Effects of leaf nitrogen concentration versus CND nutritional balance on shoot density and foliage colour of an established Kentucky bluegrass (Poa pratensis L.) turf

A 3-yr field study was conducted in a loam site and a sand site to determine the effects of N, P, and K application rates on N status, shoot density and foliage colour of Kentucky bluegrass clippings and to derive critical N values. The experiment was arranged in a completely randomized block design with four replicates and three nutrients, N (0 or 50 to 300 kg ha(-1) yr(-1)), P (0 or 21.8 to 87.3 kg P ha(-1) yr(-1)), and K (0 or 41.7 to 250 kg K ha-1 yr-1), equally broadcast six times per growing season. We examined N concentration values, log-transformed N concentration values, and CND values for compositional nutrient simplexes comprising one (V-N1), three (V-N3), five (V-N5), or 11 (V-N1) macro- and micro-nutrients. Critical values for N expressions were obtained from linear relationships with targeted visual shoot density and foliage colour ratings that were closely related to each other (R-2 = 0.92). The highest coefficients of determination (0.975 to 0.980) were obtained with V-N5 and V NIP The VN5 values were consistent with literature data and across experimental sites, and were the most successful (99% success) in diagnosing N sufficiency in 328 qualified specimens from commercial sod farms. Compared with raw concentrations, the CND transformation reduced from 0.6 to 0.1 the degree of inter-correlation among nutrients in principal component analysis and was amenable to a chi(2) distribution of CND indices. Using a critical imbalance index (CNDr(2)) of 5.6 as chi(2) value, and a critical CND I-N(2) index of 1.5 for a 5-nutrient simplex, we diagnosed as imbalanced 179 qualified specimens of which 110 specimens presented excessively high N level among the 328 qualified specimens in commercial stands. The proposed five-nutrient CND norms proved to be effective in diagnosing N status in Kentucky bluegrass clippings across experimental, literature, and survey data sets.