Tackling climate change with machine learning

D Rolnick, PL Donti, LH Kaack, K Kochanski… - ACM Computing …, 2022 - dl.acm.org
Climate change is one of the greatest challenges facing humanity, and we, as machine
learning (ML) experts, may wonder how we can help. Here we describe how ML can be a …

Geo-bench: Toward foundation models for earth monitoring

A Lacoste, N Lehmann, P Rodriguez… - Advances in …, 2024 - proceedings.neurips.cc
Recent progress in self-supervision has shown that pre-training large neural networks on
vast amounts of unsupervised data can lead to substantial increases in generalization to …

ReforesTree: A dataset for estimating tropical forest carbon stock with deep learning and aerial imagery

G Reiersen, D Dao, B Lütjens, K Klemmer… - Proceedings of the …, 2022 - ojs.aaai.org
Forest biomass is a key influence for future climate, and the world urgently needs highly
scalable financing schemes, such as carbon offsetting certifications, to protect and restore …

Quantification of carbon sequestration in urban forests

LJ Klein, W Zhou, CM Albrecht - arxiv preprint arxiv:2106.00182, 2021 - arxiv.org
Vegetation, trees in particular, sequester carbon by absorbing carbon dioxide from the
atmosphere. However, the lack of efficient quantification methods of carbon stored in trees …

Simple and approximately optimal contracts for payment for ecosystem services

WD Li, I Ashlagi, I Lo - Management Science, 2023 - pubsonline.informs.org
Many countries have adopted payment for ecosystem services (PES) programs to reduce
deforestation. Empirical evaluations find such programs, which pay forest owners to …

High-accuracy Machine Learning Models to Estimate above Ground Biomass over Tropical Closed Evergreen Forest Areas from Satellite Data

K Tappayuthpijarn, BS Vindevogel - IOP Conference Series …, 2022 - iopscience.iop.org
Quantifying the amount of biomass stored in forested areas has been traditionally done with
manual field measurements, which is costly, time consuming and doesn't scale well over …

Generating Physically-Consistent Satellite Imagery for Climate Visualizations

B Lütjens, B Leshchinskiy, O Boulais… - … on Geoscience and …, 2024 - ieeexplore.ieee.org
Deep generative vision models are now able to synthesize realistic-looking satellite
imagery. However, the possibility of hallucinations prevents their adoption of risk-sensitive …

Improving the accuracy of automated labeling of specimen images datasets via a confidence-based process

Q Bateux, J Koss, PW Sweeney, E Edwards… - arxiv preprint arxiv …, 2024 - arxiv.org
The digitization of natural history collections over the past three decades has unlocked a
treasure trove of specimen imagery and metadata. There is great interest in making this data …

Learning Geospatial Region Embedding with Heterogeneous Graph

X Zou, J Huang, X Hao, Y Yang, H Wen, Y Yan… - arxiv preprint arxiv …, 2024 - arxiv.org
Learning effective geospatial embeddings is crucial for a series of geospatial applications
such as city analytics and earth monitoring. However, learning comprehensive region …

Tackling the overestimation of forest carbon with deep learning and aerial imagery

G Reiersen, D Dao, B Lütjens, K Klemmer… - arxiv preprint arxiv …, 2021 - arxiv.org
Forest carbon offsets are increasingly popular and can play a significant role in financing
climate mitigation, forest conservation, and reforestation. Measuring how much carbon is …