Do they at least make things more efficient or better in some other way, most of the time?
The R-QBAM main thread
- Thread starter Tanystropheus42
- Start date
I feel like they are pretty useless, with no benefits and the harm of additional paperwork.
I wanted to got this one done a little quicker, but RL got in the way a little, so here we are. Today's patch adds almost all of Chile (Easter Island will be included with the next patch), plus the final chunks of Argentina, in addition to a moderate general overhaul of the rest of Argentina.
What does it say that I'm not even going to complain about the fjords, as Canada and Greenland have numbed me to it. It was tricky and fiddly, but I got it done. One thing I will note, contrary to many maps I'm not showing Lake Colhuehuapi, as I'm pretty certain it was a dry lakebed in January 2022, if these satellite images are to be believed. Random bit of trivia, but this patch adds a new southernmost point yet added to the map. The Chilean Diego Ramirez Islands, which have finally beaten out the Norwegian Bouvet Island that has held the record since November 2022.
Next up on the schedule, Misc Islands 2, adding several islands and archipelagos in the South Atlantic and Pacific; most of Hawaii (like Alaska, some outlying islands get cut off by the world-border), Johnston Atoll, Pitcairn, Easter Island, the Falklands, South Georgia and the South Sandwich Islands. After that, French Polynesia, then the rest of the eastern Pacific. I hope to have started on Antarctica before the beginning of May.
Oh, and I almost forgot it, but here's the most up-to-date version of the ice cover layer without borders as I said I'd post previously;
Patch 119 - Chile & Patagonia;
- Added Chubut
- Added Santa Cruz
- Added Tierra Del Fuego (the Argentinian province)
- Added Chile (minus Easter Island).
What does it say that I'm not even going to complain about the fjords, as Canada and Greenland have numbed me to it. It was tricky and fiddly, but I got it done. One thing I will note, contrary to many maps I'm not showing Lake Colhuehuapi, as I'm pretty certain it was a dry lakebed in January 2022, if these satellite images are to be believed. Random bit of trivia, but this patch adds a new southernmost point yet added to the map. The Chilean Diego Ramirez Islands, which have finally beaten out the Norwegian Bouvet Island that has held the record since November 2022.
Next up on the schedule, Misc Islands 2, adding several islands and archipelagos in the South Atlantic and Pacific; most of Hawaii (like Alaska, some outlying islands get cut off by the world-border), Johnston Atoll, Pitcairn, Easter Island, the Falklands, South Georgia and the South Sandwich Islands. After that, French Polynesia, then the rest of the eastern Pacific. I hope to have started on Antarctica before the beginning of May.
Oh, and I almost forgot it, but here's the most up-to-date version of the ice cover layer without borders as I said I'd post previously;
Patch 119 - Chile & Patagonia;
- Added Chubut
- Added Santa Cruz
- Added Tierra Del Fuego (the Argentinian province)
- Added Chile (minus Easter Island).
Good job!
Are you waiting to do the administrative divisions of the smaller nations until after you've finish up the map?
When was the last time I got two patches done in a single day? I know it's happened before, but I can't for the life of me remember. Anyway, today's patch is a bit of a random mish-mash, featuring several islands that I otherwise didn't really know what to do with.
I also have a slight admission - I actually finished Hawaii all the way back in November 2021. It's been just out of view off the edge of the map for well over two years, I just never felt like posting it publicly. It's been so long since I did it that Hawaii needed a moderate revamp, as the original was a bit of a rush-job, I apparently had a bit of an issue with projection distortion, and I have access to better sources and basemaps these days. The old ad-hoc corals from before I had access to this source needed a full revision. Like with Alaska, several outlying Hawaiian islands and reefs get cut off by the world-border, so the state isn't completely done, but the main body of it is.
This patch also allows me to finish a country that has been incomplete since practically the beginning, as it includes the final UK overseas territories of Pitcairn, the Falklands, and South Georgia and the South Sandwich islands. This excludes the Antarctic claims, but I'm treating Antarctica separately due to the nebulous nature of the Antarctic Treaty, so I'm calling it finished. Also, I added Norway to the completed list when I added Bouvet Island, so there's precedent.
Random bit of geographical trivia, but depending on either interpretation of the Antarctic Treaty, the UK is the southernmost country in the world. If you consider the territorial claims as valid, then the UK shares the title with the other countries that claim a wedge of Antarctica, namely Chile, Argentina, Norway, Australia, France and New Zealand. If however you hold that the Antarctic Treaty voids prior territorial claims, then the southernmost country is the nation that possesses the southernmost land not covered by the Antarctic Treaty. That means an unnamed islet in the Southern Thule Group of the South Sandwich Islands, currently administered by Britain.
A bit pointless, but I thought it was interesting when I figured it out, so I thought I'd share it.
Now for some replies;
Antarctica gets bumped up the schedule for three reasons. Firstly, I agree that it doesn't get enough love, and secondly it's a big, annoying job, and I want to get it done soonish. But the main reason however is that I've had a mostly finished Antarctica WIP sitting around on my laptop for two and a half years.
