Mittwoch, 28. Oktober 2015

Mappers Blog: Bannwald Setup/ Symbolization, localization and cartography

My most recent project is about to end and therefore it is maybe time to take some notes. I mapped the area of Bannwald as part of the map Seebodenalp, a cooperation with Heinz Tüscher. My part consists essentially in a steep, forested slope with partly dense vegetation and partly abundant rocky features. The area has been mapped before in 1984 but has never been updated since.

 fig.1. Map from 1984

Assessment

I had to figure out, how to tackle the task and there where basically two questions to answer.
1) How to get there? - The slope is situated from 700 to 1000 a.s.l. and starts about 250m above the closest transport stop in Küssnacht a.R. (about 50km from where I live). There is a cable car running up to Seebodenalp but with a restricted time table and a restricted connection to the public transport. No cheap accomodation in the area. I can not loose to much time getting there and away.

2) How to localize? The regional council provides us with a laser-DEM and -DSM but it was produced in back in 2002. This means it has a relatively poor point density, is rastered, filtered (extracted leaf-trees), smoothed (extracted boulders) and in regard of the vegetation partly outdated. Thus a pretty oldtimer compared with a nowadays point cloud. In a relatively uniform slope with lots of stones and cliffs and rocks but also green features to be localized, such a base is not sufficient.  As the slope is facing north-east GPS is no big help (actually when ever was GPS  a help :-P).

New tools

Thus I was forced to update my setup and did this by:
1) Buying a scooter
fig 2. A scooter is the mappers best friend...

2) Borrowing a Range Finder 
fig.3. No! The Range Finder is the mappers best friend.

It is a TruPulse Range Finder able to measure horizontal and vertical distances, something especially handy in steep terrain.

Usual Hardware/Software-Setup 

Otherwise I the setup has not changed essentially since my last documentation. I am still very happy with the combination of OOMapper on a waterproofed Samsung Galaxy Note 2 (inductive pen technology).
fig.4. The Bannwald Setup (note the waterproofed pen)

Hybridal Mapping with OOMapper

One of the advantages of OOMapper compared with OCAD is the possibility to map hybridal: While you can digitize smaller features right away, you just can sketch features "under construction" f.ex. contours and larger areas with undergrowth.

 fig.5. Screenshot Hybridal Mapping with OOMapper.

Symbolization, localization and cartography

For me nowadays, mapping is a three step procedure, symbolization, localization and cartography (generalization & legibility) or with other words: What (aka ISOM)? Where on the map? and Does the map work (aka cartography)? 

Localization

While it was a big part of the job to localize map features correctly in the older days, I aspire to have this problem solved beforehand. Localization should generally be no big question. In the case of Bannwald I could base localization on these three base maps I could switch between.
fig.6. As done before I work with the 1.25m contours which also work as bands for contour and evtl. form-lines. Basemap calculated with OL Laser.

fig.7. The contours I combine with the vegetation height (the contour layer is semi-transparent). Basemap calculated with OCAD.

fig.8. Rarely I also used the steepness perspective. Basemap calculated with OL Laser.

Finally the mentioned range finder was the crucial tool in regard of localization. The model I borrowed would be able to directly communicate with OCAD by bluetooth. Although myself working with OOMapper I had to manually draw the measured distances on the map. For this purpose I used a helper symbol consisting in a 20m long scale.

fig.9. Helper symbol 20m scale

fig.10. Map draft with included range finder measurements.

Life cartography

Besides the possibility to switch between base maps as it suits you best, digital mapping has an other big advantage in contrast to the good old analog one: Now cartography can be considered on-site. I do this by using a set special set of symbols, I developed for the purpose of field work. This leads to that all cartographic conflicts can be solved right away.

fig.11. Some symbols of the field set.

Hence

So this was probably the hardest job I ever made spending 18 days to map 1.3 sqkm. If I would not have adapted my setup, it would have taken much longer. The map will be published next spring to be used for the 39.Innerschwyzer OL. Come and try!

Montag, 26. Oktober 2015

17th season in the Swiss National Team for Baptiste

The selections for the Swiss National Senior and Junior- Team have been published yesterday.

