Wikipedia talk:WikiProject Rivers/Maps

Archive of previous discussion, as relating to this advice page
(taken from here and edited for clarity/pertinence)

I certainly agree that clear maps of a river's course in the surrounding terrain should be added to all river articles and that too many of them lack exactly that. I don't think it should replace the infobox image, though, which IMHO should be an eye-catching photograph that is representative of the river and encourages the lay reader to read further. Of course, the choice of photo is always going to be a bit subjective and we are also limited by what photos are actually available. --Bermicourt (talk) 19:02, 13 March 2010 (UTC)
 * Generally the availability of good maps is also an issue. You worked on 3 (out of over 17,000!) articles but the map you added to Volkhov River is poorly suited to the article and the one in Northern Dvina River is labeled entirely in Cyrillic. Rmhermen (talk) 19:20, 13 March 2010 (UTC)
 * Any pointers to articles that have good maps, so we know what is required? --ClemRutter (talk) 22:36, 13 March 2010 (UTC)
 * I believe that all of of the FAs and the one FL have decent maps - see Category:FA-Class River articles Ruhrfisch &gt;&lt;&gt; &deg; &deg; 22:49, 13 March 2010 (UTC)
 * Hey all, thanks for your replies, and sorry if I sounded a bit flippant earlier. Didn't mean to diminish the work of this Project. I just felt so betrayed in my quest to patch the gaping holes in my knowledge of Russian geography... ;)
 * I know my edits are less than perfect. I just wanted to add something that gave some faint notion of where the river was... But then I thought it was better to seek some consensus on what to do and how (on that note : what do you mean by "poorly suited" ? I'd be happy to remedy any problem, or try to anyway). --Alþykkr (talk) 00:47, 14 March 2010 (UTC)
 * P.S. : I can also continue my work on northwestern russian rivers, using the same map I used for the Daugava River as a base, if you find the one I posted here acceptable. --Alþykkr (talk) 00:49, 14 March 2010 (UTC)

I have found {http://www.openstreetmap.org/} to be a useful starting point, but my graphic skills are limited as you can see on maps in maps on river articles in the Bristol area. I compare to the ordnance survey and go from there. –– Jezhotwells (talk) 04:47, 14 March 2010 (UTC)
 * I followed Ruhrfisch advice, and honestly some of the maps are pretty grisly. I followed the links through to WikiProject Maps/Conventions and made the horrifying discovery that there is no advice on rivers.I think as a matter of urgency we need to agree conventions in a page WikiProject Rivers/Maps. Aliso Creek (Orange County) is a FA (though cartographically I can't see why), it has many maps: we need to distinguish between a locator map on the Infobox, geological/topographical maps, human geography--- My principle is quite simple,-when something is  referenced in the text I want to glance at the map and see where it is. Similarly when something is  referenced on the map I want to find it in the text. But all that is up for discussion.--ClemRutter (talk) 10:53, 14 March 2010 (UTC)
 * I fully agree that principle - we should adopt it for all geographical articles and place references. It is much easier to understand what the text is saying if there are clear maps showing the places referred to. --Bermicourt (talk) 11:51, 14 March 2010 (UTC)
 * Course, major tributaries, lakes/reservoirs, cities, watershed, but try not to label things like mountain ranges, dams or other physiographic features. Also may try highlighting the course of the river on the map so it is more easily seen. See Snake river. I'm surprised Kmusser doesn't have any comments on this yet.  Shannon   talk   contribs  19:26, 14 March 2010 (UTC)
 * Watershed might be more difficult to know/map than other features. Perhaps it should be a suggested, but not a required, feature. I think highlighting the course is a must, since it will allow the reader to get an idea of the location of the river even from a small thumbnail. --Alþykkr (talk) 19:40, 14 March 2010 (UTC)
 * For mountainous regions, watershed is generally easier to map. But if you have a base map with many river courses shown, it is sometimes still possible to tell the watershed line by searching for the one 'line' where rivers from one basin don't connect to the other. For rivers in the United States, there is the USGS's NHD server, and there are also many watershed maps for other rivers.  Shannon   talk   contribs  21:54, 14 March 2010 (UTC)
 * I mainly meant that closely marking a watershed line and shading the corresponding area requires more cartographic skills/photoshop-fu than simply highlighting a river and placing main cities on a map. Perhaps it would be best to forgo the compulsory depiction of watershed at first, so that we have localisation maps more quickly. Then, gradually, we can go and add watersheds to these maps (of course, if one feels like including it right away, let him/her do so). --Alþykkr (talk) 23:20, 14 March 2010 (UTC)
 * Indeed, getting reliable information about watersheds is very difficult, even impossible for some smaller rivers. And deducing it from topography would be WP:OR. –– Jezhotwells (talk) 03:57, 15 March 2010 (UTC)
 * Watershed maps for U.S. states are available on EPA's Surf Your Watershed site. These are lo-res maps (8-digit HUC level), but it's a start. The individual watershed pages have links to USGS and state agency websites, which sometimes have more detailed info and/or maps. Moreau1 (talk) 05:14, 15 March 2010 (UTC)

