Chris Rowan
LightNews LightNews
allochthonous.bsky.social
Chris Rowan
@allochthonous.bsky.social
2.3K followers 530 following 1.4K posts
I like rocks.

I think and talk about plate tectonics, geological hazards like earthquakes, the history of the Earth system, and how we silly humans can live sustainably on our amazing planet.
Posts Replies Media Videos
allochthonous.bsky.social
🧪 For hundreds of thousand of years over the Pliocene-Pleistocene boundary, through climate change and hominid evolution itself, a technology/method for making stone tools remained the same.

A level of technological stability almost too alien for us to understand.

arstechnica.com/science/2025/1…
Text Shot: The last hominins to use Oldowan tools looked very different, and probably lived and behaved very differently, from the first; over this huge span of time, the stone tool technology itself changed less than the beings using it.
November 8, 2025 at 5:23 PM
allochthonous.bsky.social
I feel that I might have to pass on this exciting opportunity, given that (i) I am not the Chris Rowan who authored this paper, and (ii) even so, I know enough to know that the 'Indian hedgehog' it refers to is a gene and not a nocturnal animal.
Start of an e-mail that reads "Dear Dr Rowan, Due to your involvement in the field and the research you published in your paper, "Indian Hedgehog Signalling in Kidney Development", IntechOpen invites you to contribute a chapter to "Nocturnal Animals - Adapted or Inherited Plasticity" an Open Access book edited by Dr. Heimo Mikkola from University of Eastern Finland. Work with an internationally recognized peer group and gain increased visibility for your published work."
October 22, 2025 at 2:33 PM
allochthonous.bsky.social
Reading William Whewell's 1832 review of Lyell's Principles of Geology, Vol 2, in which he coined "uniformitarianism", and I just love this passage summarizing what in many modern geological textbooks would describe as the principle of uniformitarianism.

babel.hathitrust.org/cgi/pt?id=hv...
Screenshot of part of a scanned page from a 19th century scientific journal. Text reads: "Now it appears that, in the present order of things, certain changes do go on, both in the position of portions of the earth, and in the forms of certain organised beings. The volcano and the earthquake are seldom long idle; the ocean is an unremitting assailant of the solid earth; the countless host of streams and showers second his attack, on the other side, by efforts formidable from their multitude and perseverance. The coast yields; the crest of the mountain descends; large tracts of the earth tremble and change their elevation: the volcanic island lifts its head above the waves. Here, at least, are some elements of mutations in the form of the earth and of the bed of the sea. The more we examine such causes, the more constant the more extensive, the more powerful, does their operation appear. In the course of progressive ages, what effects may they not produce? And what limit are were to place to the time during which their work has proceeded? We know that this past period must be long; we know not how long. Who shall prove to use that the forces which we ourselves witness are too weak, or unfit, to produce all the facts of position which the earth's crust exhibits? Such is the reasoning of the advocate of the geological adequacy of the existing dynamical laws of the world."
October 15, 2025 at 4:29 PM
allochthonous.bsky.social
But I think the coolest example I've found is ikaite pseudomorphs within metamorphosed Dalradian rocks in Scotland, where they appear to have grown *after* an episode of burial and metamorphism that generated a slatey cleavage.

pubs.geoscienceworld.org/gsl/jgs/arti...

#MinCup25
Series of images from a scientific paper, with the caption: Kerrera Slates. A: Bed of metadolostone with aligned ikaite pseudomorphs plunging (lower arrow) in the plane of the cleavage towards the NE, crenulation lineation in the slates is marked by upper arrow (Coin is 2 cm diameter); B: Aligned ikaite pseudomorphs in calcareous slate steeply plunging (arrow) to the NE. Pseudomorph size decreases towards lower part of exposure (Hand lens is 5 cm long); C: Aligned ikaite pseudomorphs in calcareous slates, arrow shows alignment of elongate grains (Hand lens for scale); D and E: Large elongate ikaite pseudomorphs with pyramidal terminations from the Easdale Slates.
September 14, 2025 at 1:54 PM
allochthonous.bsky.social
Ikaite pseudomorphs found in the rock record are usually referred to as 'glendonite', and because ikaite is associated with sediments deposited in cold bottom waters, glendonite can be used as a paleoclimatic indicator.