Changes for the new season 

fig.1. After 12 seasons with the National Team Sara Lüscher retired. (WorldofO.com; Sara Lüschers athlete profile)

Retired/not selected: Sara Lüscher (29, A), Bettina Aebi (25,B), Marion Aebi (22, B), Andrea Roggo (20,J), Laura Diener (20,J), Sonja Borner (18,J), Remo Ruch (20,J), Jannis Schönleber (19,J)

In came: Valerie Aebischer (18,J), Katrin Müller (18,J), Deborah Stadler (18,J), Simona Aebersold (17,J), Marcia Mürner (17,J), Sebastian Baumann (19, J), Nick Gebert (18,J), Andrin Gründler (17,J)

Upgraded: Sven Hellmüller (20, J->U23), Kerstin Ullmann (20, J->B), Lisa Schubnell (20, J->B), Sandrine Müller (20, J->U23)

Years in the National Team (incl. 2016)


fig.2. The Nestor of the Swiss National Team: Baptiste Rollier looking forward to his 17th season in the red-white dress. (WorldofO.com; Baptiste Rolliers athlete profile)

Baptiste Rollier 17 seasons
Daniel Hubmann 16
Fabian Hertner 14
Andreas Rüdlinger 14
Rahel Friederich 13
Judith Wyder 11
Andreas Kyburz 11
Raffael Huber 11
Sabine Hauswirth 10
Martin Hubmann 10
Matthias Kyburz 9
Julia Gross 8
Sarina Jenzer 8
Florian Howald 8
Alain Denzler 7
Florian Schneider 6
Elena Roos 5
Lisa Holer 5

Overview Swiss National Team Since 1999


I updated the chart with the selection dynamics since 1999.

Dienstag, 20. Oktober 2015

Mappers Blog: It was not your fault! Blame the map.

A short analysis on the Glarus Schwändi Map, used for the Swiss Middle Championships last weekend. I was the mapper and I ran the H 40 course for testing purposes. What I am doing now is go through the course and check where the mistakes where made and look for possible map related contributions.

The map

So first, here is the competition map. Have a look at it. Which controls might be tricky?
 

 The rates of mistakes

Here is what winsplits says, if we apply the default 20 seconds/20 percent rule.

And btw. he total number of participants is 46.

Control by control 

The first control is a short one and actually pretty easy. I can be spotted from 50m. Still there is a 30% mistake rate (MR) here. Partly it can be explained with the stressed starting situation. But otherwise one has to admit, that the control ring mask the bigger land form. Also there is a missing dash point on the eastern small knoll.
Same excerpt without overprint.

The second control was easy on the leg, but it seems some (20%) had problems to find the right side of the right stone here. Note how the contours interfere with the black objects.

Then a series with easier controls to 4 ....
... 5, 6, 7....
... 8 and 9. The easiest control of the course.
10 also looked easy but it some took it too easy. Might be a consequence of the climb to 8.
In this section the overprint again masks the terrain. Mostly the dell on the way into 11 and along the gully towards 12. At 12 the MR starts to rise again to over 20%. The mapping inside the control circle is not that clear.

And now to the most tricky section. At 13 MR is 30%, 14 tops with 43% and 15 follows with 21%. 13 and 14 are pretty easy legs, if you look on the map excerpt without overprint below...  13: follow the ditch, 14: follow the valley. A smaller mapping issue can also be spotted south-east  of 13 where the contour has a gap crossing the ditch.

After a pretty physical leg to 16 with a low MR 17 and 18 came back with 26% MR both. The cartographic issues here: The small knoll south of 17, suiting as attack-point is very prominent, but was drawn with a form-line due to over-correctness. A possible source of irritation.  Then on running speed it was not possible to see that the 17th was no boulder but a knoll. Small brown surfaces just melt together with black.
The cliff for the 18th control was, as you can see from the excerpt below, added to make a control point by the course setter...  and therefore was the least prominent cliff mapped on this map.

