Welcome Pratacultural Science,Today is
LI X, LIU X M, SUN B, JIANG J C, YU H Y, WU D D, DU X C, WANG H X, JIA J J, YANG H M. Machine learning–based assessment of grassland aboveground biomass in Gansu Province. Pratacultural Science, 2024, 41(2): 1-11. doi: 10.11829/j.issn.1001-0629.2022-0214
Citation: LI X, LIU X M, SUN B, JIANG J C, YU H Y, WU D D, DU X C, WANG H X, JIA J J, YANG H M. Machine learning–based assessment of grassland aboveground biomass in Gansu Province. Pratacultural Science, 2024, 41(2): 1-11. doi: 10.11829/j.issn.1001-0629.2022-0214

Machine learning–based assessment of grassland aboveground biomass in Gansu Province

  • To evaluate the aboveground biomass of grassland in Gansu Province, several grassland biomass inversion models based on machine learning were constructed by combining the ground sample data for grassland aboveground biomass from 2005 to 2018 in Gansu Province with vegetation index and meteorological factors. Comparison of prediction accuracies for the different models indicated that the Random Forest model had good applicability in the estimation of grassland aboveground biomass in Gansu Province. The main results were as follows: 1) Among the constructed machine learning models, the Rborist model demonstrated the highest accuracy, with an R2 of 0.758 based on screened variables. 2) Grassland aboveground biomass for Gansu Province estimated from 2000 to 2018 using Rborist (Random Forest, Rborist) model and 17 selected variables indicated an annual increase over the past 20 years, and the average grassland aboveground biomass ranged from 828.21 kg·ha−1 to 1 118.71 kg·ha−1. The average annual increase was 8.13 kg·ha−1 (P < 0.05). 3) For the grasslands in Gansu Province, 47.41% showed a recovery trend, 26% remained stable, and 26.59% showed a deterioration trend of varying degrees.
  • loading

Catalog

    Turn off MathJax
    Article Contents

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return