Celeste Mcfarlane

💗 Fun, Femme, Fresh 💕 Data analytics, air quality and environmental justice @LamontEarth 💓 they/them

[Celeste Mcfarlane @Celeste_MM_]

@vfmcneill 😊🙈💜

[Celeste Mcfarlane @Celeste_MM_]

@vfmcneill 🥹😊💜

[Garima Raheja (she/hers) @rahejagarima]

Thanks @columbiaclimate for featuring our PM2.5 monitoring work in Lomé, Togo! ft. @KokouSabi @dekadzewofia @Celeste_MM_ @d_westy

[Garima Raheja (she/hers) @rahejagarima]

My 1st 1st-authored paper just dropped! We report the first-ever ambient PM2.5 data from Togo. Daily/weekly patterns are governed by (anthro) local influences, but the baseline is driven up by the (natural) Harmattan which makes AQ levels very unhealthy.
https://t.co/XpTB1pRL3F

[Celeste Mcfarlane @Celeste_MM_]

@vfmcneill 😍😍😍

[Garima Raheja (she/hers) @rahejagarima]

@vfmcneill @Celeste_MM_ AGU rebrand incoming

[Garima Raheja (she/hers) @rahejagarima]

Thanks for the feature, @columbiaclimate! I’m so honored to be here at @LamontEarth working with partners near and far on solving our most pressing environmental problems 🌎🌍🌏 https://t.co/XuBOSKmnLi

[Celeste Mcfarlane @Celeste_MM_]

@vfmcneill Oooh 😍🤩, indeed it is!!!

[Celeste Mcfarlane @Celeste_MM_]

@vfmcneill Omg so fun femme AND fresh 😍

[Celeste Mcfarlane @Celeste_MM_]

@geocuriosity The LCSs in this study were used to analyze PM concentrations. Looking at temporal patterns of PM concentrations we indirectly made assumptions about emissions sources. I myself, however, am not familiar with work using LCSs to directly attribute PM to a particular source.

[Celeste Mcfarlane @Celeste_MM_]

I made an adult Twitter to tweet about adult things like my new paper: “Application of Gaussian Mixture Regression for the Correction of Low-Cost PM2.5 monitoring data” so here I am doing that. Cool statistical model aside, aren’t my figures pretty?!?

https://t.co/KUNqkLEfdx

[Celeste Mcfarlane @Celeste_MM_]

@geocuriosity This method essentially learns multiple calibration methods and uses probability to decide which/how to apply the given methods it’s formed. This brings increased hope to reliable data across site transfers and more accessible high quality data

[Celeste Mcfarlane @Celeste_MM_]

@geocuriosity There is the issue of model reliability across sites, but the cool thing about this method is it can estimate its own uncertainty.

[Ivan Alcantara @ivancalcantara]

“pronouns are confusing” bro wtf is a conda environment