The last control was no problem from this side. But the elder Hubmann lost some precious seconds going straight over the hill from south-west. Also here the over-print mask the contours.

Resolution

1) Check for the effect of the overprint on the terrain representation, not only the control ring but also the connecting line.
2) Check for legibility issues not only for features with the same color but also between features with similar colors (brown-black)

Dienstag, 23. Juni 2015

Leading orienteering nations since 1966

fig.1.: Ranking of leading countries at successive WOCs 1966-2014 (host countries highlighted).

Fig. 1 is taken from an article by Dave Nevell from the April issue of Britain's National Orienteering Magazine CompassSport by courtesy of both author and magazine. Its content is described the following:
The countries have been ranked on how well their overall teams have performed each time round, based on the average gap beteween their runners and the average medal winning times for each discipline. Adjustments have been made for lower numbers of runners (e.g. non-qualification for the final). This then gives an idea of the current world order...
The graph is impressing as it visualizes the ongoing and total dominance of the top 4 since the beginning of modern orienteering. Sweden, Norway, Finland and Switzerland took 93.5% of all top four. While these nations all have between 27 and 30 top four positions over the years, former Czechoslovakia -> Czech Rep. got 5, Denmark, Russia and France each got 1.



Dienstag, 5. Mai 2015

Mappers Blog: Fürstenwald setup

fig.1. Fürstenwald setup.
Now have even I changed side and started to map fully digital. The main reason besides I like to challenge myself was that Open Orienteering Mapper opened up a way to step into digital mapping before and without spending a lot of money on hardware. So here I document the setup developed for revisioning the map of Fürstenwald, Chur, Switzerland.
The point with digital mapping is, that you can choose your base map according to the actual task.

Hardware

fig.2. The  overview
  • Samsung Galaxy Note 2, 5.5 inch screen
    • with S Pen (inductive) and extra S Pen (waterproofed)
    • with 1 original battery 3100 mAh and 1 extra battery 3400 mAh (total 13h with GPS on and screen on max)
    • inside a waterproof case by redpepper
    • protective film (anti-glare)
  • compass Silva
Pics:
fig.3. Normal and waterproof S Pen (Latex & shrinking hose)

Comments: Seems to me pretty ideal for mappers. Inductive pen, allweather and easy energy supply handling (no external battery). Total weight: 250gr. Total costs: about 200 CHF /190 Euro (Phone bought second hand)

Software

fig.4. Screenshot OOM 0.5.96 with old map as basemap, scribble plus activated GPS tracking and positioning.

  • Open Orienteering Mapper 0.5.96 for Android

Comments: One of the advantages of the OOM is, that you actually can continue working like you did in the analogue days. You just can scribble your features on the map. This is useful for features can not finalize directly (f.ex. wide greens). Of course, the rest you directly draw as map features. Other advantages: it is free and it is under development. You can contribute to its future.
A disadvantage of OOM is that it is not (yet) 100% compatible with recent OCAD formats (bascially some difference in dashing and no virtual gaps in OOM). In the actual case, revising an existing OCAD map, I therefore imported scribble tiles (georeferenced pngs) and digitized objects into OCAD11 for the finishing.

Basemaps

  • Snapping Grid
  • Old Map in gray and magenta tones
  • LiDAR Terrain Model contours 1m
  • LiDAR Surface Model (Both slightly outdated (!) therefore an recent additional Aerial)
  • Aerial Orthophoto
fig.5. Live extraction of natural borders from the orthorectified aerial photo
fig.6. Old map overlayed with 1m contours. Also visible the snapping grid (thin hairlines)
fig.7. Screenshot defining vegetation borders based on the DSM.

Comments: To fit in changes into the existing map, I use a "snapping grid" consisting in the old map features vector lines. Basically very thin/optionally invisible lines (in protected mode) I can snap to while drawing new features.
In this case the other base maps or "locator maps" where  provided. As  stated in the quote above the strength of digital mapping is, that you can choose base´maps after the actual situation. In open forest you can identify every single tree with the surface map. In more dense forest the contours are a big help. The internal GPS is indeed the very last measure.