First, a little context. I didn't decide one day to make a QBAM replacement. Instead I always thought that it was too big of a challenge, and that the best option was to keep on incrementally patching and improving the QBAM. However, it gradually sunk in just how broken the QBAM is, and I drifted away from making patches around 2019. I still did some stuff behind the scenes, but never posted anything.
Wind forward two years. In the autumn of 2021, I came up with a fun AH scenario that I wanted to explore, whipping up a map for it on the QBAM in late August. I wanted to do the rest of the world, but being well aware of the QBAM's many shortcomings, I decided to go on a new push to improve things. The first job I decided to tackle was perhaps the biggest one - Antarctica.
The QBAM Antarctica is just dead wrong. There is basically nothing redeemable about it, its is simply horrendously inaccurate, and this has been a known problem for a while. [Quick tangent, I think this is because it was added after the rest of the QBAM had been published. Whereas the coasts and borders for the earliest QBAM's appear to be in a mildly squished Robinson projection, some important details like the world-border and Antarctica were added later, in an incredibly shoddy job that has done a lot to cripple the QBAM going forward. But I digress.]
Because the traditional QBAM Antarctica is so irredeemably flawed, I knew I had to make something from scratch. I knew that the community had long suspected that the QBAM was in or adjacent to Robinson, so I took my copy of the 8K-BAM, shrunk it down a little so that G.Projector would play with it (it doesn't appear to work well with large file-sizes alas), then reprojected it. I figured that actual Robinson was a good enough crutch that I could fudge it to fit with the QBAM's projection once it was done, safe in the knowledge that anything I did would be a massive improvement over the original, even if it was a little wonky.
Instead what happened was, over the month of September 2021, as I slowly traced over the Antarctic coasts, I gradually came to realise that a complete QBAM re-do was actually possible given the tools I had on hand - the 8K-BAM backlog, G.Projector and Paint.net. So Once I was done with Antarctica, I changed tack and decided to try for a full-world QBAM overhaul that would become the R-QBAM project.
The first few days were a bit chaotic, as I tried my hand at mapping various bits of the globe before settling on the Contiguous United States and diving in. A couple months later someone pinged me on the Map Thread, I replied with some of my current progress as a teaser, and the response was positive enough that I created the R-QBAM thread. That is also the origin of the Misc WIP layer I've been using to dump areas I've mapped but not published yet, as there were seveal areas I had done some work on that were outside the scope of the published patches that din't feel completely polished.
I have, over the last two years, slowly chipped away at adding the old WIP's, most notably when I added South Africa, and in the form of bits of Canada that proved very useful over the last few months. Today's patch adds the penultimate WIP, Hawaii, leaving just Antarctica, the original, to be added. Here's the thing though - I did those WIP's when I was still finding my feet making the R-QBAM, and I hadn't found many of the sources I now rely on heavily. As such, while the old WIP's are pretty good, they are still full of minor inaccuracies, and need quite a bit of double-checking before I'd be willing to post them.
While the 8K-BAM has some flaws in the polar areas *cough-Greenland-cough*, it appears to have been using a good source for Antarctica. Having since found good datasets myself, I can confirm that the old WIP is fairly accurate in general outline, albeit lacking in detail. At the time I was mostly tracing over the 8K-BAM verbatim, whereas today I would use it as a base while comparing it with half a dozen other trusted sources to produce something a bit more fine-tuned. I didn't want to add Antarctica till I had done most of the rest of the southern continents done, and considering how long the Raj patch took and how I wanted to do South America and Canada in parallel, that ended up taking quite a long time to do.
If I was doing Antarctica from scratch, I would probably divide it up into two or three chunks to do separately. As I've done a lot of the work already, I feel confident doing the entire continent in one patch. Based on how long it took to incorporate those bits of Canada I mentioned earlier, it'll probably take me somewhere between a week and a fortnight to polish up what I have to my current standards.
A bit of a digression I'll admit, but I kinda feel like I need to explain why I eventually post an entire continent without too much apparent effort.
Once the main map is done, I'll have two priorities. Firstly, adding first level divisions to every country where it is feasible to do so (I can't really add them to such countries as Nauru, San Marino or Antigua and Barbuda for example).
The second is continuing work on the 1914 map, both for the useful historical borders and to make a 1914 blank map with reservoirs and other human changes removed or at least accounted for. That should in turn make producing further historical patches easier.
Patch 120 - Misc Islands 2;
- Added (most of) Hawaii
- Added Johnston Atoll
- Added Pitcairn
- Added Easter Island
- Made the Juan Fernandez Islands autonomous (I din't realise they were first time)
- Added the Falkland Islands
- Added South Georgia and the South Sandwich Islands, finishing the UK
I also have a slight admission - I actually finished Hawaii all the way back in November 2021. It's been just out of view off the edge of the map for well over two years, I just never felt like posting it publicly. It's been so long since I did it that Hawaii needed a moderate revamp, as the original was a bit of a rush-job, I apparently had a bit of an issue with projection distortion, and I have access to better sources and basemaps these days. The old ad-hoc corals from before I had access to this source needed a full revision. Like with Alaska, several outlying Hawaiian islands and reefs get cut off by the world-border, so the state isn't completely done, but the main body of it is.
