Skip to main content
ARS Home » Plains Area » Miles City, Montana » Livestock and Range Research Laboratory » Research » Publications at this Location » Publication #185025
Title: PREDICTING FORAGE QUALITY IN THE NORTHERN MIXED-GRASS PRAIRIE

Author
item Haferkamp, Marshall
item Macneil, Michael
item Grings, Elaine

Submitted to: Research Update for Fort Keogh Livestock and Range Research Laboratory
Publication Type: Popular Publication
Publication Acceptance Date: 6/30/2005
Publication Date: 7/15/2005
Citation: Haferkamp, M.R., Macneil, M.D., Grings, E.E. 2005. Predicting forage quality in the northern mixed-grass prairie. Research Update for Fort Keogh Livestock and Range Research Laboratory. p. 37.

Interpretive Summary: Quality and quantity of forage produced are important factors affecting livestock production from grazinglands. 'Greenness' has been proposed as an indicator of herbage quality particularly N content. In this study, standing crop samples were collected during the period April through October over 3 yr from each of 3 replicated grazing regimes on silty range site in eastern Montana. Samples were sorted into live, current dead, and old dead components, ground, and analyzed for N content. Degree days (base 45º F) were calculated from monthly average temperatures for Miles City, MT obtained from the NOAA. An equation to predict percent N from percent dead and cumulative degree days was developed using multiple linear regression procedures of SAS. This equation accounted for 75.9% of variation in percent N and prediction error variance was 0.026. To validate this equation, data were obtained from an independent study of 8 areas on silty and clay-pan range sites grazed during fall and winter, but not grazed during the growing season, and sampled April through September. Samples from these sites had been separated into live and dead components and percent N was determined. When the equation developed was used to predict percent N in the validation data, the resulting correlation between predicted and actual values was 0.79. Regression of actual values on predicted values was 0.995±0.125, intercept was not different from 0.0 (P=0.22), and prediction error variance was 0.042. This equation may have utility for predicting quality of forage from Northern Great Plains range sites.

Technical Abstract: Quality and quantity of forage produced are important factors affecting livestock production from grazinglands. 'Greenness' has been proposed as an indicator of herbage quality particularly N content. In this study, standing crop samples were collected during the period April through October over 3 yr from each of 3 replicated grazing regimes on silty range site in eastern Montana. Samples were sorted into live, current dead, and old dead components, ground, and analyzed for N content. Degree days (base 45º F) were calculated from monthly average temperatures for Miles City, MT obtained from the NOAA. An equation to predict percent N from percent dead and cumulative degree days was developed using multiple linear regression procedures of SAS. This equation accounted for 75.9% of variation in percent N and prediction error variance was 0.026. To validate this equation, data were obtained from an independent study of 8 areas on silty and clay-pan range sites grazed during fall and winter, but not grazed during the growing season, and sampled April through September. Samples from these sites had been separated into live and dead components and percent N was determined. When the equation developed was used to predict percent N in the validation data, the resulting correlation between predicted and actual values was 0.79. Regression of actual values on predicted values was 0.995±0.125, intercept was not different from 0.0 (P=0.22), and prediction error variance was 0.042. This equation may have utility for predicting quality of forage from Northern Great Plains range sites.