近红外光谱法测定羊草干草的9项品质指标

Nine Quality Indexes Determination of Leymus chinensis Hay by Near Infrared Reflectance Spectroscopy

以120份不同产地、不同生长期的羊草干草样品为试验材料,利用近红外光谱(NIRS)技术,采用偏最小二乘法(PLS),全面建立了羊草的干物质(DM)、粗蛋白(CP)、粗纤维(CF)、中性洗涤纤维(NDF)、酸性洗涤纤维(ADF)、酸性洗涤木质素(ADL)、粗脂肪(EE)、粗灰分(CA)、总能(GE)等9个羊草品质指标的近红外检测模型。结果表明:CP、NDF、ADF、CA模型的校正决定系数(R2)均在90%以上,相对分析误差(RPD)均超过了3;DM、CF、ADL、GE模型的R2值均在80%以上,RPD值均超过了2.5;EE模型的R2值为46.11%,RPD值为1.36。由此可见,除EE外,羊草品质相关各指标均可以通过近红外光谱技术实现准确检测,这对于实现我国羊草品质的快速检测评价及优良种质资源的筛选具有重要意义。

Leymus chinensis is one of the important forages in northern China.The feasibility of rapid detection of quality of Leymus chinensis was evaluated by near-infrared reflectance spectroscopy (NIRS) technique,which can provide a fast and reliable analysis for the quality evaluation of Leymus chinensis.A total of 120 samples of Leymus chinensis from different habitats and different growing period were selected as experimental materials to establish detection models of 9 Leymus chinensis quality indicators,including dry matter (DM),crude protein (CP),crude fiber (CF),neutral detergent fiber (NDF),acid detergent fiber (ADF),acid detergent lignin (ADL),ether extract (EE),crude ash (CA) and gross energy (GE),by NIRS and partial least squares (PLS) method.The results showed that the correction coefficient (R^2) of CP,NDF,ADF and CA models were all above 90%,and relative predict deviation (RPD) values were all more than 3.R^2 values of DM,CF,ADL and GE models were all above 80%,while RPD values all exceeded 2.5.EE model had a R^2 value of 46.11% and a RPD value of 1.36.In conclusion,indicators related to the quality of Leymus chinensis can be accurately detected by NIRS,except EE,which is of great significance for achieving the rapid quality detection and selecting excellent genetic resources of Leymus chinensis.