Exploring Historical Geology with Wolfram Language

Okay. Hello, everyone. My name is Jeff Bryant and I am a member of the WolframAlpha team. I'm a research programmer with that team. Today I'm going to be presenting some kind of recreational stuff that I've been able to dabble in as part of my work here with curating data, as well as some tools that's going to help us to explore historical geology. So this is a little bit different than just, you know, picking up rocks and analyzing them, but it's more about how the Earth has changed. So with that, let me get started here. All right, so as you've seen in the abstract reconstructions, the geological history of the Earth can be carried out using a number of tools in the Wolfram language. We're going to be using throughout this talk, Geographics, some entity value stuff, as well as some Wolfram function repository items that I've written. So that's just a little bit of a preview of some of the things that you're going to see. So what is historical geology? It's a branch of geology that studies the changes to the Earth and its systems over geologic time using preserved evidence. So this is everything from continental placement, even atmospheric evolution over time, fossils, all that kind of stuff all falls under historical geology. And, you know, animals that have kind of come and gone, you know, evolved and then gone extinct, such so on and so forth. And we're going to start with just some basic entity value stuff. So entity value has been around since version 10, and one of the primary domains that we're going to explore for historical geology is the geological period entity type. So you can actually see quite a few periods that we have within here. These are not just periods, periods, kind of a double use word here. But we have not just periods, but we have eons and eras, periods, epochs and ages. So as you scroll down here and you can see all these entities, these are all the different entities that are within this entity type here. And they're all hierarchical, some are subsets of some of the others. And we'll be able to explore some of that with some of the things coming up here. Within this domain, we've got a number of properties. Everything from. Some of these are just textual. They kind of give you some descriptions of some things that happened, for example, biological events. We can see that here. So during the Ordovician period, for example, you had, at the very end, you had, or I'm sorry, at the very beginning, you had the Cambrian Ordovician extinction event. You had the appearance of some of the first vertebrates, appearance of coral reefs. So quite a few things happened during the Ordovician period and it was also ended by another mass extinction. We have a few properties here. One of them is biodiversity. So you can see for the Phanerozoic eon, the Ordovician period and the Mesozoic era, I've got three different time series and they're all being plotted together on one date list plot. So you can see the blue in the background gives you the entire Phanerozoic eon. And within that you can see the Ordovician period here in orange, and the Mesozoic era here in green. So this is a way that you can combine these and you can kind of see this is the biodiversity. So you can see where the mass extinctions were. Every time you see the number of genera, so genus, the number of those dropped, you get mass extinctions. So many of the geological periods in history were defined based on mass extinctions. So you can see some really big ones here. So here you can see at the very end of the Mesozoic, that's when the dinosaurs and all the other life, the Cretaceous Paleogene extinction, there was also one just before the dinosaurs. Here you can see this is called the Great Dying, also known as the Permian extinction. So you can see that in such a plot. We've also got some time series data on sea level. So here you can see I've plotted the sea level changes throughout the Ediacaran period and then followed by the Cambrian period. So this is just before you've probably heard of the Cambrian explosion. This is just before that. You can see sea levels were in general on the rise. And then at the start of the Cambrian period, shown in orange here, you can see that it started to kind of plateau here. But there was a sea level rise just before the Cambrian period, globally. There's also another. Here's another time series that I'm kind of showing here. Atmospheric CO2. Some of the atmospheric stuff is a lot. This is kind of related to climatology. It's kind of debatable. And so you often get a lot of different studies that try to analyze what the atmosphere was like back then. And as a result, you can get multiple curves showing as the time during the Carboniferous period, as it changed, you get different curves showing what the trend was depending on how they derived their estimates. So this is four different types of sources for atmospheric carbon dioxide. They vary quite a bit. So you have to be very careful with these types of studies. So that was with entity value directly. Now I'm going to switch a little bit to the Wolfram function repository. Now, the Wolfram function repository has lots of different functions. Some of them are, you know, not quite polished enough to make it into the system. They were written by external people. Sometimes they're just very niche. So they're maybe a little too niche to make it into the system level. So that's probably the case with