高寒草甸牧草常规营养成分与总能预测模型的建立
Establishment of Prediction Model between Total Energy and Conventional Nutrients in Alpine Meadow Forage

建立高寒草甸牧区牧草养分与牧草总能之间的能值预测模型，为高寒草地健康评价及家畜管理提供理论及技术支持。以牧草干物质（DM, x1）、粗蛋白（CP, x2）、粗灰分（Ash, x3）、酸性洗涤纤维（ADF, x4）、中性洗涤纤维（NDF, x5）、钙（Ca,x6）、磷（P, x7） 7 种常规营养成分作为自变量，牧草总能（GE，y）作为依变量，建立高寒草甸暖季和冷季牧草不同自变量组合与总能之间的回归方程。结果表明：暖季牧草的DM、ADF和NDF与GE呈正相关，CP、Ash、Ca、P与GE呈负相关；冷季牧草的CP、Ash、ADF、Ca、P与GE呈正相关，DM和NDF与GE呈负相关，其中P与GE有较强的相关性；预测准确性较高的模型为：y=9 469-57.71x1+1.34x2-11.06x3+7.70x4+11.38x6+2 306x7，R=0.906 3，CV=0.69%；y=-6 117+93.81x1+39.82x2+230.61x3+15.54x4-65.96x6-8 000x7，R=0.815 9，CV=2.44%。综合得出，在高寒草甸草场可选择以牧草DM、CP、ADF、Ash、P建立与GE相关且准确性更高的预测模型。
This experiment provides a theoretical and technical support for the evaluation of alpine grassland health and livestock management by establishing a prediction model between total energy and forage nutrients in alpine meadow pastoral areas. In regression equations of forage total energy(GE,y), seven independent variables include dry matter(DM, x1), crude protein(CP, x2), ash(Ash, x3), acid detergent fiber(ADF, x4), neutral detergent fiber(NDF, x5), calcium(Ca, x6), and phosphorus(P, x7) . The result indicates that DM, ADF, NDF of warm season forage are positively correlated with GE, while negative to CP, Ash, Ca, and P; however, CP, Ash, ADF, Ca, and P of cool season forage are positively correlated with GE, while negative to DM and NDF, among which P is strongly correlated with GE. The model with higher prediction accuracy are y=9 469-57.71x1+1.34x2-11.06x3+7.70x4+11.38x6+2 306x7，R=0.906 3，CV=0.69%；y=-6 117+93.81x1+39.82x2+230.61x3+15.54x4-65.96x6-800 0x7，R=0.815 9，CV=2.44%.It is concluded that DM, CP, ADF, Ash and P of the forage can be selected to establish a prediction model with higher correlation and accuracy with GE in alpine meadow pasture.