Advancing a Standard for Carbon Flux Data Across Scales
BenchFlux is a research initiative with the goal of creating a new machine learning benchmark for carbon sequestration data.
BenchFlux is a research initiative with the goal of creating a new machine learning benchmark for carbon sequestration data.
BenchFlux is a consortium of research teams working on methods to integrate atmospheric data and remote sensing data across scales. Our goal is to improve on the current paradigm of scale-agnostic carbon flux data by creating a new machine learning benchmark that can consistently marry top-down remote sensing data with bottom-up eddy covariance data. See our proposed work⤵
Society faces the urgent challenge of developing sustainable, equitable, and scalable solutions to climate change, driven by the critical need for evidence-based management of life support systems like carbon, water, and heat.
To address this need, we are developing scale-aware benchmarks for measuring land carbon sequestration by leveraging atmospheric carbon flux data alongside in-situ and remotely sensed earth observations.
We aim to provide code tutorials and datasets to the climate community to help scientists and developers add scale awareness into their climate solutions.
Paul is a Professor in the Department of Biological Systems Engineering in the College of Agricultural and Life Sciences at the University of Wisconsin. He studies surface-atmosphere exchange (especially with eddy covariance) and the role of land use, land management, and conservation in the climate system.
Paul Stoy 🇺🇸
Álvaro Moreno-Martinez Álvaro Moreno is a senior researcher at the University of Valencia's Image & Signal Processing Group (ISP), specializing in remote sensing and machine learning for ecosystem monitoring at large scales. Dr. Moreno is a Google Developer Expert in Earth Engine and participates in NASA’s CEOS Land Product Validation subgroup, focusing on carbon fluxes estimations.
Álvaro Moreno-Martinez 🇪🇸
Emma Izquierdo-Verdiguier Emma Izquierdo-Verdiguier is senior Scientist at Institute of Geomatics in the University of Natural Resources and Life Sciences (BOKU), Vienna. Her expertise is focused on using Machine Learning to analyze Earth Observation data. In 2012, her paper was ranked second in the student’s competition of the IEEE Geoscience and Remote Sensing Symposium (IGARSS). She is a Google Developer Expert in Earth Engine (from 2022) and a member of the ELLIS AI excellence network (since 2024).
Emma Izquierdo-Verdiguier 🇦🇹
Sadegh Ranjbar is a research scientist at at the University of Wisconsin with experience in AI/ML, remote sensing, and geospatial analysis for environmental, agricultural, and urban systems. Skilled in large-scale data analysis, predictive modeling, and satellite data automation. Proficient in Python, cloud computing, and interdisciplinary research.
Sadegh Ranjbar 🇺🇸
Sam works on project coordination, outreach, and tutorials for the BenchFlux project. He has a research background in geology and geography, and currently serves as the geospatial coordinator at AtmoFacts.
Samuel Bower 🇺🇸
Maoya Bassiouni Maoya Bassiouni is an ecohydrologist in the Department of Environmental Science, Policy and Management at UC Berkeley. Her research combine physical and eco-evolutionary theory with data science to understand how vegetation interacts with environmental change and how these interactions influence water, carbon, and energy cycles from soil, plant, to the biosphere.
Maoya Bassiouni 🇺🇸
Yanghui is an assistant professor in the Department of Biological Systems Engineering at Virginia Tech. Her research quantifies terrestrial ecosystem dynamics in response to environmental changes and management practices, using a combination of satellite remote sensing, machine learning, and process-based models.
Yanghui Kang 🇺🇸
Trevor is an Associate Professor at UC Berkeley and Faculty Scientist at the Lawrence Berkeley National Laboratory. He is also the Chief Scientist at Earthshot Labs. His research interests include Terrestrial ecosystems and global change, ecosystem ecology, nature-based climate solutions, carbon sequestration, biosphere-atmosphere interactions and feedbacks, data science, and machine learning.
Trevor Keenan 🇺🇸
Ankur Desai is Vilas Distinguished Achievement Professor and Department Chair in the Department of Atmospheric and Oceanic Sciences at the University of Wisconsin-Madison. His lab studies and has authored nearly 200 articles regarding ecosystems, weather, and climate. This work spans from Wisconsin to the globe and uses both long term observations in nature and advanced computer simulations.
Ankur Desai 🇺🇸
Stefan is the founder of AtmoFacts, a climate technology startup that focuses on spatially attributing fluxes. Stefan previously led the Surface-Atmosphere Component of the US NSF National Ecological Observatory Network (NEON), focusing on data quality and availability. He serves on various national and international committees including AmeriFlux and FLUXNET, and in the alignment of the NEON and EU ICOS observing systems.
Stefan Metzger 🇺🇸