Sign in

Michael Scroggie

@thescrogster.bsky.social
639 followers 846 following 6 posts

Ecologist, skeptic, batrachophile and stats geek. Scientist by day 🐸📈💻, volunteer firefighter 🚒🧑‍🚒 by night. This is a personal account. Follow/like/retweet ≠ endorsement.

PostsRepliesMedia
Reposted by Michael Scroggie
Anthony B, @swearyanthony.bsky.social · 04/10/2026
Once again - the problem isn't "ooo AI models are too scary and powerful". It's that the developers of these products are irresponsible dolts, who have absolutely no intention of changing how they behave. Stop making excuses, use existing laws and make them responsible for their own actions.
1024398
Reposted by Michael Scroggie
James Balamuta @coatless.bsky.social · 04/10/2026
webrarian is a new R package. It turns a folder of R scripts and data into a static website where R runs in the visitor's browser, with no compute server and nothing to install. blog.thecoatlessprofessor.com/posts/introd... #rstats #webr #webassembly
A card for the R package webrarian. Large text reads "A folder in. A website out." On the left, a small card lists a folder named lab-01 holding lab.R, setup.R and a data folder. A red arrow points from it to a browser window that shows a scatterplot of car weight against miles per gallon with brick-red points. The address coatless-wasm.github.io/webrarian is on the card.
314359
Reposted by Michael Scroggie
John Birmingham @birmo.bsky.social · 26/09/2026
You know one way of stopping AI agents from accessing Australian government data would be by making them have MyGov accounts.
1333784
Reposted by Michael Scroggie
Michael Dorman @michaeldorman84.bsky.social · 19/09/2026
I wrote a book on spatial networks, using Python, networkx, and osmnx Online version can be found here: bgu-geography.com/net/ Comments and ideas are welcome!
055
Reposted by Michael Scroggie
Daniel Knowles @dlknowles.bsky.social · 19/09/2026
I really hate the framing of these stories as "an AI agent broke out and hacked something." No, an inadequate sandbox meant that a powerful and unpredictable computer program was set loose on the open internet by the company that designed it. The agency isn't with the "AI agent", it's with the firm
431633459
Reposted by Michael Scroggie
Corey Bradshaw @conservbytes.bsky.social · 16/09/2026
More than 2 million Indigenous people lived in Australia in 1788: new research theconversation.com/more-than-2-...
theconversation.com
More than 2 million Indigenous people lived in Australia in 1788: new research
Genetic, archaeological and environmental evidence suggest the Indigenous population fell by more than 90% in the 70 years after the British invaded.
075
Reposted by Michael Scroggie
Robert (Bob) Kubinec @rmkubinec.bsky.social · 12/09/2026
This "going rogue" framing is nonsense. These are still just text analysis models. It's openAI that is "going rogue" by giving a computer program complete autonomy to make decisions without human oversight. That's a danger whether it's a fancy new LLM or an old-fashioned SQL database.
58617
Reposted by Michael Scroggie
Michelle Wille @duckswabber.bsky.social · 11/09/2026
Part of the HPAI modelling challenge, leveraging a stochastic mechanistic model to manage fictional outbreak on Jolly Island. Lessons learned from exercise useful for Australia 👉 www.biorxiv.org/content/10.6...
132
Michael Scroggie @thescrogster.bsky.social · 02/09/2026
www.abc.net.au/news/2026-09...
abc.net.au
Dugong population drops 50pc in 'real shock' for researchers
The number of dugongs in the Gulf of Carpentaria, one of the animal's most globally important populations, has halved in the past two decades, a decline scientists have called a "real shock".
033
Reposted by Michael Scroggie
Nick Evershed!? @nickevershed.bsky.social · 31/08/2026
‘Scary’: how misinformation and AI hallucinations are infiltrating Australia’s parliament - Guardian analysis reveals that dozens of policy submissions from across the political spectrum incorrectly summarise real research and invent or wrongly cite sources www.theguardian.com/australia-ne...
theguardian.com
‘Scary’: how misinformation and AI hallucinations are infiltrating Australia’s parliament
Exclusive: Guardian analysis reveals that dozens of policy submissions from across the political spectrum incorrectly summarise real research and invent or wrongly cite sources
9255168
Reposted by Michael Scroggie
Ian Fraser @ianfraser.bsky.social · 17/08/2026
Black Cambridge professor Jason Arday endured 249 times more media coverage over an *accusation* of plagiarism than did the White Cambridge professor William O'Reilly on being *found guilty* of plagiarism. And the accusation came from a “known racist propagandist”. @ibramxk.bsky.social
theemancipator.org
The Media Lynched Jason Arday
The death Friday of Jason Arday, a 41-year-old Black scholar who resigned from the University of Cambridge after waves of attacks over the last few weeks, hits close to home. His death left me so deva...
