Some argue protected areas in fire-prone landscapes can undermine forest carbon (fuel build-up → higher fire risk). This paper shows Spanish protected areas consistently outperform comparable unprotected lands, strongest in National Parks. iopscience.iop.org/article/10.1...

I don’t know what this is, but I am here for the team names

Essential reading for conservation scientists & ecologists (in fact anyone doing science across borders!) 🌍🧪https://onlinelibrary.wiley.com/doi/10.1111/ele.70227?af=R Remote-control science is an internet-enabled form of "parachute" science FWIW, I see a lot of this in AI and macroecology research

"Podemos vivir sin lobos (...) como estuvimos a punto de vivir sin linces, como posiblemente podamos vivir en pocos años sin urogallos, salmones y anguilas (...). Pero somos muchas las personas que no queremos vivir sin ellos" Léete estoy de @ernestodmartin.bsky.social elpais.com/clima-y-medi...

📣 NEW! I’ve just released the BIGGEST and perhaps most creative project I’ve ever worked on! “Searching for Birds” searchingforbirds.visualcinnamon.com 🐤 A project, an article, an exploration that dives into the data that connects humans with birds, by looking at how we search for birds.

February: “Even the land is tired. Most of the birds that can afford it have gone to Florida” 🤓

Would be interesting to compare the results on more recent models - but this problem won’t go away. LLMs are always going to be extrapolating from what has already, and often, been thought, which is why they aren’t windows to the future but anchors to the past.

And, bonus fact, it actually spells pretty good (unlike, say, cat pee!). A musky, sweet and sour smell. In the dry climate of the Gobi (where our long-term study is), the smell doesn’t last as long as in the wet Himalayas, so it’s extra special when you find a strong scent mark - it must be fresh!

This has been my experience too. When I run out of brainpower, I sometimes turn to LLMs. But then comes the endless back-and-forth with the thing to actually make the code work; I guess that’s the vibe coding part 🥺 I can’t imagine what this is doing to the skills of those newly learning coding…

The 10th Annual #Tech4Wildlife Challenge is starting soon!📸 Help us celebrate a decade of conservation technology by sharing how you use #Tech4Wildlife throughout next week. Learn how to participate 👉 wildlabs.net/article/join... #conservation #conservationtech #naturetech

Cool! Carnivores help mycorrhizal fungi to disperse. Paper is here for those interested (although paywalled so I can’t actually read it): onlinelibrary.wiley.com/doi/10.1111/...

Coffee! ☕️ 🫘 For the UK: Kiss the Hippo has great credentials (direct sourcing, carbon neutral, organic) kissthehippo.com/pages/kiss-t... For the Netherlands: Wakuli does some great work direct with suppliers www.wakuli.com/pages/mission

Love to see this: a women’s camera trapping group in the high mountains of Himachal Pradesh, western Himalaya. All power to them. Exemplary data collection in the field shown as well, including proper testing of camera setup using “walk test” 😇 #cameratrap 🌏🧪 youtu.be/98Tigg7WXCg?...

Beautiful and incredible scenes from the centre of the Netherlands, a country with 1.5 times the density of people in the UK. A pack of 11 wolves roaming through the snow!

Alongside “AI colonialism”, the other big worry with this is that modelling results are never confronted with reality, they exist in some kind of desk-based bubble. All models are wrong, only some are useful. It’s by going out & touching the grass that we can sort the useful from the nonsensical.

A drone’s-eye view of the South Gobi, Mongolia, at sunset. Home to our long-term study of snow leopards & the social-ecological system of ungulates, carnivores & livestock herders. You might wonder if anything was living down there. But there are 20 snow leopards & 2000 ibex. Life finds a way! 🌏

“It is difficult to get a man to understand something when his salary depends on his not understanding it.” – Upton Sinclair I do over-share this quote, but it's just so apt for so many things wrong with the world (pretty timeless too).

We wrote the Strain on scientific publishing to highlight the problems of time & trust. With a fantastic group of co-authors, we present The Drain of Scientific Publishing: a 🧵 1/n Drain: arxiv.org/abs/2511.04820 Strain: direct.mit.edu/qss/article/... Oligopoly: direct.mit.edu/qss/article/...

Hey everyone, is there a literature (can be grey) discussing the ethics of using #generativeAI in #conservation and-or #ecology? I have to persuade some folks to consider the “darker” side of it and some authoritative/academic sources would really help. Thanks for any links. 🌍#conservationscience

Enjoyed reading about this partnership between the Kuikuro and Brazilian/US archaeologists. ✅Trust built over decades ✅Shared research objectives ✅Indigenous data sovereignty ✅Credit sharing ✅Respect & humility (“I didn’t “discover” anything”) #participatoryscience www.science.org/content/arti...

