This work showed patterns of heatshock causing discolouration that can be identified by multispectral sensors (Sentinel-2) and used this to show shock then recovery of a known seagrass bed in Quiberon France before during and after a Heatwave.
Check out the full paper here: doi.org/10.1016/j.rs...
When intertidal Seagrass experiences heat shock, it can cause discolouration to the leaves. We often use colour of vegetation to assess percentage cover of leaves from optical remote sensing, so understanding heatwave effects to colouration is highly important for our estimates of seagrass cover!
🚨🚨 New Intertidal Seagrass Paper Alert🚨 🚨
This time led by Simon Oiry who utilised a compination of laboratory 👨🏻🔬 and satelitte 🛰️ experiments to assess the effects of Heatwaves on the optical properties of Seagrass and developed the novel (and widely applicable) Seagrass Heat Shock Index!
It’s happeninggggg! ☺️☺️☺️
My go to for stats learning, refreshing and rethinking! m.youtube.com/watch?v=ztbY... get ready for lots of memes and hopefully owls being drawn
@rmcelreath.bsky.social
github.com/rmcelreath/s...
"You can agree or disagree with him, but Polanski does not pretend to be something he is not.
He is unusually willing for a politician to give a straight answer to questions.
Does he want to rejoin the EU? “Yes."
observer.co.uk/news/nationa...
I cannot agree more!
That makes sense, in academia it is the same struggle of balance! Just with less nature for my liking! But maybe that is just the grass always being greener
Oh wow yes, i can only imagine! I come from the other extreme of perhaps (if i am lucky from my point of view) 50 days of fieldwork a year! (But that doesnt include data processing/analysis/repotrting!)
60% in the field?!?! Damn, that is the dream!!!
After a few decades as an ecologist, with >60% (~20yrs) spent in the field undertaking terrestrial vert fauna surveys, my biggest take-aways inc:
1. You don't know everything,
2.You will forever keep learning,
3. Indigenous cultural knowledge = landscape ecology = awareness,
4 awareness = knowledge.
This is an education problem, not a tool problem; and we don't want people simply moving from thinking p-values are magic to thinking confidence intervals are.
Testing a YOLO-based computer vision model to ID 13 benthic species from underwater images. It boosts automated species recognition, aiding large-scale biodiversity monitoring & better marine conservation. @azti.bsky.social
#AI4Oceans #BenthicEcology #MarineScience #BioBoost+
Paper: doi.org/10.1016/j.rs...
User-Friendly Application: s2maxextenticecreams.projects.earthengine.app/view/europea...
ICE CREAMS: github.com/bedeffinian/...
We also found some general latitudinal patterns in intertidal seagrass, with the proportion of the intertidal area covered by seagrass increasing as you go south, and not only that, the cover/density of that seagrass tends to increase the further south you go.
From this total area we found hotspots of intertidal seagrass across the whole of Europe, with three notable hotspots coming out as the biggest in the North Frisian Wadden Sea (Germany), Arcachon Bay (France) and Ria Formosa (Portugal).
Then using a cloud-computing approach we classified the over 15,100 km ² of intertidal area in Europe, finding an area larger than both Paris and Lisbon combined appeared to be seagrass at over 50 % cover (212 +/- 18.8 km ²)
By updating the ICE CREAMS model, we improved its accuracy from 82 % in previous work to 91 % by decreased its’ false positive rate on over 60,000 independent and openly accessible validation data.
5) lead a Call-to-Action for the availability of data following the FAIR principles to improve future iterations of this European scale map.
Through this work we feel we successfully achieved all of the first four aims and hope that this paper will achieve the fifth: a Call-to-Action!
2) use ICE CREAMS to create a contemporary map of Intertidal Seagrass for Europe
3) provide a robust and scalable methodology to allow others to map intertidal seagrass at large scales
4) provide an open-access and hopefully easy to use tool to view this map
We had 5 main aims:
1) improve and fine tune the open access model Intertidal Classification of Europe: Categorising Reflectance of Emerged Areas of Marine vegetation with Sentinel-2 (ICE CREAMS 🍨)
We aimed to do this by using products or data that were openly available and consistent across Europe from the Copernicus Sentinel-2 imagery, the Google Earth Engine cloud-computing used to sort this imagery, the ICE CREAMS model used to classify the imagery even down to the outputs themselves.
We have managed to create *An Intitial Map of European Intertidal Seagrass*. This is, to our knowledge, the first ever attempt to map Intertidal Seagrass at the scale of Europe (or any continental scale for that matter)!
s2maxextenticecreams.projects.earthengine.app/view/europea...
🚨🚨🚨 New Open-Access Paper, Dataset, Application and Call-to-Action 🚨🚨🚨This one is for the fans of Intertidal Seagrass 🌱🌊 and Findable, Accessible, Interpretable and Reusable (FAIR) data. 🧵 doi.org/10.1016/j.rs...
probably my most deeply-held belief: it is not possible for a slide deck to be both (1) a good presentation aide; and (2) a standalone document that can be read and understood in isolation.
"In a landmark study, OpenAI researchers reveal that large language models will always produce plausible but false outputs, even with perfect data, due to fundamental statistical and computational limits."
www.computerworld.com/article/4059...
Hey you wonderful EO scientists. Why not nominate your amazing colleague or team or nominate yourself for the next ESA EO excellence award. Nominations have just opened and close 10th Dec. Research cash prize for lucky winners:
www.eoxcellence.com
A few more days to apply!!!
We are seeking a postdoctoral researcher in statistics to develop methods for analyzing large scale biodiversity and ecosystem function data at the University of Helsinki.
This is a great opportunity for someone who wants to do advanced Bayesian modeling with direct link to real world applications.
Not one needs context for this
A brief glance at the github seems exactly the style of multivariate analysis that needs to become common place! 👏🏻 will try incorporate into my own teaching if i can, and fingers crossed for continuity!
