Having created aerial maps and diagram maps for the Moutohora Branch we are currently generating the same for the main line (PN to Gisborne). So far there are 43 individual maps or diagrams for the 54 km from Palmerston North to Dannevirke.
Palmerston North Gisborne Line [0G]: Volume 5 Progress Update 7
It takes a lot of time to generate each map out of the GIS as each view has to be individually checked for details since it is common to find mistakes at this point. One of the most common issues is for alignments of roads in each map. Linz supplied the source layers for a lot of the information included in the maps and many alignments of transport corridors are significantly different from those depicted in the aerial photos. We assume when these were created their cartographer’s brief was to create these large numbers of lines as quickly as possible and were done on a very small scale rendering of the aerial photography due to the alignments being so far off in many cases. Obviously we went to considerable lengths to ensure the alignments matched for the railway but are finding many road alignments need to be corrected as well.
This one fact illustrates why we go to great lengths to download and have available for map use the Linz aerial imagery for each corridor. The experience with Google Earth was the alignments of succeeding generations of aerial images could not be guaranteed or controlled. This led to the earlier Google Earth maps we produced being misaligned and was one of the key factors in progressing past the use of Google Earth. It now appears necessary to factor in downloading the aerial images for each corridor which was something we hoped to avoid in the short term for Basic level maps but is now going to be included and will add probably another week to the production of the Basic maps for each corridor.
The other question that may be raised is the number of maps being produced for each volume. As you can work out, at the current rate and scale there will be hundreds for Volume 5 and people naturally ask questions about how economic a printable version of the map volume would be. After all the first release of maps which is available from our WordPress.com site was produced in 2014, covers only Napier-Gisborne and it has 118 pages, 107 have maps on them with 2 maps per page so that was 214 maps which is a similar ratio of maps to distance as for the current production so there could be quite a big volume for this corridor.
After considering that conundrum it has been decided the PDF volume (or Intermediate) version of each volume will be a condensed version using smaller scales wherever possible. The maps we are publishing online are generally at a maximum of 1:10000 scale and use larger than this for individual stations and areas where more detail is useful. This is a luxury we can afford for the online edition as we are not paying anything for hosting. For Intermediate level map PDF volumes, smaller scales will be used to a greater extent to reduce the number of maps output, with consequent loss of detail. The other improvement which is also reflected in the online maps is being able to rotate the map view, with a north arrow in the bottom left corner that automatically rotates to show where North is for any angle, which means we can make more efficient use of the physical view captured in each map. Whilst we are not paying anything to host PDF volumes either, the market they are designed for is hardcopy printout which does have an associated cost as well as the physical size of each volume and this is one key reason why there has been no interest in commercial publication from the railfan press.
All in all, there will be a large volume of maps being produced for this volume and as current output so far has only got us over 2 days to 54 km we can see it is quite slow, we would hope to see a lot more output over the next week. As noted above each map view has to be individually composed, which includes rotating to the correct angle, checking the detail and which scale is needed, then after the image is generated a second level of checking at a higher resolution. Fixing a lot of mistakes at this level is adding to the slow progress and we think we will have to spend more time addressing the road alignment issues at the earlier stages of map production to speed this up in future volumes. It could also prove to be the case that 12 basic volumes are going to take up all of the next 12 months with no time left for any other work in the project.