| 施凯泽,徐婷婷,罗 胤,袁启慧,李 泽,李 维,冷鸿天.云南松天然林生物量模型构建[J].林业调查规划,2026,51(3):1-9 |
| 云南松天然林生物量模型构建 |
| Establishment of Biomass Models for Natural Pinus yunnanensis Forests |
| |
| DOI： |
| 中文关键词: 云南松天然林 地上生物量模型 地下生物量模型 相容性生物量模型 |
| 英文关键词: natural Pinus yunnanensis forests aboveground biomass model underground biomass model compatible biomass model |
| 基金项目:云南省科技厅科技计划项目(202404CB090005). |
|
| 摘要点击次数: 545 |
| 全文下载次数: 63 |
| 中文摘要: |
| 为构建云南松天然林二元及一元地上生物量模型、地下生物量模型、不同维量生物量模型,选取云南省云南松天然林为研究对象,二元模型以全省作为一个统一建模单元,一元模型按照云南省内生长状况分为5个大区,并采用决定系数( R2)、估计值标准差(SSE)、总相对误差(TRE)、平均系统误差(ASE)、平均预估误差(MPE)和平均百分标准误差(MPSE)共6个变量作为模型评价指标。结果表明,地上生物量模型R2均大于0.9、MPE均小于1%;地下生物量模型除滇南、滇东南地区外,其余R2均大于0.9,二元模型MPE小于1%,一元模型MPE均小于10%;二元不同维量生物量模型以木材、树皮R2较高,均大于0.9,树枝大于0.8,树叶大于0.7,MPE 均小于1%;一元不同维量生物量模型木材R2均约0.9,其余均在0.6~0.9区间,MPE 均小于5%。各模型精度达到使用要求,可作为成果使用。 |
| 英文摘要: |
| This paper took the natural Pinus yunnanensis forests in Yunnan Province as the research object,and constructed the binary and unary aboveground biomass (AGB) models, underground biomass(UGB) models, and component biomass models of the natural Pinus yunnanensis forests. The binary model took the entire province as a unified modeling unit, while the unary model was divided into five major regions based on the growth conditions within Yunnan Province. Six variables, namely the coefficient of determination (R2), the standard deviation of estimate (SSE), the total relative error (TRE),the average systematic error (ASE), the mean prediction error (MPE), and the mean percentage standard error (MPSE), were used as model evaluation indicators. The results showed that for the AGB models, R2 values exceeded 0.9 and MPE values were below 1%. For the UGB models, R2 values were above 0.9 for all regions except southern and southeastern Yunnan, and MPE was below 1% for the binary models and below 10% for the unary models. For the binary component biomass models, R2 values were highest for wood and bark (both>0.9), followed by branches (>0.8) and foliage (>0.7), with MPE values all below 1%. For the unary component biomass models, R2 values were approximately 0.9 for wood, and ranged between 0.6 and 0.9 for other components, with MPE values below 5%. In summary, all models achieved satisfactory accuracy and can be applied for practical use. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
