Talk:Lake Tauca/Archive 1

Latest comment: 3 years ago by Jo-Jo Eumerus in topic Frontiers
Archive 1

Pre-FA review

Since Jo-Jo Eumerus has indicated they wish to nominate this article at WP:FAC, I'm doing a review to help identify what issues might come up there. I'll add notes as I go through the article; it might take me a couple of days to complete the review.

  • Why is there a citation on the second sentence of the lead? It's not wrong to have one there, but it's usual to remove cites in the lead except for controversial statements or direct quotes.
  • Suggest changing the order of the constituent lakes in the second sentence to correspond to the Po Co Yu elements of Pocoyu.
  • "80,000–48,000 kilometres": I think it's more usual to have things like this in increasing order of size, not decreasing.
  • Giving the water level in terms of altitude instead of depth is a little odd, since we don't know the altitude of the bed of the lake. Can we give this some context?
  • "The lake was saline, and whether Lake Titicaca contributed ...": I don't think the lake's salinity has anything to do with the information about Lake Titicaca, so I wouldn't connect these two sentences.
  • "The drying of the lake may have been caused by the Bølling-Allerød climate oscillation, leaving the salt deposits of Salar de Uyuni": the way this is phrased implies that if the drying wasn't caused by the Bølling-Allerød climate oscillation, then it didn't leave the salt deposits; I assume that's not true so I'd suggest rephrasing.
  • Suggest mentioning in the caption for the lead image that the blue surface is Lake Titicaca.
  • Not sure what to do about it, but the last half of the first paragraph of the Overview section is a staccato list of short declarative sentences; it reads like a grab bag of facts rather than a narrative description. I'll come back to this and see if I can copyedit it a bit.
  • The four images down the right hand edge at the top of the article are probably too much; I think we need to do something about these. Can a couple of them be thumbnailed? I see there are custom sizes set; I'm not a big fan of this rule, but I believe MoS says these should be left at default sizes unless there's a very good reason not to do so.
  • The map showing the "extent of Lake Tauca, clearly visible in the topography": I assume the pale brown area is Lake Tauca? Given that the lake level is thought to have varied, can we be more precise here, or give more information in the caption? Can the map be annotated? Perhaps arrows identifying Lake Titicaca and the three remnant elements: Lake Poopó, Salar de Uyuni and Salar de Coipasa, even if it's just the approximate location that's identified by the arrows. That would give the reader a better chance of connecting this to the lead map.
  • The 3D map is attractive but to be honest I'm not sure what I learn from it that I don't get from the text. If we keep it I think it needs to be captioned in such a way that the reader can connect it to the rest of the maps; for example, it appears that we're looking south, from about the middle of the Salar de Uyuni -- is that right? And I don't think we need "remainder of Lake Tauca" in the caption; that's made clear in the text by this point.
  • The list of estimated surface areas in the first paragraph of "Geography" makes for eye-glazing reading. I think this might be better presented in a table. And if there's any way of getting images -- even line-drawing outlines -- for any of these proposed lake sizes, that would be great.
  • "...but not Salar de Ascotán. Other sources consider the Salar de Ascotán to be part of Lake Tauca as well...": rephrase to avoid making a definite statement and then immediately contradicting it. Perhaps cut it from the first sentence, and in the second sentence give both opinions.
  • Tables might be an option for presenting some of the data about salinity and other elements; given that you quote several values, with some research identified as earlier or later, a tabular format might make this easier to scan.
  • What does "During water highstands" mean? When the lake level was high? If so, I'd just say that.
  • Suggest using the full name of an element when first mentioning it in running text, and giving the abbreviation in parentheses. This wouldn't be the style in an academic paper, but general readers will find it easier.

