Quantifying Greenhouse Gas Emissions And The Mitigation Potential In Agriculture With Literature Statistics Method And Case Study

Reliable assessment of greenhouse gas emission of different agricultural activities and analysis of the technical potential of emission reduction are the keys to achieve the rational management of agricultural resources as well as climate change mitigate in agriculture.By multi-methods(in-situ monitoring,linear model,process-based model and carbon footprinting),multi-scales(site scale,regional scale and national scale)and multi-sector(farm production,food consumption)analysis,the study quantified agricultural greenhouse gas emissions for China in order to provide sound scientific base for persuing the mitigation from farm production to food consumption.1.Based on a database with 823 field measurements of N2O emission from 104 studies across China's croplands was compiled.Fertilizer induced N2O emission factors(N2O_EF)were established by three methods(M1:average values of individual N2O_EF;M2:a linear model with intercept;M3:a linear model without intercept).The study showed that N2O_EF varied with regions,crops,and fertilizer types.The N2O_EF of maize was relatively high from the Huang Huai Hai area.The N2O_EF of rice was lower from the region of Yangtze River.The N2O_EF of wheat was higher from the Yangtze River.Through 30%data validation by three methods and IPCC default EF,modeled values by M2 and IPCC were significantly different from the measured values.The uncertainties caused by methods were 10.2%,12.5%and 9.4%,respectively.Based on the best N2O_EF,the total amount of N2O emission caused by mineral nitrogen application for China's staple crops was 36.2 Gg N of rice,33.0 Gg N of maize,and 23.2 Gg N of wheat in 2015.2.Data associated with environmental attributes and management practices were used to fit empirical regression models for both N2O emissions and crop yield.The rate of N2O emissions was explained by fertilizer rate,crop type,temperature,soil clay content,and the interaction between N rate and fertilizer type.This model explained 48%of the dataset variance.We observed that with all other variables fixed,N2O emissions increased from rice,to legume and then other upland crops,and was higher for mineral than organic fertilizer.The variables significant in the yield model were rate of fertilizer,temperature,crop type,and soil clay content,which explained 35%of the dataset variance.We used both empirical models to explore the optimum N rates(N rate at which the lowest N2O emissions per ton of yield)for combinations of crop and fertilizer types.Optimum values of N application rate varied between 100 and 190 kg N per hectare for mineral and organic fertilizer,respectively,and between crop types.As for case studies,we used the model to estimate emissions from staple crops in China.And the total annual emissions from rice,wheat,and maize in China were to be around 27.3,29.2,and 37.2 Gg N2O-N.3.With comparative analysis on the capability of seasonal cumulative N2O fluxes simulated with 425 field measurements of N2O emission from 67 studies across China's croplands with four different models.The models adopted were 1)daily time-step version of CENTURY(DAYCENT),2)DeNitrification-Decomposition model(DNDC),3)Linear regression model by our previous study(LRM),and 4)IPCC emission factors.The DAYCENT and DNDC models were reasonably accurate for crop yield predictions versus measurements with R2 values of 0.60 and 0.66;and the DNDC showed significantly underestimation according to bias and t-test analysis.For seasonal cumulative N2O emission prediction,the correlation of modelled and measured N2O emissions had the R2 of 0.14,0.14,0.23 and 0.15 for DAYCENT,DNDC,LRM,and IPCC,respectively.The DNDC model significantly underested with 0.52 kg N2O-N ha-1.The modelled daily N2O emission against observations from the tested fields indicated that the DAYCENT and DNDC models had a good performance on temporal patterns though they could not capture the emission peaks perfectly.LRM performed well on N2O emission prediction for paddy rice,while DAYCENT for wheat and IPCC for maize.Besides,all models simulated N2O fluxes well for soybean,but not well for cotton or fallow.Moreover,the DAYCENT and LRM performed well under different fertilizer management(no fertilizer,mineral