Month: January 2025

Darby, Yujie, and Andrew are coauthors on a paper led by Jeffrey Uyekawa and Ben Lucas of NAU’s Department of Mathematics and Statistics in a special Landscape Ecology section of the journal Land. The study uses the Extreme Gradient Boosting machine learning model to model land-atmosphere CO₂ fluxes, at 30 minute temporal resolution, across 44 NEON sites. In addition to standard k-fold cross-validation techniques, the study applies a “leave-one-site-out” (L1SO) approach to test predictions at a site which had not been used, in any way, to train the model. The results show strong potential for machine learning-based models to make more skillful predictions than state-of-the-art process-based models, being able to estimate the multi-year mean carbon balance to within an error ±50 gCm⁻²y⁻¹ for 29 of 44 test sites. L1SO model performance was better when ecologically similar sites were included in the training data, and worse when there were no similar sites in the training data (e.g., more unique ecosystems in the Pacific Northwest, Florida, and Puerto Rico). Results also point to the enormous potential of machine learning to predict not only the long-term carbon balance of an unknown site, but even the inter-annual variation in that carbon balance.  These results have significant implications for being able to accurately predict the carbon flux or gap-fill an extended outage at any AmeriFlux site, and for being able to make skillful predictions of ecosystem-scale carbon balance in support of natural climate solutions.

A companion paper on water fluxes is in preparation.

NEON tower at Niwot Ridge, Colorado (courtesy of NEON).

Mark D. Schwartz, University of Wisconsin–Milwaukee, is well-known for his pioneering efforts—beginning in the 1980s—to bring the science of phenology into the modern era. He is also highly regarded as the editor of the book Phenology: An Integrative Environmental Science. The third edition of this book, published by Springer, was recently released. Across 644 pages, the book’s 27 chapters span a wide range of topics. You’ll see lab members’ names on several of those chapters:

• Perry and Andrew wrote the chapter on Mesic Temperate Deciduous Forest Phenology. Special congratulations to Perry on his first first-author publication!

• Oscar and Andrew wrote an updated version of the chapter on Near-Surface Sensor-Derived Phenology.

• Darby and Andrew contributed to the chapter on Phenology in Higher Education, which was led by long-time collaborator Theresa Crimmins of the USA-National Phenology Network.

Our lab remains committed to fostering an inclusive environment, and we kicked off the year by participating in IE 103: Inclusive Language, led by Lee Griffin (Assistant Director of Learning and Development) and Dr. Justin Mallet (Vice President of Inclusive Excellence). This engaging session highlighted the power of language in creating spaces where everyone feels valued and heard while strengthening our lab culture. We’re grateful to Lee and Dr. Mallet for their continued guidance and inspiration. We look forward to continuing this journey later in the spring with IE 104: Elevating Inclusive Voices.

You can read more about NAU’s commitment towards Inclusive Excellence here, where you will also find a calendar of upcoming events .