畜牧兽医学报 ›› 2015, Vol. 46 ›› Issue (9): 1574-1583.doi: 10.11843/j.issn.0366-6964.2015.09.012

• 动物营养 • 上一篇    下一篇

奶牛胃肠道甲烷排放模型估算精度的评估分析

谢天宇1,2,王敏1*,王荣1,4,颜志成1,5,石惠宇2,高民3*,谭支良1   

  1. (1.中国科学院亚热带农业生态研究所,长沙 410128;2.内蒙古农业大学动物科学学院,呼和浩特 010018;3.内蒙古农牧业科学院动物营养与饲料研究所,呼和浩特 010031;4.湖南农业大学动物科学技术学院,长沙 410128;5.湖南农业大学园艺园林学院,长沙 410128)
  • 收稿日期:2014-10-16 出版日期:2015-09-23 发布日期:2015-09-23
  • 通讯作者: 王敏,副研究员,E-mail:wing_mail@hotmail.com;高民,研究员, E-mail:gmyh1588@126.com
  • 作者简介:谢天宇(1989-),男,内蒙古乌兰浩特人,硕士生,主要从事反刍动物营养与调控研究,E-mail:xie8803262@163.com
  • 基金资助:

    国家自然科学基金(31472133);中国科学院战略性先导专项(XDA05020700);国际原子能项目(16315);现代农业(奶牛)产业技术体系建设专项资金(CARS-37)

Evaluation of Accuracy of Models to Predict Enteric Methane Emissions in Dairy Cows

XIE Tian-yu1,2,WANG Min1* ,WANG Rong1,4,YAN Zhi-cheng1,5,SHI Hui-yu2,GAO Min3* ,TAN Zhi-liang1   

  1. (1.Institute of Subtropical Agriculture,the Chinese Academy of Sciences,Changsha 410128,China;2.College of Animal Science,Inner Mongolia Agricultural University,Hohhot 010018,China;3.Institute of Animal Nutrition and Feed,Inner Mongolia Academy of Agricultural and Animal Husbandry Sciences,Hohhot 010031,China;4.College of Animal Science and Technology,Hunan Agricultural University,Changsha 410128,China;5.College of Horticulture Landscape,Hunan Agricultural University,Changsha 410128,China)
  • Received:2014-10-16 Online:2015-09-23 Published:2015-09-23

摘要:

本研究旨在评估和比较8个常用奶牛胃肠道甲烷排放模型的估算精度,分析影响模型估算精度的原因。在湖南望城白箬铺奶牛场选用28头体况良好的荷斯坦奶牛,测定奶牛胃肠道甲烷排放量、体重、产奶量、采食量和营养日采食量(酸性洗涤纤维、中性洗涤纤维和总能)等。利用预测误差均方(Mean squared prediction error,MSPE)和一致性相关系数(Concordance correlation coefficient,CCC)两种分析方法衡量奶牛胃肠道甲烷排放量估算值与实测值间的差异,进而评估和比较8个模型的估算精度,分析影响8个模型估算精度的原因。结果表明:模型8的估算精度最高,其次模型1、2、3和6的估算精度处中等水平,模型4、5和7的估算精度较差;模型1和2的估算误差主要来自回归斜率偏离,模型3、6和7的估算误差主要来自整体偏差,模型4和5的估算误差来自回归偏离和整体偏差。结果解释:模型1和2的回归斜率偏离主要是由于试验测得的Ym值与模型推荐的默认值相差较多;模型3的整体偏差主要是由于在相同干物质采食量的条件下,模型建立时的奶牛胃肠道甲烷排放水平远高于本试验;模型4和5的回归斜率偏离和整体偏差以及模型6的整体偏差主要是由于模型本身没有考虑到日粮碳水化合物的瘤胃降解率、日粮成分瘤胃的流通速率等因子的影响;模型7的整体偏差主要是由于稻草采食量不能等于实际日粮粗饲料采食量。以产奶量(kg•d-1)和体重(kg)为变量的模型8是本试验估算精度最高的模型。有必要继续开展试验扩充奶牛胃肠道甲烷排放数据,建立更加准确、可靠的甲烷排放模型估算方法。

Abstract:

This experiment was conducted to evaluate and compare the accuracy of 8 published models to predict the enteric methane emissions in dairy cows,and analyze the factors that could affect the accuracy of prediction.28 Chinese Holstein dairy cows were selected in the Wangcheng Bairuopu dairy farm,Hunan province,to determine the enteric methane emissions,body weight,milk production,feed intake and other nutrients intake(acid detergent fiber,neutral detergent fiber and gross energy) and etc.The difference between the predicted and observed values of enteric methane emissions from dairy cows was estimated and compared based on the mean squared prediction error(MSPE)and concordance correlation coefficient(CCC)methods,and the factors influencing the accuracy of predictions of 8 published models were analyzed.The results showed that the highest accuracy of prediction for model 8,medium accuracy of prediction for model 1,2,3 and 6,and the lowest accuracy of prediction for model 4,5 and 7.The errors influencing the accuracy of model 1 and 2 were mainly caused by deviation of regression slope from unity;the errors influencing the accuracy of model 3,6 and 7 were mainly caused by the overall bias;the errors influencing the accuracy of model 4 and 5 were caused by both deviation of regression slope from unity and overall bias.The predicting errors of model 1 and 2 were attributed to the difference of calculated Ym and IPCC default Ym;the predicting errors of model 3 was due to that enteric methane emissions for the model 3 was higher than that in this study under the same DMI;the predicting errors of the model 4,5 and 6 were mainly caused for not considering the rumen feed digestion of carbohydrate,ruminal passage rate of dietary nutrient and etc;the predicting errors of model 7 was due to that rice straw intake couldn’t represent real dietary forage intake.Model 8 with 2 variables of milk yield(kg•d-1)and BW(kg)had the highest accurancy of prediction in this trial.Experiments are still needed to collect more data to develop accurate and reliable methods of models to predict enteric methane emissions in dairy cows.

中图分类号: