Taken a liking to the ford=intermittent given the many intermittent streams and created a custom preset in JOSM to add this value not in the standard preset which only has yes & stepping_stones. JOSM does not moan, shows a blue dot where the tag is used and QAs like Osmose are fine with it too, so far.
Checking Taginfo…
ford=yes intermittent=yeshas 11,231 usesford=yes seasonal=yeshas 7,547 uses (slightly different I know, just including here for the sake of comparison)ford=yes depth=0has 2,308ford=seasonalhas 1,389ford=intermittent1,369
depth=0 is what the Wiki suggests:
Useful combinations
depth=*Normal depth you might encounter. Probably deeper during flood or when raining heavily. Perhaps even set to 0 for fords that are dry most of the year.
3 Likes
The author of the documentation used the least intuitive choice among those.
2 Likes
One of the more straightforward solutions is simply not to map ford=yes in places where there is no ford. The validators will complain, but there are lots of things that the validators aren’t right about.
3 Likes
The absence of a ford would be ford=no, but that’s a little different than saying there’s a stream crossing that you don’t have to mind under normal circumstances.
I’m reminded of how a walkway can lead right up to a street on both sides; that location might or might not be a place where you can cross the street. But we also have a consensus that a sidewalk crossing a driveway, where you certainly can cross the driveway, is so unremarkable that we don’t bother tagging it explicitly as a crossing.
If we allow that an intermittent stream crossing is similarly unremarkable, there’s a risk of bias if the road or track can be used by different modes of transportation. Perhaps a road crossing a small creek would be no match to a 4×4, but it would matter more to a casual hiker. If we don’t attempt to characterize “intermittent” any further, the next mapper who comes along won’t necessarily know whether it affects the ford’s significance.
2 Likes
Not if we use the least capable mode of travel as the measuring stick. If its someplace that a non-athletic person could hike to, I’d use that person as that measuring stick.
Inherent in the nature of a ford is variability. The example you gave is a ford. It might be intermittent, perhaps not. The fact that it’s trivial for a 4x4 is irrelevant (unless other modes of travel are not permitted). I’m not sure we can get very detailed on the difficulty of fords. If we add some difficulty metrics, such as depth, width, velocity, season, … it gets too complicated
What happened to the ephemeral tag? It seems very appropriate for waterways
Use the tag natural=gully instead of the intermittent waterway tagging. This is well suited for dry erosion features encountered western areas such as New Mexico and Arizona.
5 Likes
Nice, I didn’t know about that tag. The documentation points out some disagreement between translations about how big a gully can be, but I’ve seen some quite large washes the size of rivers that are normally used as bike-hike trails or even informal roads. Not sure if there would be a problem with stretching the tag in that manner.
1 Like
Someone mapped our desert aera with thousands of intermittent streams. It looks like we live in a marsh area. It makes much more sense to tag them as gulley’s. And gulley’s don’t require a ford.
The larger washes can be left as intermittent streams. It is a judgement call. But the large washes can carry water during a big rain storm.
2 Likes
We also have Proposal:Natural=wadi - OpenStreetMap Wiki for mapping wider washes.
If this is not ford=no, how would you tag it?
The “do not enter” is ford=no: a place where you cannot/should not ford a waterway. This discussion is about places where you not only can ford a waterway but wouldn’t even notice any water as you do so.
The “when flooded” part is less clear.
Ok. We’re more or less on the same page. I just read that as being the same as the “absence” of a ford.
Edit: Which honestly still makes sense. If there’s nothing to “ford” why are we using ford=*?
1 Like
Right, it sounds like, between natural=gully and the proposed natural=wadi, we might not even need to map a ford, because these aren’t merely “intermittent” creeks. So is this discussion basically converging on the one about introducing the latter tag?
1 Like
I like the idea of distinguishing between “active” waterways and ones that rarely have water. But I’m not sure natural=gully is the solution. That’s already a documented tag with over 8,000 uses.
I haven’t surveyed the usage, but I suspect that many of the uses of natural=gully imply a feature with sharp erosion banks. That’s not always the case with normally dry waterways.
+1 gully is a better description for many of these ways that are currently waterway=stream, intermittent=yes. even better than ephemeral. Most of these are the product of NHD import. Without local knowledge it’s tricky to retag
1 Like
Not from the US: I recently read an account of a mountain rescue emergency call: People got into a thunderstorm, they found themselves caught between two streamlets that swell beyond crossing. They could have taken another route and arrived safe at the hut.
For my liking, I still see CS tag closed:osmose:1070 too often. Osmose thinks there are 8680 fords missing in the area with 2431 mapped ones. Wouldn’t tagging the stream “intermittent” be the better solution?
Something should be done to make tagging more precise. ford=intermittent for fords that are dry most of the year might be the easiest way out.
The “Do Not Enter When Flooded” sign implies that the Department of Transportation identifies this is an intermittent stream. Based on their words, it could be tagged as such and include a ford. They suggest that travelers should not enter the ford (or they may be washed away). This could be the case with any ford. Stop and proceed carefully. Or find a safer route. Or if caught between two flooded fords, wait for the rain to stop.
Because of the associated reference photo, natural=gully may imply a feature with sharp erosion banks. However, in the desert these features vary between sharp and soft banks as you walk along them. Perhaps a photo with a soft bank should be included with the natural=gully description. Just for better clarification.
After much research, I could find no better existing tags to map these desert erosion features.
For me, a gully is an erosion feature caused by water flow. However, the water flow is rare and only occurs during heavy rain. Heavy rain is rare in the desert. In my desert and mountain foothill area, an intermittent waterway implies a seasonal water flow. Such as a spring and stream that flows during spring snow melt season and is dry the rest of the year.
flood_prone=yes fits well for a place that is usually dry and when there is water it’s not safe to pass. It’s not like these spots become fords when the water shows up. They become hazards.
https://wiki.openstreetmap.org/wiki/Key:flood_prone
4 Likes
Where I live, it is common even for intermittent streams to cross trails through a small tunnel=culvert (like File:Kölcse, Hungary - panoramio (5).jpg - OpenStreetMap Wiki ).
.jpg){kind=link}
If there is no tunnel=culvert, then it is generally a ford, even if intermittent.
A highway=path that crosses a waterway=stream without tunnel nor bridge should be mapped as ford. If the crossing has no tag, it just means that there is a missing tag.
1 Like
The thing is that the “waterways” the OP is talking about aren’t really even intermittent streams. They are desert features called washes or gullies that are dry the vast majority of the time. Very occasionally water flows in them after heavy rain. When this happens it is a flash flood and the wash could be dry again in a matter of hours. So most of the time the crossing (if it can even be called that) is dry and nobody would call it a ford. When it’s flooded it may be too dangerous to cross so also not a ford.
5 Likes