Epidemic epistemic trespassing. Knows a lot about graphene. Monash Uni. Directed fleet.org.au. Fellow @scienceacademy.bsky.social. Plays bass for www.instagram.com/push_the_trigger Bird photos at www.flickr.com/photos/off-world
2,481 followers 1,029 following 2,528 posts
view profile on Bluesky 
🚨Border Patrol arrested two firefighters who were actively fighting a wildfire at the time. When the other members of the crew asked for a chance to say goodbye, a Border Patrol agent told the crew member “you need to get the f*ck out of here. I’m going to make you leave.”
🇧🇷 Tiê-de-topete ♂ 🇺🇸 Black-goggled Tanager ♂ 🔬 Trichothraupis melanops (Vieillot, 1818) 📍 Miracatu/SP/Brasil #Birds #BirdWatching #BirdsOfTheWorld
I love hummingbirds. As to what kind you've got, I encourage you to figure it out for yourself! 😁
Every person who spends a bit of time actually watching and listening to nature knows things that no app knows, and in fact aren't written down anywhere. And never will be, there will never be enough bytes. Anyway, rant over. 11/11
Nature is vast and contains multitudes. There's stuff going on in your backyard that the app can't tell you about. What do the Song Sparrows on your block eat? How big is their territory? Is it bigger or smaller than the Song Sparrows in the park? Where do they nest? When do they sing? 10/
The orchid people are amazing. Maybe one day I will learn their ways. They are lucky, there is no app that will find an orchid for you, yet. And they are smart, they keep the location of their orchids secret (usually). 9/
You can learn to find a Scarlet Tanager on your own. Everyone can. I have no special physical powers - my eyesight and hearing are fairly crap. But I learned to find birds and you can too. Or you can learn to find orchids, or bugs, or bats, or frogs, or dragonflies, or mushrooms... 8/
If you let an app find you your first Scarlet Tanager, you are letting that app take that joy from you *forever*. (In exchange for some digital confetti on a screen. Yay.) 7/
When you're new to watching and listening to nature, it is hard work learning to find a Scarlet Tanager. I've put in the work, and I remember finding my first Scarlet Tanager, in Amherst, MA, with great joy! 6/
If I tell them they've just heard a Large-billed Scrubwren, or an Olive Whistler, which are rare in the area, they'll remember that fact for 30 seconds. 5/
People who visit my part of the world and go for a walk and happen to see a Superb Lyrebird singing with tail fanned over its head, or come across a flock of Yellow-tailed Black Cockatoos cracking acacia branches to get the grubs inside, remember it for a lifetime. 4/
I don't do it anymore, because it is annoying. The joy that people find joy when they make a discovery in nature on their own lasts a lifetime. But telling people what is around them has almost no impact at all. 3/
There's a parlour trick I used to do (I don't anymore) when I'd go for a walk in the woods with people who aren't birders. For a few minutes I'd say out loud the name of each new species I heard or saw (sometimes 20 or 30). 2/
Want to learn to write songs and play guitar? Now you can have an app do it for you! Apologies in advance, rant incoming. Think hard about what you are doing when you let AI watch nature for you. 1/ apnews.com/article/ai-n...
Man, like I said. Starting that graph at January 2025 is more or less the same as lying.
Or (1B) intrinsic change in severity (no particular evidence for this). Or (2) a change in testing, or coding of deaths (no particular evidence for this either. After the wave we'll have a better picture, and can break out by age, geography, etc.
I don't know the answer. Some guesses: (1) lower severity affects the more severe outcomes (death, ICU, hosp) relatively more. Severity could be lower because (1A) immune escape variants infect those with acquired immunity to a greater extent. Also infections skew younger for same reason.
Data sources: gis.cdc.gov/grasp/fluvie... covid.cdc.gov/covid-data-t... www.cdc.gov/nwss/rv/COVI... 3/3
Here are US covid ED visits, hospitalizations, deaths, and wastewater viral activity level all relative to their 2024 averages. Deaths have been low relative to other metrics for months. 2/
Three weeks later, this is still holding up--percentage of deaths due to pneumonia/influenza/covid in the US is still at 2019 levels, as we approach the peak of the summer covid wave. Lines mark Week 34 in each year. 1/
I'll also note that acquired immunity is the biggest driver here, but there has also been a drop in severe outcomes in those without acquired immunity, which suggests to me a drop in prevalence as well (an indirect effect of acquired immunity at the population level). bsky.app/profile/mich...
I agree, and I've written as much man, many times. So the people that follow me likely know this and have heard it from me already. (You can't write everything in every thread!) bsky.app/profile/mich...
