Tao Huang

@BCAL_Boise researcher interested in remote sensing, hydrology, and machine learning @UNH_GradSchool @caryinstitute alumna 🇺🇸 🇹🇼

[Tao Huang @WildcatTao]

PyArrow is faster for reading large CSV files: https://t.co/83rE81TKUT

[Tao Huang @WildcatTao]

Grid-based management turns Idaho’s forests into a giant chessboard: https://t.co/G8LZwIRGQB via @NASA

[Tao Huang @WildcatTao]

@Ologies think about this episode after seeing the exhibit at @burkemuseum

[Tao Huang @WildcatTao]

AlphaEarth Foundations delivers a global, 10m-resolution dataset of analysis-ready embeddings for every year since 2017, ready to use with minimal effort.

[Qiusheng Wu @giswqs]

Google just released the AlphaEarth Foundations with 64 dimensions of satellite embeddings at 10-m resolution at the global scale! It is very interesting! Check it out

Blog post: https://t.co/RviUl5TRIc

Dataset: https://t.co/mCMafemeE2

Paper: https://t.co/isoE4zOXfZ

#AI #geospatial #remotesensing #geoai

[Tao Huang @WildcatTao]

Interesting podcast! How satellite data and Vision Transformers are creating the 10-meter global natural forest map. Also dive into tackling noisy data & uncertainty.

[Google DeepMind @GoogleDeepMind]

What if we could confidently predict the future of our ecosystems - including how humans are impacting them? 🌳

Our host @fryrsquared and Nature Lead @DrewPurves discuss how AI could unlock accurate simulations of our environment, allowing us to better protect it.

Watch now ↓

[Tao Huang @WildcatTao]

However, it sometimes struggles to predict out-of-range values, such as extreme temperatures not present in the training dataset. Even with advancements in machine learning, domain knowledge and process-based models remain crucial for interpreting ML in some cases.

[Tao Huang @WildcatTao]

The FireSat satellite has launched to help detect wildfires as small as 5x5 meters

[Google AI @GoogleAI]

FireSat, a new wildfire detection proto satellite, has launched on SpaceX's Falcon 9! High-res imagery updated every 20 mins will soon help detect 5x5m wildfires globally. Congrats to @MuonSpace, @MooreFound, @EarthFireAll: https://t.co/uyHdBH0LFh

[Tao Huang @WildcatTao]

High-resolution UAS imagery offers ecologically relevant information at a much finer scale than moderate-resolution satellite data. Check out this review of UAS applications in landscape ecology: https://t.co/V0U62RLF3m

[Tao Huang @WildcatTao]

Our new paper shows how high-resolution multispectral drone imagery can effectively map vegetation cover in spatially heterogeneous landscapes https://t.co/OGX6ud2dE5

[Tao Huang @WildcatTao]

High-resolution imagery is useful for estimating plant abundance and informing management decisions. However, users must be conscientious of false negative (missed plants) and false positive (misclassified) errors.

[Tao Huang @WildcatTao]

Happy new water year! #MississippiRiver #AWRA2024

[Tao Huang @WildcatTao]

Annual Watershed Watch to help school kids gather data of the Boise River

[Tao Huang @WildcatTao]

Soil burn severity, crucial for predicting post-fire hydrology, requires laborious field assessments. This paper demonstrates machine Learning models and the explainable AI approach to translate remotely sensed vegetation burn severity to soil burn severity #XGBoost #Shapley

[M @SadeghMojtaba]

Led by Teymoor Seydi, we developed Physics-informed #MachineLearning models to infer soil burn severity (SBS) from Vegetation Burn Severity (VBS)

Published in @EarthsFutureEiC, available #OpenAccess at https://t.co/ZhGV5t9dtR
#wildfire #remotesensing #Weather #soil #climate

[Tao Huang @WildcatTao]

@giswqs Can't wait to test the model on my drone footage!

[Tao Huang @WildcatTao]

There are thousands of temples in Taiwan, but this one is unique. The entire building can be raised during floods!

[Tao Huang @WildcatTao]

how to create custom functions in R and the importance of documenting your function

[USGS Data Science @USGS_DataSci]

Looking to level up your data analysis workflows? Read all about “Writing functions that work for you!”
This is the first in a new blog series, “Reproducible Data Science with R”

https://t.co/XPAkbP40Hu

#rstats #WDFN #WaterDataForTheNation #Blog

[Tao Huang @WildcatTao]

If you are interested in post-fire vegetation, runoff&erosion, please come to my lightning talk tomorrow! #FBF2024

[Tao Huang @WildcatTao]

I am looking forward to #AGU23 this week! One of my poster presentations will be tomorrow afternoon.

[Tao Huang @WildcatTao]

Happy new water year! #hydrology #BoiseRiver