17532
Reposted by Michael Scroggie
Dr Alex Bond @thelabandfield.bsky.social · 15/08/2026
New paper in @bou.org.uk's Ibis: there is still no comprehensive database of bird type specimens. Many major collections have no published type catalogue at all. Where they exist, they're often print-only, partial, or decades out of date. 🧵
16527
Reposted by Michael Scroggie
Michelle Wille @duckswabber.bsky.social · 13/08/2026
Peregrine Falcons, San Fransisco. Within 8 mo of the outbreak, occupancy had dropped to 65.1%, and 2 yr later (2025), only 36.2% of territories remained occupied 👉 rapt.kglmeridian.com/view/journal...
094
Reposted by Michael Scroggie
Brenton von Takach @certhionyx.bsky.social · 12/08/2026
Patterns in the pond: the role of colour, pattern and texture in species delimitation of polymorphic froglet species of south-western Australia doi.org/10.1071/ZO25...
043
Reposted by Michael Scroggie
Robbi Bishop-Taylor 🛰️🌏 @satellitesci.bsky.social · 11/08/2026
This is a first for me: a citation for a paper I never wrote... in a new book published by @SpringerNature.com no less! 😱 Geospatial Application of Coastal Hazards: Monitoring, Mitigation, Adaption and Prediction Springer Nature, 2026
A screenshot of Google Books, showing a citation highlighted in yellow. This citation does not exist.
63315
Reposted by Michael Scroggie
International Journal of Wildland Fire @ijwildlandfire.bsky.social · 08/08/2026
🔥New in IJWF: Price et al. show that 94% of houses destroyed in the 2019–20 NSW fires were within 100 m of flammable vegetation, supporting current bushfire risk mapping—although extreme fire weather pushed house losses further into surrounding landscapes. 🔗 doi.org/10.1071/WF25... #IJWildlandFire
Map of New South Wales showing Bush Fire Prone Land (BFPL) with the footprint of the 2019–20 bushfires overlaid. The locations of the 2475 destroyed houses is indicated.Flowchart showing how each of the two building datasets (Building Impact Assessment (BIA) and Microsoft Building Footprints (MBF)) were subsetted and used in the four analyses.Two examples of buildings destroyed more than 400 m from flammable vegetation. (a) The destroyed buildings most distant from flammable vegetation. This area is 20 km west of Rappville, New South Wales (NSW) (maximum 1464 m from flammable vegetation), burnt by the Busbys Flat Rd fire, 8 October 2019 at Forest Fire Danger Index (FFDI) 16. The smaller red dots are destroyed outbuildings and the larger dot is a destroyed house (840 m from flammable vegetation). These buildings were burnt on 8 October 2019 by the Busby’s Flat Rd fire. The overlays are vegetation categories. (b) Cobargo, 488 m from flammable vegetation, burnt by the Badja fire, 31 December 2019 under FFDI 73. The eucalypt (Euc.)-Forest category is the Bush Fire Prone Land (BFPL) category one and the non Euc.-Forest category are the BFPL categories 2 and 3 combined. (c) Shows the location of the two communities and the two fires in NSW.
096
Reposted by Michael Scroggie
David Ho @davidho.bsky.social · 26/07/2026
Facing federal science budget cuts, biologist Daniel Blumstein resorted to uploading marmot videos to OnlyFans to raise funds for his marmot research. However, when he uploaded his driver’s license for verification, OnlyFans informed him that his appearance didn’t match the marmots in the videos.
nytimes.com
‘OnlyMarms’: Marmots Are on OnlyFans to Raise Money for Research (Gift Article)
A team of scientists uploaded videos of the squirrel-like rodents in the wild on the adult-content site OnlyFans to raise money amid government budget cuts.
172299682
Reposted by Michael Scroggie
Marco Festa-Bianchet 🇨🇦 @marcofesta.bsky.social · 22/07/2026
Contagious melanoma in catfish from Lac Memphrémagog, about 30 km from where I live. Fourth documented transmissible cancer 🧪 www.nature.com/articles/s41...
nature.com
Brown bullhead catfish melanoma represents a novel transmissible cancer - Nature
Whole-genome sequencing analysis of tumour and matched non-tumour host tissues of the brown bullhead catfish reveals a type of naturally occurring transmissible cancer.