-- More tonight or tomorrow. Mike Christie (talk - contribs - library) 11:57, 2 December 2016 (UTC)

Good points Mike. The two images you talk about I added, not Jo-Jo as main author of the article. The topographic map to visualize the extent of the lake in the topography. An annotated map would certainly improve the article. The 3D image I added because it was unused (!) in Commons, it shows the drastic change in topography on one of the boundaries of the lake (and as such the provenance area for the lake sediments) and to fill the white space next to the TOC. I did the GAN review and for an FA what I definitely want to see included is the human settlement around the (remainders of) the lake. Humans settled in that area as of 12,000+ years BP and there must be archaeological and anthropological finds that support that (and indicate lake level retreat with habitated sites). That would be needed imho to broaden the scope of the article that is now very much geology and paleogeography focused but misses the human factor still. Tisquesusa (talk) 14:00, 2 December 2016 (UTC)
As far as I know, there is no archeological site that seems to relate to Lake Tauca. Given the dates indicated in List of countries and islands by first human settlement and that the settlement of the Americas occurred around the time Tauca disappeared, I suspect that there may not been any people around to leave archeological traces. I am a bit less certain on the Coipasa sub-stage, though, although a quick Google Scholar search doesn't turn up anything clear. Jo-Jo Eumerus (talk, contributions) 16:59, 2 December 2016 (UTC)
Indeed, according to http://www.sciencedirect.com/science/article/pii/S0278416516301131 human occupation of the Puna did happen around the end of the Ticaña low lake level, and that's at a fair distance from Tauca. https://www.researchgate.net/profile/Jose_Capriles3/publication/258811214_The_earliest_human_occupations_in_Bolivia_A_review_of_the_archaeological_evidence/links/0046352eaa89b7e91b000000.pdf indicates some sites may have been closer - I'll see to work some information from these sources in. Otherwise:
  • I am a bit unsure on how to integrate information on lake bed altitude, especially since it is not uniform. Maybe indicate the deepest surface of the ancient lakebed?
  • Fixed some of the lead section issues, and wording issues elsewhere.
  • Unfortunately, I did not find any freely licensed map that shows these surface areas, and a request on the Commons Map Lab has been left unanswered.
  • I agree that the lake size estimates belong into a table, as do some of the other information. I'll see into this tomorrow.
  • Come to think about this, it may be worth specifying what height "during water highstands" refers to - 3830 m is the height indicated in the source for the lowest spillway.
  • Spelled out some element names; should the others also be spelled out?

That's so far. Jo-Jo Eumerus (talk, contributions) 17:42, 2 December 2016 (UTC)

Thanks for the research on the archaeology. Then I think it would be good to include "human settlement in the Lake Tauca area started around 3500 BP, 3700 years after the latest lake phase", or so, Tiwanaku is world famous, to the east on the Llanos there have been evidences from much earlier; still in the Tauca phase. To show that indeed it wasn't populated when the lake was still around.
The end of the Lake Tauca level is mentioned at 7200 BP (so Holocene), but the main phases were earlier (Late Pleistocene). I've put the largest rank in the infobox (72,600-7200 BP). But those ages should be better constrained and explained and also mentioned in the intro section that can and should be expanded anyway.
What is possible using "image overlay" templates is showing the extent of Lake Tauca roughly or by plotting points on a map. Imho for a GA it's ok like this, but an FA needs at least a good map to show what the topic is about. That's also why I added that light-coloured topographic map, as there it really jumps out.
If the lake levels are analysed with respect to the present-day elevations, it will be hard to re-write them in absolute values (depths), because of all the isostasy in the area. I've put the maximum depth of the lake at 142 m, that was mentioned in the text and the depth of 100 m. If the references contain more specific information would be good to improve the depth estimates. I don't have access to more than the abstracts of (most of) the publications, so unfortunately cannot help with that. Tisquesusa (talk) 18:30, 2 December 2016 (UTC)
Update on some things I've done:
  • A bit uncertain what, if any, information to add about the lake depth. There is some material about the lake depth vis-a-vis the Uyuni basin already.
  • I sort of like the current image arrangement.
  • I've moved the surface area information into a table.
  • Added a blurb about when human habitation occurred during the Tauca phase. There is preciously little connecting Homo sapiens with the lake so I am not inclined to add much more, unless specific sources can be found (most of what "Tiwanaku""Tauca" brings up simply discusses them separately)
  • I am a bit wary of some of the infobox data. For starters, while it's indeed possible that if the lake level reached about 3830 m it would have overflowed into the Rio Pilcomayo drainage basin via 20°16′14″S 66°26′37″W / 20.27050°S 66.44357°W / -20.27050; -66.44357, Quebrada Tajra Palca, Rio Olleria, Rio Ticatica/Tumusla River (this info is based on both present-day topography and the information in the source), for the most part the lake level was not high enough. And by all accounts, some water from the Carcote/Ascotan basins would have seeped into the Rio Loa (I don't remember the source for this). So I am not sure if we should mention an outflow at all, or perhaps potentially through the Rio Pilcomayo[1] during high lake levels or seepage into the Rio Loa[2] or some similar formulation.
  • Also, specified a bit when spillover is to be expected.
  • I'll see about the depth of 142 meters; lake ages will be a PITA.
That's what I have for now. Jo-Jo Eumerus (talk, contributions) 11:50, 3 December 2016 (UTC)
  1. ^ Citation
  2. ^ Other source

