cansk: Saskatchewan Provincial Highways (1-399)

[oscar]

I'm starting development of the Saskatchewan Provincial Highways set. Serious work to start once usaak is activated, and usaca goes into peer review.

cansk would cover the primary routes 1-399. There are four levels within that group (1-99 most major highways; 100s major northern highways; 200s spurs to parks; 300s assorted other spurs). But all have similar signage, unlike the distinctive signage for secondary routes (600-799) and northern secondary routes (900s).

There are seven draft routes already in the HB, transferred from the old cansph route set. They give me a headstart on the initial 1-99 route group I'll begin with.

There are a few known issues to resolve, in the 1-99 routes. Shapefiles might help with questions of where the routes end, though I might not otherwise need much help from them.

-- SK 34 south end is unclear. It once connected to MT 511 at the border, but that crossing has been closed for several years. The road is barricaded only at the border, but the pavement seems to have been ripped up in the former customs area, south of an unnamed side road 0.09km north of the border. Moreover, junction signage on SK 18, where I saw the notices about the border closing, implies that SK 34 ends at SK 18. But I just drove through on SK 18 without checking out signage on SK 34, and GMSV has only pre-closing imagery of that part of SK 34.

-- Routing of SK 11, and several other routes (SK 5, SK 7, and SK 14), through the Saskatoon clusterfork -- unclear whether any of these routes still exist within Circle Drive. The province's 2015/16 official highway map shows major roads in Saskatoon, but not the route designations (if any) on or inside Circle Drive.

[yakra]

But all have similar signage

You say "similar" -- do you mean "identical", or just "similar"?
This example looks pretty identical to me.

Shapefiles might help with questions of where the routes end, though I might not otherwise need much help from them.

Here's what I've found, with the standard disclaimer that the shapefiles, while a very good guide, can't be taken 100% as gospel.

-- SK 34 south end is unclear. It once connected to MT 511 at the border, but that crossing has been closed for several years. The road is barricaded only at the border, but the pavement seems to have been ripped up in the former customs area, south of an unnamed side road 0.09km north of the border. Moreover, junction signage on SK 18, where I saw the notices about the border closing, implies that SK 34 ends at SK 18. But I just drove through on SK 18 without checking out signage on SK 34, and GMSV has only pre-closing imagery of that part of SK 34.

The 9.0 revision of the SK shapefiles, dated 2017-06-05, has the south end at the border.

-- Routing of SK 11, and several other routes (SK 5, SK 7, and SK 14), through the Saskatoon clusterfork -- unclear whether any of these routes still exist within Circle Drive. The province's 2015/16 official highway map shows major roads in Saskatoon, but not the route designations (if any) on or inside Circle Drive.

SK11 looks to be as much a headache as ever (some earlier discussion in the cannf thread)...
Earlier versions of the shapefiles (6.0, dated 2012-09-25; and 7.0, dated 2014-07-22) had TCH16 on Circle Dr's southeastern leg and Idylwyld, and SK11 along the eastern leg of Circle Dr, counter to what's in the HB. (I didn't mention this in the cannf thread, instead focusing on GMSV and the confusing signage on the ground.)
6.0 did not include Circle Dr's southwestern leg yet.
7.0 included the SW leg, new geometry from Clancy Dr to SK11's "ToSK219" point. This, oddly enough, had RTNUMBER1 set to 11. Only as far as Clancy Dr, mind you.
9.0 has now switched this around: TCH16 on the eastern leg & SK11 on the SW leg & Idylwyld, as in the HB. The new section of Circle Dr now has RTNUMBER1 = "None".
Stumbled across this when looking for SK5 signage, FWIW. Note the TO SOUTH 11.
SK5 is as shown in Google: West on College Dr, University Bridge, and 25th St all the way to Idylwyld. This includes a new (ly constructed?) segment of 25th St (compare ESRI or BingMap views), not shown in the 7.0 shapefiles. So apparently SK5 was important enough to update to this new routing, but I son't see any signage in GMSV indicating its existence west of Circle Dr.
I have to wonder whether, similar to AB, SK has some sort of City Connecting Link (Yes, I just used Kansas terminology, and I am not sorry) setup that could lead to wacky hijinks. @julmac, any insight here?
SK7 continues northeast to SK14, as we'd expect. As far as a 7/14 overlap is concerned, the trail goes cold: RTNUMBER1 = 14 from here on in. What I've seen working with the GeoBase files nationwide is that RTNUMBER1 is not always set to the lowest route number in case of a concurrency; for example The 16/83 overlap in Manitoba has RTNUMBER1 = 83. 14 is the more cromulent thru route here; that makes sense. The RTNUMBER2 attribute is often ignored, simply set to "None" when it should probably have a number. OTOH, I see RTNUMBER2 used on the 16/11 overlap, so it seems that this datum is used in Saskatchewan? (Though -- universally? Oy...)
SK14 continues east to Idylwyld.