Now live! www.ethicalconservation.org/toolkit/ -Offers a fresh perspective on “human-wildlife conflict” -Suggests shifting the narrative from demonisation of wildlife towards an empathetic & nuanced approach to conflict, in partnership w/ local communities #ethicalconservation #conservationscience🌍

I find it’s helpful to always start from the premise: “All models are wrong, but some are useful” (Box, 1979)

23mins running uphill to celebrate October 23rd, International Snow Leopard Day. #23for23 Because conservation often feels like an uphill battle. But I guess you never know when the crest will suddenly appear and you’re on the downhill… #SnowLeopardDay #MoveForSnowLeopards

We (Snow Leopard Trust) will have a consultancy available soon to do climate-smart protected area planning in Kyrgyzstan. Will involve #cameratrap data & Zonation. Suitable for PhD/post-doc level or a lab group. Get in touch if interested! Will post link once it’s live! 🧪🌍#conservationscience

Important reflections on the power dynamics in large-scale syntheses of ecological data. Many of them perpetuate colonial structures. An example that comes to mind is Google & Wildlife Insights, who used camera trap data to publish the SpeciesNet algorithm, with no credit given to field researchers

Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia: doi.org/10.5281/zeno... We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical thinking, expertise, academic freedom, & scientific integrity. 1/n

Over on the Other Place I would occasionally put this thread together, most recently during lockdown, four years ago in fact, and I realised I'd not done it here. So, bear with, and feel free to mute as this is an epic (genuinely, I’m not sure we won’t reach hitherto non-invoked thread limits tbh).

*Seeing double* Nope, these images are not the same, or even from the same image sequence! They were taken 23 days apart, with raging snow storms & high winds between them. #Snowleopards are creatures of habit, regularly coming back to the same tiny patches of ground to see who's been around.

Next Thursday (11th), join us to discover how 300+ local community members & camera traps generate vital ecosystem data, transforming biodiversity monitoring in the Sanjiangyuan. 📸🌿 Learn about innovative incentive models for #conservation in #China. 👉 bit.ly/SLNBioMonitoring

Confused writing is usually a symptom of confused thinking. As we struggle to clarify writing, we clarify our thoughts. AI writing aids rob us of that struggle, leaving clean-looking text and thoughts still confused for lack of inspection. Writing is not just a product; it is a diagnostic tool.

Only a pre-print for now, but after 4 years of hard work I couldn't resist sharing this! The Global Canopy Atlas: analysis-ready maps of 3D structure for the world's woody ecosystems 📜: doi.org/10.1101/2025... Huge team effort led by the brilliant Fabian Fischer!

History appears to be repeating itself, with a new "mega-rice" project underway in South Papua, a repeat of the social and ecological disaster that was Mega Rice I in Kalimantan in the late 1990s. The UN says that > 50,000 Indigenous people will be directly affected e360.yale.edu/features/ind...

Consider the first headline -- "Brazil secures Amazon allies for $125 billion global forest fund" -- against the backdrop of everything else Brazil is doing. The hypocrisy from the host of the COP30 climate conference is staggering.

Incredible graph, and pretty much sums it up. Teeing off as the world burns.

Here’s me kind of hoping it doesn’t shut down Libgen though, a vital lifeline for many who don’t have fancy pantsy university journal access (= vast number of researchers in Global South, plus almost everyone in the NGO world)

From last week spotting the turtle doves in the Knepp rewilding project, to having the same species at our garden pond here in Ronda, Spain!

Can we all agree macroecology is also lawful evil 😈 Spot on for Conservation ecology

Interesting 🤔 Seems like a good way forward to ensure rigour and accuracy in work. I suspect there is a grey area, though, in deciding when genAI is acting as an assistant (which is allowed under their rules) and when it has stepped over the line into creator (not allowed).

AI challenge to find lost Amazonian civilizations draws critics | Science | AAAS www.science.org/content/arti... There are echoes here of how macroecologists mine satellite & other global data to make maps of conservation priorities, also often without consultation or validation on the ground.

Are drones fancy gadgets, or can they effectively guide management strategies and align with the overarching goals of protected areas? Rossi & Wiesmann 2025 argue the latter 👌 besjournals.onlinelibrary.wiley.com/doi/10.1002/...

My first @uk.theconversation.com article with @jackantbam.bsky.social and @lambin-ecology.bsky.social and @kennyafc.bsky.social "Surprisingly effective way to save the capercaillie: keep its predators well-fed" theconversation.com/a-surprising...

A no-brainer if ever I saw one.

A new paper in The Lancet, one of the world’s foremost peer reviewed medical journals, estimates that USAID prevented 91 million deaths across 133 countries over 20 years. The paper estimates that Elon Musk’s DOGE funding cuts to USAID will lead to 14 million deaths by 2030 (4.5 million children).

Related issues in #cameratrap AI exist, still unresolved. E.g. Google’s SpeciesNet, ostensibly trained on public data (techcrunch.com/2025/03/03/g...) but actually trained on partly embargoed data, w/ little recognition for the 100’s of people in the Global South who collected & curated the data 😧

A really clear and practical guide to implementing SMART and improving its effectiveness over time (written by the SMART Partnership and @faunaflora.bsky.social): docs.google.com/presentation...

Nepal has launched its first-ever species-specific action plan for dholes (Cuon alpinus), allocating 262.9 million rupees ($1.9 million) over five years to address key threats such as habitat loss, prey depletion, disease and competition with larger predators.

No mention of the disease risks. Or the welfare concerns of releasing birds into places that don’t suit them. This is pretend conservation, dressed up as the real thing and given a free ride by a serious environmental journalist. Hugely damaging. www.theguardian.com/environment/...

Our friendly neighbourhood fox passes just under our bedroom window almost every night. Taken with a @naturespy.bsky.social Ursus solar-powered ☀️ #cameratrap, which has been successfully powering itself across Dutch forests and Mongolia deserts recently. Goodbye battery changes! 👋

we're living through one of the greatest technological revolutions in the history of mankind and it's not AI

Sometimes R just beats you up for no reason.