Much of what I commented on above is now fixed; I'll carry on through the article and we can take another look after the first pass is done. I plan to do some more this afternoon; meanwhile, I noticed one point worth mentioning. I don't know if it's the consequence of the recent edits, but there's now some repeated material in the geography section: "Lake Tauca was the largest paleolake in the Altiplano,[3] and its predecessors are known as Lake Minchin and Lake Escar." and "Lake Tauca was the largest lake on the Altiplano in 120,000[49] or 100,000 years." More this afternoon. Mike Christie (talk - contribs - library) 14:35, 4 December 2016 (UTC)

I've now read through the whole article, and rather than make detailed comments I think there's a general issue that needs to be addressed first. There's a lot of detailed data in the article, but it's not presented in a way that makes it easy for the reader to get at it. The table introduced for the various estimates of surface is a definite improvement, but I think something similar is needed in several more places. The presentation is made more difficult by the multiple different conclusions drawn by the various research papers cited.

I think a good start would be to try to tabularize more of the data, perhaps in a sandbox. Here are some examples of data that might be easier to read in a table:

  • Geography: shoreline estimates
  • Hydrology: element concentrations
  • Duration: age estimates and the associated estimated water levels
  • Climate: possibly some of the temperature estimates, though there's more narrative here and I'm not sure this would work

I also wonder if the reader would benefit from a slightly different high-level organization. How about following the overview section with a historical overview of the history of Lake Tauca? Something like this (in reality this would be longer and more detailed; this is just to clarify what I'm talking about):

Although there is disagreement over the exact dates, water depths, and related data, there is a consensus on the broad outlines of Lake Tauca's history. It formed at least N years BP, probably as a result of X climate change. It may have been fed partially or wholly by outflow from the nearby Lake Titicaca. In the first phase of its existence it is thought to have extended across X, Y and Z; estimates of its surface area for this period vary from A years to B years BP. The depth is difficult to determine exactly because of isostatic rebound but it is likely to have been at least C metres deep at the deepest point. Fossils and other evidence indicate that the lake supported gastropods, algae, and planktonic diatoms. By no earlier than D years BP, and possibly as late as E years BP, the lake had shrunk to about F square km; Lake Poopó was probably separate from the main body of Lake Tauca by this time.

I just made that up based on the reading of the article I've done so far, but I hope you see what I'm driving at -- essentially this would expand the equivalent paragraph in the lead to include a much more detailed narrative, but keep it almost entirely as narrative, without the insertion of large quantities of indigestible data. If that data can then be placed in supporting tables, I think it turns the article structure into a main line of narrative text, with branches out to the data in tables that support the reader's interest in particular statements in the narrative. Mike Christie (talk - contribs - library) 18:34, 4 December 2016 (UTC)