[yakra]

Here's a list of RTNUMBER1 values mined from the 9.0 shapefiles:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 51 52 54 55 56 57 58 60 80 99

102 106 120 123 135 155 165 167

201 202 204 209 210 211 212 219 220 221 224 225 229 240 247 255 261 263 264 265 271

301 302 303 304 305 306 307 308 309 310 312 316 317 318 320 321 322 324 332 334 335 339 340 342 343 349 350 354 355 357 358 361 363 364 365 367 368 369 371 373 374 376 377 378 379 381 383 397

[julmac]

Based on the published maps I've seen (post 2013), SK 34 appears to still be in the provincial inventory south of SK 18, despite the sign removals. The Highway Hotline also includes a description of SK 34 as "US Border to Jct Hwy 13".

My suggestions for Saskatoon:
-route TC 16 on the east/north leg of Circle Drive (traditional routing of TC 16 pre SW leg)
-route SK 11 on south/west leg of Circle Drive (traditional routing of SK 11 up to Idylwyld)
-route SK 5 along 25th Street to Idylwyld.
-route SK 14 to Idlywyld.
-SK 7 ends at SK 14 (SK 7 is signed out-bound from Idylwyld, but not inbound, as in... https://goo.gl/maps/EHH7DREgGuG2)
-Include Idylwyld (Circle Dr S to Circle Drive N) in cannf

SK5 is as shown in Google: West on College Dr, University Bridge, and 25th St all the way to Idylwyld. This includes a new (ly constructed?) segment of 25th St (compare ESRI or BingMap views), not shown in the 7.0 shapefiles. So apparently SK5 was important enough to update to this new routing, but I son't see any signage in GMSV indicating its existence west of Circle Dr.
I have to wonder whether, similar to AB, SK has some sort of City Connecting Link (Yes, I just used Kansas terminology, and I am not sorry) setup that could lead to wacky hijinks. @julmac, any insight here?

Yes, in SK, highways within cities are under city jurisdiction, with some exceptions. It's similar to Alberta, except that city controlled routes in Alberta often have a "TO" banner, whereas in SK they don't. SK 5 has an interesting history. Prior to the extension of 25th Street to Idylwyld Drive, it was routed rather awkwardly on 2nd Ave and 23rd Street. It was well signed, though, due to the fact that it used to extend to Lloydminster (along present day TC 16) prior to the re-numbering of TC 16 across the western provinces, and prior to the construction of Circle Drive. Needless to say, the value (outside of our circles of course) of a well signed SK 5 in downtown Saskatoon diminished significantly by the time the 25th Street extension was constructed.