0159
Reposted by Michael Scroggie
Nature Reviews Methods Primers @natrevmethodsprimers.nature.com · 14/07/2026
Explore how generalized additive models for location, scale, and shape (GAMLSS) extend traditional regression by modeling mean, variance, and more. This #Primer highlights the role of distributional regression in risk assessment and predicting uncertainty. www.nature.com/articles/s43...
nature.com
Distributional regression using generalized additive models for location, scale and shape - Nature Reviews Methods Primers
Distributional regression models entire conditional response distributions. In this Primer, Merder et al. discuss how distributional regression is used to predict uncertainty and extreme conditions in...
192
Reposted by Michael Scroggie
𝚖𝚊𝚌𝚔𝚊 𝚜𝚣𝚎𝚌𝚑𝚗𝚘 @mackaszechno.bsky.social · 14/07/2026
TIL to use singleSide with #rstats {sf} st_buffer. spencerschien.info/post/spatial...
spencerschien.info
Getting Creative with Spatial Buffers | Spencer Schien
Use spatial buffering to create new spatial objects.
063
Reposted by Michael Scroggie
Annette Gordon-Reed @agordonreed.bsky.social · 13/07/2026
Don’t know who to credit for this. But it is how I feel.
5497203011
Michael Scroggie @thescrogster.bsky.social · 13/07/2026
Mind blown 🤯
010
Reposted by Michael Scroggie
Ecography @ecography.bsky.social · 10/07/2026
Scrutinizing the Wallacean shortfall: global gaps in snake occurrence data across space and environment vist.ly/5aqsx #GBIF #Inventory #Serpentes
World map showing rarity distribution with a color gradient from yellow to purple.Eight histograms showing frequency distribution of records by continent with small maps.
032
Reposted by Michael Scroggie
Anthony Ricciardi @ecoinvasions.bsky.social · 10/07/2026
"Spread around the world by long-distance bird migrations, H5N1 has already infected at least 500 wild bird species and 50 mammalian species since it first emerged in 1996. It has caused catastrophic die-offs..." #bioinvasions www.science.org/doi/10.1126/...
science.org
As bird flu threatens, New Zealand vaccinates endangered birds
Arrival of H5N1 virus on Australian mainland triggers ambitious vaccination program
0139
Reposted by Michael Scroggie
Ben Bolker @bbolker.bsky.social · 10/06/2026
new diamond open-access Journal of Statistical Ecology launched 🎉🎉🎉 jse.centre-mersenne.org @fredbarraquand.bsky.social @oaggimenez.bsky.social @marieaugermethe.bsky.social @vianeylb.bsky.social (apologies to any Bsky editors I missed)
jse.centre-mersenne.org
Journal of Statistical Ecology Journal of Statistical Ecology
011870
Reposted by Michael Scroggie
High Country News @highcountrynews.org · 29/06/2026
Fire survivors say “it’s morally reprehensible,” and experts worry the wagering could lead to arson.
hcn.org
People are betting on wildfires. Should they? - High Country News
Fire survivors say “it’s morally reprehensible,” and experts worry the wagering could lead to arson.
48542198
Reposted by Michael Scroggie
Esteban Botero-Delgadillo @eboterod.bsky.social · 22/06/2026
Identification of camera trap images by artificial intelligence and human experts produces similar multi-species occupancy models @jappliedecology.bsky.social besjournals.onlinelibrary.wiley.com/share/4GAHKTCP…
144
Reposted by Michael Scroggie
Aki Vehtari @avehtari.bsky.social · 26/06/2026
posterior R package has now JOSS paper you can cite, too joss.theoj.org/papers/10.21... with @paulbuerkner.com , Jonah Gabry, @mjskay.com
joss.theoj.org
posterior: Tools for Working with Posterior Distributions in R
Bürkner et al., (2026). posterior: Tools for Working with Posterior Distributions in R. Journal of Open Source Software, 11(122), 10526, https://doi.org/10.21105/joss.10526
25416
Reposted by Michael Scroggie
Henrik Bengtsson @henrikbengtsson.bsky.social · 21/06/2026
🚀 Want to run purrr::map() in #parallel? Use y <- map(x, fcn) |> futurize() from the futurize package #rstats futurize.futurever.se
The 'futurize' hexlogo - Magic Touch ParallelizationThe 'purrr' hexlogo - a curled up cat
0275
Reposted by Michael Scroggie
Emily Grubert @gruberte.bsky.social · 20/06/2026
This isn’t actually funny but I keep cackling to myself when people say we should pursue AI because it will tell us how to solve climate change. Like, we *know* how, we’re just not. ::burns so much fossil fuel to ask:: hey AI, what do? AI: “stop burning fossil fuel”
24818205
Reposted by Michael Scroggie
Ben Harrap @bharrap.bsky.social · 04/06/2026
You probably saw `filter_out()` in {dplyr} 1.2.0 but did you notice `when_any()` and `when_all()`? Not only is it easier to write conditions, but the usage of commas instead of | and & means formatters like air will align all conditions the same 🤩 #rstats dplyr.tidyverse.org/reference/wh...
