Field Testing a Rapid Method for Estimating Alfalfa Quality

Abstract

Predictive equations for alfalfa quality (PEAQ) based on height of the tallest stem and maturity stage of the most mature stem in a sample were developed to estimate neutral‐detergent fiber (NDF) and acid‐detergent fiber (ADF) concentrations in alfalfa (Medicago sativa L.). Field testing of these equations is limited outside the state of Wisconsin where they were developed. Our objectives were to test these equations for estimating alfalfa NDF and ADF across a wide geographic area and to evaluate the performance of PEAQ on a whole‐field basis by using within‐field subsampling. Alfalfa samples varying in height and maturity were collected throughout the growing season from fields in New York (n = 28), Pennsylvania (n = 23), Ohio (n = 48), California (n = 45), and Wisconsin (n = 48) in 1994 to 1996. Additional samples were collected in Ohio and Wisconsin from producer‐managed fields in which 5 to 10 subsamples per field were taken on each sampling date (n = 296 subsamples from 51 fields). Observed NDF and ADF values were regressed on estimated values. The accuracy of PEAQ in other states was at least equal to that observed in Wisconsin. Across all states, regression equations for NDF and ADF were slightly biased (b ≠ 1.0 and/or y‐intercept ≠ 0 at P < 0.01); however, prediction errors were sufficiently low to allow use of PEAQ as a preharvest management tool. Root mean square error values ranged from 19.1 to 23.9 g kg−1 for NDF and 15.0 to 19.0 g kg−1 for ADF. Prediction errors were 16.2 g kg−1 for NDF and 13.2 g kg−1 for ADF across Ohio and Wisconsin when regressing observed means on estimated means of five subsamples per fieldsampling date combination. We conclude that predictive equations for alfalfa quality based on a combination of stem height and maturity were robust across a wide range of environments.

The use of indirect methods for the prediction of lucerne quality in the first cut under the conditions of Central Europe

“…The prediction equations across all the years were more accurate than the GDD models. The original equations according to Hintz and Albrecht (1991) Sulc et al (1997) evaluated the performance of PEAQ across five states and presented the R 2 value 0.74 for NDF (= 23.67 + 0.41 S + 0.21 MSL) and 0.73 for ADF prediction. According to Owens et al (1995), the CP was not predicted accurately using PEAQ.…”

“…These parameters are related not only to NDF or ADF but also to CP, DOM, and IVDDM (Sulc et al, 1997;Rinne and Nykanen, 2000;Mitchell et al, 2001). Near infrared reflectance spectroscopy (NIRS) belongs to indirect methods (Mika et al, 2003) but this method is not so suitable for field conditions, moreover, special instrumentation is needed.…”

“…This method is based on the length of the tallest stem and the stage of the most mature stem in the sample (Hintz and Albrecht, 1991). The equations also performed well across the spring and summer growth cycles (Sulc et al, 1997) although lucerne quality varied among growth cycles during the year (Niwinska et al, 2005), which is connected with changes in the leaf stem ratio (Tyrolová and Výborná, 2008). Similarly Skládanka et al (2008) documented changes in fibre content in forage of grasses at the end of the growing season.…”

“…The above discussed methods are quick, easy and their accuracy is acceptable under the USA conditions (Owens et al, 1995;Allen and Beck, 1996;Sulc et al, 1997). Since no single method will ever result in a perfect prediction of lucerne quality in the future, a combination of methods may be most acceptable (Sulc et al, 1999).…”

Yield, Morphological Characteristics, and Chemical Composition of European‐ and Mediterranean‐Derived Birdsfoot Trefoil Cultivars Grown in the Colder Continental United States

“…Spredor alfalfa had relatively high CP concentrations that were similar to Norcen, but it had much greater NDF and ADF concentrations that, for example, exceeded Bokor by about 22%. Thus, our study did not generate an appropriate dataset for evaluating the usefulness of morphological characteristics for predicting BFT forage quality, as was done previously for alfalfa (Sulc et al, 1997). By contrast, variation in NDF, ADF, and ADL were weakly or not significantly associated with CT in BFT (R 2 ≤ 0.08).…”