Ok, having made most of the existing river maps I'll chime in. Absolutely all river articles should have a map, if using the Geobox template for the infobox there is already a space reserved for where the map should go - I suggest using it. I think the general map conventions are applicable to map of rivers as far as colors are concerned. As for what should go on the map is going to depend on the river and what data is available. I don't think consistency is as critical as showing the features important to the river being discussed. Drainage basins would ideally be on all of them, but for many rivers they simply aren't available, especially outside the U.S. Topography is nice, but can be tricky to get to look good and for rivers in non-mountainous areas might not be important. Very simple maps such as File:Susq.png get the basic job of locating the river done and are nice because other language sites can use them without modification. Ideally these would be used in conjunction with a more detailed map that labels important features that are discussed in the article. File:Stjohnsriver detailmap.png is the most complicated one I've done and perhaps goes a little overboard. Examples that strike a nice middle ground include File:Juniatamap.png, File:YellowstoneRiverMap.jpg, File:Seine drainage basin.png, File:Columbiarivermap.png. Recently I've begun including locator inset maps to give the rivers more context as in File:Crotonrivermap.png - I think they help especially for smaller rivers. Kmusser (talk) 14:02, 15 March 2010 (UTC)
 * I was going to say exactly that : locator insets are, in my opinion, a really, really good idea (even perhaps an imperative). It helps not only with small rivers, but also with the fact that Wikipedia is supposed to be internatial-friendly. As a European (there I said it), I'm not able to recognise most U.S. coastlines or state boundaries ; conversely, most Americans would probably have the same problem for European rivers (and both Europeans and Americans, for Russian rivers, etc.). Thus, I'd suggest either having locator insets (for smaller rivers), or, for bigger rivers, to frame the map so as to include very recognisable coastlines. For instance, for File:Seine drainage basin.png, I'd perhaps have included southeast England in the frame so as to be sure that people can locate the river and its estuary more easily (but I only say that to nitpick, and because I'm jealous of your cartographic skills,Kmusser. Your maps are really good, I think they indeed would made for a nice "standard model") ).
 * Using a reserved space in the infobox for the map is also, I think, a really good idea. --Alþykkr (talk) 02:20, 16 March 2010 (UTC)
 * Actually, I think the Seine River map is also really good.  Shannon   talk   contribs  04:02, 16 March 2010 (UTC)
 * Regarding the insets - you can accomplish the same thing using an entirely separate map as well - as is done in the Aliso Creek (Orange County) infobox. I think locating the river in a broader context is important, but it doesn't necessarily all have to be done with the same map. Kmusser (talk) 13:33, 16 March 2010 (UTC)

Establishment of guidelines
I agree with Kmusser above that strict consistency is not a requirement. However, I think it would be nice to have some broad consensus on what we would like to see on a river map, so as to provide some guidance to those who will make these maps (as it's the purpose of this page). Thus, here's an attempt at a rough synthesis of what was said so far : Just a draft, of course. Everything's still up for discussion... --Alþykkr (talk) 04:36, 16 March 2010 (UTC)
 * Regarding tributaries and lakes, my general guideline is to include those that are discussed in the article. Kmusser (talk) 13:36, 16 March 2010 (UTC)
 * Okay then, it seems like there's consensus on this : if it's in the article, put it on the map. The problem is that the article will probably evolve more rapidly than the map. So, what elements should be added on the map, even if they are not in the article ? --Alþykkr (talk) 15:22, 16 March 2010 (UTC)
 * It should include anything that should be in the article? Any map is going to take a little research to see if the article is missing anything important.  I have done rivers where say the article was missing any mention of a major trib, I'll go ahead and put the trib in the map, and then add it to the article as well. Kmusser (talk) 15:58, 16 March 2010 (UTC)
 * I personally think that all major bodies of water (lakes, reservoirs) should be included, if they would occupy a significant portion of the map, or if they are just important to the hydrography, etc. (However, not necessarily marshlands, as on the St. Johns River map.) Anything that is significant to the watershed should be on the map regardless if it is in the article or not. For example for the map I made for the Karun River, I included the courses of the tributaries Vanak, Khersan, Koohrang, and Bazuft even though they aren’t mentioned on the article (before I edited it to, which was after the map was made) nor do they have articles of their own.


 * Regarding topography, you can get public domain topography from Demis Mapserver (global) or USGS Seamless (US only), as I do. You could also get more detailed topography from Planiglobe (global), but the window for that is really small so it’s not really recommended. However it has the quality of Demis but in greater detail, so it’s better to use for maps of smaller rivers.   Shannon   talk   contribs  20:55, 16 March 2010 (UTC)


 * I'd like more feedback on topography. I think it can make the map a little harder to make, but Shannon is right in that it is readily available.  If there is consensus that it is wanted on all/most maps I'll change how I'm doing mine to add it more often. Kmusser (talk) 14:21, 18 March 2010 (UTC)