Correlation and regression relationships between quantitative and qualitative indicators of perennial grass and legume mixtures

Tatyana Bozhanska and Boryana Churkova

Abstract: The growth and development of the following six double mixtures of legumes and grasses were examined: bird‘s-foottrefoil + red fescue; white clover + perennial ryegrass; white clover + kenthucky bluegrass; red clover + timothy-grass; blue hybrid alfalfa + cock‘s foot and red clover + meadow fescue, to establish correlation and regression relationships between their quantitative and qualitative indicators determining their nutritional value.
High correlation relationships were found in the chemical composition among some main indicators. The amount of crude protein with the weight percentage of legume crops had a high correlation coefficient (r = 0.870) and the crude fiber content correlated positively with the height of grasses (r = 0.820) and legumes (r = 0.806).
There was a high positive correlation between crude fiber content with ADF (r = 0.827) and the cellulose (r = 0.814) and a high negative correlation between the mean values of crude fiber and those of the feed unit for milk – (r = -0.920) and growth (r = -0.860).
The graphical regression models allow determining the crude protein content by the weight percentage of legumes and the crude fiber content by the height of grass plants in the mixture.
The feed unite (FUM and FUG) can be predicted and determined by the graphical models based on crude fiber content with sufficient accuracy for practical purposes.

Keywords: chemical composition; correlations; grass mixtures; regression relationships

Date published: 2020-06-29