This patch also allows me to finish a country that has been incomplete since practically the beginning, as it includes the final UK overseas territories of Pitcairn, the Falklands, and South Georgia and the South Sandwich islands. This excludes the Antarctic claims, but I'm treating Antarctica separately due to the nebulous nature of the Antarctic Treaty, so I'm calling it finished. Also, I added Norway to the completed list when I added Bouvet Island, so there's precedent.
Random bit of geographical trivia, but depending on either interpretation of the Antarctic Treaty, the UK is the southernmost country in the world. If you consider the territorial claims as valid, then the UK shares the title with the other countries that claim a wedge of Antarctica, namely Chile, Argentina, Norway, Australia, France and New Zealand. If however you hold that the Antarctic Treaty voids prior territorial claims, then the southernmost country is the nation that possesses the southernmost land not covered by the Antarctic Treaty. That means an unnamed islet in the Southern Thule Group of the South Sandwich Islands, currently administered by Britain.
A bit pointless, but I thought it was interesting when I figured it out, so I thought I'd share it.
Now for some replies;
Antarctica gets bumped up the schedule for three reasons. Firstly, I agree that it doesn't get enough love, and secondly it's a big, annoying job, and I want to get it done soonish. But the main reason however is that I've had a mostly finished Antarctica WIP sitting around on my laptop for two and a half years.
First, a little context. I didn't decide one day to make a QBAM replacement. Instead I always thought that it was too big of a challenge, and that the best option was to keep on incrementally patching and improving the QBAM. However, it gradually sunk in just how broken the QBAM is, and I drifted away from making patches around 2019. I still did some stuff behind the scenes, but never posted anything.
Wind forward two years. In the autumn of 2021, I came up with a fun AH scenario that I wanted to explore, whipping up a map for it on the QBAM in late August. I wanted to do the rest of the world, but being well aware of the QBAM's many shortcomings, I decided to go on a new push to improve things. The first job I decided to tackle was perhaps the biggest one - Antarctica.
The QBAM Antarctica is just dead wrong. There is basically nothing redeemable about it, its is simply horrendously inaccurate, and this has been a known problem for a while. [Quick tangent, I think this is because it was added after the rest of the QBAM had been published. Whereas the coasts and borders for the earliest QBAM's appear to be in a mildly squished Robinson projection, some important details like the world-border and Antarctica were added later, in an incredibly shoddy job that has done a lot to cripple the QBAM going forward. But I digress.]
Because the traditional QBAM Antarctica is so irredeemably flawed, I knew I had to make something from scratch. I knew that the community had long suspected that the QBAM was in or adjacent to Robinson, so I took my copy of the 8K-BAM, shrunk it down a little so that G.Projector would play with it (it doesn't appear to work well with large file-sizes alas), then reprojected it. I figured that actual Robinson was a good enough crutch that I could fudge it to fit with the QBAM's projection once it was done, safe in the knowledge that anything I did would be a massive improvement over the original, even if it was a little wonky.
Instead what happened was, over the month of September 2021, as I slowly traced over the Antarctic coasts, I gradually came to realise that a complete QBAM re-do was actually possible given the tools I had on hand - the 8K-BAM backlog, G.Projector and Paint.net. So Once I was done with Antarctica, I changed tack and decided to try for a full-world QBAM overhaul that would become the R-QBAM project.
The first few days were a bit chaotic, as I tried my hand at mapping various bits of the globe before settling on the Contiguous United States and diving in. A couple months later someone pinged me on the Map Thread, I replied with some of my current progress as a teaser, and the response was positive enough that I created the R-QBAM thread. That is also the origin of the Misc WIP layer I've been using to dump areas I've mapped but not published yet, as there were seveal areas I had done some work on that were outside the scope of the published patches that din't feel completely polished.
I have, over the last two years, slowly chipped away at adding the old WIP's, most notably when I added South Africa, and in the form of bits of Canada that proved very useful over the last few months. Today's patch adds the penultimate WIP, Hawaii, leaving just Antarctica, the original, to be added. Here's the thing though - I did those WIP's when I was still finding my feet making the R-QBAM, and I hadn't found many of the sources I now rely on heavily. As such, while the old WIP's are pretty good, they are still full of minor inaccuracies, and need quite a bit of double-checking before I'd be willing to post them.
While the 8K-BAM has some flaws in the polar areas *cough-Greenland-cough*, it appears to have been using a good source for Antarctica. Having since found good datasets myself, I can confirm that the old WIP is fairly accurate in general outline, albeit lacking in detail. At the time I was mostly tracing over the 8K-BAM verbatim, whereas today I would use it as a base while comparing it with half a dozen other trusted sources to produce something a bit more fine-tuned. I didn't want to add Antarctica till I had done most of the rest of the southern continents done, and considering how long the Raj patch took and how I wanted to do South America and Canada in parallel, that ended up taking quite a long time to do.
