Objectives The sandy soil in the old Yellow River belt is low in capacity of water and nutrient retention. We studied the application technique of controlled-release urea to achieve high yield and quality of rice.

Methods A 2-year field experiment was conducted in soil collected from the old Yellow River belt of northern Jiangsu Province. The medium-maturing medium-japonica rice cultivar Suxiu 867 was used in this study. The experiment was a split-plot design. The main plot was two fertilizer application methods: 1) controlled-release urea and conventional urea were mixed in 6∶4, and applied as basal fertilizer in one time (BC), and 2) with the same mix ratio as in BC, controlled-release urea and 50% of conventional urea applied as basal fertilizer, and the left 50% of conventional urea was top-dressed in the later panicle initiation (BC+PU). The subplot was resin-coated urea with controlled-release period of 80 (CR80), 100 (CR100) and 120 (CR120) days. 100% conventional urea with 4 times of application was used as the control (CK). Rice yield, number of stems and tillers, leaf area index (LAI), and dry matter accumulation of rice were investigated. The quality of processing and appearance, nutrition, cooking quality, eating quality, and RVA profile characteristics of rice were measured.

Results The yields in both the treatments under BC and BC+PU were in order of CR80 > CR100 > CR120. With the same controlled release urea, the rice yield under BC+PU was higher than that under BC treatment. The CR80 achieved the highest yield under BC+PU plot, with the yield of 9.64 t/hm² in 2018 and 10.73 t/hm² in 2019, and 5.32% and 5.81% higher than that of CK (P < 0.05), respectively, which attributed to the increase of total number of spikelets and seed-setting rate on the basis of stabilizing panicle number. The LAI and dry matter accumulation of rice in CR80 were higher than those of the other treatments in the middle and late growth stage, especially at the maturity stage, the LAI was significantly higher than that of CK. No matter under BC or BC+PU plot, there was no significant difference among three types of resin coated urea in rice processing quality. There was no significant difference in appearance quality, cooking and eating quality, and the RVA profile characteristics of rice between CR80 in BC+PU method and conventional urea application. The CR80 in BC method increased the amylose content, gel consistency and taste value, and increased peak viscosity, trough viscosity, final viscosity and breakdown value, and improved the cooking and eating quality. In both BC and BC+PU methods, CR100 and CR120 treatments increased the protein content, reduced chalkiness percentage and chalkiness degree, and decreased the amylose content and taste value, which also caused a steady decline in peak viscosity, trough viscosity, breakdown, final viscosity, and finally obtained a poor cooking and eating quality. Compared with BC treatment, BC+PU treatment improved appearance quality and nutrition quality, and reduced rice cooking and eating quality.

Conclusions In the sandy soil of the old Yellow River belt, the resin-coated urea with 80 days releasing period showed optimum effect in yield and quality of rice. Application of all the resin-coated urea and half conventional urea as basal fertilizer and top dressing of the left urea at panicle initiation facilitates better coordination between quality and yield but reduces cooking and eating quality of the rice. Application of all the resin-coated and conventional urea as basal fertilizer does not affect rice processing quality, however, improves cooking and eating quality of rice, at the expense of yield slightly.