摘要
为了解宁南山区典型植被恢复模式之一——人工林地土壤碳氮转化的特征以及二者的关系,运用PVC顶盖埋管法进行1 a的原位矿化培养实验,每隔2个月采样,研究柠条、山桃、山杏林地中土壤有机碳、可溶性有机碳和微生物量碳与土壤有机氮、无机氮以及净氨化速率、净硝化速率、净矿化速率和微生物固定速率在一年中的变化特征以及碳氮耦合关系.结果表明在原位培养过程中,61~120 d碳氮的变化最明显,主要受到土壤水分的影响;土壤有机碳和全氮极显著正相关,土壤微生物量碳氮、可溶性有机碳氮显著正相关;土壤有机碳转化速率显著影响净氨化速率、净硝化速率和MBN转化速率,且符合一元线性回归方程;柠条地培养一年后土壤微生物商(MBC/SOC)、MBN/SON显著升高,而且净硝化速率、净矿化速率显著大于山桃和山杏.
The study aimed to investigate the characteristics and relationship between soil carbon and nitrogen transformation of artificial forestlands,which is one type of vegetation restoration in the mountain area of Southern Ningxia. Soil samples were collected every two months in a year from three forestlands,and the characteristics of soil organic carbon,dissolved carbon,microbial biomass carbon,organic nitrogen,inorganic nitrogen,soil ammonification,nitrification and mineralization rates,microbial immobilization rates and coupling of soil carbon and nitrogen were studied by the in-situ closed-top PVC tube incubation methods. The results showed that:in the process of in-situ incubation,the most obvious changes of carbon and nitrogen were in 61-120 days which was mainly affected by soil moisture; There were significantly positive correlations between the soil organic carbon and the total nitrogen,microbial biomass carbon and microbial biomass nitrogen,dissolved carbon and dissolved nitrogen; Transformation rates of soil organic carbon had significant effects on the soil ammonification,nitrification and microbial immobilization rates. It can be well simulated by model of linear regression equation; Microbial quotient,MBN/SON were significantly increased in soil of Caragana korshinskii land. Net nitrification rates,net mineralization rates in Caragana korshinskii land were significantly higher than that in Prunus davidiana and Prunus mandshurica lands.
出处
《环境科学》
EI
CAS
CSCD
北大核心
2015年第9期3401-3410,共10页
Environmental Science
基金
国家自然科学基金项目(41101254)
关键词
碳氮变化
人工林地
原位培养
碳氮耦合
线性回归
change of soil carbon and nitrogen
artificial forestlands
in-situ incubation
coupling of carbon and nitrogen
linear regression