If I was doing Antarctica from scratch, I would probably divide it up into two or three chunks to do separately. As I've done a lot of the work already, I feel confident doing the entire continent in one patch. Based on how long it took to incorporate those bits of Canada I mentioned earlier, it'll probably take me somewhere between a week and a fortnight to polish up what I have to my current standards.
A bit of a digression I'll admit, but I kinda feel like I need to explain why I eventually post an entire continent without too much apparent effort.
Once the main map is done, I'll have two priorities. Firstly, adding first level divisions to every country where it is feasible to do so (I can't really add them to such countries as Nauru, San Marino or Antigua and Barbuda for example).
The second is continuing work on the 1914 map, both for the useful historical borders and to make a 1914 blank map with reservoirs and other human changes removed or at least accounted for. That should in turn make producing further historical patches easier.
Patch 120 - Misc Islands 2;
- Added (most of) Hawaii
- Added Johnston Atoll
- Added Pitcairn
- Added Easter Island
- Made the Juan Fernandez Islands autonomous (I din't realise they were first time)
- Added the Falkland Islands
- Added South Georgia and the South Sandwich Islands, finishing the UK
So let me join the party today guys since the base whole Americas is complete! It's trully wonderfull achievement
Also... Hi... Uhn again, kinda ackward huh, sorry for my hiatus, my life has been quite messy lately, so...
My plan was to complete the US before posting again but I had gotten so little time as well Lousiana is annoyingly complex that stuck me a few times, but it's been so long since I last posted so here we go:
- Wyoming, Nebraska, Colorado, Kansas, Oklahoma, Texas and Lousiana rivers fully done;
- Rivers from Mississippi and Alabama are half done while Florida and Georgia are a third done;
So yes, I'll sitll get this done... Eventually
Also here's the current whole world map of what has been done rivers for who got lost in the patchs
So yeah, @Tanystropheus42 congrats with the Americas, you ideed endured well the Chilean and Canadian landscapes, and well let me be a bit nitpicky, I know they are incrdible small but they do affect EEZ's so, west of the South Georgia Islands there's the Shag Rocks, and on thesoutheast (EDIT: it's southwest, I messed up the direction) of Kerguelen Islands there the Îlot Solitaire, these 2 minuscle features are missing... Should they be shown? I really not sure, if wasn't for EEZ questions I would not but well.
Also... Hi... Uhn again, kinda ackward huh, sorry for my hiatus, my life has been quite messy lately, so...
My plan was to complete the US before posting again but I had gotten so little time as well Lousiana is annoyingly complex that stuck me a few times, but it's been so long since I last posted so here we go:
- Wyoming, Nebraska, Colorado, Kansas, Oklahoma, Texas and Lousiana rivers fully done;
- Rivers from Mississippi and Alabama are half done while Florida and Georgia are a third done;
So yes, I'll sitll get this done... Eventually
Also here's the current whole world map of what has been done rivers for who got lost in the patchs
So yeah, @Tanystropheus42 congrats with the Americas, you ideed endured well the Chilean and Canadian landscapes, and well let me be a bit nitpicky, I know they are incrdible small but they do affect EEZ's so, west of the South Georgia Islands there's the Shag Rocks, and on the
Last edited:
New river patch!!! =DSo let me join the party today guys since the base whole Americas is complete! It's trully wonderfull achievement
Also... Hi... Uhn again, kinda ackward huh, sorry for my hiatus, my life has been quite messy lately, so...
My plan was to complete the US before posting again but I had gotten so little time as well Lousiana is annoyingly complex that stuck me a few times, but it's been so long since I last posted so here we go:
- Wyoming, Nebraska, Colorado, Kansas, Oklahoma, Texas and Lousiana rivers fully done;
- Rivers from Mississippi and Alabama are half done while Florida and Georgia are a third done;
So yes, I'll sitll get this done... Eventually
Also here's the current whole world map of what has been done rivers for who got lost in the patchs
So yeah, @Tanystropheus42 congrats with the Americas, you ideed endured well the Chilean and Canadian landscapes, and well let me be a bit nitpicky, I know they are incrdible small but they do affect EEZ's so, west of the South Georgia Islands there's the Shag Rocks, and on the southeast of Kerguelen Islands there the Îlot Solitaire, these 2 minuscle features are missing... Should they be shown? I really not sure, if wasn't for EEZ questions I would no but well.
Welcome back :]
Hell yeah good job kingSo let me join the party today guys since the base whole Americas is complete! It's trully wonderfull achievement
Also... Hi... Uhn again, kinda ackward huh, sorry for my hiatus, my life has been quite messy lately, so...
My plan was to complete the US before posting again but I had gotten so little time as well Lousiana is annoyingly complex that stuck me a few times, but it's been so long since I last posted so here we go:
- Wyoming, Nebraska, Colorado, Kansas, Oklahoma, Texas and Lousiana rivers fully done;
- Rivers from Mississippi and Alabama are half done while Florida and Georgia are a third done;
So yes, I'll sitll get this done... Eventually
Also here's the current whole world map of what has been done rivers for who got lost in the patchs
So yeah, @Tanystropheus42 congrats with the Americas, you ideed endured well the Chilean and Canadian landscapes, and well let me be a bit nitpicky, I know they are incrdible small but they do affect EEZ's so, west of the South Georgia Islands there's the Shag Rocks, and on thesoutheast(EDIT: it's southwest, I messed up the direction) of Kerguelen Islands there the Îlot Solitaire, these 2 minuscle features are missing... Should they be shown? I really not sure, if wasn't for EEZ questions I would not but well.