something like this. It's not to say it couldn't be incorporated at some point, but for now it's just a resource function. It's called geological period chronology chart. This particular chart is useful for kind of placing the various geological periods. I mentioned that they're hierarchical. This allows you to see that hierarchy. So here, for example, you can see I've done the entire Paleozoic era. You can see that the Paleozoic era right here was a subset of the Phanerozoic, and it was followed by the Mesozoic and the Cenozoic, and you can see all the subdivisions within that. So the Paleozoic began with the Cambrian down here, followed by the Ordovician, so on and so forth. And then of course, all the way up to the top, we are actually in. We're in the Phanerozoic eon. And here you can see the Cenozoic era, but the one that we asked for, the Paleozoic, is highlighted in red here. It's kind of a subtle red, but it's there. That's the one you asked for. And then we can explore some other periods here. So here's the Quaternary period. So we've essentially. You can see Quaternary is up here at the very top on this top chart. And down here we've zoomed in a little bit. So here you can see the Quaternary, it's up here. That's the period that we are in right now. It's part of the Cenozoic. It was preceded by the Neogene and the Paleogene. You can see we've got a time scale off to the right here, telling you the age and millions of years. And you can dig in even further here. So here we've got the Pleistocene epoch, and these are the ages within the Pleistocene. So here you can see the Calabrian age, which I asked for, is a subset of the Pleistocene. It's followed by the Chibanian and then the late Pleistocene, and then right after that was us, the Holocene. In addition to that, WFR we're going to come back to the Wolfram language itself and we're going to explore some stuff with geographics, yet another tool for exploring the historical geology here. So, just as a primer, Geographics basically is set up to handle modern continents and data for the most part. And so most of what you see here is part of the geo background. Here you get the modern continents, North America, Asia, Africa. For geological history, though, we want to go back much further so these become less relevant. So, as you may know, with geographics, you can overlie things like the United States is a polygon here. We've got it styled red, and you can see that's overlaid on top of the geo background. So here's an example where we've taken a polygon and put it on top of the background and highlighted in red. But if your polygons that you want to show represent the position of the continental plates in some distant past, then if you overlay those polygons here, here, you can see that I've asked for the continental plates property from the Ordovician period. I can evaluate that real quick. So you can see what you get. You can see that for a given data object in the Ordovician period, we get an association. We have snapshots throughout the Ordovician of polygon sets. So you can see 445 million years ago, we have one set of polygons. If you keep scrolling down, you can see we have another one at 460 million years ago. And you keep scrolling down further. It's a bit much to look at all at once. Here you can see 480 million years ago. So that was the beginning of the Ordovician. So you can take all these polygons that are returned by that property, go down here, and you can say we can plot them on geographics, but you can see these modern continents kind of get in the way. So what you can do by that is you turn off the default geo background, since that modern stuff's all part of the background. And then you can see just the polygons that come from the property. And here we've selected a particular set of polygons, the ones from the snapshot 445 million years ago. So this is essentially where the continental plates were that far back. And then once you've done that, you can apply whatever styling and stuff that you want to those polygons here, you can see we've styled them brown. We're still plotting the entire world, but we've Set the GEO background to light blue, so that kind of gives you an ocean color. We've selected a different projection and we've turned on the grid. So this gives you something more like what you might expect from an ancient continental arrangement. So you can see here, I don't know if you can recognize this or not. This is actually modern North America, although it's turned on its side, so the east coast is actually facing south. So this was called Laurentia and it's actually kind of rotated and much further south than it is in modern day. But that's where we were during the Ordovician period for those of us here in North America today anyway, so that's all fine, but it's kind of a little bit of work that you might have to do if you want to do that kind of thing. So I've created another Wolfram function repository item called Continental Plate Maps. That makes it a little bit easier. You can just feed in the entity into this WFR and once again you get these. In this case they come back rendered already for you. So rather than you just getting raw polygons back for each of the snapshots in the association, you actually now get a geographics back. So just it's kind of an ease of use type, function, repository, item. If you're interested in a particular snapshot of time, you can just put in a year instead of an entity in. And assuming it's