That seems a bit more like text that should belong into the lead section. Adding more tables for data is definitively doable. I don't think the conflicting information can be fixed as there isn't any preponderant opinion on these items. Jo-Jo Eumerus (talk, contributions) 19:14, 4 December 2016 (UTC)
Well, let's see how it looks once a bit more of the data is in tables. I think something along those lines might be doable, but we'll see. Mike Christie (talk - contribs - library) 20:06, 4 December 2016 (UTC)
These are things that probably work in table format:
  • Heights of the lake above sea level, in its several phases.
  • Mineral content.
  • Temperature estimates.
Possibly also dates, and I did originally use a table for them. I originally backed it out because it was oversimplifying the information quite a bit. Other data that are problematic in prose form? Jo-Jo Eumerus (talk, contributions) 20:42, 4 December 2016 (UTC)
Alright, I've tabulated some more of the information in question. I wonder if lake levels could be rendered by a graph more effectively than with text; unfortunately, the graphs used in the various papers are non-free. Jo-Jo Eumerus (talk, contributions) 17:29, 6 December 2016 (UTC)
What do you think of the reformat of the table I did today? I think it looks a little more concise. Would it be useful to add another column to give the range of dates over which Lake Tauca is thought to have existed by each of these sources? Mike Christie (talk - contribs - library) 00:40, 7 December 2016 (UTC)

@Mike Christie: The table change is fine. I think that showing the time course via table may be a problem as it's a "three-dimensional" information while tables only allow two of them. Finally, for when I have more time adding a graph based on {{Graph:Chart}} with the information on the various dated shorelines may be a better way to illustrate water levels. Jo-Jo Eumerus (talk, contributions) 16:01, 7 December 2016 (UTC)

I have a couple of things I need to work on but plan to come back here in a day or two and follow up; I will see if there's more that can be done by way of putting information in tables before we go back to the prose. Mike Christie (talk - contribs - library) 02:37, 8 December 2016 (UTC)

Other palaeolakes

The data on the various palaeolakes is a bit confusing, so I'm assembling some of it here for reference.

  • (From the lead): duration is 18.5-11.5 kya, possibly as late as 8.5 kya
  • Tauca existed at around 14,100 BP
  • Predecessors include Minchin and Escara
  • Other lakes include:
    • Ouki 120-98 kya (may be subdivided in the future)
    • Salinas 95-80 kya
    • Inca Huasi 46 kya (this is sometimes considered to be the Minchin phase)
    • Sajsi ** 24-20.5 kya
    • Coipasa ** 13-11 kya
  • Coipasa is a late phase of Tauca
  • At 11.5 kya Lake Tauca was drying up
  • Minor lake events are recorded in Salar de Uyuni at 28.2-30.8 kya and 31.8-33.4 kya
  • 28 kya precedes Tauca by 2 ky
  • Tauca's radiometric age ranges from 72.6-7.2 kya
  • Duration estimates are given in detail in the duration section, all within the range given in the lead
  • Duration section indicates Coipasa is sometimes considered a phase of Tauca; so is Sajsi

Looking at this, as far as I can see you never give the dates of Lake Minchin. Per the Lake Minchin article it seems it was 22-21 kya. I think a table would be useful here too; I'll draft something. Mike Christie (talk - contribs - library) 11:20, 8 December 2016 (UTC)

Here's a table summarizing the above, with one or two additional bits of information. I think something like this would be worth adding to the article, but there are some inconsistencies -- when is Lake Minchin, for example? Mike Christie (talk - contribs - library) 10:51, 9 December 2016 (UTC)

Duration (kya) Name Notes
191 Escara
120-98 Ouki May be subdivided in the future
95-80 Salinas
46 Inca Huasi Sometimes considered to be the Minchin phase
30.8-28.2 Minor lake event at Salar de Uyuni
33.4-31.8 Minor lake event at Salar de Uyuni
24-20.5 Tauca/Sajsi An early phase of Tauca
18.5-8.5 Tauca
13-11 Tauca/Coipasa A late phase of Tauca
Perhaps worth adding some citations for the table data, but it seems fine. The problem with Lake Minchin is that the concept was coined a bit before reliable dating techniques existed. When they came up it was found that a) the highest shoreline actually belongs to Tauca and b) that Minchin may not have been one single lake phase, instead being an artificial combination of Ouki and other lake cycles. Bit of a WEIGHT issue seeing as there does not seem to be a consensus on the use yet. Jo-Jo Eumerus (talk, contributions) 12:28, 9 December 2016 (UTC)
Yes, citations of course; I didn't want to fiddle with that if you don't think this is worth adding. Perhaps not including Minchin is the best idea, and make sure the text around the table gives the reader the information you just mentioned? Mike Christie (talk - contribs - library) 12:33, 9 December 2016 (UTC)
I've concised the tables using abbr=on, added a table comparing other Pleistocene paleolakes on Latin American Altiplanos.
What would be good for the infobox and coords in general is to position the red dot in the depocentre of the lake (by approximation). Now it is very off centre. If that reflects the paleogeography, it would be right, but does it?
Intro needs expansion to fully reflect the contents and to push the TOC down to the end of the infobox. Tisquesusa (talk) 15:39, 9 December 2016 (UTC)