Yeah, I know I said that the next patch would just be French Polynesia, but things kinda snowballed. So instead, have the entire eastern pacific, featuring the aforementioned French Polynesia, the Cook Islands, Niue, a third of Kiribati, Half of American Samoa (Like with Alaska and Hawaii, cut in half by the map border) and half a dozen tiny islands claimed by the United States in the 19th century because a bird once crapped there.
There are however quite a few things that need mentioning, both geographical and political. Apologies in advance for the wall of text.
First up, I'm showing Niue and the Cook Islands as de-facto independent nations. A strong case can be made that as dependent territories of New Zealand they should be classed as Kiwi protectorates and shown accordingly. However, both Niue and the Cook Islands really do function as if they were independent nations, with domestic governments and foreign relations, relying on New Zealand mostly for military protection. While the pair are less sovereign than most nations, and are not member states of the United Nations, they have a similar relationship to new Zealand as the Marshall Islands, Micronesia and Palau have to the United states, are effectively fully self-governing, and participate in multiple UN agencies. In addition, wikipedia lists them in the 'Other states' section in its list of sovereign states, alongside such other edge cases as Somaliland or North Cyprus.
Also, it means I can colour them distinctly rather than have them both in the same shade of light blue.
Not entirely sovereign nations aside, the bigger problem compiling this patch was figuring out how to show the various isolated underwater reefs that dot the South Pacific. While reefs in the immediate proximity of islands are fairly comprehensively mapped, there are several reefs (and, as we'll see, "reefs") quite far away from the nearest land that, while shallow, don't quite peak above the surface. I was planning on just following the world coral-cover dataset I've been relying on so far, however I spotted a few notable inconsistencies in the data that forced me to do some deeper digging.
One of the best resources for this actually turned out to be the Operational Navigation Charts produced by the United States and allied militaries during the latter years of the cold war, digitised here. I last mentioned them in regards to Greenland, for showing the pre-warming coastline and for providing a few fixed locations I used to figure out which datasets are useful and which aren't. The series also covers much of the rest of the globe, with the maps of the South Pacific often highlighting and naming isolated reefs and shoals, which was incredibly useful when trying to figure out whether to show a feature or not. In addition, this book on global coral cover (one of the primary sources for the corals GIS dataset I've been using) digitised by the internet archive here, the big hisatlas maps of the Pacific (which shows islands and reefs in a fair amount of detail) and finally this book on Sailing Directions in the Eastern Pacific from 1952 (that lists the locations of many known or suspected reefs and shoals), all proved useful sources.
First up, I scaled back the reefs northwest of the Hawaiian archipelago. While there are banks there, and fairly shallow ones at that, they aren't full coral reefs. Basically no source aside from the corals GIS data shows them as such, including Spalding et al. (2001). The one exception is Raita Bank, where which crops up in some sources, so it stayed in.
Next an easy exclusion, which Wikipedia agrees is likely a phantom island, Filippo Reef. The only confirmed 'sighting' was in 1926, the sea depth at the reported location is greater than five kilometres and it is omitted by both the coral GIS data, Spalding et al. (2001) and the ONC maps (which don't even cover the patch of sea where the reef would be if it existed). A relatively clear-cut case.
Now we turn to the Cook Islands, which can be divided into two distinct groups of islands, northern and southern. There was only one, fairly easy submerged reef in the northern group, Tima Reef, just southeast of Pukapuka Atoll (not to be confused with Puka-Puka Atoll in French Polynesia). It has a wikipedia entry, the bathymetry data is favourable, it's shown by both hisatlas and the relevant ONC map, it appears on a map in Spalding et al. (2001), and gets a mention in the 1952 Sailing Directions, with a fairly in-depth description. It's there alright, even if the GIS dataset doesn't show it.
The southern Cook Islands in contrast have two, extremely annoying to pin down reefs. First up, Winslow Reef, which I suspect but can't prove is a phantom reef. It has an incredibly sparse wikipedia entry and appears sporadically in the sources. Spalding et al. (2001) shows it, as does the GIS data, and it is mentioned in the Sailing Directions (1952), as well as appearing on the relevant ONC map. But here's the thing, as far as I can tell, it remains largely unexplored. The Directions mention is incredibly brief, its GIS entry is copied verbatim from Spalding et al. (2001), and this report from an expedition to the area in 2013 found nothing at the reef's supposed coordinates. It is however fairly isolated and supposedly understudied, and absence of evidence is not necessarily evidence of absence, so I very tentatively included it.
