Abstract

On the premise that fertilizer needs might be better understood if plant removal were known, the yield of N, P, and K was recorded for 15 harvests of field plots over 3 years. Tall rescue (Festuca arundinacea Schreb.), Kentucky bluegrass (Poa pratensis L.), and orchardgrass (Dactylis glomerata L.) plots were replicated three dines and were split to receive annual applications of A, 224-98-186 (kg/ha of N-P-K) half in spring, half in summer); B, 224-0-0 (scheduled as A); and C, 112-49-93 (spring only).

In individual harvests these grasses frequently differed in yield; percent N, percent P, and weight of N, P, and K removed. Combined analyses (five harvests/year for 3 years) confirmed differences among species in yield and percent N and P; in weight of elements removed by species, however, only the lower quantity of P removed by Kentucky blyuegrass was significantly different, averaging 50% less than the other grasses.

The decrease in K removal from NH₄NO₃ at 224-0-0 vs 224-98-186 was greater than decreases in removal of N, P, or dry matter. Seasonal and annual decreases in K were consistent.

When element removal was calculated as annual yield ✕ mean percentage of the element rather than the sum of the products of plot yield ✕ plot composition, significantly lower mean values for N and K (4.2 and 5.0% lower, respectively) were obtained. N and K percentages were significantly correlated with DM yields. Phosphorus percentage values did not correlate with DM yields. The discrepancy between the actual weight of P removed (by a grass at a fertilizer treatment in a given year) and the short-cut calculation might be either positive or negative; a comparison between such extreme values would lead to large errors.

Copyright © . .