Objectives Drip-irrigated rice demonstrates good water-saving potential and favorable yield performance. However, spikelet degeneration remains a major constraint, and the underlying mechanisms are note yet fully understood. In this study, we evaluated the impacts of water and nitrogen management on root growth, N uptake and utilization, and yield formation in drip-irrigated rice. The findings would provide a theoretical basis for understanding spikelet degeneration and optimizing water-nitrogen strategies.

Methods A split-plot field experiment was conducted during the 2022 and 2023 growing seasons. The main plots were assigned three irrigation regimes: continuous flooding (FI); conventional drip irrigation (DIO, maintaining soil moisture at 90%−100% of field capacity); and water-stressed drip irrigation (DIS, maintaining soil moisture at 80−100% of field capacity). The subplots consisted of three N application levels: low (LN, N 240 kg/hm²), moderate (MN, N 300 kg/hm²), and high (HN, N 360 kg/hm²). At the early, middle, and late panicle differentiation stages (58, 68, and 78 days after emergence, respectively), photosynthetic parameters were measured, and the above ground plant parts were sampled for analyse of biomass and nitrogen content, and the root growth was investigated in 0−50 cm soil profiles, segmented into 10-cm layers. At the late panicle differentiation stage (78 days after emergence), the primary and secondary panicle branches and the spikelet were sampled to investigate the degeneration rate. The grain yield was investigated at maturity.

Results Compared with FI, DIO and DIS treatments exhibited 5.95% and 11.97% higher branch degeneration rates, and 1.63% and 2.98% higher spikelet abortion rates, respectively, resulting average grain yield reductions of 24.75% and 53.32% over the two-year period. During the panicle differentiation stage, the DIO treatment significantly reduced leaf N content, photosynthetic rate (Pn), deep root biomass (10−50 cm), and partial factor productivity of nitrogen (PFPN) by 17.23%, 13.44%, 57.95%, and 28.03%, respectively, relative to FI. The DIS treatment caused even greater reductions of 24.56%, 26.67%, 65.01%, and 55.55%, respectively. Under the same irrigation regime, the HN treatment increased leaf Pn, leaf N content, and grain N content by an average of 8.67%, 10.62%, and 6.58%, compared to MN, whereas the LN treatment showed a decreasing trend. Deep root biomass under DIO and DIS was significantly lower than under FI. Increasing nitrogen input promoted deep root biomass and yield of rice while significantly reducing the degeneration rates of branches and florets. Partial Least Squares Path Modeling (PLS-PM) demonstrated that improved water-nitrogen management (increased nitrogen application and drip irrigation) enhanced the proportion of deep roots and nitrogen uptake efficiency, thereby improving Pn and PFPN, which in turn reduced panicle branch degeneration and spikelet abortion rates, and ultimately mitigated yield losses in drip-irrigated rice.

Conclusions Drip irrigation promotes shallow root growth but reduces deep root biomass of rice. This shift in root architecture poses a risk of inefficient nitrogen utilization and constitutes a key factor leading to the degeneration of branches and spikelets. Increasing nitrogen fertilizer can enhance deep root biomass and nitrogen use efficiency, thereby mitigating spikelet degeneration in drip-irrigated rice, however, may lead to decreased nitrogen use efficiency. Further study is needed to regulate the drip irrigation and the distribution of rice root system in soil.