Quantifying driving factors of vegetation carbon stocks of Moso bamboo forests using machine learning algorithm combined with structural equation model

Shi, Yongjun; Xu, Lin; Zhou, Yongqiang; Ji, Biyong; Zhou, Guomo; Fang, Huiyun; Yin, Jiayang; Deng, Xu

doi:10.1016/j.foreco.2018.07.035

Cited by 56 publications

(15 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The best‐fitting model was selected to calculate the mangrove AGC, then the spatial distribution map of the TC for Hainan Island mangroves was generated based on the best‐fitting mathematical model of AGC and BGC. Finally, we used structural equation modelling (SEM) (Angelini et al, 2016; Shi et al, 2018) to analyse the effects of various driving factors on carbon stocks (AGC, BGC and TC) for the Hainan Island mangroves (Figure 2).…”

Section: Methodsmentioning

confidence: 99%

“…The standardized direct and indirect total effects of the driving factors on mangrove carbon stocks were evaluated using SEM analysis. Initially, we tested the collinearity of all influential factors, then the initial model was established based on previous related studies (Angelini et al, 2016; Ouyang et al, 2017; Shi et al, 2018). Next, we parameterized the initial SEM using a standardized data set to test its goodness‐of‐fit using the χ 2 test, p ‐value, standardized root means square residual (SRMR), comparative fit index (CFI) and goodness‐of‐fit index (GFI).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Spatial patterns and driving factors of carbon stocks in mangrove forests on Hainan Island, China

Gou

Bai

Moreno‐Mateos

et al. 2022

Global Ecol. Biogeogr.

View full text Add to dashboard Cite

Aim Mangrove forests are important coastal wetlands for the blue carbon budget and play a significant role in mitigating global climate change. However, spatial patterns of carbon stocks in mangrove forests on an island scale have not been quantified owing to methodological limitations and lack of understanding of controlling factors. We took the entire Hainan Island as a case study and aimed to carry out a comprehensive investigation of the spatial patterns and driving factors of carbon stocks in mangrove forests. Location Southern China. Time period 2017–2020. Major taxa studied Mangrove forest. Methods The upscaling method combined with field surveys and Sentinel‐2 imagery analysis were used to compare different models for optimization of mangrove ecosystem carbon stock estimations. We also used structural equation modelling (SEM) to evaluate the factors driving the distributional patterns of mangrove carbon stocks on an island scale. Results The current total mangrove carbon stock of the entire Hainan Island was estimated to be 703,181 Mg C (with a mean density of 192 Mg C/ha), with the above‐ and below‐ground carbon stocks averaging at 44.7 and 147.3 Mg C/ha, respectively. The mangrove carbon storage in the north‐eastern region of the island was the highest and in the west region the lowest. Sediment nitrogen content and plant species diversity had the most positive driving effects on the distribution of total carbon stock for Hainan Island mangroves. Main conclusions The combination of field surveys and Sentinel‐2 imagery analysis can be applied to regional‐scale estimations of carbon stocks for mangrove forests. Spatial pattens of mangrove carbon stocks vary among locations on Hainan Island, and soil nutrient (especially nitrogen) availability is the dominant factor regulating carbon stock variations along the land‐to‐sea gradient. Our findings have significant implications for better understanding the distribution of mangrove carbon stocks and quantification of the global blue carbon budget.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Spatial patterns and driving factors of carbon stocks in mangrove forests on Hainan Island, China

Gou

Bai

Moreno‐Mateos

et al. 2022

Global Ecol. Biogeogr.

View full text Add to dashboard Cite

show abstract

“…The increased availability of large databases such as national forest inventories has led to the increasing use of random forests algorithms to identify the factors driving a particular variable of interest, often associated with forest productivity. For example, Shi et al (2018) quantified the driving factors of carbon stock in bamboo forests, and Thom and Keeton (2019) identified stand structure as the main driver of carbon storage disparities between different forest types in North America. However, when comparing the predictive performance of various techniques to predict aboveground biomass from field observations, Corona-Nuñez et al (2017) found that general linear models (GLM) and general additive models (GAM) outperformed more complex approaches, including random forests.…”

Section: Methodological Developments: Statistics Moving Into New Realmsmentioning

confidence: 99%

Modelling Forest Ecosystems: a crossroad between scales, techniques and applications

