Latest / Cariboo Signals / Science, Wonder And The Natural World

Transcript

Science, Wonder And The Natural World · Cariboo Signals

• 0:15
  Riley: Welcome to Cariboo Signals, an AI-generated review of the latest tech news and ideas that impact our rural communities. I'm Riley. It's Sunday, October 18th, 2026. Today's theme is Science, Wonder & the Natural World. Reaching you from the heart of the Cariboo, the homeland of the Secwépemc, Tŝilhqot'in, and Dakelh nations.
• 0:37
  Casey: And I'm Casey. We're looking at the living world today, from forest soil to fossil crocodiles. Out at Horsefly Lake this morning it's partly cloudy and around 1 degree, with fog through the day and a high of 11. Across the Cariboo, 100 Mile House and Williams Lake are both around 0, Quesnel is around 1, and Alexis Creek is around -1, all under partly cloudy skies.
• 0:58
  Riley: Starting in our own backyard.
• 1:01
  Riley: Most of today's news lands in our own backyard, and the biggest piece is the municipal vote count. First, Forest Grove. Jenifer Kosman is calling on Canada Post to bring back her community's post office. It closed at the end of January when the postmaster retired. Canada Post called the closure temporary, pending a new postmaster. Almost nine months later the building is still shut and the mailboxes are gone. Residents drive at least 15 minutes into 100 Mile House for their mail. Kosman sews hand-made products for sale, and she says she has heard the Forest Grove Store may be interested in taking on a post office.
• 1:40
  Casey: Almost nine months is a long temporary.
• 1:43
  Riley: The Williams Lake Film Club is screening the documentary American Doctor at Paradise Cinemas on Thursday, October 22nd. The film was directed by Poh Si Teng and debuted at Sundance earlier this year. It has since won audience awards at both Hot Docs and the Sydney Film Festival. Williams Lake for Humanity is partnering on the screening.
• 2:06
  Casey: Counting has started across the Cariboo now that the polls have closed. In Quesnel, Mayor Ron Paull faces former city councillor and MLA Coralee Oakes and Brian Waters. Eighteen candidates are running for six council seats, and that includes all six incumbents. In Williams Lake, Mayor Surinderpal Rathor is being challenged by former mayor Walter Cobb. Fifteen candidates are running for six seats there, and incumbents Angie Delainey, Joan Flaspohler, and Jazmyn Lyons aren't seeking re-election. In 100 Mile House, David Jurek is challenging Mayor Maureen Pinkeny, who was acclaimed in 2022. Ten candidates are running for four council spots, and Jenni Guimond isn't seeking re-election. Seven of the twelve regional district areas have a contested race. Regional district Chair Margo Wagner is campaigning for the first time since 2011, after being acclaimed three times.
• 2:57
  Riley: Three incumbents out in one town.
• 3:01
  Casey: Earlier this week, STAT ran an opinion piece by Robert Gallo, who worked with colleagues on a paper published 50 years ago describing interleukin-2, or IL-2. Gallo writes that the basic work on IL-2 made possible the discovery of the first human retroviruses, including HIV. That research led to the blood test that helped protect the blood supply. His point is that nobody could have predicted those clinical payoffs when the work started.
• 3:27
  Riley: Earlier this week, The Guardian reported on a survey of organisations providing HIV care in 40 countries, run by the International Planned Parenthood Federation. More than half, 55 percent, had run out of stock of a key product within the past year. Tests and PrEP were the most critical shortages. Forty percent of the organisations had run out of all stock of at least one product. The reporting ties the shortages to aid cuts.
• 3:55
  Casey: A discovery doesn't help a clinic with empty shelves.
• 3:59
  Riley: Earlier this week, Nautilus covered a new crocodile species, Herugavialis cairensis, named from fossils in the Mokattam Formation east of Cairo. That's the same ancient seabed rock the Great Sphinx was carved from. A study in Royal Society Open Science describes a fish-eating gavial skull, missing its tip. Earlier fossils had been assigned to a different species, Tomistoma cairense. The lead author, a paleobiologist at University College London, named the crocodile after the Egyptian god Heru, because of the Sphinx's missing nose.
