Correction: Yu, Q., et al. Narrowband Bio-Indicator Monitoring of Temperate Forest Carbon Fluxes in Northeastern China. Remote Sens. 2014, 6, 8986-9013

Abstract. In many forest ecosystems, nitrogen (N) deposition enhances plant uptake of carbon dioxide, thus reducing climate warming from fossil fuel emissions. Therefore, accurately modeling how forest carbon (C) sequestration responds to N deposition is critical for understanding how future changes in N availability will influence climate. Here, we use observations of forest C response to N inputs along N deposition gradients and at five temperate forest sites with fertilization experiments to test and improve a global biogeochemical model (CLM-CN 4.0). We show that the CLM-CN plant C growth response to N deposition was smaller than observed and the modeled response to N fertilization was larger than observed. A set of modifications to the CLM-CN improved the correspondence between model predictions and observational data (1) by increasing the aboveground C storage in response to historical N deposition (1850–2004) from 14 to 34 kg C per additional kg N added through deposition and (2) by decreasing the aboveground net primary productivity response to N fertilization experiments from 91 to 57 g C m−2 yr−1. Modeled growth response to N deposition was most sensitive to altering the processes that control plant N uptake and the pathways of N loss. The response to N deposition also increased with a more closed N cycle (reduced N fixation and N gas loss) and decreased when prioritizing microbial over plant uptake of soil inorganic N. The net effect of all the modifications to the CLM-CN resulted in greater retention of N deposition and a greater role of synergy between N deposition and rising atmospheric CO2 as a mechanism governing increases in temperate forest primary production over the 20th century. Overall, testing models with both the response to gradual increases in N inputs over decades (N deposition) and N pulse additions of N over multiple years (N fertilization) allows for greater understanding of the mechanisms governing C–N coupling.

Download Full-text

Greenup dates change across a temperate forest-grassland ecotone in northeastern China driven by spring temperature and tree cover

Agricultural and Forest Meteorology ◽

10.1016/j.agrformet.2021.108780 ◽

2022 ◽

Vol 314 ◽

pp. 108780

Author(s):

Chao Ding ◽

Wenjiang Huang ◽

Shuang Zhao ◽

Biyao Zhang ◽

Yao Li ◽

...

Keyword(s):

Temperate Forest ◽

Northeastern China ◽

Tree Cover ◽

Spring Temperature

Download Full-text

Estimating Forest Carbon Fluxes Using Machine Learning Techniques Based on Eddy Covariance Measurements

Sustainability ◽

10.3390/su10010203 ◽

2018 ◽

Vol 10 (1) ◽

pp. 203 ◽

Cited By ~ 8

Author(s):

Xianming Dou ◽

Yongguo Yang ◽

Jinhui Luo

Keyword(s):

Neural Network ◽

Machine Learning ◽

Eddy Covariance ◽

Carbon Flux ◽

Carbon Fluxes ◽

Forest Carbon ◽

Machine Learning Techniques ◽

Support Vector ◽

Inference System ◽

Learning Techniques

Approximating the complex nonlinear relationships that dominate the exchange of carbon dioxide fluxes between the biosphere and atmosphere is fundamentally important for addressing the issue of climate change. The progress of machine learning techniques has offered a number of useful tools for the scientific community aiming to gain new insights into the temporal and spatial variation of different carbon fluxes in terrestrial ecosystems. In this study, adaptive neuro-fuzzy inference system (ANFIS) and generalized regression neural network (GRNN) models were developed to predict the daily carbon fluxes in three boreal forest ecosystems based on eddy covariance (EC) measurements. Moreover, a comparison was made between the modeled values derived from these models and those of traditional artificial neural network (ANN) and support vector machine (SVM) models. These models were also compared with multiple linear regression (MLR). Several statistical indicators, including coefficient of determination (R2), Nash-Sutcliffe efficiency (NSE), bias error (Bias) and root mean square error (RMSE) were utilized to evaluate the performance of the applied models. The results showed that the developed machine learning models were able to account for the most variance in the carbon fluxes at both daily and hourly time scales in the three stands and they consistently and substantially outperformed the MLR model for both daily and hourly carbon flux estimates. It was demonstrated that the ANFIS and ANN models provided similar estimates in the testing period with an approximate value of R2 = 0.93, NSE = 0.91, Bias = 0.11 g C m−2 day−1 and RMSE = 1.04 g C m−2 day−1 for daily gross primary productivity, 0.94, 0.82, 0.24 g C m−2 day−1 and 0.72 g C m−2 day−1 for daily ecosystem respiration, and 0.79, 0.75, 0.14 g C m−2 day−1 and 0.89 g C m−2 day−1 for daily net ecosystem exchange, and slightly outperformed the GRNN and SVM models. In practical terms, however, the newly developed models (ANFIS and GRNN) are more robust and flexible, and have less parameters needed for selection and optimization in comparison with traditional ANN and SVM models. Consequently, they can be used as valuable tools to estimate forest carbon fluxes and fill the missing carbon flux data during the long-term EC measurements.

Download Full-text