Yes, treating Lake Minchin in the text (perhaps first discuss Minchin and Tauca, then the other lakes mentioned in the table, then the table?). @Tisquesusa: A bit bothered by the wholly unsourced table of altiplanoes you've added. I don't think there is any information on the depocentre, the coordinates probably should be removed - I've no clue which coordinates to use otherwise. Jo-Jo Eumerus (talk, contributions) 15:48, 9 December 2016 (UTC)

The information from the newly added table comes from Wikipedia itself for Mexico and this article. Sources would be good to add, I can do it after the weekend, out of time now. It is illustrative to show the altiplanos in Latin America, the paleolakes formed there and the relation with the settlement of the Americas.
The coords cannot be removed; they need to show a location. Then the lowest present-day topographic point? Now it is off to the west. Tisquesusa (talk) 15:53, 9 December 2016 (UTC)
I am not seeing sources for most of the claims there... Jo-Jo Eumerus (talk, contributions) 10:48, 11 December 2016 (UTC)

Article structure and organization

I made an offline outline of the article and have been staring at it for a while. I think it would be good to move some of the contextual data higher up in the article, and arrange things so that the context is given before details in every case. This is a bit tricky because all the information is so interlinked (glaciation/precipitation/climate/known climatic events/other palaeolakes), but I think we should try. The article is very information-dense, and I think a more top-down narrative approach might help.

How about starting the body of the article (i.e. the existing Overview section) like this (please pardon any inaccuracies in the content; I'm trying to freely paraphrase):

The Altiplano, a high plateau in the central Andes extending over 1,000 km (620 mi) north to south, has a dry climate today, but has supported a series of lakes over the last 200,000 years, the most recent phase of which is known as Lake Tauca. The largest surviving lake is Lake Titicaca, at the northern end of the plateau; there is also Lake Poopó, further south, which is a remnant of the ancient lakes. The exact duration of each phase is still a matter of debate, but overall the sequence is thought to be as follows.
(insert table of palaeolakes, with dates/durations)
Lake Tauca was initially thought to be a separate phase from Sajsa and Coipasu, but the three are now considered to be a single lake event, with Sajsa and Coipasu regarded as phases of Lake Tauca. Earlier research posited a Lake Minchin as well, but the term, while still debated, does not have an agreed-upon definition. At its greatest extent, Tauca covered most of the southern Altiplano, including Lake Poopó and what are now the salt flats of Salar de Uyuni and Salar de Coipasa; these salt flats were deposited when Lake Tauca disappeared.

The current structure goes into considerable detail about the lake itself, before zooming out, so to speak, to give the climatological context and the research on the duration of the lake. This is logical but it makes for some difficulties because, for example, the section on duration has to repeat material already given in the first section, such as the discussion of the phases. It might be better to have less material in the overview section, and reserve the detail for the subsections. For example, suppose we were to use the introductory paragraphs I give above, or some version of them, along with two or three more paragraphs mentioning a few more key facts -- climate background during that time, and perhaps the hydrological contributions from Titicaca/glaciers/precip. Then go to the Climate section next, with an initial paragraph setting the climate context -- what stadials/interstadials happened during this time period, mention the Lake Glacial Maximum, the Central Andean Pluvial Event, and so on. Then go into detail on the specific climate changes in the Altiplano. At the end of this section, the reader should understand that lakes like this often formed during glacial periods, and what the evidence shows for the climate at the time. Some of what's currently in the Context section might go here too. Then geology/hydrology, and perhaps duration after that, then biology.