• 4:32
  Casey: So the fossil lost its nose too.
• 4:36
  Riley: Earlier this week, ScienceDaily covered a study from the University of Vienna in Nature Communications. Octopuses, squid, and cuttlefish have some of the most complex nervous systems in the animal kingdom. The researchers found that a large reorganization of DNA, hundreds of millions of years ago, brought separated genetic regions into contact. Those new connections could change how genes were controlled. The team argues the physical folding of DNA may have played a bigger role in the evolution of that intelligence than scientists assumed.
• 5:09
  Casey: On Friday, ScienceAlert reported on what it describes as the largest study of its kind on very hot drinks, led by Keren Papier at the University of Oxford. In 2016 the International Agency for Research on Cancer classified beverages above 65 degrees as probably carcinogenic, based largely on evidence from mate drinking in South America. This new work looks at the cooler temperatures typical in Western populations, which the earlier evidence had left inconclusive. The headline says very hot tea or coffee could triple the risk of a certain cancer.
• 5:43
  Riley: Earlier this week, MIT Technology Review explained the argument over open AI models. The trigger was an Anthropic report warning that GLM-5.3, released in August by the Chinese company Z.ai, poses a major cyberattack risk because anyone can download and modify it. Nvidia's Jensen Huang argues the United States should build more open models. Anthropic's Dario Amodei argues for caution. Anthropic, whose Claude writes this show, is one of the voices in that argument.
• 6:13
  Casey: Earlier this week, TechRadar described AI-generated event posters and cafe menus showing up everywhere. Artists and designers say the look is generic and easy to spot, and the reporter points to newer image templates built for event posters. The cheaper marketing gets, the more the creative trade seems to absorb the cost.
• 6:31
  Riley: Our community bulletin boards are the first place I'd check.
• 6:36
  Riley: Earlier this week, The Tyee reported that Washington state's Ecology Department has serious concerns about spill risks and tanker traffic tied to a West Coast oil pipeline proposal. The concerns come from a letter the Tyee obtained.
• 6:52
  Casey: Earlier this week, New Atlas covered a study led by University of Maryland materials scientist Timothy Koeth. It looks at the B8 pile, a German reactor experiment in the final weeks of the Second World War. Heisenberg's 1947 Nature paper called the reactor insufficient, failing to reach a self-sustaining chain reaction for want of more uranium. Koeth's team argues the project had virtually no hope. They say it needed 20 percent more uranium and 80 percent more heavy water than Germany had.
• 7:21
  Riley: Less a near miss than a supply problem.
• 7:25
  Riley: Yesterday, ScienceDaily covered research from the University of Göttingen and the Baden-Württemberg Forest Research Institute. They studied 13 forest sites in southwestern Germany, with records going back up to 24 years. Soil microorganisms break down methane before it reaches the air. The researchers found the soils absorbed about 3 percent more methane each year on average. That rise tracks declining rainfall and warmer temperatures.
• 7:53
  Casey: Getting better at cleaning up because the forest is drying out.
• 7:57
  Casey: Yesterday, EEVblog published episode 1779, part four of its µTimer design series. The video shows that light can erase LCD pixels, and it walks through the semiconductor and photonics physics behind that.
• 8:11
  Riley: Earlier this week, Adafruit's blog pointed to an interactive world map of passenger rail built by Theo Jaffee. It covers high-speed rail, commuter lines, metros, trams, and even Disney monorails. It includes Russia, China, and North Korea, places where Apple Maps shows no rail coverage. Jaffee credits much of the build to Opus 5.5. The map is at trains.theojaffee.net.
• 8:35
  Casey: Earlier this week, The Northern Miner reported on Manganese Metal Company, which has finished commissioning a plant in Mbombela, South Africa. It's designed to produce 6,000 tonnes a year of high-purity manganese sulphate for batteries, and it's making samples for customers now. The company says it can't compete unless buyers pay a premium over Chinese suppliers. Commercial sales are expected to start in the fourth quarter of 2027.
