Emily Lines
@emilyrlines.bsky.social
1.7K followers 910 following 99 posts
Massive fan of trees, fairweather fan of data. UKRIFLF, Assoc Prof, University of Cambridge. #womeninstem she/her linesresearchgroup.com
Posts Media Videos Starter Packs
emilyrlines.bsky.social
oh I would. if all the deadlines weren't tomorrow....
emilyrlines.bsky.social
yeah sure I can write a paper and a proposal on two totally different topics today, on 3 hours sleep and massively jetlagged.

my brain is not happy with me.
emilyrlines.bsky.social
Having a great time at Silvilaser this week, full of inspiring talks&research on forest lidar. But it is heartbreaking to hear how many presenters are absent bcos they couldn't get Canada visas. We're missing out on great science and collaboration bcos of short sighted&cruel geopolitical decisions.
Reposted by Emily Lines
Reposted by Emily Lines
mikegalsworthy.bsky.social
Bravo Banksy. 👏

“A new mural by elusive street artist Banksy has appeared on the side of the Royal Courts of Justice building in central London.”
emilyrlines.bsky.social
Watching the wind... #lidar
Reposted by Emily Lines
thierryaaron.bsky.social
Pleased to announce that @tristramwyatt.bsky.social & I have a new article out in @uk.theconversation.com, based on work with colleagues from @scientistsforxr.earth

We outline "How scientists can contribute to social movements and climate action"

Here's a thread with some background & key points 🧵
How scientists can contribute to social movements and climate action
Collective action on climate is crucial. Scientists play an important role in driving change.
theconversation.com
Reposted by Emily Lines
davidho.bsky.social
We have to be purposeful about ending the use of fossil fuels because they’re not going to end themselves anytime soon. While wind and solar are cheaper, they are not as profitable. Therein lies the problem.
Reposted by Emily Lines
juliajones.bsky.social
If you know researchers with expertise in forestry/agro forestry please share this permanent lectureship @bangoruniversity.bsky.social. The quality of life in north wales is fabulous: great colleagues & a stunning location.
jobs.bangor.ac.uk/details.php....
emilyrlines.bsky.social
Manifesting your future life as a pop star. Bravo 👏👏👏
emilyrlines.bsky.social
Mercury is in retrograde, but the lions gate is open. Cosmic energy is powerful atm @markuseichhorn.bsky.social. be careful
Reposted by Emily Lines
21group.bsky.social
It's never bullying says the university

It's a misunderstanding, personality clash, cultural difference, performance management issue, ill-advised but unintentional

A he-said-she-said situation, the accused is under a lot of pressure

It's Hanlon's Razor, incompetence not bullying

@jcsj.uk
emilyrlines.bsky.social
Just turned down a #manel. A 2 day event with only male speakers? To the kind woman who invited me - quit helping this kind of event, it won't serve you. Those in charge make their priorities clear when they set these programmes. Token diversity & words of reassurance do nothing. Just don't engage.
emilyrlines.bsky.social
My university is applying a 1.4% pay rise this year, our only buffer against the worsening cost of living crisis. Why aren't our leaders aren't embarrassed & ashamed to treat staff this way? #UKHE is broken in so many ways, and these pathetic'pay deals' push staff morale beyond rock bottom.
emilyrlines.bsky.social
Our study demonstrates the potential of leveraging existing available datasets and a generalised deep learning framework with only a relatively small amount of location-specific data to create a tailored model for Cameroon.
emilyrlines.bsky.social
We find that:
1. High performance is possible for even detailed classes (macro-average F1 0.77).
2. Landsat-8 outperformed higher resolution PlanetScope data for classification.
3. Country-specific reference datasets, classification schemes and expertise are essential to maximise usability.
emilyrlines.bsky.social
Unlike many large scale studies, we use local data & context-specific classes. We collated a dataset of detailed drivers from a wide range of sources (nature.com/articles/s41597-024-03384-z) and used this to build & test a model for classifying drivers in #GFC polygons.
A labelled dataset to classify direct deforestation drivers from Earth Observation imagery in Cameroon - Scientific Data
Scientific Data - A labelled dataset to classify direct deforestation drivers from Earth Observation imagery in Cameroon
www.nature.com
Reposted by Emily Lines
markuseichhorn.bsky.social
Last week was the final conference of the @3dforecotech.bsky.social network. Over 100 researchers showcasing new methods for measuring and modelling forests. We've achieved a huge amount in four years and there's much more to come.
Group photo from the 3DForEcoTech meeting in Poland
emilyrlines.bsky.social
That does sound more sensible
emilyrlines.bsky.social
who are these photos for? probably three will make their way onto a news article on the uni website that will disappear in a day. not worth the disruption and intrusion
emilyrlines.bsky.social
Why are conferences so heavily photographed? I'm at my second one in two weeks with photographers and I'm so distracted by the moving & clicking. I understand the desire to promote events but it is disruptive when you're trying to concentrate and someone's taking the 500th photo of the audience...