{"title":"Study on Monitoring Methods for Net CO2 Exchange Rate of Individual Standing Tree","authors":"Z. H. Xu, Y. D. Zhao","doi":"10.1134/s1021443724603938","DOIUrl":null,"url":null,"abstract":"<h3 data-test=\"abstract-sub-heading\">Abstract</h3><p>The net CO<sub>2</sub> exchange rate, a pivotal plant physiology metric representing the carbon sequestration and release capacity of individual trees, is crucial for unraveling the underlying mechanisms of plant growth, carbon balance dynamics and environmental adaptability. This study focused on optimizing the static assimilation chamber to facilitate automated and long-term acquisition of the net CO<sub>2</sub> exchange rate in individual standing tree, with the entire <i>Radermachera sinica</i> as the research object. Concurrently, we monitored environmental factors and stem water content; Notably, a proprietary stem water content sensor was innovatively employed to capture the internal water dynamics within stem tissue; While the Internet of Things (IoT) technology was leveraged to establish a monitoring system for the net CO<sub>2</sub> exchange rate of individual standing tree. Initially, we conducted an exploratory analysis on the characteristics of the net CO<sub>2</sub> exchange rate by integrating stem water content under distinct watering conditions, and uncovered interplay between plant carbon sequestration capacity and internal water dynamics. Subsequently, machine learning models, including the support vector machine (SVM), backpropagation (BP) neural network, and random forest (RF) algorithms, were developed to predict the net CO<sub>2</sub> exchange rate. The results revealed that under normal watering conditions, the net CO<sub>2</sub> exchange rate exhibited diurnal U-shaped variations, generally transitioning from positive to negative in the morning and vice versa in the evening, with daily carbon sequestration remaining negative. Under drought stress and subsequent rehydration, the net CO<sub>2</sub> exchange rate demonstrated a gradual reduction, followed by disruption, and eventual recovery, resulting in the daily carbon sequestration transitioning from negative to positive, then back to negative. A significant positive correlation was observed between the net CO<sub>2</sub> exchange rate and stem water content change rate; In most cases, positive or zero stem water content change rate corresponded to carbon release, whereas negative change rate indicated carbon absorption. The RF model exhibited superior predictive accuracy, displaying strong agreement between predicted and actual values. Specifically, under normal watering conditions, the RF model achieved Root mean square error (RMSE), coefficient of determination (<i>R</i><sup>2</sup>), and mean absolute error (MAE) values of 1.356, 0.8576 and 0.9257%, respectively; Under drought stress and subsequent rehydration, corresponding values were 1.4567, 0.8436, and 1.0258%, respectively.</p>","PeriodicalId":21477,"journal":{"name":"Russian Journal of Plant Physiology","volume":"187 1","pages":""},"PeriodicalIF":1.1000,"publicationDate":"2024-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Russian Journal of Plant Physiology","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1134/s1021443724603938","RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"PLANT SCIENCES","Score":null,"Total":0}
引用次数: 0
Abstract
The net CO2 exchange rate, a pivotal plant physiology metric representing the carbon sequestration and release capacity of individual trees, is crucial for unraveling the underlying mechanisms of plant growth, carbon balance dynamics and environmental adaptability. This study focused on optimizing the static assimilation chamber to facilitate automated and long-term acquisition of the net CO2 exchange rate in individual standing tree, with the entire Radermachera sinica as the research object. Concurrently, we monitored environmental factors and stem water content; Notably, a proprietary stem water content sensor was innovatively employed to capture the internal water dynamics within stem tissue; While the Internet of Things (IoT) technology was leveraged to establish a monitoring system for the net CO2 exchange rate of individual standing tree. Initially, we conducted an exploratory analysis on the characteristics of the net CO2 exchange rate by integrating stem water content under distinct watering conditions, and uncovered interplay between plant carbon sequestration capacity and internal water dynamics. Subsequently, machine learning models, including the support vector machine (SVM), backpropagation (BP) neural network, and random forest (RF) algorithms, were developed to predict the net CO2 exchange rate. The results revealed that under normal watering conditions, the net CO2 exchange rate exhibited diurnal U-shaped variations, generally transitioning from positive to negative in the morning and vice versa in the evening, with daily carbon sequestration remaining negative. Under drought stress and subsequent rehydration, the net CO2 exchange rate demonstrated a gradual reduction, followed by disruption, and eventual recovery, resulting in the daily carbon sequestration transitioning from negative to positive, then back to negative. A significant positive correlation was observed between the net CO2 exchange rate and stem water content change rate; In most cases, positive or zero stem water content change rate corresponded to carbon release, whereas negative change rate indicated carbon absorption. The RF model exhibited superior predictive accuracy, displaying strong agreement between predicted and actual values. Specifically, under normal watering conditions, the RF model achieved Root mean square error (RMSE), coefficient of determination (R2), and mean absolute error (MAE) values of 1.356, 0.8576 and 0.9257%, respectively; Under drought stress and subsequent rehydration, corresponding values were 1.4567, 0.8436, and 1.0258%, respectively.
期刊介绍：
Russian Journal of Plant Physiology is a leading journal in phytophysiology. It embraces the full spectrum of plant physiology and brings together the related aspects of biophysics, biochemistry, cytology, anatomy, genetics, etc. The journal publishes experimental and theoretical articles, reviews, short communications, and descriptions of new methods. Some issues cover special problems of plant physiology, thus presenting collections of articles and providing information in rapidly growing fields. The editorial board is highly interested in publishing research from all countries and accepts manuscripts in English.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