This is quite a message! 👇
Yes, same here, I can turn up at work and get a free flu jab but need to book my covid jab at the pharmacy. I don't understand that.
In US for 6-23 month-olds the covid vaccination rate is 4.5% and for flu it is 63% (14 times higher!) This is a very vulnerable group for covid (and flu). I find that astounding, and yes, I think there must be a problem with what pediatricians are advising.
There's also Novavax. But yeah, I guess it is all due to the soup of misinformation we swim in.
Sorry, I re-read your response and realised you were probably asking about Australia! Deaths data in Australia is painfully slow to get published. So we won't know about this winter's covid deaths until US is going into its winter.
Who are these vaccine skeptics that won't get a covid shot but happily turn up each year for a flu shot? 4/4
Implosion of the CDC and vaccine skepticism in the US are disasters. The odd thing to me is that annual vaccination rates for flu are more than twice as high as for covid, and still fairly substantial in US (46-47% of adults) though they've dropped some. Australia is similar. 3/
For ages less than 65 (probably even less than 75) covid has been less deadly than influenza for a while. (Also note that influenza deaths also undercounted more than covid, and if you consider *per infection*, influenza is far deadlier of course!) 2/
It's going to be close, but I think covid deaths may drop under flu deaths in 2025. Or may remain just slightly more. There's a lot of factors involved there, discussed in the thread below. 1/ bsky.app/profile/mich...
(Eventually we will have data on covid deaths in Australia but it takes a long time.) 3/3
Hard to say if this is just a slow year for covid or this is a continuation of the long-term trend of reduced severity since 2022. But a factor of 2 drop in one year is noteworthy! Data from: www.health.nsw.gov.au/Infectious/c... 2/
Severe covid outcomes (ED, hospitalizations, deaths) in US have been dropping--down ~2X in last year vs. previous. What about Australia? Harder to get data, but NSW publishes emergency dept visits and admissions, both also down 2X in past year vs. previous. Thick dark red line is year avg. 1/
😬
The salami sub throwing guy WAS NOT indicted by a DC federal grand jury. Guess they didn’t see it being a felony assault on a LEO. But the Banksy-inspired murals have been fabulous!
the CDC had quite a run, its history was one of remarkable success My heartfelt thanks to all the devoted professionals who helped make the world a safer, healthier place Its willful destruction is a monstrous act
Down here it is called kino. en.m.wikipedia.org/wiki/Kino_(b...
HOLD THE PRESSES: Even though HHS says Monarez is out (see tweet) her lawyer says otherwise. What a shitshow. bsky.app/profile/mark...
In a functional government, this would be a five alarm fire where Congress calls in Kennedy and asks him what the fuck is going on
yeah I don’t think RFK can fire her, she’s a presidential appointee
Abbe Lowell and I represent CDC Director #SusanMonarez. Contrary to govt statements, Dr. Monarez has neither resigned nor yet been fired. She will not resign. We have issued the following statement:
been waiting for someone, anyone, to threaten these monsters with future consequences. it’s one of the only cards they can play and it has taken far too long. now they need to say it over and over.
Illinois Gov. Pritzker vows to pursue Trump officials who participate in an illegal National Guard deployment to Chicago: "If you hurt my people, nothing will stop me - not time or political circumstance - from making sure that you face justice under our constitutional rule of law."
Arguably one of the greatest shortcomings of today's intellectual environment is the widespread failure to understand how digital power works. You can reclaim a bar/town hall/public square by showing up. You ABSOLUTELY CANNOT do that with a digital platform.
Its a worthwhile question to ask if knowingly compromising your ethics is worth 40k followers on a dying social media platform, and I'm perfectly within my rights to be confused why you're still doing the thing you know is bad for nothing
Generative AI loves it, and so the em dash — elegant, gentle friend to writers — is under attack (don't @ me...) www.theringer.com/2025/08/20/p...
That was one of the most batshit press conferences of Trump's public life. He brandished a photo of Putin and promised to deploy the US military to occupy Chicago, New York, and San Francisco. A sane country would be moving toward impeachment and removal right now.
At least equally concerning, istm, is where the cesium-137 on the shipping containers came from.
This past year was mostly Flu A, and about half H1N1, half H3N2. 2017-18 had unusually high A/H3N2 (thought to be more severe especially in those born before 1968), and also fairly low vaccine effectiveness.