Figure from supplementary information of www.science.org/doi/10.1126/...
Heat map showing Phanerozoic distribution of glendonite (pseudomorph after the hydrated carbonate ikaite) over time (horizontal axis, from about 450 million years ago on the left to the present on the right, divided into what looks like 1 million year bins) and space (vertical axis, from 0 degrees absolute latitude on the bottom to 90 degrees at the top, 5 degree bins). Where glendonite has been identified within a particular age/latitude bin, it is shaded a light pink for a low number of recorded occurrences, green for a moderate number of recorded occurrences, and blue for a high number of recorded occurrences. The distribution of glendonite occurrences is shown to be very non-uniform over geological time, with low to intermediate counts between 40 degrees and 75 degrees latitude in the past 50 million years; largely moderate to high counts between 60 and 90 degrees latitude from about 180 to 100 million years ago; and low to intermediate counts between about 45 and 85 degrees latitude from about 300 to 240 million years ago.
September 14, 2025 at 1:54 PM
allochthonous.bsky.social
👀 pushing people to rate answers on subjects they have limited knowledge of seems almost designed to push the models towards truthiness.

https://www.theguardian.com/technology/2025/sep/11/google-gemini-ai-training-humans
Text Shot: In December, Google sent an internal guideline to its contractors working on Gemini that they were no longer allowed to “skip” prompts for lack of domain expertise, including on healthcare topics, which they were allowed to do previously, according to a TechCrunch report. Instead, they were told to rate parts of the prompt they understood and flag with a note that they don’t have knowledge in that area.
September 12, 2025 at 3:20 PM
allochthonous.bsky.social
So much to cringe at in this interview: "fusion on the grid in 5 years because AI"; "don't believe your geological survey about fracking, believe me"; but “renewables have had enough subsidies” is top-level chutzpah from a former fossil fuel exec.

https://www.bbc.com/news/articles/cqlz5p314z0o
Text Shot: Mr Wright defended the billions of dollars of cuts the Trump Administration has made to renewable energy subsidies. He said wind power has been subsidised for 33 years and solar for 25 years. "Isn't that enough?" the Energy Secretary asked: "You've got to be able to walk on your own after 25 to 30 years of subsidies."
September 12, 2025 at 12:31 PM
allochthonous.bsky.social
A thick sliver of ultramafic mantle rock, which has a much mechanical strength at 25-50 km depth due to its higher melting temperature, helps to explain how the Himalayas have gotten as high as they are without collapsing in on themselves.
September 11, 2025 at 12:29 PM
allochthonous.bsky.social
⚒️🧪 New modelling suggests there might be a "mantle sandwich" beneath the Himalayas, with the underlying slice of Indian crust not underthrust as a rigid block, but flowing as a partially melt beneath the Asian plate, and underplating it. agupubs.onlinelibrary.wiley.com/doi/10.1029/...
Panel of a figure from a scientific paper, showing an approximately 500 km south to north cross-section through the structure of the lithosphere beneath the Himalayan mountain chain, as interpreted by the authors of this paper. In the south, 40 km-thick Indian crust (coloured dark grey), with various north-dipping thrust faults in the top 20-25 km, is underlain by Indian lithosphere (coloured light purple). At about 200-250 km along the section, the transition to 30 km-thick Asian crust (coloured light grey), is marked by another north dipping thrust fault, labelled as the "Indus-Yarlang-Tsangpo suture. Beneath this, a roughly 30 km purple layer of Asian mantle lithosphere is underlain by a shallow slab of highly deformed Indian crust, still connected to the surface, and deformed mantle lithosphere beneath. The Asian mantle lithosphere is a contrast to the standard model of the collision, which has the Indian crust directly underthrusting the Asian crust.
September 11, 2025 at 12:29 PM
allochthonous.bsky.social
But it turns out the "formed by geologic process" part of the definition comes into play, according to the International Mineral Association - if crystallization is occurring within a living cell, it doesn't count.