one of these snapshots that's available, you'll get back just the geographics result. And you can even combine those together and you can create a movie. And I'm going to play this real quick here. Hopefully this comes through smoothly. You can see how the continents kind of moved over time, then broke up and eventually led to our modern arrangement. I'm going to stop the autoplay here. So here you can see probably about 300 million years ago, you've got Pangea. As you keep going forward, Pangea breaks up, splits apart and starts moving into their modern configuration, which we get to basically today. Once again, these are the continental plates. It doesn't show anything about sea level, nothing like that. This is just the position of the continents. There's no elevation or anything involved here. Alright, so something else that we have that's actually very new, we don't even have linguistics on the production site yet. That'll be coming in the next alpha release. So here in the next couple weeks. But the data is already there. It's a new entity type called Geological Formation Data. This was actually a predecessor to we're actually working on overhauling species data. I know Stephen mentioned this early on, we're going to have this much more complex taxonomic species level. One of the things that data included was dinosaurs and the geological formation that those dinosaurs were found in. And we wanted a way to symbolically represent those. So we created a new entity type called geological formation. Comes with a number of properties that you can see here. Some of these are very technical, like S plate and G plate. I'm not exactly sure what those are used for, but they're kind of an identifier for particular type of data. And for a given. Here's an example for you may have heard them. If you're into dinosaurs, you might have heard of the Morrison Formation. Excuse me, you can ask for the lithology of that particular formation, and it tells you basically the types of rocks that are found in that formation. So mudstone, sandstones, limestones, lots of sedimentary rock. And here, you can see here an example where I've asked, what geological periods does that formation correspond to? And you can see that it corresponds to several different ages. And then going further with that, we can ask to plot from the geological period of those of that formation. We sort them by their time range, and we can call the geological period chronology chart that we did before to see. You know, we can see that the geological periods that they were part of are part of the Late Jurassic. So the Morrison Formation comes from the Late Jurassic and the Early Cretaceous. So for each one of the entities that we saw in the previous slide, you get a chronology chart showing where each of those time periods are. So it's Late Jurassic, Early Cretaceous. So that's why dinosaur fossils are often found there, because it's during the Mesozoic, when the dinosaurs were around. So you can also plot where those can be found. So here's an example where just from my own personal experience, I'm from originally Indiana, and so I used to do some fossil hunting in southeast Indiana, down in this area. And here you can see we've actually used geolisplot to plot where these formations are found. And in this case, it's the Liberty Formation, the Whitewater Formation, and the Fairview Formation. And they all kind of overlap down here in southeast Indiana, Ohio and Kentucky. And I used to. That's Ordovician and age. And I used to actually find all kinds of fossils just right there on the surface. It's kind of a fun little geology trip if you want to take your kids there. Just lots of road cuts, very old fossils Though not dinosaurs predates fossils about 450 million years old. You can actually query dinosaurs for what geological formations they belong to. And so here you can see those. So T Rex belongs in all these different formations. And you want to plot where are those formations found. You can actually do that with Geolisplot. So T Rex is this result, and I've done those. And you can see they're all basically found in the United States, some up here in Canada and stuff, but Montana, the Dakotas, all that kind of area. So you get the Hell's Creek formation. That's one I've heard of. But, yeah, it's like if you want to find T Rex fossils, these are where they've typically been found, where all. You see all these little orange areas. So, useful information. You can go a little bit further here and ask for all dinosaurs and all formations that dinosaur fossils are found in. And here I've plotted a little dot on the world showing you are all the places in the world where known dinosaurs have been found. Notice they're all on land. Dinosaurs are all land animals. There's lots of aquatic reptiles and all kinds of things like that, and flying reptiles, but those were not dinosaurs. Dinosaurs are specifically land animals that had a particular hip structure, and they're actually very well defined taxonomically. And then here's an example where I've actually used a georegion value plot to show where the highest density of dinosaurs fossils are found. So you can see, once again, they're kind of in the American west here. There's a couple places over in China that are pretty high. But, yeah, dinosaurs are found in quite a few places, but many of them are found in particular areas just where the conditions were best. All right, and we've got another one here, another