In terms of actual burden of disease, it's very likely that Covid-19 deaths dropped well below flu deaths in 2025. But we won't have an official estimate from CDC for several years. www.cdc.gov/covid/php/su... www.cdc.gov/flu-burden/p... 3/3
- 2025 was worst year for flu since 2017 - better identification of flu as cause of death - Covid-19 causing less severe disease due to population immunity - lower Covid-19 prevalence due to population immunity - worse identification of Covid-19 as cause of death (though still better than flu) 2/
It's strange to say this, but I think there's a fair chance that Covid-19 may just squeak below influenza as a cause of death in the US in 2025. Quite a change, as Covid-19 was almost 5X higher than influenza in 2024. A few reasons: 1/
Data sources: covid.cdc.gov/covid-data-t... covid.cdc.gov/covid-data-t... 3/3
Though, if the top is near, it's a very small summer wave by historic standards, and much smaller than last year. 2/
Looks like the latest covid wave in the US will crest in the next week or so, at least for the adults (right). Under-18s are still on the rise, will likely crash around school start. If so, that would be a close repeat of last year. 1/
This oped builds off this news from last week that I want to share because that's a boss headline www.bostonglobe.com/2025/08/08/m...
Practically though, it appears to have increased the short-time fluctuations in the data. Which probably makes it less useful for detecting changes in the recent trend. So I am a bit surprised that they made the change. 5/5
The change to non-normalized data *might* make sense given that they are interested in producing a number that tracks trends in wastewater RNA. It removes any trend that might be present due to changes in the normalization (e.g. seasonal changes in PMMoV). 4/
They continue to aggregate the data by calculating a z-score, and normalizing to a rolling one-year baseline, before aggregation. Thus the WVAL is still not proportional to RNA copies detected, but rather a number which reflects recent trends. 3/
The new methodology does not use normalized source data (normalized data was previously used when available). Source data is often normalized by flow rate and/or population (using a fecal indicator such as pepper mild mottle virus RNA). 2/
A quick addendum to this. The CDC updated their methodology for calculating the Wastewater Viral Activity Level today. 1/ www.cdc.gov/nwss/data-me...
In which case, severe disease is down for those with acquired immunity (vaccinations and/or prior infections). This is a hypothesis at this point, and needs more observations. But worth watching. 8/8
The hospitalization rate in infants is similar to (or a little lower than) last summer. It seems plausible that this reflects prevalence which is normal for a summer wave. 7/
What could be going on? Infants (ages 0-1) are a new cohort each year, with little acquired immunity. They also aren't drivers of transmission. It's likely that the infection rate in infants simply reflects prevalence in older children and younger adults (families). 6/
Important: There is no evidence for an increase in severity in this latest covid-19 wave. Among ages 75+ (the worst affected by severe covid-19) emergency department visits and hospitalizations remain far lower than last summer. 5/
I note that the rise in hospitalizations in ages 0-1 is unprecedented in previous waves. It suggests to me that they may be driving ED numbers in ages 0-11. (Maybe on top of the seasonal effect due to holidays.) 4/
There's been some discussion on the other site already about the rise in emergency department visits in young children, which are seeming to lead this latest wave, and whether this is an annual effect due to changes in transmission around the holidays. 3/
Here I've normalized each age group's hospitalizations to its 2024 average. The overall rate (all ages) is in black. The different ages follow each other fairly well up and down the waves. But the latest data (only up to 26 July 2025) shows 0-1s increasing well above the pack. 2/
Something interesting is going on with the latest covid-19 wave in the US. Hospitalizations in ages 0-1 are up much more than in other ages. (More accurately: other ages remain unusually low, ages 0-1 look normal.) 1/
A rally in Alaska ahead of the meeting between President Trump and Putin.
The CDC is still publishing the same products with the same methodology as they did in the Biden administration. Sure, we should be skeptical, but there's no sign that there's been a change to how covid data are reported.
He has blocked me. But he's been tagged in a few dozen posts on this over the last year on X and Bluesky, so I presume he knows.
Not a stupid question. Many would like to know how many people are getting infected with covid, and it seems like wastewater could provide the answers. And maybe one day it will! But the science hasn't gotten that far yet.
So the result is that amateurs like Weiland, who masquerade as scientists, end up destroying trust in public health.
Worse, the implication is that there is a simple answer to "how many people are getting infected with covid each day?" which the CDC could provide, but doesn't. (This is incorrect; they can't provide a scientific answer to this.)
What is the harm if Weiland decides to multiply the CDC's WVAL by a number and calling it "daily infections"? The harm comes because some people (including mainstream media) will mistake it for science.
