Yong Li1, Louise Barton, Deli Chen. 1. Key Laboratory of Agro-ecological Processes in Subtropical Region, Institute of Subtropical Agriculture, The Chinese Academy of Sciences, Hunan, China. yli@isa.ac.cn
Abstract
BACKGROUND: Besides land management and soil properties, nitrous oxide (N(2)O) emissions from the soil may be responsive to climatic variation. In this study the Water and Nitrogen Management Model (WNMM) was calibrated and validated to simulate N(2)O emissions from a rain-fed and wheat-cropped system on a sandy duplex soil at Cunderdin, Western Australia, from May 2005 to May 2007, then it was deployed to simulate N(2)O emissions for seven scenarios of fertiliser N application under various climatic conditions (1970-2006). RESULTS: The WNMM satisfactorily simulated crop growth, soil water content and mineral N contents of the surface soil (0-10 cm), soil temperatures at depths and N(2)O emissions from the soil compared with field observations in two fertiliser treatments during calibration and validation. About 70% of total N(2)O emissions were estimated as nitrification-induced. The scenario analysis indicated that the WNMM-simulated annual N(2)O emissions for this rain-fed and wheat-cropped system were significantly correlated with annual average minimum air temperature (r = 0.21), annual pan evaporation (r = 0.20) and fertiliser N application rate (r = 0.80). Both annual rainfall and wheat yield had weak and negative correlations with annual N(2)O emissions. Multiple linear regression models for estimating annual N(2)O emissions were developed to account for the impacts of climatic variation (including temperature and rainfall), fertiliser N application and crop yield for this rain-fed and wheat-cropped system in Western Australia, which explained 64-74% of yearly variations of the WNMM-estimated annual N(2) O emissions. CONCLUSION: The WNMM was tested and capable of simulating N(2) O emissions from the rain-fed and wheat-cropped system. The inclusion of climatic variables as predictors in multiple linear regression models improved their accuracy in predicting inter-annual N(2)O emissions.
BACKGROUND: Besides land management and soil properties, nitrous oxide (N(2)O) emissions from the soil may be responsive to climatic variation. In this study the Water and Nitrogen Management Model (WNMM) was calibrated and validated to simulate N(2)O emissions from a rain-fed and wheat-cropped system on a sandy duplex soil at Cunderdin, Western Australia, from May 2005 to May 2007, then it was deployed to simulate N(2)O emissions for seven scenarios of fertiliser N application under various climatic conditions (1970-2006). RESULTS: The WNMM satisfactorily simulated crop growth, soil water content and mineral N contents of the surface soil (0-10 cm), soil temperatures at depths and N(2)O emissions from the soil compared with field observations in two fertiliser treatments during calibration and validation. About 70% of total N(2)O emissions were estimated as nitrification-induced. The scenario analysis indicated that the WNMM-simulated annual N(2)O emissions for this rain-fed and wheat-cropped system were significantly correlated with annual average minimum air temperature (r = 0.21), annual pan evaporation (r = 0.20) and fertiliser N application rate (r = 0.80). Both annual rainfall and wheat yield had weak and negative correlations with annual N(2)O emissions. Multiple linear regression models for estimating annual N(2)O emissions were developed to account for the impacts of climatic variation (including temperature and rainfall), fertiliser N application and crop yield for this rain-fed and wheat-cropped system in Western Australia, which explained 64-74% of yearly variations of the WNMM-estimated annual N(2) O emissions. CONCLUSION: The WNMM was tested and capable of simulating N(2) O emissions from the rain-fed and wheat-cropped system. The inclusion of climatic variables as predictors in multiple linear regression models improved their accuracy in predicting inter-annual N(2)O emissions.
Authors: L Barton; B Wolf; D Rowlings; C Scheer; R Kiese; P Grace; K Stefanova; K Butterbach-Bahl Journal: Sci Rep Date: 2015-11-02 Impact factor: 4.379