What do you think? I think something needs to be done to make the information here more digestible. Mike Christie (talk - contribs - library) 22:21, 12 December 2016 (UTC)

I'll take a gander tomorrow. Sorry for not being so active here - Lascar (volcano) and Llullaillaco are distracting me at the moment. Jo-Jo Eumerus (talk, contributions) 23:34, 12 December 2016 (UTC)
No problem; don't feel pressured by me -- I'm happy to work at your piece, or stop for a while if you're busy. I always have other projects I can be working on in the meantime! Mike Christie (talk - contribs - library) 01:26, 13 December 2016 (UTC)
OK, taking a look at your proposed arrangement, it seems fine for me. There are several inaccuracies but you know that already. Jo-Jo Eumerus (talk, contributions) 20:03, 13 December 2016 (UTC)
If you like, I can do a sandbox rewrite, without the citations, of the overview section. I'm hesitant to do it in mainspace because I have no access to the sources and I'd need you to review for citation locations after we agree on the rewrite. Or would you prefer to do a draft rewrite yourself? Either way works for me. Mike Christie (talk - contribs - library) 00:29, 14 December 2016 (UTC)
I think a sandbox rewrite (at Talk:Lake_Tauca/FA1 , just like with the GA review?) would be good, with Jo-Jo as primary collector of source material (unfortunately I also don't have access to Sciencedirect et al.). It would help to structure the now academic and listed after the other comments written article better.
Let there be no misunderstanding; I like Jo-Jo's work, otherwise I wouldn't promote the article to GA or send him an award for it. But FA needs a bit of rewriting to serve the less informed readers some more. The more eyes, the better. Tisquesusa (talk) 02:44, 14 December 2016 (UTC)
Sorry, but after Lascar and Llullaillaco I need a short break from large scale edits. If someone wants to write up a sandbox with a more "for regular readers" text and with citations carried over, they can do so. It's easier to rearrange the current text while keeping the citations than having to transfer the whole citation body to new text, I believe. Jo-Jo Eumerus (talk, contributions) 11:11, 14 December 2016 (UTC)
No worries; there's no deadline. I'm happy to try a sandbox rewrite, but I've got a couple of other things going on too, so let me know when you think you're ready to take another look at this, and I'll do a rewrite at that point, and link it here for discussion. Mike Christie (talk - contribs - library) 14:30, 14 December 2016 (UTC)
I'd be ready now, if it's fine for you. Jo-Jo Eumerus (talk, contributions) 10:33, 18 December 2016 (UTC)
I just became one of the TFA coordinators and I need to schedule a few days of TFA to make sure I know what I'm doing, so I'll do that first, and then come back to this, probably today. Mike Christie (talk - contribs - library) 11:53, 18 December 2016 (UTC)
I'm just getting started on this; will report back here when I have something I think is worth looking at. Mike Christie (talk - contribs - library) 21:26, 20 December 2016 (UTC)

I started working on an expansion of the overview text I wrote above and quickly ran into another problem. Take a look at User:Mike Christie/Sandbox3, which is a collection of every cited statement in the article that refers to an age, depth, surface area or altitude of Lake Tauca or one of the associated phases. Currently it's in the order that the statements are presented in the article (that's the first column).

The main problem seems to be that the dates (and other data) given by the various sources vary so much that there's no way to say "Lake Tauca" and have it mean the same thing in every context. Just a few examples: Blard (2013) gives 17-14.5 kya as the duration; that's the most recent source. Fornari et al (2001) have 26.1 - 15.6; Sylvestre (1999) has 15.4-12; Clayton & Clapperton have 13.9-11.5; Grosjean (1994) says it probably started 17.4-12.4; and earlier estimates are 12.5-11 kya, 14.1-11 kya, 14.8-7 kya, and 13-10 kya.