Wolfram function repository item, a geological period bedrock plot. So something you might want to know is if you're in a given location and you strip away all all the topsoil, all the farming, all the soil that's built up over time, and you get down to the bedrock. How old is that bedrock? So here's an example of a function repository item that I've created here called geological period bedrock plot, where you can put in an entity, I can format that. So if I do shift command E or shift control E on Windows, you can see that that's the Permian period. And all I've done is I've put the GeoCenter on Oklahoma, and I've said plot the Permian period bedrock and go out to a range of 500 miles. And you can see that all this area you see in orange here and over here a little bit on the east side of the Rocky Mountains, especially to the south, this is where you can find rocks that are of Permian in age. By default, they are colored according to the Commission for the Global Map of the World. So there's standardized colors that are used for each age. Here's an example here in the Midwest where I've plotted multiple time periods. So Ordovician, Silurian, Devonian, Pennsylvania and Mississippian. And I've centered on Cincinnati. Once again, this is the area that I used to go fossil hunting in. So right about down in here, you can see that you see Ordovician period rocks, all this kind of dark, kind of cyan colored, blue, green color. That's all Ordovician age. Outside of that, you've got Silurian. Outside of that, you've got Devonian. Outside of that you've got Mississippian. And outside of that you've got Pennsylvanian. So it's like you've got really old rock and it gets gradually younger as you get away from Cincinnati. This is called the Cincinnati Arch because this was an area of uplift that was then sheared off through geological period. And so you end up with kind of a bullseye where the older rock and then it gets gradually younger as you get outwards from that. You can kind of see that structure here, but you can see if you're in a given area here in Illinois, you're probably going to get more things from Mississippian and Pennsylvanian type. Whereas if you go over to southeast Indiana, you're going to get older stuff. All right, moving on. So here's an applied use here. So typically when you have rock layers, you get just a general trend in geology. You get older rocks at the bottom and the rock layers get gradually younger as you go upward as they get deposited later. So younger rocks are on top, older rocks are at the bottom. Typically that's going to be, you know, a continuous trend unless something happens. And so here, for example, we go back to the geological period chronology chart that I showed before. You can see we've got the Precambrian. And then at the bottom, the oldest is the Hadean. That's when the Earth was still molten. After that you got the Archean, when the first continents first started to form. And after that, Proterozoic. And as in general, you move upwards. So this is kind of like the rock layer. You get the oldest rocks at the bottom and younger as you move upwards. So that's why the time chart is organized the way that it is. It's kind of, it mirrors the way RO are deposited. But sometimes due to things like erosion or lack of deposition, you can end up with a gap in the rock record. So normally you would see, you know, you don't get rocks from this time period because everything was molten. The first actual rocks, the oldest rocks in the Earth, appeared during the Archaean. And then as you move up, you should find younger rocks. But sometimes after these rocks get deposited, you can get erosion and it'll shear those off. When that happens, you get a discontinuity in the rock record and that's known as an unconformity. I'm not going to read through all this. You can download through the handout there if you want to find out a little bit more about that. But those unconformities can actually be seen visually. And so, for example, here's a case where I've used geological period bedrock plot to plot rocks that are Cambrian Age and Meso Proterozoic. This is about a billion years difference in time. And I've plotted them at the same time and I've centered on Frederickstown, Missouri. And you can see right down here we've got those types of rocks, two rocks of wildly different ages right next to each other. That is evidence that you're going to find a discontinuity there, one of these unconformities. So if I zoom in a little bit. So here it's the same, same data. I've just zoomed in and I've gone to about 10 miles from Frederickstown and you can see we've got the Cambrian rock and the Mesoproterozoic rock right next to each other. And it goes right through here. What that's evidence of is you could probably find this in a road cut. So actually me and my wife a little while back, we actually went on a road trip to see if we could find that. And sure enough, we found a road cut where we found this. This is an example of Mesopoterozoic 1 1/2 billion year old rock. And right on top of it you see this rock that's distinctly distinctively different in texture, very cobbly. This is Cambrian age. So what you have here is an unconformity right where the two meet. So you've got a billion years of rock missing in the rock record. And what's interesting is that this is actually something that's found globally. So globally it's actually really Easy to find throughout the entire world. About a billion years of