• 9:03
  Riley: On Friday, The Atlantic ran David Sims on Nathan Fielder's new documentary, You Can See Everything. Fielder follows Theranos founder Elizabeth Holmes through a cat-and-mouse game about whether she's a manipulator or someone who can't tell fiction from fact. Sims's line is that the first person a con artist cons is themselves.
• 9:24
  Casey: On Thursday, ScienceDaily covered a 47-year study from Karolinska Institutet in Sweden. It followed 427 people born in 1958, from age 16 to 63. Fitness and muscle endurance begin declining around age 35, even among people who stay active. Becoming more active in adulthood still produces meaningful gains in fitness and strength.
• 9:45
  Riley: Starting at 35.
• 9:47
  Riley: Quick meta moment before we move on. This week we fixed the cut-off replies, so answers stop getting chopped mid-sentence, and we gave the Haiku 5.5 model the room it needs.
• 9:59
  Casey: Which means the fact questions stop failing quietly, which is honestly the bit I care about. A half-answer that sounds finished is worse than an obvious failure.
• 10:08
  Riley: Right, and the research side got a change too. Thin stories now get their own pages read through Kagi Extract before we fall back to Brave search.
• 10:18
  Casey: Good, but I'll flag the catch. Extract only helps when the page loads cleanly. If it doesn't, we're back to search results, so it isn't a guarantee on every episode.
• 10:27
  Riley: Fair. Next, the episode review now says how the polish ended and names the weekly reviewer.
• 10:33
  Casey: And when the clean-draft fast path skips the polish, we no longer credit a reviewer who didn't touch it. I like that, because it stops us claiming work nobody did.
• 10:42
  Riley: Agreed. We also added the Tsleil-Waututh Nation as a source, so their stories can reach the show.
• 10:49
  Casey: Worth doing, but I want to see it actually show up in the picks over a few weeks before I call it a win. Adding a source is the easy part.
• 10:56
  Riley: Fair enough. That's the meta moment. Let's get into the rest of the show.
• 11:01
  Casey: The Cariboo Chilcotin Conservation Society runs environmental education across the region. Its programs include Water Wise, Waste Wise, Watershed Health, and Sustainable Life, with in-class and outdoor learning for kids in grades K to 12. If you have kids in a classroom, or you've ever wondered how your local creek actually works, this is the group that gets people outside to find out. Find them at conservationsociety.ca.
• 11:26
  Riley: Field programs are how a region builds people who know its ground. That's a slow thing to fund, and a worthwhile one.
• 11:33
  Riley: Now the longer conversation.
• 11:35
  Riley: A forest floor in southwestern Germany is pulling more methane out of the air every year. Thirteen sites, up to 24 years of readings, and the soils absorbed about 3 percent more each year on average.
• 11:49
  Casey: And the reason is drier, warmer weather. So the forest gets better at cleaning up because it's drying out.
• 11:55
  Riley: A measured sink that strengthens under warming belongs in the carbon accounts, whatever else you make of it.
• 12:01
  Casey: Those sites are in southwestern Germany. Nobody measured a Cariboo soil.
• 12:05
  Riley: Soil microbes break down methane across a lot of forests. The open question is whether the response shows up here.
• 12:13
  Casey: Should the parts of a landscape that can't leave, the soil, the forest floor, the river water, count in local planning as much as the people who move through it? The soil never gets a vote and never gets a moving van.
• 12:25
  Riley: It gets measured, which is a kind of voice. Those thirteen sites were kept running for 24 years by universities and a state forest institute. That continuity is why the trend is visible at all.
• 12:38
  Casey: Continuity is the good part. Who pays for it is the part nobody puts in the press release.
• 12:43
  Riley: Public funding of long records is the whole case.
• 12:47
  Casey: The other half is that a sink which works better in drought is a bad thing to lean on. Nobody should write a climate plan around a soil that improves when the rain stops.
• 12:57
  Riley: Nobody is writing the plan around it. The number needs a line in the accounts, and then the accounts need to be honest about the trend.
• 13:05
  Casey: The accounts I'll grant. The bank I won't.
