanned. I am planning to set up the request for this next week. Meanwhile tidying up the traces from the previously obtained aerial photography continues.

Monday 27 July 2015

Aerial Photo Order 3 (Otago Central Railway) - Cromwell to Alexandra

I am currently setting up an order for 21 more aerial photos as follows:
  • Survey 1452 of 1962
  • Run S. Photo 7 - this should give me the very end of the line at Cromwell which so far has eluded me.

    Run W, Photos 19-20 and  Run X, Photos 1-2: Alexandra

  • Survey 5073 of 1977
    • Run D, Photo 15; Run G, Photo 7; Run H, Photo 15 cover the 1907 Clyde station and just to its south and should therefore show the NZED siding that was in use until 1978.
  • Survey 5880 of 1981
    • Run A, Photo 1 should show the new Clyde station yard just opened (hopefully - this is a single photo in this series as no others exist in the lists). It would be interesting to speculate if this was a special one taken by NZR to base an aerial map of the Clyde yard, as they did use aerial photographs to make diagrams for staff use.
  • Survey 11936 of 1992. This is a big survey that must have been taken quite soon before the first filling of the Clyde Dam started. I am using this series to get a look at selected stations along the route rather than the whole route.
  • Run L, Photo 2; Run M, Photo 3; Run I, Photo 1/3; Run Q, Photo 2/3; Run R. Photo 2/3; Run U, Photo 1/2; Run V, Photo 4.
  •  Survey 5361 of 1979. This is to be scanned at high resolution because it is 1:25000 instead of the considerably larger scales of the other prints. Because of this it will end up being at a cost of $30 for the single image. I can't afford to have too many high res scans along the way because of the cost (I also have to pay the $50 base fee so I may as well get my money's worth and put in another 20 images at 600 dpi which can be included in the $50).
So that is 21 images for $80 - just under $4. And we are finally moving down to Alexandra. As I expect 4 images should give enough detail of Alexandra yard and from here we will go fairly quickly down the line to Wingatui. I am estimating another 6 batches of photos will be needed to finish all the remaining yards down to Middlemarch. It is possible I could continue looking at aerials for the Taieri Gorge section as well as there are many stations which have disappeared since that time.

So there is some scope to look again at the publication timeline. I also intend to obtain the chainage chart for the line and I don't know how many pages it is. Hopefully 20 pages will cover the Cromwell Gorge because the bridges and features in that section are not documented elsewhere whereas all the bridges and other things on the Rail Trail section have been documented by DOC in their Conservation and Scientific Notes publications. Once all of that is put together the publication date ends up getting pushed out a bit. The earliest as far as I can see will be the middle of next year.

Once Volume 17 is finished the aim will be to push ahead with first editions of every other volume (2nds of North Auckland and Napier-Gisborne) and this might go on concurrently with the Volume 17 work if I end up waiting for further photo orders to come in. There won't be any other volume as detailed as Volume 17 in the near future, All those other volumes will have to wait for 2nd (or 3rd in the case of NA/NGL) editions to get any aerial sourced data into them (except for the Christchurch area and a few other cases where free aerial footage has been used).

Sunday 26 July 2015

NZ Rail Maps: Otago Central Railway

Well this blog has now been launched as being separate from NZ Rail Maps and I have copied across all of the content from Enzed Transport that is maps related. The final layout of this blog and various other settings have yet to be completed and may well change over the next few days.

What is now relevant is that this blog will be the public face of the project. The actual development work is carried out in a private Facebook group with restricted membership. Since the posts in that group are not visible to non-members, the blog will be published and will also send a post feed to the Enzed Transport & NZ Rail Maps Facebook page.

As some previous posts have described, the Otago Central Railway is the current focus of the Maps project, specifically in relation to Volume 17, 2nd edition which is being put together. The first part of the route to be looked at has been the Cromwell Gorge. By using old aerial photography the route of the former railway has been traced in relation to present day features such as Lake Dunstan and the current highway. This includes the sites of the Cromwell, Waenga and Doigs stations.

The main focus of this has now reached Clyde and at present a request for further aerial photographs from Archives New Zealand is being put together to get some pictures of the Clyde and Alexandra areas. And so it will continue down the line towards Wingatui. Archives New Zealand recently doubled their charges, so that for 20 images it now costs $50 instead of $25. This means I will be limiting my purchases to 20 per month or $50 in general, except where extra scans are needed, but that will not really be an issue because it's hard work to get through them very quickly anyway. Bearing in mind that it can take Archives up to two weeks to process a request then one batch per month seems entirely reasonable. I actually received the last batch about three weeks ago so it does seem reasonable to assume that one batch a month is in keeping with my productive output during term time, rather than the school holidays.