rock is missing in most areas. And it's likely due to the fact that in the intervening time period, you had what was called the Cryogenian Period, and you had large glaciers covering the entire world. And it actually likely eroded much of that rock in many, many areas. But this was nice visual evidence that you could actually see that thing called the Great Unconformity. It was also called the Cryogenian Period, was also often called Snowball Earth because the glaciers probably went all the way to the equator. So deviating a little bit. There's also some stuff you can do with third party data. So not just stuff that's within the Wolfram language, but you can get some very sometimes large data. Sometimes it's very niche, very specialized, but you can do some pretty cool things with it. So I'm just going to show some examples of that, some things that you can do. So for example, you want to find out how the geologic, how the United States was assembled. And I found a data set that actually shows if you go down below the surface, bedrock to the basement, the very bottom of that, you want to find out when was that rock first put in place. And I actually was able to create this geographics example here. I put tool tips here where you can see the ages of these. So, for example, here in Illinois, a lot of that comes from the Masatzal Formation, and that's about 1.7 to 1.6 billion years old. So it's quite old. And you can see that in general, if you look here, some of the oldest ones, here's if you go up here to Montana, in Wyoming, in that area, you can see you can find rock that's 3.5 billion years old. And as you move this way kind of downwards, the rocks get younger. So it appears that North America. So here's 1.8 billion years old, 1.7, 1.5. So it generally gets younger as you go down into the right here. So it appears that as North America was forming, you had the, these kind of suture zones that formed right along here where it was slowly assembled over billions of years. Here's an example where I've done the same thing, but I colored it differently. So this is using the standard colors that I've used before. But if you want more distinct colors, you can do that as well. It's the same data, just colored differently. Notice here we have this little brown, little squiggle that kind of cuts through some of this. This is actually called the Mid continent, My mind is escaping me here. We're actually the United States tried to rip itself apart basically about a billion years ago. And what you're seeing here is the intercontinental rift. That's what I'm thinking of, the Mid Continent rift. And this basically shows a failed rift that tried to rip North America apart and then failed. But it left behind this actually a large rift zone that's deeply buried. You can actually see that in this data. Here's a different way of annotating the data where I've actually put some call outs here also up here in Alaska. And if you wanted to visualize the textual data, you could also do it in tabular form as well. I was able to form a data set out of the data that's in the external data set. And I want to mention that there is a link up here to Community where I go into more detail about all that. If you're interested in this particular data set, including where the data comes from. Okay, continuing on another one here and also another community post here. What if you want to do like we did before, where you show global maps of the world but you want to include sea level changes, things like that. Such data is available, it's image in nature rather than polygons. And I was able to find a data set that shows that right here. I'm going to speed up here because we're getting short on time. I want to make sure there's time for some questions here. So here you can see that I've compared the two side by side. So here's the old polygon data that we had that's in the system. And then here's the image data. So you can see some of this. Just because the continental plates were here doesn't mean that it was above sea level. So the images are better at showing where the sea level was at that time. Once again, it's a community post, so feel free to check that out if you're interested in how that's done. Another one here, a little bit more applied. Where are coal beds found here in the United States? The community link right there you can see that a lot of it's found over here in kind of the West Virginia area, but there's some over here in Illinois as well as well as in the Southeast and a few spots throughout the Rockies and things like that. So here you can see I've got the ages as tooltips. So here it's tertiary in age, here it's Pennsylvanian and Permian in age. And down here I've got these are coal beds by type, so this tells you not the age but the type of coal. So here you can low volume bituminous to high volume bituminous lignite down here, so it's lower quality coal, much younger, hasn't officially turned into true coal. And here's another example here. So I've zoomed in a little bit here. So I've zoomed in on Pittsburgh. So I've got some additional resources here you guys might want to check out. Some of these are just links to some of the ref pages that tools that are useful entity value, dinosaur entity type, geological period and geographics several Wolfram Function repository items that I've used throughout this presentation. And I've also got not just some of the ones that I showed here, but some additional ones as well. Some community