[oscar]

My suggestions for Saskatoon:
-route TC 16 on the east/north leg of Circle Drive (traditional routing of TC 16 pre SW leg)
-route SK 11 on south/west leg of Circle Drive (traditional routing of SK 11 up to Idylwyld)
-route SK 5 along 25th Street to Idylwyld.
-route SK 14 to Idlywyld.
-SK 7 ends at SK 14 (SK 7 is signed out-bound from Idylwyld, but not inbound, as in... https://goo.gl/maps/EHH7DREgGuG2)
-Include Idylwyld (Circle Dr S to Circle Drive N) in cannf

It sounds, from yakra's post about shapefiles, that Idylwild is part of SK 11, and so wouldn't belong in cannf. That would have the advantage of keeping SK 11 as a continuous route, from Regina to near Prince Albert. Circle Dr., of course, needs to be updated for cannf, since about half of it is neither in SK 11 nor TCH 16.

As for the note on part of SK 5 being city-maintained, I lean to treating that as a non-issue, so long as the province designates or recognizes (such as in shapefiles) the route within the city. That resembles the issues I'm dealing with for California state routes, which normally remain state-maintained within cities, but the state lately has been fobbing off on cities many route segments previously maintained by CalTrans.

While re-reviewing the Saskatchewan official highway map, I noticed that secondary routes in the 600-799 range are called "municipal roads", and appear to be maintained by the province's (all too) numerous rural municipalities. However, they might be province-designated, with some long 6xx and 7xx routes spanning multiple RMs. But northern secondary routes (900s) seem to all be called "provincial roads" and to be provincially-maintained, since the province provides municipal services in the sparsely-populated northern region. It will be a long time if ever before we map routes in the 600s, 700s, and 900s. However, we refer to such routes as SK___ in the TCH route files, and the handful of draft cansk route files. I'm inclined to leave those route labels alone, consistent with how we're labeling secondary routes in Alberta and Manitoba.

[oscar]

I've updated the cansk route files moved out of cansph, and also added a few routes here and there, including samples from the 1xx, 2xx, and 3xx tiers. In cannf, Circle Dr. has been finally updated, since several cansk routes intersect that loop around Saskatoon.

Some of these routes have unclear endpoints. I expect that to be a problem especially with the 2xx routes (loops and spurs to parks), which tend to be short but poorly mapped. I'll hold off on addressing those endpoints, until I've finished drafting the route set (still about 120 routes to draft, so that'll take awhile) and can compile a complete list of what needs followup.

[oscar]

Here's a list of RTNUMBER1 values mined from the 9.0 shapefiles:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 51 52 54 55 56 57 58 60 80 99.

One of the new routes I've added is 39A, consisting of old 39 mileage through Estevan, assuming (as Mapnik does) that 39 has been moved to the new truck bypass opened in 2015 or 2016. Does 39A show in the 2017 shapefiles? Any other suffixed routes?

The provincial Highway Hotline map shows 39 on both the bypass and the old route through downtown Estevan. Other maps and official info don't address. And neither GMSV's camera cars nor I have been to Estevan lately (it was the first city in Saskatchewan I've been to, in 1994, but did not inspire a re-visit).

[yakra]

In cannf, Circle Dr. has been finally updated, since several cansk routes intersect that loop around Saskatoon.

• IMO the three SK11/16 points would be better served as SK11_A, SK11_B, and SK11_C. Compare TXLp7.
• SK11 could then change to SK11_N, denoting the multiplex split.
• SK762 is only signed for Valley Rd, with enough blank space for a shield to be added. Shapefiles have 762 ending at the Corman Park / Saskatoon line. 'City Connecting Link" hijinks, as discussed upthread?
• WamRd -> WarRd

Some of these routes have unclear endpoints. I expect that to be a problem especially with the 2xx routes (loops and spurs to parks), which tend to be short but poorly mapped. I'll hold off on addressing those endpoints, until I've finished drafting the route set (still about 120 routes to draft, so that'll take awhile) and can compile a complete list of what needs followup.

Shapefiles should be helpful for this when the time comes.

Does 39A show in the 2017 shapefiles? Any other suffixed routes?