A snippet of R code:

mtcars |> filter_out(when_any(cyl == 6, hp < 100))
18811
Reposted by Michael Scroggie
Jarrett Byrnes @jebyrnes.bsky.social · 01/06/2026
Really nice piece on statistical workflows. Most of us do 2 of the 4 steps, but, really, it's time to start thinking about the rest! royalsocietypublishing.org/rsta/article...
royalsocietypublishing.org
A four-step simulation-based workflow for ecological analysis and science
Abstract. Ecology is a discipline that has faced increasing challenges as the disconnect between its scientific and statistical methods has become more obv
066
Reposted by Michael Scroggie
Jakub Nowosad @nowosad.fosstodon.org.ap.brid.gy · 31/05/2026
📄 New our preprint: Moving Beyond Spatial and Random Cross-Validation in Environmental Modelling: A Call for Prediction-Domain Adaptive Evaluation arxiv.org/abs/2605.13689v1 #SpatialStats #Ecology #MachineLearning
063
Reposted by Michael Scroggie
Michelle Wille @duckswabber.bsky.social · 26/05/2026
Summary of our surveillance during spring 2025, including recently arriving shorebirds and seabirds, as well as some marine mammals. All negative for HPAI. 👉 www.biorxiv.org/content/10.6...
173
Reposted by Michael Scroggie
Michelle Wille @duckswabber.bsky.social · 26/05/2026
Great to see this one published. An incredible overview of the impact of HPAI on wildlife, with data from WAHIS, EMPRES and the literature. Also case study on Sandwich tern integrating mortality, ecological, serological + genetic data 👉 onlinelibrary.wiley.com/doi/10.1002/...
175
Reposted by Michael Scroggie
The Great Pumpkin Papers 🎃🎃🎃 @professormusgrave.bsky.social · 24/05/2026
Hard day at the office. Inspecting trolleys, double checking switches, tying people to the track. But that’s life at the moral philosophy factory
1446661
Reposted by Michael Scroggie
John Delmenico @thebigjohnnyd.bsky.social · 22/05/2026
wait youre telling me a 'Mexican' fast food chain made by two white aussie investor bros using other people's names and likeness to pretend to be authentic failed in the american market where their competition was actually authentic Mexican food? there was simply no way of predicting that
1019531
Reposted by Michael Scroggie
GRASS @grassgis.fosstodon.org.ap.brid.gy · 09/05/2026
GRASS 8.5.0 released: two years of community work. 🐍 New grass․tools subpackage: GRASS tools as Python functions with NumPy I/O and proper exceptions 📊 JSON output across dozens of tools (r․info, r․univar, …) 📖 Docs rewritten in Markdown: tool pages show CLI, grass․tools, and grass․script tabs ⚡ […]
fosstodon.org
Original post on fosstodon.org
094
Reposted by Michael Scroggie
Matthijs Hollanders @matthollanders.bsky.social · 07/05/2026
occARU now supports sites within regions, e.g., multiple arrays of camera traps or microphones. Random (zero-sum) region-level intercepts are estimated for both occupancy and detection, and spatial site effects are computed per-region, which is much faster! 1/2 mhollanders.github.io/occARU/
mhollanders.github.io
Occupancy Models for Automated Recording Unit (ARU) Data
Bayesian (multispecies) occupancy models for ARU data using Stan.
161
Reposted by Michael Scroggie
Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 05/05/2026
JAGS has catalyzed so much sober, reusable model development. Really an amazing project that just keeps going.