The second reef drove me slightly mad for most of Sunday. It appears as a large, circular structure, very reminiscent of a coral atoll on the GIS dataset. I thought it was a submerged reef, but was stymied at every turn trying to find the identity of this unknown 'reef'. As far as I can tell, it doesn't even have a name. I already knew that the corals GIS used Spalding et al. (2001) as a primary source, but after spending way too long looking for a digitised copy, I was annoyed to find that that source makes no mention of it in its section on the Cook Islands. Dead-end. I then spent an awful lot of time looking for named reefs in various old atlases on DavidRumsey.com, and drew a blank. I even broke out my own physical vintage atlas to see if I could find anything in the area, with no luck (complete aside, but every cartography nerd needs a cumbersome atlas older than they are).
I eventually found something when I stumbled on the right ONC chart. There, in the right position and exactly the same shape as in the GIS data (accounting for projection distortion) was my mystery reef. And it still told me basically nothing. The 'reef' is labelled as "breaker position approximate (reported 1945)", which is basically useless because it doesn't give a name. Once I stumbled on the Sailing Directions (1952) I checked there, and found this equally unhelpful entry; "Breakers were reported about 230 miles westward of Rarotonga in approximately 21°05' S., 163°57' W.". I know this pertains to the mystery reef as the coordinates line up with its position on the ONC chart, and it gives basically the same incredibly brief description that adds no new information. I have found no other leads, after an awful lot of searching.
In summary, breakers were reported there in 1945 and, as far as I can tell, never again since. It was mentioned in a handful of places and charted on a few maps as a possible shipping hazard but with no further corroboration of its existence. At some point the compilers of the GIS data stumbled on one such map, possibly the very same ONC map I just linked to copying the shapes without doing any further corroboration. The baffling thing is that they added this reef even though it's probably a phantom, but missed the blindingly obvious Beveridge Reef four degrees to the west (more on that below). I suspect it might have something to do with the two features being either side of the cutoff between two maps in the ONC series. The unnamed reef got in, the very real Beveridge reef off to the side on another sheet didn't.
Having reviewed what scant evidence I could draw together, I strongly suspect that this unnamed reef is indeed a phantom reef, and duly omitted it from the R-QBAM.
On another note, there are two major offshore reefs under the Aegis of Niue. The first is the aforementioned Beveridge Reef, which like Tima Reef absolutely exists. It has a wikipedia article, appears on the ONC maps, has an entry in the Sailing Directions (1951), has been discussed in a couple of places on the internet, gets mentioned by Spalding et al. (2001) and even had a ship wreck there as recently as 2017. But oddly it doesn't show up on the GIS data. I included it anyway as the evidence is so overwhelmingly strong, but it was a notable omission. The second Niuan reef is Antiope Reef, which is more of a marginal edge-case, but is mentioned in every source that also covers Beveridge Reef, so I decided to include it. A couple of other reefs in the same general area as Niue such as Harans Reef are more questionable, and are usually considered phantom islands.
And now we reach French Polynesia. The first one to cover is Portland Reef in the very Southeast corner of the Gambier Islands. It appears on the corals GIS data, lines up with the bathymetry, has a wikipedia entry and appears on one of the ONC charts as "Banc Portland". It got in. There are a handful of questionable reefs in the Austral Islands to the far south. Firstly, the GIS data has a small reef to the northeast of Rapa Iti island. I excluded this as it appears in no other source I have found. To name just the most prominent omissions, the Sailing Directons (1952), the ONC maps and Spalding et al. (2001) all either lack this feature or do not mention it, so it was omitted. I judged that two final banks did indeed make the cut. Neilson or Lancaster Reef to the northwest of Rapa Iti, and Moses Bank south of Rurutu. Both are missing from the GIS data but are present in Spalding et al. (2001). Both appear on the ONC maps in the right places and get mentions in the Sailing Directions (1952). Moses Bank is even shown by hisatlas. The combined evidence was enough that I added both.
Well that was a slog. It was annoying, but I lost the better part of a weekend researching this, so I felt it needed explaining in full. Aside from the opaque mess that are the isolated coral reefs, there is a little bit of geographical trivia I wish to impart. Kritimati island is the atoll with the largest land area in the world, making up 70% of the dry land in the Republic of Kiribati, as well as its single land pixel on the R-QBAM. Also, there's an atoll in the Solomon Islands where the locals claim that they were annexed by the Kingdom of Hawaii in the 1850's, and thus they should be US citizens. Certainly an interesting way to try and wrangle an American visa.
Next stop, Antarctica ...
Patch 121 - Eastern Pacific;
- Added French Polynesia
- Added the Cook Islands
- Added Niue
- Added the eastern half of American Samoa (damn map border)
- Added Kingman Reef
- Added Palmyra Atoll
- Added Jarvis Island
- Added the rest of the Line Islands, making up the eastern third of Kiribati
There are however quite a few things that need mentioning, both geographical and political. Apologies in advance for the wall of text.
