Modeling Terrestrial Net Ecosystem Exchange Based on Deep Learning in China DOI Creative Commons
Zeqiang Chen, Lei Wu, Nengcheng Chen

и другие.

Remote Sensing, Год журнала: 2024, Номер 17(1), С. 92 - 92

Опубликована: Дек. 30, 2024

In estimating the global carbon cycle, net ecosystem exchange (NEE) is crucial. The understanding of mechanism interaction between NEE and various environmental factors ecosystems has been very limited, interactions are intricate complex, which leads to difficulties in accurately NEE. this study, we propose A-DMLP (attention-deep multilayer perceptron)-deep learning model for simulation as well an interpretability study using SHapley Additive exPlanations (SHAP) model. attention was introduced into deep perceptual machine, important information original input data extracted mechanism. Good results were obtained on nine eddy covariance sites China. also compared with random forest, long short-term memory, neural network, convolutional networks (1D) models distinguish it from previous shallow machine estimate NEE, show that have great potential modeling. SHAP method used investigate relationship features simulated enhance normalized difference vegetation index, enhanced leaf area index play a dominant role at most sites. This provides new ideas methods analyzing by introducing interpretable These advancements crucial achieving reduction targets.

Язык: Английский

Regulation Mechanisms of CO2 Fluxes in Subtropical Mountain Peatlands Based on Long‐Term In Situ Observations at the Dajiuhu Peatland DOI

Shiyu Yang,

Jiwen Ge,

Xiangnan Xu

и другие.

Journal of Geophysical Research Biogeosciences, Год журнала: 2025, Номер 130(1)

Опубликована: Янв. 1, 2025

Abstract The Dajiuhu peatland in Shennongjia, China, is a highly representative subalpine peatland, emblematic of subtropical mountainous peatlands. Due to the lack long‐term situ continuous observations and in‐depth studies on CO 2 absorption emission patterns, regulation mechanisms flux peatlands remain unclear. Since July 2015, we have conducted over five years fluxes major environmental factors ecosystem. We calculated annual average net ecosystem carbon exchange (NEE) decomposed NEE into gross primary productivity (GPP) respiration (Reco), thus examining ecosystem's separately. results indicate that from 2016 2020 was −283.6 g C m −2 yr −1 , reflecting strong sink. Our study indicates peatland. Temperature most direct factor affecting emission, serving as important driver short time scales. Precipitation only affects but has significant impact NEE, being key maintaining peatland's sink function. Variations precipitation also led differences between years. illustrate an role sub‐tropical mountain mitigating greenhouse effect moisture conditions crucial for protecting its ecological functions.

Язык: Английский

Процитировано

0
The "Fe-S Wheel": A New Perspective on Methylmercury Production Dynamics in Subalpine Peatlands DOI
Mingyuan Gao,

Yongqiang Ning,

Chutong Liu

и другие.

Journal of Hazardous Materials, Год журнала: 2025, Номер 493, С. 138401 - 138401

Опубликована: Апрель 25, 2025

Язык: Английский

Процитировано

0
Characteristics of Carbon Fluxes and Their Environmental Control in Chenhu Wetland, China DOI Open Access
Ya Zhang, Li Liu, Hua Luo

и другие.

Water, Год журнала: 2024, Номер 16(22), С. 3169 - 3169

Опубликована: Ноя. 6, 2024

Carbon dioxide (CO2) flux measurements were conducted throughout the year 2022 utilizing eddy covariance technique in this study to investigate characteristics of carbon fluxes and their influencing factors Chenhu wetland, a representative subtropical lake-marsh wetland located middle reaches Yangtze River China. The results revealed that mean daily variation CO2 during growing season exhibited U-shaped pattern, with ranging from −12.42 4.28 μmolCO2·m−2·s−1. ecosystem functions as sink season, subsequently transitioning source non-growing evidenced by observations made 2022. annual absorption was quantified at 21.20 molCO2·m−2, figure is lower than those documented for specific lake wetlands, such Dongting Lake Poyang Lake. However, measurement aligns closely average net exchange (NEE) reported wetlands across Asia. correlation between daytime photosynthetically active radiation (PAR) can be accurately represented through rectangular hyperbola equations season. Vapor pressure deficit (VPD) acts constraining factor NEE, an optimal range established 0.5 1.5 kPa. Furthermore, air temperature (Ta), relative humidity (RH), vapor difference are recognized principal determinants affecting NEE nocturnal period. association Ta conforms van’t Hoff model, suggesting increases response elevated timeframe.

Язык: Английский

Процитировано

0
Niche of woody plant populations in Picea purpurea community in the upper Taohe River DOI Creative Commons
Yang Zhao, Rui Qi, Bo Li

и другие.

Ecological Indicators, Год журнала: 2024, Номер 166, С. 112557 - 112557

Опубликована: Авг. 31, 2024

The niche of plant populations is affected by the environment, species characteristics and anthropogenic disturbance. Picea purpurea, as a major constructive in northeastern Qinghai-Tibetan Plateau, had been severely damaged. Although national project for protection natural forests has promoted recovery its community, structure, survival status, development trend, factors affecting it are still unclear. We selected P. purpurea communities Zecha, Dayugou Yeliguan forest zones at different altitudes disturbance levels. analyzed woody plants relationship between altitude, absolute advantages species, population dominance width tree layer show decreasing trend with there overlap all pairs. In shrub layer, dominant mostly Caprifoliaceae Rosaceae besides seedlings, proportion pairs YLG>ZC>DYG, appeared divergence convergence addition, seedlings most this index was highest. mean values layers were YLG>DYG>ZC. greater than that same zones, indicating more stable providing evidence maintains community stability. Regression analyses showed minimum temperature main factor dominance, niche, richness population. Disturbance did not significantly affect seedling populations, but differentiation species. conclude mainly influenced altitude Altitude-induced climatic variation fundamentally determines distinct composition niche. Anthropogenic altered habitat heterogeneity enriched structure. Furthermore, towards expansion. Understanding structure on environmental gradients enriches our ideas implementing vegetation restoration sustainable management subalpine context climate change, conducive to improving conservation capacity or type.

Язык: Английский

Процитировано

0
Modeling Terrestrial Net Ecosystem Exchange Based on Deep Learning in China DOI Creative Commons
Zeqiang Chen, Lei Wu, Nengcheng Chen

и другие.

Remote Sensing, Год журнала: 2024, Номер 17(1), С. 92 - 92

Опубликована: Дек. 30, 2024

In estimating the global carbon cycle, net ecosystem exchange (NEE) is crucial. The understanding of mechanism interaction between NEE and various environmental factors ecosystems has been very limited, interactions are intricate complex, which leads to difficulties in accurately NEE. this study, we propose A-DMLP (attention-deep multilayer perceptron)-deep learning model for simulation as well an interpretability study using SHapley Additive exPlanations (SHAP) model. attention was introduced into deep perceptual machine, important information original input data extracted mechanism. Good results were obtained on nine eddy covariance sites China. also compared with random forest, long short-term memory, neural network, convolutional networks (1D) models distinguish it from previous shallow machine estimate NEE, show that have great potential modeling. SHAP method used investigate relationship features simulated enhance normalized difference vegetation index, enhanced leaf area index play a dominant role at most sites. This provides new ideas methods analyzing by introducing interpretable These advancements crucial achieving reduction targets.

Язык: Английский

Процитировано

0