Characterization of dairy productive systems in the Tenth Region of Chile using multivariate analysis

Ricardo Smith R.1, Víctor Moreira L.1 y Luis Latrille L.1

This study characterizes and classifies dairy farms from the Tenth Region of Chile using technical, productive, and human capital information. Two hundred and ninety dairy farms were surveyed and characterized by ten quantitative and nine qualitative variables in 1997. The collected information was analyzed using three multivariate exploratory statistical methods (principal component analysis, multiple correspondence analysis, and cluster analysis) to generate four dairy productive systems. The first system included farms characterized by low technological level, spring calving, no stall-housing, a stocking rate of 0.62 animal units per hectare, and milk production per cow between 741 and 1,547 L year-1. The second productive system included farms with fall-spring calving, 1.25 months per year of stall-housing, 0.72 animal units per hectare, and milk production per cow between 1,617 and 3,112 L year-1. Farms in the third productive system had year-round or fall-spring calving, almost no stall housing, 1.16 animal units per hectare, and an annual milk production per cow varying from 2,018 to 3,671 L. The fourth system included farms with the highest technological levels, year-round calving, stall-housing time equivalent to 2 or more months per year, 0.96 animal units per hectare, and annual cow milk production ranging from 3,925 to 5,348 L.

Keywords: dairy farms, dairy productive systems, multivariate statistics, multiple correspondence analysis, cluster analysis.
1 Universidad Austral de Chile, Facultad de Ciencias Agrarias, Casilla 567, Valdivia, Chile. E - mail: rsramirez@arec.umd.edu; vmoreira@uach.cl