(from www.sciencedirect.com/science/arti...)
Screenshot, with text reading "Substances formed from man-made materials by geological agents, as well as purely biogenic substances that have no geological counterparts or whose origins owe essentially nothing to geological processes, are not regarded as minerals. Therefore, if a crystalline phase was formed entirely within the organism, it could not be approved as a mineral; however, if an organism excreted a liquid that then crystallized on a natural rock surface owing to a geological process, it could be approved as a mineral."
September 6, 2025 at 3:49 PM
allochthonous.bsky.social
Namibia is one of the very few places it has been discovered: this pegmatite I visited in 2008 is in South Africa, close to the Namibian border. Too far south, but possibly formed at a similar time.... not that I would have recognized Jeremejevite if I'd seen it.!
View of a near-vertical cliff of light pink rock, in moderate shadow under a blue sky. In the center of the picture, a complex interfingering network of light grey rock, containing large white and shiny crystals, can be seen stretching over the rock face. In the bottom foreground, viewed from behind, are the head and shoulders of a dark-haired, light skinned person who is examining the outcrop.
September 3, 2025 at 3:07 PM
allochthonous.bsky.social
Even 15 years later, I remember how impressed I was by this massive stibnite specimen in the American Museum of Natural History in New York. A very Fortress of Solitude/crown of the Witch King aesthetic with all those interlocking columns.

#MinCup25
A large mineral specimen, more than a foot wide in at least one dimension, sits behind a glass case, illuminated at each end by spotlights from above. The specimen is an aggregate of multiple elongated, columnar crystals, generally of the order of a centimeter wide and ten centimeters long projecting outwards at various angles and interlocking with each other in complex patterns. The crystals have a grey, metallic sheen.
September 2, 2025 at 1:22 PM
allochthonous.bsky.social
If you are positing a seven day creation, then we know one of the creator's two picks was "fast"; I guess the coastline plagiarism is good evidence that their other pick was "cheap".

xkcd.com/3132/
Three panel comic with stick figure people. First panel: a stick figure is gesturing with a pointer to a map of the world centred on the Atlantic, with the east South American and west African coast highlighted in red, and saying "People had long noticed that South America and Africa had similarly shaped coastlines." Second panel shows the figure is giving a lecture to other stick figures sitting behind desks. They continue: "In the 20th century, geologists finally found the explanation:" In the third panel, the stick figure sitting at the first desk pipes up, saying "Plagiaris-" "Continental drift", says the lecturer, interjecting. "Oh.", finishes the person at the desk. Alt text on original comic reads: "Hey, a bunch of the early Cretaceous fossils on each side seem to have been plagiarized, too!"
August 23, 2025 at 4:01 PM
allochthonous.bsky.social
Felt I should alt-text this, although it kind of destroyed my soul to do so.

Anyway, see thread by @cbdawson.bsky.social: making your class make conceptual diagrams with LLMs in the first class of the semester is possibly your best bet for convincing them they don't actually understand anything.
Pseudo-3D cross-section of a generic landscape with blue ocean on the left and a flat green plain in front of a steep-sided, snow-capped mountain on the right. In the blue sky, a sun is on the left and a white cloud on the right. On the plain, a puddle-shaped lake is directly juxtaposed against by a vaguely meandering river-shaped water body that flows into the subsurface when it reaches the front edge of the picture. Further to the right, a single small plant growing on the plain is scaled up to gargantuan proportions. Superimposed on this is text and arrows meant to show the water cycle. Starting from the sun in the top left, an arrow labelled "evaporation 97%" heads down to the oceans. Then an arrow points from the oceans to a "surface water 0.01% label" in the subsurface underneath the alleged lake and river. And arrow with only half a head then finished on a "transpiration" label underneath the giant mutant plant. Meanwhile, an arrow labelled "atmosphere 0.001%" heads from the sun to the cloud in the top right of the figure. The cloud and giant mutant plant are linked by a non-directional line labelled "Condination". Two "groundwater 0.6%" labels hang around, unconnected and unloved, in the bottom left and bottom right.
August 20, 2025 at 12:53 PM
allochthonous.bsky.social
I had to try this myself. I first asked it to describe the water cycle, then it offered to draw it.

