Frontiers of Data and Computing ›› 2023, Vol. 5 ›› Issue (2): 60-72.

CSTR: 32002.14.jfdc.CN10-1649/TP.2023.02.005

doi: 10.11871/jfdc.issn.2096-742X.2023.02.005

• Special Issue: AI for Science • Previous Articles     Next Articles

Estimation of Carbon Flux of a Temperate Forest Ecosystem Based on Next-Generation Information Technologies

GAO Tian1,2,3,4,5(),ZHU Jiaojun1,2,3,4,5,*(),ZHANG Jinxin1,2,3,4,SUN Yirong1,2,3,4,YU Fengyuan1,2,3,TENG Dexiong1,2,3,LU Deliang1,2,3,4,YU Lizhong1,2,3,4,WANG Zongguo6   

  1. 1. Qingyuan Forest CERN, National Observation and Research Station, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
    2. CAS Key Laboratory of Forest Ecology and Management, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
    3. CAS-CSI Joint Laboratory of Research and Development for Monitoring Forest Fluxes of Trace Gases and Isotope Elements, Shenyang, Liaoning 110016, China
    4. Liaoning Key Laboratory for Management of Non-commercial Forest, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
    5. Key Laboratory of Terrestrial Ecosystem Carbon Neutrality, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
    6. Computer Network Information Center, Chinese Academy of Sciences, Beijing 100083, China
  • Received:2023-03-05 Online:2023-04-20 Published:2023-04-24
  • Contact: ZHU Jiaojun E-mail:tiangao@iae.ac.cn;jiaojunzhu@iae.ac.cn

RichHTML

12

PDF

92

Abstract:

[Objective] The large uncertainty in forest ecosystem carbon flux estimation challenges the application of information technology in ecology. This study aims to change the traditional inefficient way of collecting and processing forest information and promotes the fusion of research on forest ecosystem carbon flux and the next-generation information technology, which explores paradigm shift in research and implementation of forest ecology and forestry sciences. [Methods] Using information technology such as the Internet of Things, near-ground remote sensing, artificial intelligence, and big data, the carbon flux estimation of a temperate forest ecosystem is demonstrated. [Results] The indices such as fractal dimension are introduced to propose a new parameter for quantifying terrain complexity. A new unmanned aerial vehicle Light Detection And Ranging dataset that fused leaf-off and leaf-on point cloud is introduced to extract the tree structural information for a total of 330,000 trees. The wind regime is numerically simulated, and the three wind patterns are identified. The frequency of CO2 concentration signals is analyzed, and forest ecosystem carbon flux is estimated by a machine learning framework with uncertainty reduced to 15.9%. [Limitations] The estimation of forest ecosystem carbon flux is still based on the classical eddy covariance method model. The estimation of forest ecosystem carbon flux at the regional scale has not been carried out yet. These require the systematic integration of boundary layer meteorology, forest ecology, remote sensing science and other disciplines and cutting-edge information technology. [Conclusions] The integrated framework of forest ecosystem carbon flux estimation with information technology has been developed to realize the shift of the paradigm of traditional studies on forest ecology and forestry sciences. The framework has successfully demonstrated its application in typical temperate forests in the Eastern Mountain of Liaoning Province.

Key words: carbon neutrality, forest ecosystems, complex terrain, carbon fluxes, machine learning algorithms, Internet of Things, Light Detection And Ranging (LiDAR)