No suffixed routes. All RTNUMBER1 values are completely numeric, no alpha (Aside from "None", which I left off the above list for obvious reasons").
The Estevan bypass is shown, just with RTNUMBER1 = None. 39 is still sown on the old alignment thru town.
Take with a grain of salt; it may not mean much. It could be that things are just taking a while getting updated.

The provincial Highway Hotline map shows 39 on both the bypass and the old route through downtown Estevan. Other maps and official info don't address.

*slow clap*

[oscar]

I'm continuing to draft route files, with an emphasis on routes following the province's western, southern, and eastern borders, and short routes connecting them to AB, MT, ND, and MB routes. I noticed the new canabs routes in the HB, which connect to some two and three-digit primary SK routes, and give me more border points to pilfer. I'll submit a pull request as more routes accumulate.

I've also done SK 42 in central Saskatchewan, AFAIK the only primary SK route split by a ferry crossing (across Lake Diefenbaker). That crossing is replaced by an ice bridge in the winter, though the bridge is not yet open this winter (usually open in February through mid-March). There is precedent in the Northwest Territories to not split routes connected by ferries in the summer and ice bridges in the winter. However, I'm disinclined to push that precedent, and would instead have separate Lucky Lake and Eyebrow route files for the western and eastern SK 42 route segments.

[yakra]

cannt was activated Oct 14, 2009. This predates the split-by-ferries guideline.

[oscar]

Draft route files for all major southern primary routes < 100, and EDIT: 1xx northern primary routes, are now in the HB. Still working on 2xx park access routes (7 of 21 already in the HB), and 3xx minor southern primary routes (19 of 50).

Does 39A show in the 2017 shapefiles? Any other suffixed routes?

No suffixed routes. All RTNUMBER1 values are completely numeric, no alpha (Aside from "None", which I left off the above list for obvious reasons").
The Estevan bypass is shown, just with RTNUMBER1 = None. 39 is still sown on the old alignment thru town.
Take with a grain of salt; it may not mean much. It could be that things are just taking a while getting updated.

There are definitely other suffixed routes that have been around longer than the Estevan bypass -- 10A and 52A in the Yorkton area, 41A near Melfort, plus TCH routes 16A and 16B in Yorkton and North Battleford respectively.

[yakra]

Checking these out in the 9.0 shapefiles, for curiosity's sake, to see how things are done in SK.

TCHNBa (16B): On the SW<->NE leg, RTNUMBER1 = 4 & RTNUMBER2 = 16. On Railway Ave, no route numbers are listed.
TCHYor (16A): On Smith St & Gladstone Ave, no route numbers are listed. On Broadway St, RTNUMBER1 = "None" & RTNUMEBR2 = 16A on the eastbound & undivided alignments. Westbound lists no route #s.
10A: 16A overlap as noted above. The rest lists no route #s.
41A: No route #s listed.
52A: No route #s listed.

Is this indicative of sloppy record-keeping in Saskatchewan WRT just the letter-suffixed routes, or maybe with route numbering overall?
I couldn't say. I haven't looked at the SK files very much at all.
The more I work with the Geobase files, I find that while there's a national framework set up with how the files are arranged, each province has its own mojo, and does things differently within that framework.
GeoBase files are no magic bullet, and are best supplemented by other info sources.

[yakra]

SK42:
SK646 -> SK646_S
add in SK646_N, although quite close to SK45

[oscar]

Got it, fix to be submitted with other stuff later this evening (done).

I'm picking up similar missing waypoints on other routes, as I rough out some of the 2xx park access routes. I'll need to do my own double-checks, before putting cansk into preview.

Many of those 2xx routes are turning out to be a mess, including two that are not in any online mapping (one exists, I found its general location in a park guide and GMSV confirms it's signed; the other I have no idea if or where it exists), and others with unclear endpoints. Expect slow going on that tier.

[yakra]

I can use shapefiles to help track down those if/where cases, and their endpoints.