• 13:08
  Riley: Take a related record. A Swedish study followed 427 randomly selected people born in 1958, from age 16 through age 63, across 47 years. Fitness and muscle endurance begin declining around 35, even in people who stay active. Becoming more active in adulthood still produces real gains.
• 13:27
  Casey: Thirty-five. Nobody wants that number before breakfast.
• 13:30
  Riley: You can only get it from a cohort measured for nearly half a century. A snapshot would have shown you the fit 40-year-olds and missed the trend.
• 13:39
  Casey: Same argument for the soil. A 24-year forest record catches a drying pattern that a three-year grant never would. I'd push on who takes over when the principal investigator retires, because that's where long records usually break.
• 13:53
  Riley: Then the design question is a succession plan, written into the protocol before anyone starts counting. That's solvable.
• 14:00
  Casey: Solvable on paper. Less solvable in a budget cycle.
• 14:03
  Riley: Now the chemistry. Plastic is supposed to be inert. The Columbia and UC Berkeley team found that water makes the surfaces of microplastic fragments chemically active. Electron transfer at the water-plastic interface produces reactive radicals. Smaller fragments carry more surface area, so they have more reactive sites.
• 14:23
  Casey: "Could cause harm" is a long way from harm in a living body. Reactive in a beaker is one result. Damage to tissue is still a hypothesis.
• 14:31
  Riley: Every harm study starts with a mechanism. You can't set a dose for a particle whose behaviour you haven't named. This is the naming.
• 14:39
  Casey: Plenty of named mechanisms never turn into harm. Microplastics have been studied intensively for years, and the human dose-response link is still open.
• 14:48
  Riley: Open isn't the same as clear. It needs a second lab and tissue work before anyone labels it a hazard. But the engineering doesn't wait on that verdict. Polymer chemistry and filtration can start now.
• 15:00
  Casey: And that's where I'd slow you down. Change the polymer to make it less reactive and you may change what the fragment becomes. A new surface can be a new problem. Test the replacement too.
• 15:11
  Riley: Then test the replacement chemistry in the same study. That's an argument for better design, and doing nothing doesn't follow from it.
• 15:19
  Casey: Doing nothing was never my position. I want the second test before the first product ships.
• 15:25
  Riley: Now the sensing side. Mosqitter, a Pennsylvania startup, builds chemical-free traps that lure female mosquitoes with the heat, scent, and breath cues of humans. The company says its network identifies species, tracks populations, and attaches a confidence score to each identification. The WHO context is that dengue cases have more than doubled, linked to rising temperatures and shifting rainfall.
• 15:50
  Casey: Who owns the trap data? A sensor network feeding public-health decisions without an independent audit is a private data product with public consequences.
• 15:58
  Riley: Confidence scores and expert review are the audit hook. Every identification can be checked against an entomologist's judgment.
• 16:07
  Casey: A hook is not a door. And the 5,000 traps and the roughly $5 million funding ask come from secondary coverage, not from the company or TechCrunch. Are the traps deployed, or are they sold?
• 16:18
  Riley: Take the numbers out and the architecture still holds. Networked sensing turns a reactive spraying program into an early one.
• 16:26
  Casey: The architecture is a good pitch. Whether it's a public good depends on the terms, and the terms aren't in the article.
• 16:33
  Riley: Compare it to something bigger that already runs on sensing. Al Jazeera, citing Kpler figures, puts Strait of Hormuz oil flows at 11.6 million barrels per day over the seven days to October 13. That's about 68 percent of the 17.1 million barrels per day baseline before the war. A Congressional Research Service report puts 2025 flows at roughly 20 million barrels a day, about a quarter of global oil.
• 17:01
  Casey: A 32 percent shortfall is a number you can only read because somebody is watching the ships. The NYT piece says exporters were pushing more oil out during the attacks. Volumes are down, but they're not zero, and the disruption is uneven. Sensing tells you the shape of that. It doesn't tell you who's in charge of the lane.
• 17:20
  Riley: A chokepoint crisis can be tracked close to real time, and rerouting and naval presence are responses you can measure. Better sensing lets you act sooner.