I initially thought that the right way to deal with this was to build a table of durations, as I suggested above, because I thought that would clarify things. Unfortunately what it clarifies is that there is no way to say "Lake Tauca" in one context and have it mean the same thing as in another context. For example, the article has "Around 28,000 BP [...] preceding Lake Tauca by about two millennia", followed three sentences later by "The radiometric age of Lake Tauca ranges from 72,600 to 7200 BP". "Lakes Tauca and Minchin have been considered the same lake system": but which "Tauca" are we talking about here? Does this include Coipasa? Sajsi? Ticaña? There are at least half-a-dozen different statements that the lake peaked at a certain level or at a certain time; these all vary and the article quotes them without privileging any one of them above the others.

I haven't read the sources used here, but in most cases they seem to be research papers rather than secondary summaries. Is the problem that there is no source that provides an overview; that summarizes the research and gives the "state of play"? If, for example, we knew that the last two or three sources (Blard 2011, Blard 2013, maybe Placzek, Quade & Pathcett 2006) were regarding as superseding the work that had gone on before, then we could take their numbers and relegate the older sources to footnotes, or only use them when they provide additional information that does not seem in conflict with the more recent sources. But I suspect we don't know that to be the case, so we can't privilege any of the studies over the others.

I'll keep thinking about this but I confess I'm puzzled for the moment. I can think of several recommendations, none of which is very attractive:

  1. Strip out all the details that vary from source to source, giving only the broad outlines of e.g. dates, depths, phase names, and so on, and perhaps going into more detail on some of the specifics where not contradicted by others sources -- mineral data, or contemporary climatic events, for example. This avoids the inability to privilege one source, but it loses a lot of information.
  2. Treat the article as a review of the sources, and organize it by source/date, so that you review the changing research over time. This might be good for a historian of the field, but is a poor way to present data to the reader.
  3. Assume the most recent one or two sources have included information from the older sources, and write the article on that basis. I don't think we're justified in making this assumption.
  4. Reduce much of the article to tables, like the sandbox table I link to above. This avoids writing a narrative we can't source (e.g. we have little basis for assessing the discrepancies) but is very dry.

If forced I'd go with option 4 above, but I'm still thinking about it. If you're still willing to have me involved in trying to help improve this article (i.e. if you're not mad at me for stirring up difficulties) I'd like to find another editor with the appropriate background and get a third opinion. Actually my brother-in-law, who is a professor of archaeology and has a background in geology, is staying here for a week starting tomorrow, so I might pick his brains on this. Mike Christie (talk - contribs - library) 23:58, 20 December 2016 (UTC)

@Mike Christie: You are not stirring up difficulties, all of the concerns you have raised so far are potentially legit issues. However, I don't think that we can put the dates into a table at all. The reason why the article relies on primary research for the timing is because most sources which aren't primary research give a large number of estimated ages (their own interpretation of the primary data) that are in disagreement with each other and tend to lack context - I did already try to put that into a table earlier and realized it won't work. That's a common issue with using secondary sources for raw information. That's why I sort of aimed for #2 when I wrote that section. Honestly, I don't think science is progressed enough on this to establish a definitive timeline. Jo-Jo Eumerus (talk, contributions) 09:11, 21 December 2016 (UTC)
https://en.wikipedia.org/w/index.php?title=Lake_Tauca&oldid=739016764 was the last revision with the previous table. I suspect that not all summarizations are equally reliable; if so, we could maybe use a table after all and prune out all the less-reliable summarizations. Jo-Jo Eumerus (talk, contributions) 09:28, 21 December 2016 (UTC)

I spoke to my brother-in-law about this article; he doesn't know anything about Wikipedia article norms but has an academic background in geology, so take his comments on that basis. His feeling was that assembling the sources on Lake Tauca is something that might be done by a graduate student for a publishable paper, but that to make a Wikipedia article out of the material available requires too much synthesis. If no such synthesis already exists, then I agree with him that the material can't be synthesized here. However, there must be a source somewhere with some summary-level discussion of Andean palaeolakes, though perhaps nothing that focuses specifically on Lake Tauca. So perhaps the material here, along with more material on the other lakes, could be used to write Andean palaeolakes, with the secondary source used to help organize the article.