Two for two on water cycle diagrams without a real cycle on them…
August 20, 2025 at 12:28 PM
allochthonous.bsky.social
⚒️ A #Strataday with a view from central Vermont exactly one week ago: Owl's Head atop the Devonian Knox Mountain Pluton, a granite riddled with multiple generations of cross-cutting pegmatite dikes.
View from an exposed elevated rock outcrop across a forested, mountainous landscape beneath a blue sky. Large trees in the foreground frame an elongated lake between two forested peaks in the middle ground, with multiple undulating mountain ridges on the horizon. The rock exposure is a patchy grey and white granite, and contains two diagonal dikes that cross in the centre. View of an undulating, mostly horizontal surface of granite bedrock, with greyish and whitish patches representing more quartz rich and feldspar-rich regions of the granite. The image is about 1m across. The surface is cut by multiple vertical dikes of material with similar composition but coarser crystal size, which are slightly raised from the surface. Different orientations (mid-left to mid-right, bottom left to top right, bottom right to top left) represent different cross-cutting generations of these pegmatite dikes.
August 9, 2025 at 2:38 PM
allochthonous.bsky.social
Mention of Verdnasky and tsunami reminds me of this interesting tidbit I read about at nearby Port Lockroy: a tsunami generated by an Aleutians earthquake in 1946 washed away boats and a base hut when it reached Antarctica.
Photo of an information board with text on top of the picture of an iceberg, located at Port Lockroy, Antarctica. Two columns of text on the left under the title "The Port Lockroy tsunami" reads: On 2 April 1946 the sea suddenly ebbed away exposing the seabed between Goudier and Doumer Islands. The men realised that, if the sea returned as fast as it had gone away, the base might be flooded. So they climbed onto the roof of the Nissen hut. When the sea returned, it rose 1m above the high tide mark and washed away one of the three old Norwegian boats. The men blamed the wave on an iceberg rolling over but, on the previous day, a tsunami had been generated off the Aleutian Islands. It caused widespread havoc as it spread down the Pacific but it might have gone unnoticed at Port Lockroy except for the constricted channels accentuating the wave's height. In the rightmost column, a grainy black and white photo of a hut in an icy landscape has the caption "A hut built only 1.5 m above high water in a very narrow channel at the Argentine Islands (50km south) was swept away by this tsunami."
July 31, 2025 at 2:57 PM
allochthonous.bsky.social
Thought it was worth looking up the DART ocean buoy data - here's the record for the buoy closest to the earthquake, which occurred around 11:25 GMT on the 29 July. Important information for estimating tsunami arrival times and magnitudes around the Pacific Basin.

www.ndbc.noaa.gov/station_page...
A plot of water level measured by an ocean buoy close to the Kamchatka Peninsula, between about 10 pm GMT on the 29 July and 10 am GMT on the 30 July. The green dots show measurements taken every minute, and the red line at around the 11:30 GMT mark shows measurements taken every 15 seconds. A series of higher frequency oscillations, peaking at around around plus/minus 1.5m, are clearly seen on top of the diurnal/tidal cycle between about 11:30 pm GMT on the 29th July and 2am GMT on the 30th July.
July 30, 2025 at 11:33 AM
allochthonous.bsky.social
20 years after Hurricane Katrina, have any lessons been learnt? Since the levees were rebuilt without accounting for sea-level rise, it seems that sadly, the answer is no.

https://arstechnica.com/science/2025/07/20-years-after-katrina-natgeo-and-new-orleans-remember/
Text Shot: Van Heerden has run storm surge models assuming a three-foot rise in sea level. "What we saw was the levees wouldn't be high enough in New Orleans," he said. "I hate to say it, but it looks like another ticking time bomb.
July 28, 2025 at 11:13 AM
allochthonous.bsky.social
This image of a phytoplankton bloom in the Baltic Sea looks like a giant tracer experiment! 🤩