First up, I'm showing Niue and the Cook Islands as de-facto independent nations. A strong case can be made that as dependent territories of New Zealand they should be classed as Kiwi protectorates and shown accordingly. However, both Niue and the Cook Islands really do function as if they were independent nations, with domestic governments and foreign relations, relying on New Zealand mostly for military protection. While the pair are less sovereign than most nations, and are not member states of the United Nations, they have a similar relationship to new Zealand as the Marshall Islands, Micronesia and Palau have to the United states, are effectively fully self-governing, and participate in multiple UN agencies. In addition, wikipedia lists them in the 'Other states' section in its list of sovereign states, alongside such other edge cases as Somaliland or North Cyprus.
Also, it means I can colour them distinctly rather than have them both in the same shade of light blue.
Not entirely sovereign nations aside, the bigger problem compiling this patch was figuring out how to show the various isolated underwater reefs that dot the South Pacific. While reefs in the immediate proximity of islands are fairly comprehensively mapped, there are several reefs (and, as we'll see, "reefs") quite far away from the nearest land that, while shallow, don't quite peak above the surface. I was planning on just following the world coral-cover dataset I've been relying on so far, however I spotted a few notable inconsistencies in the data that forced me to do some deeper digging.
One of the best resources for this actually turned out to be the Operational Navigation Charts produced by the United States and allied militaries during the latter years of the cold war, digitised here. I last mentioned them in regards to Greenland, for showing the pre-warming coastline and for providing a few fixed locations I used to figure out which datasets are useful and which aren't. The series also covers much of the rest of the globe, with the maps of the South Pacific often highlighting and naming isolated reefs and shoals, which was incredibly useful when trying to figure out whether to show a feature or not. In addition, this book on global coral cover (one of the primary sources for the corals GIS dataset I've been using) digitised by the internet archive here, the big hisatlas maps of the Pacific (which shows islands and reefs in a fair amount of detail) and finally this book on Sailing Directions in the Eastern Pacific from 1952 (that lists the locations of many known or suspected reefs and shoals), all proved useful sources.
First up, I scaled back the reefs northwest of the Hawaiian archipelago. While there are banks there, and fairly shallow ones at that, they aren't full coral reefs. Basically no source aside from the corals GIS data shows them as such, including Spalding et al. (2001). The one exception is Raita Bank, where which crops up in some sources, so it stayed in.
Next an easy exclusion, which Wikipedia agrees is likely a phantom island, Filippo Reef. The only confirmed 'sighting' was in 1926, the sea depth at the reported location is greater than five kilometres and it is omitted by both the coral GIS data, Spalding et al. (2001) and the ONC maps (which don't even cover the patch of sea where the reef would be if it existed). A relatively clear-cut case.
Now we turn to the Cook Islands, which can be divided into two distinct groups of islands, northern and southern. There was only one, fairly easy submerged reef in the northern group, Tima Reef, just southeast of Pukapuka Atoll (not to be confused with Puka-Puka Atoll in French Polynesia). It has a wikipedia entry, the bathymetry data is favourable, it's shown by both hisatlas and the relevant ONC map, it appears on a map in Spalding et al. (2001), and gets a mention in the 1952 Sailing Directions, with a fairly in-depth description. It's there alright, even if the GIS dataset doesn't show it.
The southern Cook Islands in contrast have two, extremely annoying to pin down reefs. First up, Winslow Reef, which I suspect but can't prove is a phantom reef. It has an incredibly sparse wikipedia entry and appears sporadically in the sources. Spalding et al. (2001) shows it, as does the GIS data, and it is mentioned in the Sailing Directions (1952), as well as appearing on the relevant ONC map. But here's the thing, as far as I can tell, it remains largely unexplored. The Directions mention is incredibly brief, its GIS entry is copied verbatim from Spalding et al. (2001), and this report from an expedition to the area in 2013 found nothing at the reef's supposed coordinates. It is however fairly isolated and supposedly understudied, and absence of evidence is not necessarily evidence of absence, so I very tentatively included it.
The second reef drove me slightly mad for most of Sunday. It appears as a large, circular structure, very reminiscent of a coral atoll on the GIS dataset. I thought it was a submerged reef, but was stymied at every turn trying to find the identity of this unknown 'reef'. As far as I can tell, it doesn't even have a name. I already knew that the corals GIS used Spalding et al. (2001) as a primary source, but after spending way too long looking for a digitised copy, I was annoyed to find that that source makes no mention of it in its section on the Cook Islands. Dead-end. I then spent an awful lot of time looking for named reefs in various old atlases on DavidRumsey.com, and drew a blank. I even broke out my own physical vintage atlas to see if I could find anything in the area, with no luck (complete aside, but every cartography nerd needs a cumbersome atlas older than they are).
I eventually found something when I stumbled on the right ONC chart. There, in the right position and exactly the same shape as in the GIS data (accounting for projection distortion) was my mystery reef. And it still told me basically nothing. The 'reef' is labelled as "breaker position approximate (reported 1945)", which is basically useless because it doesn't give a name. Once I stumbled on the Sailing Directions (1952) I checked there, and found this equally unhelpful entry; "Breakers were reported about 230 miles westward of Rarotonga in approximately 21°05' S., 163°57' W.". I know this pertains to the mystery reef as the coordinates line up with its position on the ONC chart, and it gives basically the same incredibly brief description that adds no new information. I have found no other leads, after an awful lot of searching.