I've asked another Wikipedia editor who is a geologist to comment, and I will also leave a notice at WT:GEOL. Mike Christie (talk - contribs - library) 12:50, 31 December 2016 (UTC)

Huh. My impression is that writing Wikipedia articles is all about assembling sources, provided that a) they are reliable, b) one does not take conclusions from them that are not explicitly stated and c) the sources are actually all about the same thing. Some of the sources here are in-depth discussion of this particular last paleolake, others discuss more general aspects and mention some information on it as well. Jo-Jo Eumerus (talk, contributions) 16:46, 31 December 2016 (UTC)
Fair point. You have the sources and I don't, so I can't be too definite. We don't have a survey article, though, right? Something that pulls together the various papers and presents a summary of the state of understanding? Without that I think it's going to be very hard to put a coherent structure together for the article that's more than a listing of results from the various sources.
If nobody responds in the next few days from WT:GEOL, I'll ask at WT:FAC for an editor with science experience to comment. Mike Christie (talk - contribs - library) 17:30, 31 December 2016 (UTC)
I am a geologist (MSc.) and am working on similar (paleo)lakes in the Andean chain (Colombia). I see both your points of view; Mike asking for a summary article to use as source to avoid the rather fragmented style of sourcing various articles and Jo-Jo's point that such a summary source shouldn't be needed to write a proper Wikipedia article. In essence, this Lake Tauca wiki article is that first summary of sources, honouring the principle of Wikipedia. It is not "original research" (not saying anyone claimed that), but a compilation of research that hasn't been done elsewhere before. I am seeing the same with Colombia, where summary sources are -surprisingly- rare.
Imho that shouldn't hinder the FA review process; the goal of Wikipedia is to present referenced material from reliable sources in a coherent manner (synthesis). That coherent manner is at the moment maybe not ideal, but then it should be made more coherent. The introduced tables definitely are a step in that direction, the text may need rewriting in some areas. Tisquesusa (talk) 18:18, 31 December 2016 (UTC)
I think the issue is whether the synthesis required to make the article coherent goes further than WP:SYNTH allows. I'm going to post at WT:FAC asking for other opinions, rather than wait; I don't think I can add more here. Mike Christie (talk - contribs - library) 22:12, 31 December 2016 (UTC)
Note left here. Mike Christie (talk - contribs - library) 22:22, 31 December 2016 (UTC)

I am starting to wonder if Tauca may be too big and complex a topic to make a good FAC about. There is not much science done on these Central Andes topics and good summary sources are thus rare. This also leaves a lot of unsettled questions such as "when did Lake Tauca really exist" or "when did Nevado Sajama last erupt" and you need to write "not yet settled upon; X, Y and more recently Z are the proposals" and "only radiometric date is A". Jo-Jo Eumerus (talk, contributions) 22:35, 31 December 2016 (UTC)

I've only glanced over this, but it seems a very similar type of problem to the video game/actor biography problem that we frequently get when there are no over-arching sources, only a lot of fragmentary ones. This may be different though: am I correct in saying that we pretty much represent all the available material here? This immediately puts it ahead of the other type of article when we can never be sure that we are representing the sources. Here, we almost certainly are. Then we have the problem of synthesis. I wonder if the best suggestion might be the one of Mike's, where we treat the article as a review of sources, then we don't have to draw any conclusions or give contradictory information. Give whatever facts are certain, then present an overview of the academic work. I think this would be how a history article might treat a similar problem. There remains the danger of reader accessibility, but perhaps not if we include the certain information early on, and only then go into detail about the uncertainties. I'm not sure if I'm making sense here, and as I say I've only glanced at the article. I'll get back in my box now. Sarastro1 (talk) 12:34, 2 January 2017 (UTC)

Central Andean Pluvial Event vs Central Atacama Pluvial Event

Central Andean Pluvial Event and Central Atacama Pluvial Event may or may not be the same thing; the former is covered in this article. Jo-Jo Eumerus (talk) 13:23, 6 September 2020 (UTC)

Frontiers

Parking this here as I am not sure it's reliable enough - Frontiers has a somewhat dodgy reputation. Jo-Jo Eumerus (talk) 18:06, 7 September 2021 (UTC)