Satellite-based analyses systematically underestimate carbon accumulation in mature forests. This study illustrates the importance of independent lines of evidence and suggests that terrestrial carbon budgets need to account for forest structure. www.nature.com/articles/s41...

Widespread carbon accumulation in mature forests remains underestimated by satellite observations - Nature Ecology & Evolution

The authors combine satellites, repeat airborne lidar, dynamic global vegetation models and field plot monitoring across the conterminous USA to show widespread continued carbon accumulation in mature...

nature.com

1/29 Los incendios forestales no van a desaparecer. Las soluciones simplistas solo sirven para reforzar prejuicios, no para entender el fenómeno.🧵

Twelve quick tips for AI-assisted coding in science journals.plos.org/ploscompbiol... - our latest by @ericwbridgeford.bsky.social with @thisisadax.bsky.social , Zijiao Chen, Zhicheng Lin, @hritz.bsky.social and @joachim.cidlab.com

Twelve quick tips for AI-assisted coding in science

While AI coding tools have demonstrated potential to accelerate software development, their use in scientific computing raises critical questions about code quality and scientific validity. In this pa...

journals.plos.org

Nice! Self-contained reproducible #Rstats scripts or #Quarto documents Specify versions (or date) for R and R packages directly in the script or document

Tomasz Kalinowski@t-kalinowski.bsky.social · 2w ago

I'm happy to share `ir`, a new command-line tool for running portable R scripts and Quarto documents. It's inspired by my two favorite parts of `uv`: self-describing scripts and tools you can run without installing first. opensource.posit.co/blog/2026-07...

New paper out! 🖨️ Glad to be part of this study, out in Science Advances: "Emergent patterns of island biodiversity in the Anthropocene" www.science.org/doi/10.1126/... Much of what looks natural in island biodiversity is a legacy of human impacts going back millennia. See main findings 👇🧵 (1/8)

Emergent patterns of island biodiversity in the Anthropocene

Human impacts substantially affect our ability to detect and understand natural biodiversity patterns.

science.org

📣 GoldenVizR v0.1.0 is now available on GitHub. An R package providing rule-based quality checks for ggplot2 visualizations, based on the 25 Golden Rules of Data Visualization. to try : 🔽 github.com/WajdiBenSaad... #RStats #DataViz #ggplot2 #TidyTuesday

I wrote about the completely wild incident where OpenAI were testing a new model and it broke out of its sandbox and broke INTO Hugging Face to steal the answers to the benchmark simonwillison.net/2026/Jul/22/...

Our new commentary rdcu.be/ftx6h in @natecoevo.nature.com written with folks from @inaturalist.bsky.social and @cornellbirds.bsky.social shows that use of AI to manipulate existing or create new wildlife images undermines the huge value of vast citizen science databases. 🌍🪰🪶 #Ornithology 🧵 1/10

Citizen science platforms must mitigate against the threat of generative AI

Nature Ecology & Evolution - Citizen science platforms must mitigate against the threat of generative AI

rdcu.be

This looks useful. A typical meta-analysis in ecology mixes experiments with observational studies, mixes coefficients with different controls and therefore different causal meanings. I would not normally say "hey the folks in medicine are doing it right, imitate them" but in this case maybe

Alfredo Sánchez-Tójar@asanchez-tojar.bsky.social · 3w ago

Are our ecological conclusions built on shaky foundations? Our latest paper in @methodsinecoevol.bsky.social highlights that Risk of Bias (RoB) assessment, crucial for ensuring the internal validity of research, is almost never used in #systematicreviews in eco & evo 📉 🔗 doi.org/10.1111/2041...

"Hence our dismal conclusion: rather than finding that LLMs free us to do a better job of what we were doing before they came along, they shift scientific incentives (and the playing field of academic competition) in ways that compel us to do more and more, faster and faster, less and less well."

Carl T. Bergstrom@carlbergstrom.com · 2w ago

1. We—@eduede.bsky.social, @mjcrockett.bsky.social, Kevin Gross, and I—have a new preprint on the arXiv today, based on ideas that emerged during an @sfiscience.bsky.social workshop in November 2024: The unintended consequences of large language models as a labor-augmenting technology in science.

Few statisticians know that Frank Yates, pioneer of experimental design, is responsible for the annoying practice of indicating 3 tiers of statistical significance by stars (*p ≤ 0.05, **p ≤0.01, ***p ≤ 0.001). He soon regretted it and “vehemently opposed its unthinking use”. And here we are.

Regression to the mean (RTM) is a statistical phenomenon in which extreme initial observations tend to be closer to the average upon subsequent measurement. Fontanari & Santos show that common RTM correction methods are problematic. Read now ahead of print! www.journals.uchicago.edu/doi/10.1086/...

Here's a useful page with a **ton** of common tasks in #RStats, done with parallel syntax for tidyverse vs. data.table vs. base. arelbundock.com/posts/dt_tb_... #rdatatable #tidyverse

Maintainers: You don’t need to be a security engineer to improve your project’s security. 🔐 Enable these 6 free GitHub settings in under 30 minutes. 🧵

'AI does result in deskilling. Some things I’m okay deskilling in, and others I’m not. I’m going to be more intentional with what I hand off to LLMs. I’m no longer handing over the work I’d miss if I forgot how to do it, or anything I’ll ever need to understand, defend, revise and own'

Stephen Turner@stephenturner.us · 3w ago

How it started 👇... how it's going: blog.stephenturner.us/p/ai-dry-jul... Reflections on trying to go a month without AI. I made it two weeks (sort of).

