Study On Adaptability Of 10 Alfalfa Varieties And Establishment Of NIRS Model In Agri-pastoral Ecozone Of North Shanxi Province

The test based on concluding 9 introduced varieties and 1 domestic variety(Gong Nong 4),and it was carried out for two consecutive years from 2016 to 2017.During this period,the phenological period,wintering rate,yield,and hay quality of alfalfa were measured.Applying the grey relational degree comprehensively analyzes its adaptability.The Near-Infrared Analysis Model(NIRS)was established by partial least-squares method based on the dates of an Antaris ? Fourier Transform Near Infrared Spectrometer.These are the basis and the simple and quick nutrition analysis methods for the development of the alfalfa industry in the Agri-pastoral Ecozone of northern Shanxi Province.The results showed:(1)when the plants grow the initial flowering,the early flowering stage.The longest growth period is Ganon No.4.The shortest growth period is WL366HQ and Suntol.We can determined that Ganno No.4 grows slower,and WL366HQ and Suntory grows faster in the region.The wintering rate of WL354HQ is the highest in 2016,which suggested that the breed has a stronger adaptability in this region(2)The test go on two years and harvest two stubble about two years,the hay yield of Sandy(3882.14 kg/hm2)was the largest in 2016,while the Gannon No.4(2830.95 kg/hm2)had the smallest hay yield.The hay yield of WL298HQ(5747.83 kg/hm2)was the largest,and the Gannon No.4(4201.68 kg/hm2)had also the smallest hay yield in 2017(3)The content of CP inside the plant of 10 breed anged from 15.53%to 20.26%.EE content was 3.53%to 5.19%.ADF content was 21.28%to 29.38%and NDF content was 27.52%to 38.40%.The highest CP content is WLHQ319 and the lowest is WLHQ353.The highest EE content is Suntory and the lowest is WL343HQ.The highest ADF content is WLHQ353 and the lowest is Suntory.The highest NDF content is WL354HQ and the lowest is WLHQ168.(4)The RFV of 10 alfalfa varieties was ranked:Queen>Suntory>WLHQ319>WLHQ353>WLHQ168>WL366HQ>WL298HQ>WL343HQ>WL354HQ>Gannon No.4.The RFV of all the breeds are more than 100,which indicates that all the breeds of alfalfa have a higher adaptability.(5)The comprehensive quality evaluated of the Grey Relationship Analysisthe with the adaptability,grass production and nutritional indicators 10 alfalfa varieties ranked:Suntory>WL366HQ>WL298HQ>queens>WL354HQ>WLHQ319>WL343HQ>WLHQ353>WLHQ168>Gannon No.4.(6)The NIRS model of nutrient composition of different breeds of alfalfa is established by Antaris ?Fourier transforming into near-infrared spectrometer,and the model performance index is 95.0.The correlation coefficient(Corr.Coeff.)of the CP model is 0.9975;the standard deviation of the calibration set(RMSEC)is 0.0843;the standard deviation of the prediction set(RMSEP)is 0.0873,and the correlation coefficient(Corr.Coeff.)of the calibration set is 0.9969.The correlation coefficient of the EE model is 0.9927;the calibration set standard deviation is 0.0342;the prediction set standard deviation is 0.0462,and the correction set correlation coefficient is 0.9899.The correlation coefficient of the ADF model is 0.9610;the calibration set standard deviation is 1.02;the prediction set standard deviation is 0.567,and the correction set correlation coefficient is 0.9855.The correlation coefficient of the NDF model is 0.9917;the calibration standard deviation is 0.587;the prediction set standard deviation is 0.513,and the correction set correlation coefficient is 0.9929.From the size of each coefficient,we can be judged that the quantitative analysis model has a high degree of fit.Through the verification of NIRS model,the established quantitative analysis model has high predictive ability,and it can be used for practical detection.