高级检索+
代劲松, 季碧勇, 徐达, 陶吉兴. 多尺度相容的林木年生长模型组研究[J]. 浙江林业科技, 2023, 43(5): 51-58. DOI: 10.3969/j.issn.1001-3776.2023.05.007
引用本文: 代劲松, 季碧勇, 徐达, 陶吉兴. 多尺度相容的林木年生长模型组研究[J]. 浙江林业科技, 2023, 43(5): 51-58. DOI: 10.3969/j.issn.1001-3776.2023.05.007
DAI Jinsong, JI Biyong, XU Da, TAO Jixing. Research on Model Groups for Annual Standing Volume Growth in Zhejiang[J]. Journal of Zhejiang Forestry Science and Technology, 2023, 43(5): 51-58. DOI: 10.3969/j.issn.1001-3776.2023.05.007
Citation: DAI Jinsong, JI Biyong, XU Da, TAO Jixing. Research on Model Groups for Annual Standing Volume Growth in Zhejiang[J]. Journal of Zhejiang Forestry Science and Technology, 2023, 43(5): 51-58. DOI: 10.3969/j.issn.1001-3776.2023.05.007

多尺度相容的林木年生长模型组研究

Research on Model Groups for Annual Standing Volume Growth in Zhejiang

  • 摘要: 本文以浙江省2018年和2019年森林资源连续清查的4252个固定样地数据为建模样本,其中,随机预留10%的样本用于对蓄积、胸径、树高模型进行独立验证,其余数据用于建模,分为林分和散生四旁树2类建模单元,对4个不同的树种组(松木、杉木、硬阔和软阔)建立蓄积、胸径、树高的生长率模型,并结合转换系数建立省、市、县多尺度相容的林木年生长模型组——蓄积生长模型、树高生长模型、胸径生长模型。结果表明,林分蓄积生长模型的R2均在0.63以上,rRMSE小于0.2,胸径生长模型的R2均在0.98以上,rRMSE小于0.1,树高生长模型的R2均在0.61以上,rRMSE小于0.2,模型组拟合效果良好;模型组应用于森林资源小班数据的年蓄积生长率验证效果较好(R2为0.8109,rRMSE小于0.2)。以上研究结果表明,以森林资源连续清查固定样地数据建立林木年生长模型组,能够为森林资源小班数据的模型更新提供可靠的技术支撑。

     

    Abstract: Data of fixed plots of continuous inventory for forest resources in Zhejiang province in 2018 and 2019 were collected for data modeling, among them, 10% of data were prepared for independent verification of standing volume, DBH and height. The models for standing volume growth, DBH and height were established with two modeling units and four forest groups (pine, Chinese fir, hard broad-leaved and soft broad-leaved), and that of multi-scale compatible model groups at the provincial, cities and counties level were established with conversion coefficient. The results indicated that the values of R2 for the volume growth models were greater than 0.63, with rRMSE smaller than 0.2, the values of R2 for DBH growth models were greater than 0.98, with rRMSE smaller than 0.1 , and the values of R2 for the height growth models were greater than 0.61 with rRMSE smaller than 0.2. The fitting effects of the model group were significant. The model groups were applied to the forest resources subcompartment data, the annual volume growth rates verification effect was good (R2 was 0.8109, rRMSE smaller than 0.2).

     

/

返回文章
返回