From the NASA Earth Observatory: earthobservatory.nasa.gov/images/15458...
Satellite image of a dark blue ocean surface. Streamers and whorls of green mark high concentrations of phytoplankton being stirred by wind blown currents. Islands can be seen in the bottom left and top right.
July 26, 2025 at 2:40 PM
allochthonous.bsky.social
And finally, I kind of want to know what the mass notification event that sent thousands of phones vibrating and triggered a false system alert was...
Screenshot with text: AEA has issued alerts to Android phones for a total of 1279 events detected by the Android phone network [as of 31 March 2024 (data S1)]. Of these, three were false alerts. Two were due to thunderstorms. The system detects many non-earthquake seismic events every day owing to various sources of acceleration detected by the phones. Most thunderstorms are correctly identified as such, but these two instances were outside the classifier range. Learning from these events, we expanded the range of parameters for a sonic source and increased the confidence required before an alert is used (materials and methods). Replays showed that these two events would no longer produce an alert. The third false alert was due to a mass notification event vibrating many phones. The final sentence is highlighted in blue.
July 18, 2025 at 2:14 PM
allochthonous.bsky.social
The algorithm being used at the time did seriously underestimate the magnitude of the main shock (and, to a lesser degree, the aftershock), so many in the actually affected area might not have been sent alerts.
Screenshot with text: The most challenging events for all EEW systems are the largest ones, M >7.5. The most impactful earthquakes since AEA became operational were the 6 February 2023 M 7.8 Pazarcik and M 7.5 Elbistan earthquakes in Türkiye. AEA was operational at the time in Türkiye and detected both events. For the first event, the initial magnitude estimate was 4.5 at 7.1 s after OT, and the maximum magnitude estimate was 4.9 at 18.7 s. A total of 512,411 BeAware alerts were issued out to a distance of 64 km. In the second event ~9 hours later, the initial magnitude estimate was 6.1, at 24.4 s after OT, which increased to magnitude 6.3 5.2 s later. A BeAware alert was delivered to 3,944,909 phones in this event, with warning times ranging from a few seconds to more than a minute.
July 18, 2025 at 2:14 PM
allochthonous.bsky.social
A spectacular demonstration of how the thing that Google was once famously good at - reliably finding super-obscure things on the internet - is completely orthogonal to what generative AI is “good” at. http://lablemminglounge.blogspot.com/2025/07/the-confident-insanity-of-generative-ai.html
Text Shot: This is a confidently told, gripping tale of a day in my actual life. The only problem is that it has nothing to do with my actual lived experience in that place on that day.
July 2, 2025 at 9:03 PM
allochthonous.bsky.social
Speaking of the North American ice cap, it seems even at the peak of the last ice age it wasn’t a barrier to human migration: people were definitely wandering around New Mexico 21-24,000 years ago

https://arstechnica.com/science/2025/06/study-confirms-white-sands-footprints-are-23000-years-old/
Text Shot: that makes for a grand total of 55 radiocarbon results in support of the earlier dates across the three studies. "It's a remarkably consistent record," said Holliday. "You get to the point where it's really hard to explain all this away. It would be serendipity in the extreme to have all these dates giving you a consistent picture that's in error.
June 20, 2025 at 2:05 PM
allochthonous.bsky.social
⚒️ I really feel this house should be owned by a geologist.

This is a particularly impressive example, but a number of lots around here are adorned by erratics left behind by the melting North American ice sheet 10,000 years ago; sadly, ours is not one of them.
View from the road up to a detached two storey house with pale grey siding and a dark grey roof, at the top of a gentle rise. In front of the house on the left a two to three metre-high boulder, mottled various shades of tan, orange and dark brown, blocks a clear view of the garage. To the right, the boulder blends into a rock wall built from half-metre scale blocks.
June 20, 2025 at 1:53 PM