posts that are all related to this type of thing, including some posts that talk about mountain building, orogenies, orogeny is the word for mountain building, episode in geological history and a few other things on here as well. So I encourage anyone that's interested in this type of thing to check that out and that should be about all of my presentation. If anybody has any questions, I would be happy to field those. I thought about bringing a rock in that when I did the field trip to. Let's see here. I'm trying to remember what slide I put that on here when I did the field trip to southeast Missouri here when I looked at this ancient rock. By the way, this is just about the oldest rock you can find anywhere in the Midwest. So if you're interested in that area, it's called the St. Francois Mountains. So that whole area down there you can actually find volcanic rock here in the Midwest, which was just bizarre. Normally all the rocks you find in the Midwest are going to be sedimentary, coal, limestone, that kind of thing. But it was interesting to be able to drive from Champaign, which is right here, all the way down here to southeast Missouri and actually find volcanic rock. And I actually had a sample I was going to bring in, but had this been an in person talk I could have passed around the room, but otherwise I would have just had to hold a rock up to the camera. I'm not sure how useful that would have been, but interesting to bring back a sample of volcanic rock from the Midwest though. Somebody's asking how long the talk's available. I don't know how long they're going to provide that. That's probably more of a question for the tech team. I'm not in control of that so I'm not exactly sure. I assume you mean the download, the handout. I'm not exactly sure how long the site stays up and accessible. Somebody's asking about how about fossilized fauna and flora other than dinosaurs? That's something we would like to have. I believe this is probably more of a question for Keiko. I know I can't remember if she's already given her talk. There was a talk she was going to give about species and the data set they used to overhaul species data included some dinosaur information, and I believe that's included in an update of dinosaur data that we've had historically. But I don't think it had very much, if any, ancient flora and fauna. It would be great to have that. But that would make a species data truly huge, because keep in mind that something like 99% of all species that have ever existed have gone extinct. So if you thought species data was big, now imagine if we included all of that. So it's something we would love to do. I personally just having a hobby of going fossil hunting, I would love to have a species representation or an entity representation for all the fossil creatures and stuff that have been found. But keep in mind it's even more difficult though, because trying to tax put those types of things into a taxon is tricky. Sometimes they change their mind. Biologists and taxonomists. There's also kind of a civil war going on between how you do taxonomy, especially for extinct species, because you've got different ways of classifying things. You've got the traditional kind of Linnaean taxonomy, but then you've got more evolutionary things. That's called. It's more based on what has evolved from what. And that's a different approach to how you rank things. And in that type of a model, you would see that mammals have descended from reptiles. But if you look at a Linnaean system, reptiles and mammals are two totally distinct areas. So if you go back in history and find fossils of these things that were kind of bordering between fossils of reptiles and amphibians, for example, what branch do you put them in? So the species talk, apparently somebody says, is at one o' clock today. So you might attend that talk and see some of the improvements that they've made to species data. And it's a lot more hierarchical than the old data used to be. So, yeah, hopefully with time we will get some fossil taxa. It's going to be very challenging though, because it's almost a moving target. About the time they find a fossil, a few years later, they they have a debate and they find out that it's just the young form of something that they already knew about. That's happened with, I know, some dinosaurs. So some species that used to exist kind of went away because it turns out what they had found was actually just a. A young form of a Styracosaurus or something like that, and it wasn't actually a separate species. So somebody said that they think that the notebook should be available to February 2024. So hopefully that's correct. I don't have a way of verifying that, though. Cladistics. Thank you. Somebody. Somebody mentioned that it's called cladistics. That's the other way of text of classifying historical animals. And typically, if you're studying fossils, you're probably going to follow more of a Cladistics type thing than you are more of a Linnaean taxonomy type thing, just because Linnaean stuff doesn't really address evolution. So it's kind of messy. And I'm glad I'm not involved in that civil war between the two different biological camps and having to figure out the best way to do things, things like that. All right, well, we're a couple of minutes over, so I'm going to go ahead and end this. I thank you for attending and hopefully we'll have some additional posts from me on community in the future. Thank you very much.