Fibre yield estimation in sisal (Agave sisalana perr. ex engelm.) through regression equation based on simple biometric observations

Estimation of fibre yield in sisal (Agave sisalana Perr. ex Engelm.) is absolutely necessary for well ahead planning of marketing arrangements between the actual grower or cooperatives and the fibre purchaser or mills who are situated geographically in distant locations in India. There is no yield prediction method for sisal which could estimate fibre yield well ahead of harvesting. Therefore, an attempt had been made during 2010-11 at Sisal Research Station, Bamra, Sambalpur, Odisha to develop linear regression equations of fibre yield (Y) estimation by considering number of harvestable leaf (LN), leaf length (LL), leaf breadth (LB), leaf green weight (LW) as independent variables data collected from field experiment. In total, nine regression equations were developed, of which the equation, Y = -1221.428 + (23.327) x LN+ (10.761) x LL+ 21.355 [where, Y= Fibre yield (g/plant); LN= mean Leaf number per plant (harvestable); LL= mean Leaf length (cm)] predicted the fibre yield most accurately (2.48 t/ha) with the observed actual yield data (2.39 t/ha) and the coefficient of determination (R2) value was near unity (0.981). Therefore, the developed regression equation can predict the fibre yield in sisal with acceptable accuracy for edapho-climatic condition prevailing in central plateau region of India.