| 摘要: |
| 森林土壤呼吸在陆地生态系统的碳平衡中发挥了重要作用，准确估算森林土壤呼吸量对于了
解陆地碳平衡的变化至关重要。这项研究以全球气候数据、全球森林土壤呼吸数据库为基础数据，通过
开发人工神经网络（ANN）模型建立由年平均气温（MAT）、年平均降水（MAP）、森林类型驱动的土壤
呼吸模型，预测全球森林土壤呼吸变化。模型估算的结果表明，从 1960 年到 2017 年，全球森林平均年
土壤呼吸量为 40.10±0.48 Pg C yr-1，全球森林土壤对全球土壤呼吸的贡献在 40.9% - 49.8% 之间。人工神
经网络模型预测的准确度达到 0.63，进一步改善了全球森林土壤呼吸模型预测的精度。 |
| 关键词: 全球气候数据 全球森林土壤呼吸数据库 森林生态系统 人工神经网络 |
| DOI： |
| 分类号: |
| 基金项目:广西自然科学“一个新的全球土壤呼吸模型”(20160831)。 |
|
| Study on Forest Soil Respiration Model Based on Global Databases |
|
Tian Qiuyan,Zhao Zhengyong,Yangqi,Zhuhangyong
|
|
Forestry College of Guangxi University
|
| Abstract: |
| Forest soil respiration is a critical process in the carbon cycling of terrestrial ecosystems. It
is very important to estimate the rate of annual forest soil respiration for understanding the change of carbon
balance. Based on global climate data and global forest soil respiration databases, this study developed an artificial
neural network (ANN) model driven by mean annual temperature (MAT), mean annual precipitation (MAP) and
forest types to predict the changes of global forest soil respiration. From 1960 to 2017, the average annual soil
respiration of global forest was 40.10±0.48 Pg C yr-1, which contributed between 40.9% and 49.8% to the global
soil respiration. The prediction accuracy of artificial neural network model reached 0.63 in this study, which
further improved the performance of the global forest soil respiration model. |
| Key words: global climate data global forest soil respiration database forest ecological system artificial neural network |
