Saved in:
Bibliographic Details
Main Authors: Dong, Rongchao, Sun, Yiming, Chen, Shuo, Oh, Youmi, Liu, Licheng, Xie, Yiqun, Jia, Xiaowei
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2606.00338
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914619595948032
author Dong, Rongchao
Sun, Yiming
Chen, Shuo
Oh, Youmi
Liu, Licheng
Xie, Yiqun
Jia, Xiaowei
author_facet Dong, Rongchao
Sun, Yiming
Chen, Shuo
Oh, Youmi
Liu, Licheng
Xie, Yiqun
Jia, Xiaowei
contents Methane is a potent greenhouse gas that significantly contributes to global warming. However, accurately estimating global methane emissions and consumption remains challenging due to the complex interactions among environmental drivers that may vary across spatial and temporal scales. Prior data-driven methods often overlook the inherent spatiotemporal heterogeneity of ecosystems, failing to explicitly capture site-specific characteristics and cross-year evolutionary dynamics. To address these issues, we propose the Contrastive Hierarchical Adaptive Meta-network (CHAM-net), a novel framework that explicitly learns from historical context to capture site-specific dynamics. CHAM-net employs a hierarchical encoder-decoder architecture, in which the encoder captures site-specific characteristics from historical data and then dynamically conditions the decoder to generate the final prediction. Experimental results demonstrate that CHAM-net consistently outperforms all baseline methods on both simulation and observational datasets for methane emission and consumption, achieving nRMSE values as low as 0.43 and 0.88 with corresponding R2 scores up to 0.97 and 0.68 for emission prediction.
format Preprint
id arxiv_https___arxiv_org_abs_2606_00338
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CHAM-net: A Contrastive Hierarchical Adaptive Meta-network for Robust Global Methane Flux Prediction
Dong, Rongchao
Sun, Yiming
Chen, Shuo
Oh, Youmi
Liu, Licheng
Xie, Yiqun
Jia, Xiaowei
Machine Learning
Methane is a potent greenhouse gas that significantly contributes to global warming. However, accurately estimating global methane emissions and consumption remains challenging due to the complex interactions among environmental drivers that may vary across spatial and temporal scales. Prior data-driven methods often overlook the inherent spatiotemporal heterogeneity of ecosystems, failing to explicitly capture site-specific characteristics and cross-year evolutionary dynamics. To address these issues, we propose the Contrastive Hierarchical Adaptive Meta-network (CHAM-net), a novel framework that explicitly learns from historical context to capture site-specific dynamics. CHAM-net employs a hierarchical encoder-decoder architecture, in which the encoder captures site-specific characteristics from historical data and then dynamically conditions the decoder to generate the final prediction. Experimental results demonstrate that CHAM-net consistently outperforms all baseline methods on both simulation and observational datasets for methane emission and consumption, achieving nRMSE values as low as 0.43 and 0.88 with corresponding R2 scores up to 0.97 and 0.68 for emission prediction.
title CHAM-net: A Contrastive Hierarchical Adaptive Meta-network for Robust Global Methane Flux Prediction
topic Machine Learning
url https://arxiv.org/abs/2606.00338