In summary, breakers were reported there in 1945 and, as far as I can tell, never again since. It was mentioned in a handful of places and charted on a few maps as a possible shipping hazard but with no further corroboration of its existence. At some point the compilers of the GIS data stumbled on one such map, possibly the very same ONC map I just linked to copying the shapes without doing any further corroboration. The baffling thing is that they added this reef even though it's probably a phantom, but missed the blindingly obvious Beveridge Reef four degrees to the west (more on that below). I suspect it might have something to do with the two features being either side of the cutoff between two maps in the ONC series. The unnamed reef got in, the very real Beveridge reef off to the side on another sheet didn't.
Having reviewed what scant evidence I could draw together, I strongly suspect that this unnamed reef is indeed a phantom reef, and duly omitted it from the R-QBAM.
On another note, there are two major offshore reefs under the Aegis of Niue. The first is the aforementioned Beveridge Reef, which like Tima Reef absolutely exists. It has a wikipedia article, appears on the ONC maps, has an entry in the Sailing Directions (1951), has been discussed in a couple of places on the internet, gets mentioned by Spalding et al. (2001) and even had a ship wreck there as recently as 2017. But oddly it doesn't show up on the GIS data. I included it anyway as the evidence is so overwhelmingly strong, but it was a notable omission. The second Niuan reef is Antiope Reef, which is more of a marginal edge-case, but is mentioned in every source that also covers Beveridge Reef, so I decided to include it. A couple of other reefs in the same general area as Niue such as Harans Reef are more questionable, and are usually considered phantom islands.
And now we reach French Polynesia. The first one to cover is Portland Reef in the very Southeast corner of the Gambier Islands. It appears on the corals GIS data, lines up with the bathymetry, has a wikipedia entry and appears on one of the ONC charts as "Banc Portland". It got in. There are a handful of questionable reefs in the Austral Islands to the far south. Firstly, the GIS data has a small reef to the northeast of Rapa Iti island. I excluded this as it appears in no other source I have found. To name just the most prominent omissions, the Sailing Directons (1952), the ONC maps and Spalding et al. (2001) all either lack this feature or do not mention it, so it was omitted. I judged that two final banks did indeed make the cut. Neilson or Lancaster Reef to the northwest of Rapa Iti, and Moses Bank south of Rurutu. Both are missing from the GIS data but are present in Spalding et al. (2001). Both appear on the ONC maps in the right places and get mentions in the Sailing Directions (1952). Moses Bank is even shown by hisatlas. The combined evidence was enough that I added both.
Well that was a slog. It was annoying, but I lost the better part of a weekend researching this, so I felt it needed explaining in full. Aside from the opaque mess that are the isolated coral reefs, there is a little bit of geographical trivia I wish to impart. Kritimati island is the atoll with the largest land area in the world, making up 70% of the dry land in the Republic of Kiribati, as well as its single land pixel on the R-QBAM. Also, there's an atoll in the Solomon Islands where the locals claim that they were annexed by the Kingdom of Hawaii in the 1850's, and thus they should be US citizens. Certainly an interesting way to try and wrangle an American visa.
Next stop, Antarctica ...
Patch 121 - Eastern Pacific;
- Added French Polynesia
- Added the Cook Islands
- Added Niue
- Added the eastern half of American Samoa (damn map border)
- Added Kingman Reef
- Added Palmyra Atoll
- Added Jarvis Island
- Added the rest of the Line Islands, making up the eastern third of Kiribati
Heck yeah, that's almost an entire hemisphere done! Seriously, we're almost to the home stretch, just Antarctica, the rest of Asia, and Oceania left.
About time we updated the blank map.
And just for my own sake, an alternate map without island outlines:
I apologise if this question has been asked & answered before, but is there any concrete rule being used to determine which rivers are shown as coastlines and which aren't shown at all? The implication being of course that if not, I would rather have them not shown at all, as they do kind of make the map harder to use (especially since the community has a standard of coastlines being visually indistinguisheable from borders).
If they're wide enough to be seen on this scale, then they're shown. That's basically it.
How wide is "wide enough to be seen"? At least the width of one pixel at that latitude? And where does this information come from? Is it purely based on seeing what's visible when a dataset is converted into a QBAM-resolution raster?
Yeah, that last bit is a big problem. For me, I'm able to easily distinguish between coastlines and national borders due to the completely different colors I use for both (instead of just #000000 for coastlines, I use a slightly bluer color so that in my art program of choice I can notice things easier when editing. For borders I use #111111 for national borders.)
I believe so. Any river that can be seen at this level of detail should be given at least some definition. If we were to go further down projection sizes the river borders would have to be shown as coastlines of their own due to their size. I get what you mean with preferring things being neater, though.
Well... almost all of those "rivers" are actually just really skinny reservoires, so they're not actually thin rivers.
Then again, making a map is definitely more of an art than a science. I think there is no real set-in-stone rule for which rivers are added in.
It's great to see that both the Americas are finally done. Good luck on the rest of the world, there is not that much that remains to be done.