Temporal Memory Mechanisms and Biome‐Specific Drivers of Ecosystem Carbon Flux: Insights From Explainable Deep Learning Modeling
Global Change Biology · 2026-01 · 32 (1)
DOI: 10.1111/gcb.70722 ↗Kurz gefasst
Fachartikel in der Zeitschrift Global Change Biology, online erschienen am 27.01.2026. Verfasst von Teng Ma und 11 weiteren Autor:innen. Beteiligte Einrichtung: Zhejiang Carbon Neutral Innovation Institute & Zhejiang International Cooperation Base for Science and Technology on Carbon Emission Reduction and Mo…. Hauptthemen laut MEDLINE: Carbon Cycle, Ecosystem, Deep Learning. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 104 Literaturangaben. Gefördert von: National Key Research and Development Program of China, National Natural Science Foundation of China, Basic Research Program of Qinghai Province.
Abstract (Originalsprache)
Ecosystem carbon dynamics are governed by complex temporal dependencies and environmental interactions, yet these processes remain poorly understood and quantified across diverse biomes. Here, we developed an explainable LSTM‐Attention framework that integrates LSTM networks, attention mechanisms, and gradient‐based attribution methods to reveal temporal dependencies in ecosystem carbon flux responses by analyzing eddy covariance flux measurements from 71 sites spanning eight North American biomes. Using this approach, we identified three distinct temporal memory patterns governing carbon flux responses. Grasslands exhibit short‐term memory dominance with exponentially increasing temporal contributions from distant to recent past. Deciduous broadleaf forests and wetlands show long‐term memory dominance, with deciduous broadleaf forests displaying the strongest historical dependence (attribution values declining from 0.30 at 6 months to 0.045 at 1 month). Croplands and evergreen needleleaf forests demonstrate U‐shaped dual memory patterns. Building on these temporal patterns, we identified biome‐specific environmental drivers operating within each memory framework: wetlands primarily controlled by soil moisture, evergreen needleleaf forests by radiation, and closed shrublands by vapor pressure deficit. Beyond individual drivers, we uncovered critical nonlinear interactions that diverged from linear correlations at most sites (55 of 71 with ρ < 0.5). For instance, the carbon sink capacity of deciduous broadleaf forests depends on synchronized canopy development and photosynthetic activity, while closed shrublands show strong suppression of carbon uptake by atmospheric water deficit regardless of vegetation greenness, revealing how multiple drivers jointly regulate ecosystem functioning. Validating memory's fundamental role, ablation experiments confirmed that removing memory mechanisms degraded model prediction performance and altered environmental driver identification (Kendall's Tau < 0.5), demonstrating that temporal memory is integral to accurately modeling ecosystem carbon flux responses to environmental drivers. These findings provide mechanistic insights into temporal controls of carbon exchange across different biomes. This knowledge is critical for improving terrestrial carbon‐climate feedback representations under global change.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Zhejiang Carbon Neutral Innovation Institute & Zhejiang International Cooperation Base for Science and Technology on Carbon Emission Reduction and Monitoring, Zhejiang University of Technology, Hangzhou, Zhejiang, China.
- Zhejiang Key Laboratory of Low-Carbon Control Technology for Industrial Pollution, College of Environment, Zhejiang University of Technology, Hangzhou, Zhejiang, China.
- Earth Systems Research Center, Institute for the Study of Earth, Oceans, and Space, University of New Hampshire, Durham, New Hampshire, USA.
- State Key Laboratory of Water-Disaster Prevention, College of Hydrology and Water Resources, Hohai University, Nanjing, Jiangsu, China.
- State Key Laboratory of Remote Sensing and Digital Earth, Faculty of Geographical Science, Beijing Normal University, Beijing, China.
- School of Geography and Tourism, Anhui Normal University, Wuhu, Anhui, China.
ORCID-Profile: Teng Ma, Wei He, Shuangxi Fang, Jingfeng Xiao, Ngoc Tu Nguyen, Hua Yang, Peipei Xu, Chengcheng Huang, Mengyao Zhao, Shuai Liu, Xinhui Lei, Ziyi Huang
Förderung
- National Key Research and Development Program of China
- National Natural Science Foundation of China
- Basic Research Program of Qinghai Province
Zitieren
APA
Ma, T., He, W., Fang, S., Xiao, J., Nguyen, N. T., Yang, H., Xu, P., Huang, C., Zhao, M., Liu, S., Lei, X., & Huang, Z. (2026). Temporal Memory Mechanisms and Biome‐Specific Drivers of Ecosystem Carbon Flux: Insights From Explainable Deep Learning Modeling. Global Change Biology, 32(1). https://doi.org/10.1111/gcb.70722
Vancouver
Ma T, He W, Fang S, Xiao J, Nguyen NT, Yang H, et al. Temporal Memory Mechanisms and Biome‐Specific Drivers of Ecosystem Carbon Flux: Insights From Explainable Deep Learning Modeling. Global Change Biology. 2026;32(1). doi:10.1111/gcb.70722
BibTeX
@article{Ma2026,
author = {Ma, Teng and He, Wei and Fang, Shuangxi and Xiao, Jingfeng and Nguyen, Ngoc Tu and Yang, Hua and Xu, Peipei and Huang, Chengcheng and Zhao, Mengyao and Liu, Shuai and Lei, Xinhui and Huang, Ziyi},
title = {Temporal Memory Mechanisms and Biome‐Specific Drivers of Ecosystem Carbon Flux: Insights From Explainable Deep Learning Modeling},
journal = {Global Change Biology},
year = {2026},
volume = {32},
number = {1},
doi = {10.1111/gcb.70722}
}
EUVIMED