Simulation Of Soil Respiration Characteristics In Biochar Application Fields Based On DNDC,DayCent And Empirical Models

In the process of mitigating global warming,biochar,as a soil amendment for carbon sequestration and emission reduction,significantly affects the dynamic changes of soil respiration.As an important part of the global soil carbon cycle,soil respiration significantly affects the carbon balance of ecosystem.In recent years,soil respiration prediction by model method has become a research focus.In this study,farmland soil under the winter wheat-summer maize rotation system was the main research object.Based on the data of continuous soil respiration,soil moisture,soil temperature and other physical and chemical properties monitored in the field experiment for 3 years,the conventional fertilization treatment(T1)of non-biochar application and the conventional fertilization treatment+10 t/hm2 biochar(T2)of disposable biochar application were targeted.The DNDC model,DayCent model and empirical model were used to simulate and predict soil respiration,and the model could well reflect the seasonal variation characteristics of CO2 emission in the soil of applied and non-applied biochar fields.By changing the range of external simulation conditions,the sensitivity of the simulation of soil respiration in fields with and without application of biochar to changes in climate conditions,soil conditions and farmland management measures was revealed.The main results are as follows:(1)Soil respiration:The three models can simulate the change trend of soil respiration well,and the correlation coefficients are all greater than 0.620.The simulation error of DNDC model was the smallest.The error ranges of root mean square and normalized root mean square for each treatment were 2.823～3.020 and 0.579～0.650,respectively.The efficiency of DNDC model was 0.163～0.375,indicating that DNDC model was more suitable for simulating soil respiration change characteristics.By comparing the simulation results of T1 and T2 treatments,it can be seen that the three models all show that the simulation effect of applying biochar on soil respiration is better.Soil temperature:The error between simulated values and measured values of the empirical model was the smallest.The error ranges of root mean square and normalized root mean square for each treatment were 2.587～2.953 and 0.161～0.186,respectively.The efficiency of the model was between 0.880～0.908,indicating that the empirical model was more accurate in the simulation of soil temperature.The simulation results of T1 and T2 showed that the soil temperature simulation effect of applying biochar was better.Soil moisture:DNDC,DayCent and empirical models can only basically capture the variation trend of soil water.By contrast,the simulation error of DayCent model is small,and the error ranges of root mean square and normalized root mean square of the two treatments are 0.127～0.129 and 0.535～0.547,respectively.The results showed that DayCent model was more suitable to simulate soil water and soil water when biochar was applied.(2)Model sensitivity analysis showed that:in terms of climate factors,soil respiration simulated by the three models showed a certain positive correlation with temperature and rainfall.Air temperature is the most sensitive parameter affecting soil respiration,followed by rainfall.The sensitivity of the model to soil respiration was DNDC>empirical model>DayCent under the change of climate factors.In terms of soil factors,the simulated soil respiration by DNDC and DayCent models showed a positive correlation with soil bulk density and a negative correlation with soil pH value,but the two models had little effect on soil respiration as a whole,indicating that soil respiration was almost not sensitive to it.In terms of farmland management measures,soil respiration increased with the increase of fertilizer application and irrigation amount,and the effect of fertilization amount was more significant.Compared with DNDC model,DayCent model simulated soil respiration was more sensitive to farmland management measures.Compared with no biochar application,all three models showed that soil respiration with biochar application was more sensitive to the influencing factors.(3)Under historical climate conditions,soil respiration in farmland simulated by DNDC,DayCent and empirical models showed an upward trend,but with certain fluctuation characteristics.The predicted value of soil respiration showed a consistent trend under the conditions of biochar application and no biochar application.In the two treatments,the predicted value of Daycent model showed a trend of first increasedecrease-increase-then decrease-increase,and the predicted value of DNDC model showed a trend of first increase-decrease-then increase,while the predicted value of empirical model showed a trend of gradual increase.However,the predicted values of soil respiration under biochar application were all higher than those without biochar application.Soil respiration simulation values of applied and non-applied biochar from high to low:DayCent model>DNDC model>Empirical model.The predicted values and increases of simulated soil respiration were T1>T2.According to the coefficient of variation,in the next 1-5 years,the coefficient of variation of soil respiration simulated by the model changed from large to small:DNDC model>DayCent model>empirical model;in the next 6-10 years,the coefficient of variation changed from large to small:DayCent model>DNDC model>empirical model.It is predicted that the fluctuation of soil respiration simulated by DNDC model will be smaller and smaller in the future,while the variation degree of soil respiration simulated by DayCent model will be larger and larger in the future,and the fluctuation of soil respiration simulated by empirical model will be small and irregular.In conclusion,compared with non-applied biochar treatment,the three models have smaller error and higher model efficiency in simulating application of biochar.DNDC model is more suitable for simulating farmland soil respiration after biochar application,DayCent model is more suitable for simulating soil moisture,and empirical model is more suitable for simulating soil temperature,and the characteristics of soil respiration will show an increasing trend in the future.