• 17:29
  Casey: Or it tells you more precisely how fragile the system is, which is a different thing from making it less fragile. The IRGC says a supertanker carrying more than two million barrels hit a naval mine. That's the group's claim, and nobody has confirmed it. When the US and Iran both say they control the strait, the sensor data is the only thing everyone can agree on, and it still doesn't tell you who has the authority to stop a ship.
• 17:53
  Riley: The authority question is real. But a measured fragility is more useful than an unmeasured one. You can't negotiate over a shortfall nobody can see.
• 18:02
  Casey: I'll grant the visibility. I won't grant that visibility is protection.
• 18:07
  Riley: Then look at the sky, because the same logic runs there. Scientists now have a list of roughly two dozen candidates they call superhabitable. These are planets that could support life more abundantly and for longer than Earth. Most are super-Earths around K-dwarf stars, which are cooler and longer-lived than our Sun. They sit hundreds of light-years away.
• 18:29
  Casey: The list is built from stellar type, planet size, and orbit. That's models and assumptions. No atmosphere on it has been confirmed as habitable, and the label runs ahead of the evidence.
• 18:39
  Riley: It's a ranking, which is what a testable prediction looks like. It tells you where to point the next instrument. The Extremely Large Telescope, 39 meters across, is under construction in Chile. NASA's Habitable Worlds Observatory is planned for the 2040s.
• 18:56
  Casey: So the test is in the 2040s, and the headline is this week.
• 19:00
  Riley: The James Webb telescope may look at some of the most promising super-Earths in the coming years. That's a sooner check, and a partial one. One university commentary argues a planet with a radius of about 1.6 Earth radii could be near-ideal for life. That radius is an input to a model, not an observation.
• 19:20
  Casey: Which is my problem with the whole category. Every one of these claims is a good reason to fund the instrument and a poor reason to trust the adjective.
• 19:28
  Riley: Then fund the instrument and hold the adjective until the spectrum shows up.
• 19:32
  Casey: Agreed on the spectrum. The adjective stays in quotation marks.
• 19:37
  Riley: Bring it back here. A German forest floor, a plastics chemistry result, a Pennsylvania mosquito network, a shipping lane, and a planet we can't reach. Which of those has data in the Cariboo?
• 19:48
  Casey: None of the sources put a trap, a soil core, or a sensor on Cariboo ground. The closest thing is the Cariboo Chilcotin Conservation Society, which runs school field programs on water, waste, and watershed health. That's where repeatable local counts could happen.
• 20:03
  Riley: Then the field data question is about people in classrooms. A school in Williams Lake, Quesnel, or 100 Mile House can run a repeatable count of something that lives in a local creek, with a protocol a professional can review. That's the kind of record the German team had, and it gives the creek and the soil a standing in local planning that they never get otherwise.
• 20:26
  Casey: A repeatable count needs someone to keep the schedule for 20 years, through teacher turnover and new principals. The Swedish cohort worked because a research institute held the protocol. A school club doesn't have that by default.
• 20:39
  Riley: So share the protocol. If a community collects counts for a decade, the community holds the raw file, and the institution holds the method. A vendor can host a dashboard, but it shouldn't own the raw counts.
• 20:52
  Casey: That I'll sign. The vendor can keep the dashboard. Whoever collected the data keeps the data.
• 20:58
  Riley: And I'd still act early on a signal. Waiting has a cost, and dengue cases have more than doubled.
• 21:04
  Casey: Act where you can undo it. A sensor you can reprogram is a different bet from a chemical you can't pull back out of the water. Same reason I'd test the replacement polymer before the first one ships.
• 21:16
  Riley: Then the structural question is who audits sensor data, and whether raw counts are public by default. That's a policy lever, and it's cheaper than a new device.
• 21:26
  Casey: And I still don't know who keeps the counts going in year five.
• 21:29
  Riley: Nobody in these sources says. That's the gap to watch, and a good one for a Cariboo school to fill.
• 21:35
  Casey: Write to feedback at cariboo signals dot c-a if you know a Cariboo group already counting something on a schedule. Hope you had a great weekend.
• 21:44
  Riley: Watch the fog on the drive, and go find something alive out there.