Fabian Roger 's Avatar

Fabian Roger

@latrunculia

πŸ”¬ Biodiversity. With 🧬 from the 🌬️ . πŸ„ 🐦 πŸ‡ πŸ¦‹ 🌺 🦠 Father, Scientist, Climber. Aspiring Entrepreneur.

260
Followers
69
Following
30
Posts
03.11.2023
Joined
Posts Following

Latest posts by Fabian Roger @latrunculia

1) It's about population growth rates, NOT species extinctions
2) it's only about vertebrates
3) it's average decline rates _irrespective_ of population size

I admire the intention of LPI but the index itself is terrible and widely misunderstood. Something is not always better than nothing.

22.01.2024 15:28 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

Tl;Dr the LPI averages dΓ©cline rates in vertebrate population time-series but doesn't account for population size. It also weights the time-series by taxa richness - which gives most weight to time-series with the least amount of data.

22.01.2024 15:25 πŸ‘ 0 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0
Preview
IMD on LinkedIn: Welcome to the brave new world of eDNA and ecoacoustics Navigating biodiversity accountability With over two-thirds of species lost in the past 60 years and more than a million more at risk in the next decade…

No, we have not lost two thirds of all species in the last 60 years.

I see this (or variations of) this claim all the time. It's (luckily!) a bad misunderstanding of the Living Planet Index (LPI) πŸ‘‡

www.linkedin.com/feed/update/...

22.01.2024 15:19 πŸ‘ 0 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0
Post image

The #WEF global risk support for 2023: Climate Change & Biodiversity #1-#3 for in 10 years but Biodiversity doesn't make the Top10 for in 2 years.

Biodiversity is super important but always only in the future.

This is dangerous thinking.

www.weforum.org/publications...

11.01.2024 09:49 πŸ‘ 1 πŸ” 3 πŸ’¬ 0 πŸ“Œ 0
Preview
Airborne eDNA captures three decades of ecosystem biodiversity bioRxiv - the preprint server for biology, operated by Cold Spring Harbor Laboratory, a research and educational institution

πŸ‘‡πŸ‘‡πŸ‘‡

3 decades (!) of biodiversity change captured by #airborne #eDNA 🌬️🧬

The best evidence of the power of airborne eDNA to date, long time-series, very thorough analysis.

Per Stenberg (@umeauniversity) got funding for this in 2017 (!). So glad to see it out πŸ‘πŸ‘πŸ‘

www.biorxiv.org/content/10.1...

10.01.2024 13:46 πŸ‘ 10 πŸ” 4 πŸ’¬ 0 πŸ“Œ 1

scholar.google.com/citations?us...

10.01.2024 09:13 πŸ‘ 0 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

"Das "BΓΌndnis Sahra Wagenknecht" wird von einer Doppelspitze gefΓΌhrt"

Sahra W. und S. Wagenknecht?

08.01.2024 18:44 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

Sounds like interesting work! I couldn't find in the article how the 'dust' was sampled? Is there a pre-print already?

20.12.2023 07:31 πŸ‘ 1 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

From my part, this was the brainchild of my PhD, after thinking about diversity metrics and multifunctionality metrics *a lot* and finally coming to the conclusion that they are intimately related. Jarretrobi independently thought along the same lines.

11.12.2023 21:30 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0
Hill–Chao numbers allow decomposing gamma multifunctionality into alpha and beta components Ecology Letters is a broad-scope ecology journal considering all taxa, in any biome and geographic area, and spanning community, microbial & evolutionary ecology.

I am really excited and unduly proud that the new Multifunctionality metric and framework we propose with @jebyrnes.bsky.social and Robert Bagchi was taken up and developed upon by Anne Chao herself 😊 (and a great cast of co-authors)

onlinelibrary.wiley.com/doi/full/10....

11.12.2023 21:28 πŸ‘ 2 πŸ” 1 πŸ’¬ 1 πŸ“Œ 0
Post image

I re-read @xkcd.com "what if" (2014) and came across this doodle which illustrated a task that is easy for humans but hard for computers (describe what happened in this scene?)

I am happy to report that it is _still_ difficult for computers 😁

11.12.2023 11:39 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0
Post image

Google needs to better filter its autogenerated responses.

10.12.2023 18:45 πŸ‘ 0 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

Der Diskurs ist so vergiftet wie bei keinem anderen Konflikt bevor. Mensch hat die Wahl zwischen Antisemit oder Genozid-befΓΌrworter. Wohin soll das fΓΌhren?

07.12.2023 07:39 πŸ‘ 1 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0
Preview
Biofuels: the US could switch to electric cars and solar power on just a fraction of the land Biofuels could be making climate change worse, not better. There's a way to put this land to good use.