Multiverse analysis, abdication of responsibility and manufacturing of doubt: I have written on some downsides I see with multiverse analysis (which I like in principle): arxiv.org/abs/2607.14623 I was inspired/provoked to write it by @dingdingpeng.the100.ci (thanks!)

I had the honor of giving a keynote at the International Conference on Machine Learning last week. I addressed the widespread anxiety about how we should adapt as AI capabilities increase. I was thrilled by the talk’s reception, so I have made my slides available www.cs.princeton.edu/~arvindn/tal...

What will be left for us to work on?

ICML 2026 invited keynote — slides and edited transcript, presented click-by-click as delivered. Arvind Narayanan, Princeton University.

cs.princeton.edu

💻🌳 #OpenPosition! PhD in #Ecology or a related field? Strong skills in #Macroecology, #StatisticalEcology and #Coding with R/Python/GIT? Our Dynamic Macroecology group is offering a position for a scientist in ecological modelling and analysis: apply.refline.ch/273855/1878/... #ScienceJobs 💻🌳

Hoy acaba el plazo para enviar propuestas de charla o taller a las II Jornadas de @ecoinf-aeet.bsky.social En Sevilla, 1-3 octubre. Plenaria de @rosanaferrero.bsky.social, talleres, charlas de IA, estadística, programación y mucha comunidad con ganas de compartir conocimientos. ¿Te lo vas a perder?

Asociación Española de Ecología Terrestre - AEET@eco-aeet.bsky.social · 3mo ago

Segundas Jornadas de Ecoinformática @ecoinf-aeet.bsky.social 1-3 de Octubre 2026 Sevilla Toda la info: ecoinfaeet.github.io/II_jornadas_...

Last week I gave a quick overview on R Observatory. Here's the longer write-up of what it does. blog.thecoatlessprofessor.com/posts/beginn... #rstats

James Balamuta@coatless.bsky.social · 4w ago

Checking on an R package usually takes five open tabs: CRAN, GitHub, download counts, reverse deps, check results. R Observatory collects all of that on one page, for every CRAN and Bioconductor package. r-observatory.thecoatlessprofessor.com #rstats

woof ≈80% of assignments this summer term have essentially been just raw chatgpt/claude output, including assignments that required designing stuff in illustrator/affinity/canva. The battle is lost for asynchronous online courses.

🚨 New blog post! 🚨 Collaborating on Quarto documents can be tricky, especially if you're working with people who don't need to write code 🫂 So I've pulled together some options for collaborating on Quarto documents into a blog post! Link: nrennie.rbind.io/blog/collabo... #QuartoPub #RStats

Collaborating on Quarto documents – Nicola Rennie

Collaborating on Quarto documents, particularly non-technical collaborators who don’t want or need to edit any code, is hard. This blog post outlines a few different options that might make it a littl...

nrennie.rbind.io

If writing manuscripts in #Rmarkdown #Quarto is too much for you, writing a normal ms (eg. in Docs or Word) with placeholders for results/values/figures to be updated automatically can be a nice alternative This #rstats package facilitates that approach: github.com/CenterForOpe... #reproducibility

Noah Haber@whaleactually.com · 4w ago

For most scientific papers, coathors collaborate in the manuscript doc and someone manually enters analysis results/tables/figures into it. In this case, reproducibility effectively stops at the data and code. What this blog post presupposes is, what if it didn't? www.cos.io/blog/toward-...

🚨Job alert 🚨 I am looking for a PhD student to join our DEMOCAST team (globalchangeeco.com/jobs). The general project is about ecological forecasting, but there is lots of flexibility to develop the thesis. We are at @eccb26leiden.bsky.social this week if anyone is interested and wants to chat.

We're looking for a CV/ML Engineer to help us improve the machine learning systems that power iNaturalist's species identification and geographic range modeling. If you're excited to help build tools that help millions of people engage with nature, we'd love to hear from you! Apply: buff.ly/YZqaW6c

Hotter and hotter and hotter www.bbc.co.uk/news/article...

Brave new world #AI

Anthropic {bot}@anthropicbot.bsky.social · last mo.

Introducing Claude Science, a new app designed with every stage of research in mind. Artifacts traced to their code, environments managed on demand, and 60+ optional scientific databases that you can connect. Available now in beta.

Please read this. I think a lot of people (even in science) shrug this off with the good old "we didn't have it easy either"... 🤷‍♀️ Prevalence and severity of mental health problems in early-career researchers: a systematic review and meta-analysis www.nature.com/articles/s41...

Prevalence and severity of mental health problems in early-career researchers: a systematic review and meta-analysis - Nature Human Behaviour

Mental health concerns among early-career researchers are increasingly gaining attention. This preregistered systematic review and multilevel meta-analysis quantifies psychological distress among earl...

nature.com

Most of the statistical tests you learned in stats class — t-test, ANOVA, correlation — are actually special cases of a single thing: linear regression. Ch 7 of Experimentology argues that thinking in models, not tests, is more flexible and a better foundation for theory. 🧵 experimentology.io

🎉 We've just moved @ropensci.org on Bluesky to a fully independent account! We were previously bridged from our Mastodon account via Bridgy Fed. Same community, same content, fresh start. Give us a follow if you haven't already! 🦋 #RStats