For 7 years (1997–2003), five nitrogen (N) rates (0, 60, 120, 180, and 240 kg N ha−1) were applied to sugar beets arranged in randomized complete block (RCB) design experiments with six replications and grown on light soils (sand content >50 g kg−1) in northern Greece. The aim of this work was to identify soil characteristics that affect yield, quality, and sugar beet response to N fertilization. Before sowing, soil analyses were conducted in control plots (0 kg N ha−1) at two depths (0–30 and 30–60 cm). Soils differed in their physical and chemical properties and especially in sand content, which ranged from 500 to 732 g kg−1. Quantitative (root number, RN; root yield, RY; and sugar yield, SY) and qualitative (percentage sucrose content in fresh root weight, SC; potassium, K; sodium, Na; and α-amino N) traits of control plots were used as soil fertility index. The RN was positively affected by clay content, and RY and SY were positively related with sand and negatively with silt content. The SC was negatively affected by soil (NO3)-N and sodium (Na) concentrations. Also, soil (NO3)-N concentration was positively related with root impurities (K, Na, α-amino N). In combined data over years, N rates had a negative effect on the RN. The RY was the only trait affected by years, N rates, and their interaction. The SC and SY differed significantly between years, and N rates affected significantly the former but not the latter. In combined data over years, N rates were curvilinearly related with Na concentration in roots, whereas a strong, linear relationship was found between α-amino N concentration and N rates. To study the significant years × N rates interaction evidenced for the RY, the relative response (RR) of the RY to N was introduced. Actually, the RR expresses the increase or decrease of the RY for a 150 kg N ha−1 rate compared to the control (0 kg N ha−1). The RR was strongly related with soil K concentration at the 0- to 30-cm depth (y = –0.00002x2 + 0.0082x + 0.5085, r2 = 0.92, P < 0.01, n = 7) and with total N concentration at the deeper layer (y = 1.8335x2 – 3.5312x + 2.6614, r2 = 0.88, P < 0.05, n = 6). Thus, the RY response to a rate of 150 kg N ha−1, which is the commonly applied to the sugar beet crop in Greece, can be predicted reliably by soil characteristics (K and total N concentration) determined before sowing. The strong relationship between soil K concentration and sugar beet response to N merits further research.
Read full abstract