Applicability of Predictive Equations for Alfalfa Quality to Southwestern United States and Northern Mexico

Models to estimate alfalfa (Medicago sativa L.) neutral detergent fiber (NDF) and acid detergent fiber (ADF) concentrations using easily available variables were developed and tested in northern, humid regions of the United States but have not been validated in arid areas below 35 degrees N latitude. The objective of this research was to test the performance of predictive equations for alfalfa quality (PEAQ) on nondormant alfalfa grown in irrigated desert regions in New Mexico and northern Mexico. Alfalfa with fall dormancy ratings of 8 and 9 were sampled over 1 or 2 yr at three irrigated locations. Five alfalfa samples were taken per field on each sample date between 0730 and 0830 h, leaving a stubble of approximately 4 cm. Observed NDF and ADF values were regressed on predicted values yielding r(2) and root mean square error values in the range of those previously reported for alfalfa grown in northern states. The accuracy of predictions was improved by using field means (n = 5) rather than individual samples. In most cases regression equations were biased (slope not equal 1 and/or intercept not equal 0), but the magnitude of bias was relatively small. The index of agreement (d) was greater than 0.86 for all data, indicating an acceptable goodness of fit to predict NDF and ADF concentration using PEAQ. It is concluded that PEAQ can predict, with reasonable accuracy, fiber concentrations in alfalfa grown in desert regions of the southwestern United States and northern Mexico.