fertilizer,and organic fertilizer),while DNDC and IPCC significantly underestimated the emissions under organic fertilizer applied.4.CFs of 26 kinds of crop and 6 kinds of livestock production were quantified using national statistical data.For grain crop production,CFs ranged from 0.45 kg CO2-eq kg-1(for maize)to 1.46 kg CO2-eq kg-1(for late rice).For vegetable crops,CFs ranged from 0.06 kg CO2-eq kg-1(for radish)to 0.25 kg CO2-eq kg-1(for eggplant).For fruits,CFs ranged from 0.32 kg CO2-eq kg-1(for tangerine)to 0.70 kg CO2-eq kg-1(for apple).And for industrial crops,CFs ranged from 6.80 kg CO2-eq kg-1 for long-staple cotton to 0.08 kg CO2-eq kg-1 for beetroot.The highest CFs for livestock production were observed in mutton(8.92 kg CO2-eq kg-1),while pork and chicken production had lower CFs of 6.18 and 5.26 kg CO2-eq kg-1,respectively.Methane emissions from rice paddy and emissions from fertilizer application were the largest contributors accounting about 36?93%of total CFs.Whereas GHG emissions from forage,enteric fennentation and manure treatment accounted for more than 96%for CFs of livestock and poultry production.Significant differences between CFs were found across different management patterns and farm scales.GHGs emissions from supply side food production were currently 912.5 kg CO2-eq capita-1 yr-1,which was considerably higher than that estimated from consumption being 379.6 kg CO2-eq capita-1 yr-1.5.Using floating static chamber-chromatography method,the greenhouse gases fluxes(CO2,CH4 and N2O)at the water-air interface were daily and monthly monitored from four wetlands,including natural wetland(NW),enclosure wetland for intensive aquaculture(EWIA),constructed wetland for intensive aquaculture(CWIA),and for intensive aquaculture(CWEA).As a result,annual fluxes from three artificial wetlands averaged 0.81,1.06 and 2.43 kg N2O-N ha-1,23.83,457.08 and 1360.27 kg CH4-C ha-1,1321.32,1877.04 and 2246.79 kg CO2-C ha-1 for CWEA,CWIA and EWIA,respectively.Moreover,the global warming potential(GWP)of artificial intensive aquaculture increased about ten times than the nature wetland system with the values of 60.03 and 6.08 t CO2-eq ha-1 for EWIA and NW,respectively.Additionally,the yield-scale GHG emission factor ranged from 1.21 to 5.30 kg CO2-eq kg-1;and the highest N2O and CH4 emission factor of aquaculture production were both from EWIA with the values of 0.34 g kg-1 and 0.19 g kg-1,respectively.In addition,the study also showed that GHG fluxes were positive associated with feed amount,water parameters.The results suggested that optimizing human management and improving environmental quality could achieve much mitigation regardless as an aquaculture industry or constructed wetland.6.Based on the scenarios analysis,we analysied the N2O emissions for staple crops at the year of 2020 would mainly come from corn cultivation,accounting for over 44%.In 2020,the emissions under 100%mineral fertilizer applied with the low PFPN standard was the highest(158.3 Gg N);70%organic fertilizer+30%mineral fertilizer applied with the high PFPN was lowest(98.5 Gg N).The mitigating potential could up to be 38%.Under the same PFPN level,optimal fertilization had a mitigating potential of 14%.For the same fertilization mode,the mitigating potential caused by PFPN improvement can reach to be 27%.By investigating the mitigating potential from the food consumption,we concluded that dining-out contributed a higher GHG emission than dining at home.And the Hunan cuisine had the highest GHG emssion with the value of 3.68±0.55 kg CO2-eq capita,1 meal-1,and the Guangdong cuisine showed the lowest,2.44±0.83 kg CO2-eq capita-1 meal--1.The consumption amount of livestock food greatly raised,resulting in the rapid increasing of GHG emissions from agriculture.With field monitoring,database creation,empirical regression model development and models comparison,this study updated and improved the emission factors of GHGs for China's agriculture,and improved the approach for evaluating and predicting of GHG emissions for the cropland.The mitigating technical approaches were revealed from food production to consumption,and the reduction potentials were evaluated,and proposed the methodologies and technologies for pursuing the high efficiency,low carbon and scientific production for China's agricultural development.