1273
Reposted by Michael Scroggie
Blas M. Benito @blasbenito.com · 04/05/2026
📦 My new R package {spatialData} is on CRAN! 🎉 Ecological datasets as sf data frames and their companion raster data ready for analysis and modeling. Built as example data for R packages, but great for teaching too! 🔗 blasbenito.github.io/spatialData/ ❗check the Articles section!
blasbenito.github.io
spatialData: Cool Spatial Datasets
Provides spatial datasets ready to use for ecological modelling and raster companion data for prediction: Neanderthal presence during the Last Interglacial (Benito et al. 2017 <doi:10.1111/jbi.12845>)...
26629
Reposted by Michael Scroggie
Stephen Turner @stephenturner.us · 03/05/2026
Free and open-source images, icons, and tools for creating scientific illustrations doi.org/10.59350/5zt... 🧪
doi.org
Free and open-source images, icons, and tools for creating scientific illustrations
Phylopic, NIH Bioart, Bioicons, Scidraw, Open Science Art, Health Icons, Servier Medical Art, Biodiversity Heritage Library, the Noun Project, Segment Anything, Excalidraw, draw.io, Biographics
6406227
Reposted by Michael Scroggie
Matthijs Hollanders @matthollanders.bsky.social · 14/04/2026
occARU: Bayesian (multispecies) occupancy models for ARU data using @mc-stan.org. Automated recording units (ARUs) like camera traps produce rich time series which warrant going beyond occupancy and focusing on detection rates. 1/n mhollanders.github.io/occARU/
mhollanders.github.io
Occupancy Models for Automated Recording Unit (ARU) Data
Bayesian (multispecies) occupancy models for ARU data using Stan.
184
Reposted by Michael Scroggie
Matthijs Hollanders @matthollanders.bsky.social · 07/04/2026
New pre-print! I cover a range of open capture-recapture models (single survey/robust design, multistate/multievent, in (Cormack-)Jolly-Seber variants) in Stan, and provide efficient log likelihood functions. I also introduce a method to account for unequal survey intervals in the entry process.
1117
Michael Scroggie @thescrogster.bsky.social · 02/04/2026
No, copilot. I don't want you to summarise that for me.
020
Reposted by Michael Scroggie
Luca Santini @lsantinieco.bsky.social · 28/03/2026
Our latest paper from the Urbis project investigating the spatial structuring of bird assemblage in the city of Rome 🐦🏙️
031
Reposted by Michael Scroggie
Aylwyn Scally @aylwyn-scally.bsky.social · 15/03/2026
This story is a good example of why it is so difficult to communicate genuine science to the public. Despite every scientist not on the company payroll decrying it as a bogus and unscientific project, the story presents it as some sort of both-sides issue. www.theguardian.com/us-news/2026...
theguardian.com
Can scientists really resurrect the dodo? Inside the company that says it can
Colossal Biosciences’ CEO says its work follows a ‘moral obligation’ while critics say it’s ‘tech bro’ hype that could undermine conservation
610444
Reposted by Michael Scroggie
Eucalypt Australia @eucalyptaus.bsky.social · 22/02/2026
Eucalypt Australia are pleased to present Professor Patrick Baker with the Bjarne K Dahl Medal for 2026 in recognition of his significant and sustained contribution to the science of eucalypts in the fields of forest ecology and silviculture. #NationalEucalyptDay2026 #OurEucalyptHome
0154
Reposted by Michael Scroggie
Dustin Marshall @djmmeeg.bsky.social · 07/02/2026
The principles of experimental design in ecology and evolution has 20k downloads, but unfortunately I still have to reject ~20% of papers without review because they lack replication at the right level. More discussion among mentors and mentees please. onlinelibrary.wiley.com/doi/full/10....
onlinelibrary.wiley.com
Principles of experimental design for ecology and evolution
Here I argue that we do not discuss experimental design, often until it is too late. This editorial seeks to begin a conversation about how and where to replicate appropriately.
13515
Reposted by Michael Scroggie
Rob Davis @pezoporus.bsky.social · 02/02/2026
Thrilled to be part of this new study that aims to derive a meaningful classification of distinct and recognisable Australian terrestrial bird communities. Lead by @martinemaron.bsky.social and Hannah Fraser with many other colleagues check it out here. onlinelibrary.wiley.com/doi/10.1111/...
onlinelibrary.wiley.com
A Typology of Australian Terrestrial Bird Communities
Aim Holistic measurement of the response of fauna communities to interventions requires suitable community condition metrics. However, the development of such metrics is hindered by the absence of b...
11611