The question here is whether wastewater can be used to estimate the number of infections in a community. The scientific consensus is a resounding "no" (see top post). This is why the CDC doe snot publish such estimates based on wastewater.
Just FYI, as the quote in the top post indicates, wastewater is very good for tracking *trends* in prevalence. The CDC's WVAL is designed to do precisely this, and they publish data weekly so people can track national and local trends. www.cdc.gov/nwss/rv/COVI...
I'm going to publish the entirety of an exchange I just had with the press office of the DOJ. I want you to see how they are talking about truth.
I am fairly certain that he is aware that he has made a serious error in misinterpreting the CDC WVAL as proportional to RNA concentration, which is a central assumption of his analysis. Given that he persists in publishing his estimates, I'd say that's some evidence of probable bad faith.
I think only he can answer that. I suspect he started off by providing what he thought was a useful, if amateur, estimate of covid infections. His work got misinterpreted as expert analysis by some media outlets, and he hasn't disabused them of that notion.
But the biggest problem is that Weiland is claiming to calculate and publish a number that people want to know, and that the CDC isn't publishing. To those who believe that he is trustworthy, the strong implication is that the CDC is *not* trustworthy. This tears down trust in public health.
Wastewater RNA concentration is also temperature dependent and thus seasonally dependent. It depends very strongly on disease severity in ways we don't know how to model. There's evidence it depends on variant. bsky.app/profile/mich...
I think Weiland has good insight and isn't trying to put his thumb on the scale (Hoerger is). But Weiland made the same mistake of thinking CDC's WVAL was a measure of concentration, and didn't account for the baseline correction. bsky.app/profile/mich...
Increase a band: SR71
You could read reviews that use hundreds of words to tell you how good this is, but maybe it is easier to just go to Bandcamp right now and give $9 (nine dollars!) to some cool artists and spend that time listening to some cool music. @nojoy.bsky.social - Bugland nojoy.bandcamp.com/album/bugland
New Zealand woman and six-year-old son detained for three weeks by Ice in US enduring ‘terrifying’ ordeal www.theguardian.com/world/2025/a...
But, even setting aside fundamental errors in methodology, remember that there is strong consensus among scientists is that reliable estimates of prevalence simply can't be made. What these accounts are doing is not science. If you think science and expertise matter, don't repost them. 5/5
In both the cases above, the authors are misinterpreting the CDC's Wastewater Viral Activity Level index as proportional to wastewater concentration of RNA, which it is not (it's normalized to a changing baseline). 4/ bsky.app/profile/mich...
That has not stopped amateurs on social media from producing estimates of community prevalence of covid-19 from wastewater. 3/
The report in the previous post was prepared by the US Nat'l Academies of Sciences, Engineering, and Medicine in 2024. It has over 80 authors with expertise spanning wastewater-based surveillance, water science and technology, and public health practice. nap.nationalacademies.org/catalog/2751... 2/
There is a broad scientific consensus that significant challenges remain before reliable estimates of community prevalence of disease can be made from wastewater RNA monitoring. It's a laudable goal, but the science is not there yet. 1/ nap.nationalacademies.org/catalog/2751...
Yeesh. The fact the quora post I linked to below was written five years ago but precisely details what is happening today is bumming me out quite a bit.
It's really true. People think the Nazis a) weren't ridiculous, and b) ran things well. Both things are false. (If you want to read something eye opening, please read this which argues the appearance of efficiency was really a result of a rush of the public to comply. share.google/7x8aHf9W9S88...)
More than HALF of 🇦🇺’s gas exports are given away with no royalty, no PRRT, no meaningful company tax Norway taxes oil & gas heavily so has a $2.8 trillion sovereign wealth fund. We have $1 trillion of debt. We need urgent PRRT reform to stop the great gas giveaway www.smh.com.au/business/the...
WellKiwis is an interesting longitudinal study in NZ, but they only track symptomatic infections. They do make their reports public. I think the data are interesting (there has been a huge drop in symptomatic covid-19) www.wellkiwis.co.nz
I wish we had something like the UKHSA winter coronavirus infection surveys (now no longer happening). www.gov.uk/government/s...
I'll point to a couple metrics which *might* offer a hint on trends in prevalence. In this thread I looked at hospitalizations in ages 0-1, and deaths in ages 0-1. This is a new cohort each year, with little immunity. Both metrics are down ~70% since the peak in 2022. bsky.app/profile/mich...
Of course the CDC tracks all sorts of metrics over time (test positivity, emergency department visits, hospitalizations, deaths). But the variation in these metrics reflects both prevalence and changes in immunity in the population.