2.1. Experiment site and design
The field experiment was conducted in a reclamation land in Zhangjiagang city, Jiangsu province (31°57′N, 120°46′E, Fig. 1). The soil is composed of alluvium from the Yangtze River and classified as Fluvo-aquic according to the Chinese Soil Taxonomy (Gong et al., 2001). The basic soil chemical properties are as follows: pH (H2O) 8.00±0.30; SOC 3.86±1.35 g kg−1; TN 0.41±0.16 g kg−1; NH4+-N 1.36±0.34 mg kg−1; NO3-N 11.58±5.37 mg kg−1; available phosphorus (AP) 20.57±12.67 mg kg−1; and available potassium (AK) 7.34±1.95 mg kg−1.
A plot design with replicates was adopted with 60 m × 35 m (Fig. 1). Rice-green manure (RG), rice-rape (RR), and rice-wheat (RW) rotations were included. Five fertilization treatments were set as follow: without N fertilizer (WN), farmer’s conventional urea fertilizing practice with a thrice-split application (CU), single-dose application of BBU with the conventional amount of N (CB1), twice-split application of BBU with the conventional amount of N (CB2), single-dose application of BBU with 20% reduction of N (RB1), and twice-split application of BBU with 20% reduction of N (RB2). The BBU consisted of 75% CRU and 25% conventional urea. The CRU coated with 10% waterborne polyacrylate was purchased from Jiangsu Issas New Fertilizer Engineering & Technology Co., Ltd., China and had an N content of 42% and an N release longevity of 3 months. The farmer’s conventional N application rates are 300 and 270 kg N ha−1 for rice and wheat, respectively. There was no fertilization during the rape oil and green manure seasons. Superphosphate, potassium chloride, zinc sulfate, and silicon fertilizer were incorporated into the 10−15 cm soil layer for all treatments at rates of 75 kg P2O5ha−1, 120 kg K2O ha−1, 3 kg Zn ha−1, and 0.8 kg SiO2 ha−1 as basal fertilizers in rice and wheat seasons.