首页 | 本学科首页   官方微博 | 高级检索  
   检索      

基于BP神经网络模型的全球森林土壤异养硝化过程N2O排放通量估算
引用本文:刘炳彦,张军辉,耿世聪,陶晶晶,冯月.基于BP神经网络模型的全球森林土壤异养硝化过程N2O排放通量估算[J].生态学杂志,2022(2):209-217.
作者姓名:刘炳彦  张军辉  耿世聪  陶晶晶  冯月
作者单位:中国科学院沈阳应用生态研究所森林生态与管理重点实验室;中国科学院大学
基金项目:国家自然科学基金(41675150,31800413);辽宁省自然科学基金(2019-BS-262)资助。
摘    要:N2O作为重要的温室气体之一,对地球和人类都有很大的影响。为了深入探究对有机氮异养硝化作用及其产生N2O过程的影响机制,完善全球N2O通量估算模型,本研究采用Pearson相关性分析与广义可加模型(GAM)对全球135个样点有机氮异养硝化速率及其产生N2O速率的影响因子进行分析,然后将主要影响因子作为BP神经网络的输入层来模拟全球森林土壤有机氮异养硝化速率及其产生N2O速率的空间分布。结果显示,土壤pH和土壤C/N是影响有机氮异养硝化速率的主要因素,土壤C/N、土壤孔隙含水量(WFPS)以及土壤温度是影响有机氮异养硝化产生N2O速率的主要因素。全球森林土壤异养硝化速率平均为0.4241(0.0014~0.689)μg N·g-1·d-1,异养硝化产生N2O速率平均为0.2936(0.21~1.103)μg N2O·kg-1·d-1

关 键 词:异养硝化速率  异养硝化产生N2O速率  BP神经网络  环境因子  空间分布

Estimation of N2O emission flux from heterotrophic nitrification process in global forest soils based on BP neural network model
LIU Bing-yan,ZHANG Jun-hui,GENG Shi-cong,TAO Jing-jing,FENG Yue.Estimation of N2O emission flux from heterotrophic nitrification process in global forest soils based on BP neural network model[J].Chinese Journal of Ecology,2022(2):209-217.
Authors:LIU Bing-yan  ZHANG Jun-hui  GENG Shi-cong  TAO Jing-jing  FENG Yue
Institution:(Key Laboratory of Forest Ecology and Management,Institute of Applied Ecology,Chinese Academy of Sciences,Shenyang 110016,China;University of Chinese Academy of Sciences,Beijing 100049,China)
Abstract:As one of the main greenhouse gases,N2O has a significant impact on the Earth and human beings.To explore the influence mechanisms of heterotrophic nitrification of organic N and N2O emission process and improve global N2O flux estimation model,Pearson correlation analysis and generalized additive model(GAM)were used to analyze the influencing factors of heterotrophic nitrification rates of organic N and N2O emission rates in 135 sam-pling sites worldwide.The major influencing factors were included into BP neural network model to estimate the spa-tial distribution of heterotrophic nitrification rates of organic N and N2O emission rates in global forest soils.The results showed that soil pH and C/N were the main factors affecting heterotrophic nitrification rate of organic N.Soil C/N,water-filled pore space(WFPS),and temperature are the main factors affecting N2O emission rate during heterotrophic nitrification process.The average heterotrophic nitrification rate was 0.4241(0.0014-0.689)μg N·g-1·d-1,and the average N2O emission rate in the heterotrophic nitrification process was 0.2936(0.21-1.103)μg N2O·kg-1·d-1 in global forest soils.The heterotrophic nitrification rate and N2O emission rate in the hetero-trophic nitrification process were higher in forest soils around 50°N.In contrast,the two rates were lower in forest soils from 30°S to 30°N.Although heterotrophic nitrification rate was high,N2O emission rate in the heterotrophic nitrification process was low in forest soils in Northeast Asia.N2O emission flux from the heterotrophic nitrification of organic N was(1.4584±0.3791)Tg N·a-1in global forest soils,accounting for 26.04%of N2O emission from natural soils and 8.58%of global N2O emission.
Keywords:heterotrophic nitrification rate  N2O emission rate in heterotrophic nitrification process  BP neural network  environmental factor  spatial distribution
本文献已被 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号