I new the biofuel debate but I had no idea of the scale of it 😦

tl;dr: the US produces Biofuel on an area equivalent to the *surface of the UK*, covering 10% of the fuel demand.

If photovoltaik was insatlled on 1/12th of that area, it could power an all-electric US Car fleet 😲

25.11.2023 17:38 πŸ‘ 16 πŸ” 5 πŸ’¬ 2 πŸ“Œ 2

Plos One tried to make a new phone publishing model work. MPDI saw the flaw and decided to make big bugs of it... 🀷

24.11.2023 02:20 πŸ‘ 2 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

And the journals are popular / esteemed because?

Of course IFs were deviced as proxy for 'quality' however that is defined (Zen and the Art of Motorcycle maintenance). I don't say they are a good proxy (but I do think we give them more credit than we admit). It's a different discussion though.

22.11.2023 07:58 πŸ‘ 0 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

2/2 that's an argument for preprints (which I am a great fan of) not for predatory publishers that also publish good papers. The backlash against MDPI is against the publisher l, not all articles published by it.

21.11.2023 21:58 πŸ‘ 1 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

No it doesn't and it turns the argument on its head. I argued that I don't trust MDPI to provide any wetting of the articles they publish (in my field) that I trust. Doesn't mean I blindly trust other publishers but say BES journals I trust more 🀷. And if we just go by the article 1/2

21.11.2023 21:57 πŸ‘ 1 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

And why would the librarians spend money to subscribe to journals with higher IFs? For whom did the librarians licence the journals?

But again, this is not a defence of the IF, it was just to say that journals have signaling power. And if we don't believe that, the alternatives are pre-prints.

21.11.2023 21:50 πŸ‘ 1 πŸ” 0 πŸ’¬ 2 πŸ“Œ 0

I know&I'm a big fan of preprints. I do think we need mechanisms of peer-endorsments and quality checks but I am not convinced journals are the best way to go. BUT if we go with journals, we need at least to reject the predatory ones...

21.11.2023 20:59 πŸ‘ 1 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

What I don't get is, following the 'read the paper' argument, what service does MDPI provide over (bio)arxive or any other pre-print server? It's completely free for anyone, authors and readers.

21.11.2023 20:50 πŸ‘ 2 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

Fair if it's an easy to check experimental result. If it's a review on a niche topic (of which there are many in MDPI) it's difficult to assess the quality if I am not an expert myself. If it's a reputable journal in the field I trust it was wetted, for MDPI I don't.

21.11.2023 20:47 πŸ‘ 1 πŸ” 0 πŸ’¬ 2 πŸ“Œ 0

For libraries to select what journals to subscribe to, so we scientists can read them...

21.11.2023 20:43 πŸ‘ 1 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

If any, then the role of journals is to provide a proxy to screen the Literatur. The IF was invented for that (with all its flaws). It is literally impossible to read all relevant papers and 'judge by ourselves'.

The argument against MDPI is also against the publisher, not all papers.

21.11.2023 19:46 πŸ‘ 1 πŸ” 0 πŸ’¬ 2 πŸ“Œ 0

Quality standards - such that I am reluctant to cite MDPI papers. And I think in both cases they abuse underprivileged communities and ECRs more than they serve them.

21.11.2023 13:17 πŸ‘ 3 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

Hmm, not sure I follow that logic. MDPI can very well have some
good papers but pursue a predatory (ish) model, or? Just as bioarxive can have excellent papers even if they are pre-prints. The rejection of MDPI (from me) comes from their hyper-aggressive special issue politics and the lack of 1/2

21.11.2023 13:12 πŸ‘ 2 πŸ” 0 πŸ’¬ 2 πŸ“Œ 0

Cool! As soon as you have any sequencing results I'd be very curious to chat. We also tested 8 sampler types and are about to sequence 5 marker genes :) (happy to chat before, too, of course)

20.11.2023 21:05 πŸ‘ 1 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

Oh right. Yes, Anish told me about your work, cool stuff! Would be nice to chat sometime to exchange experiences?

20.11.2023 20:40 πŸ‘ 1 πŸ” 0 πŸ’¬ 1 πŸ“Œ 0

(sorry, my Bluesky profile is barebones for now. Here is my LinkedIn www.linkedin.com/in/fabian-ro...)

20.11.2023 20:36 πŸ‘ 1 πŸ” 0 πŸ’¬ 0 πŸ“Œ 0

Would love to learn more about what you're doing!

20.11.2023 20:34 πŸ‘ 0 πŸ” 0 πŸ’¬ 2 πŸ“Œ 0