Thursday 21 May 2015

Otago Central Railway: Wedderburn Highway Realignment

Although the Otago Central Railway had few changes in alignment, roads around it were realigned many times. One of the more interesting changes is just north of Wedderburn where the map dated from 1939 confirms that the highway was then located directly south of the railway rather than north today. As part of the highway the overbridge with railway number 66a was constructed and still exists on the route today although now bypassed. The highway realignment eliminated this bridge and a level crossing, the level crossing was very close to the railway summit of 630 metres which is the highest point on the whole of the Otago Central Railway.
The Wedderburn map (an extract from an NZGS map of 1939). The section of highway realigned is in the upper section of the map where the highway crosses the railway next to the words “F.McCarthy” and then crosses again further up near “P.Darling”. But whether the latter (northernmost) crossing was on an overbridge at the time the map was drawn has yet to be determined as the alignment shown actually resembles the current alignment with a level crossing due west of where the bridge actually is.
My map showing the entire section and the old and current highway routes. Bridge No.66a clearly is shown, along with the present alignment of the road to miss the overbridge entirely. Since the bridge number indicates it was not original, it is possible that where the road goes today, due west over the overbridge, may also be the original route that the road followed, or close to it, at a level crossing. Central Otago District Council's bridge inventory lists the bridge as "[bridge no.] 108 [ward] MANIOTOTO [roadway] OLD SH DOWN FROM COAL PIT ROAD [waterway] RAILWAY" implying this may still be a public road. An interesting point is the extra land set aside at the summit which suggests provision may have been made for a summit siding.

WedderburnN-Br66The GE shot depicting Bridge No.66a which still exists today. The Department of Conservation’s CASN 137 document describes it as follows: “Bridge 66, the road overbridge for State Highway 85 north of Wedderburn, is a 1966 pre-stressed concrete bridge. The abutments of the 1900 bridge are still visible, but are plain concrete foundations which probably held wooden piles. It does not require any special historic consideration. The highest point of the line (2029 feet, 612 m) is just north of this bridge. Bridge 66a, a private concrete overbridge, is very similar to Bridge 66.” So most of the description is actually the similar 1966 bridge further south. But the comment clearly implies Bridge 66a is now in private ownership, and further to that, Bridge 66a was almost certainly built first because this section of the highway probably is much earlier than 1966 and because there was already an overbridge at Bridge 66 dating from the original line construction. How much earlier Bridge 66a may have been installed is unclear because maps between 1939 and 1970 (by which stage the highway had been realigned) are hard to find. But my guess is the S bend in the highway needed for the bridge does imply it was installed well before 1966. However the number assigned to it (66a) strongly suggests it was added sometime after the line opened, certainly after the original bridge numbering sequence was assigned to the line. If that were the case then a level crossing at this location must have preceded the bridge, and this level crossing may well be at the location of where the road currently crosses the rail trail, shown above as a black line. There are earlier Naseby Survey District maps of this location but they are ambiguous in this area. But my guess is that they show both crossings via the cadastral map land boundaries and therefore leave unanswered the question of when the bridge was actually installed. The map revision dated 1954 still shows the highway in the old position.
UPDATE: I received information today including photos of the bridge and the information is that it appears to have been installed in 1950. Clearly at this time the realignment of the highway was still some years away. Whether this part of the highway was ever sealed is another question naturally, as it is not in a sealed condition today. The deck of the bridge as it now stands does not show any sealing. The way the bridge was installed is not commensurate with the needs of a high speed highway, as with the sharp bends it has been done in the cheapest possible way to make the shortest bridge, but as the line where the bridge goes over is in a cutting, the cutting may have caused visibility problems for traffic approaching the crossing. It does appear that the bridge is in private ownership along with the road today. Also it seems more likely the black line route marked would be where the railway went prior to the bridge, as this is clearly marked out in survey boundaries and on private land it would be unlikely to be formally surveyed. The location of the crossing would have been chosen to improve visibility being well clear of the cutting which can be seen on the above map and at right angles to get the best view overall. The use of the different areas of road isn’t so clear as of today but there may be still some use of the bridge but as we can see the old level crossing route is very clear and indicates it likely is still being used also.