Blanco

Améztegui

Rodríguez

2020

Ecological Modelling

View full text Add to dashboard Cite

into different modelling approaches to try to understand forest systems, from continental, regional or landscape-scales to individual-base models. Similarly, forest models have been developed at all kind of complexity levels, from very complex, integrative models to simpler models focused on one specific ecophysiological process. Thus, continuous efforts by the forest modelling community have provided a wealth of research and expertise, but at the same time have generated a dispersion and lack of linkages between different modelling approaches and applications.Forest science is integrative by nature: understanding climate, soil and water processes is as important as understanding plant eco-physiology to understand how forests work. Therefore, forest science has evolved during the last centuries to understand how trees and forests grow, and which patterns and processes rule their development through time and space. Following the common scientific method, empirical reductionist approaches tried to isolate and study one ecological factor or variable at a time (Popper 1968), trying to reduce the complexity inherent to ecological processes. On the contrary, theoretical ecological models explore the ecological interactions among the components that exist in a forest, assembling the pieces of evidence generated through traditional field and lab work (Kimmins et al. 2010). In this special issue, we want to show some of the latest developments in forest modelling, understanding forests like working systems in which human intervention has played and is still playing a historical role, and will likely do so in the future.Following the sharp increase in data availability, modelling is now being applied in forest ecosystems at multiple scales, from continental and regional studies on forests and species distribution to leaf scales modelling plague and pathogen infestation in individual branches. Similarly, models are being applied to understand short-time processes such as nutrient transport in cells, to extremely long scales such as genetic drift in tree populations. From the application point of view, models are also being applied from purely ecological issues, such as coexistence and resource competition among tree species, to multidisciplinary socio-economic-ecological issues, such as design of tourist routes inside natural reserves, estimating timber flow to sawmills under climate change scenarios, or engineering concerns such as designing forest harvesting machinery with lower ecological footprint (in energy, rut formation, noise, etc.).When organizing this special issue, we (the guest editors) decided to carry out a quick overview of current trends in forest modelling. To do so, we used Clarivate's Web of Science to carry out a search in the top five journals with "Forestry" as their unique category

show abstract

“…Bamboo forests in China are mainly distributed in subtropical regions, such as Fujian, Jiangxi, Zhejiang, Hunan, and Sichuan Provinces [ 2 ]. Interest has grown in the ecological and socioeconomic value of bamboo forests, as bamboo forests are efficient carbon sinks that play a critical role in mitigating climate change [ 3 , 4 , 5 , 6 ] and environmental restoration [ 7 , 8 , 9 ]. For instance, Moso bamboo forests have an especially high carbon sequestration potential [ 10 , 11 ].…”

Section: Introductionmentioning

confidence: 99%

Identification of Bamboo Species Based on Extreme Gradient Boosting (XGBoost) Using Zhuhai-1 Orbita Hyperspectral Remote Sensing Imagery

Zhou

Zhao

et al. 2022

Sensors

View full text Add to dashboard Cite

Mapping the distribution of bamboo species is vital for the sustainable management of bamboo and for assessing its ecological and socioeconomic value. However, the spectral similarity between bamboo species makes this work extremely challenging through remote sensing technology. Existing related studies rarely integrate multiple feature variables and consider how to quantify the main factors affecting classification. Therefore, feature variables, such as spectra, topography, texture, and vegetation indices, were used to construct the XGBoost model to identify bamboo species using the Zhuhai-1 Orbita hyperspectral (OHS) imagery in the Southern Sichuan Bamboo Sea and its surrounding areas in Sichuan Province, China. The random forest and Spearman’s rank correlation analysis were used to sort the main variables that affect classification accuracy and minimize the effects of multicollinearity among variables. The main findings were: (1) The XGBoost model achieved accurate and reliable classification results. The XGBoost model had a higher overall accuracy (80.6%), kappa coefficient (0.708), and mean F1-score (0.805) than the spectral angle mapper (SAM) method; (2) The optimal feature variables that were important and uncorrelated for classification accuracy included the blue band (B1, 464–468 nm), near-infrared band (B27, 861–871 nm), green band (B5, 534–539 nm), elevation, texture feature mean, green band (B4, 517–523 nm), and red edge band (B17, 711–720 nm); and (3) the XGBoost model based on the optimal feature variable selection showed good adaptability to land classification and had better classification performance. Moreover, the mean F1-score indicated that the model could well balance the user’s and producer’s accuracy. Additionally, our study demonstrated that OHS imagery has great potential for land cover classification and that combining multiple features to enhance classification is an approach worth exploring. Our study provides a methodological reference for the application of OHS images for plant species identification.

show abstract

Quantifying driving factors of vegetation carbon stocks of Moso bamboo forests using machine learning algorithm combined with structural equation model

Cited by 56 publications

References 42 publications

Spatial patterns and driving factors of carbon stocks in mangrove forests on Hainan Island, China

Spatial patterns and driving factors of carbon stocks in mangrove forests on Hainan Island, China

Modelling Forest Ecosystems: a crossroad between scales, techniques and applications

Identification of Bamboo Species Based on Extreme Gradient Boosting (XGBoost) Using Zhuhai-1 Orbita Hyperspectral Remote Sensing Imagery

Contact Info

Product

Resources

About