In addition to implementing active water resource allocation measures to solve the serious groundwater overexploitation problem caused by large-scale rice planting in the Sanjiang Plain, the reasonable adjustment of rice planting areas is another method of doing so. From the perspective of groundwater recharge and discharge balance, this paper carries out a novel assessment of suitable rice planting areas in the Sanjiang Plain, which is expected to provide a new method for the implementation of land exploitation according to water resource conditions. The technical scheme is as follows: by adjusting the water resource allocation data and rice spatial distribution data in the surface water–groundwater coupled model (baseline model with dynamic land use) in the Sanjiang Plain, static land-use models under different rice planting scales were established. Through simulation and comparison, the rice area that could achieve the balance of groundwater recharge and discharge was considered the suitable rice planting scale in the Sanjiang Plain. The results showed that the average annual change in groundwater storage from 2000 to 2014 simulated by the baseline model was −0.313 billion m³, indicating that there was space for further optimization and adjustment of the rice planting scale in the Sanjiang Plain. By comparing the static land-use models of each year under the current water resource allocation pattern, the rice area of 1.021 million hm² in 2005 could effectively realize the balance of groundwater recharge and discharge. Under the new water resource allocation pattern of 2035, the water resource conditions in the Sanjiang Plain will be greatly improved, which can support a rice planting scale of 3.058 million hm² on the basis of ensuring the balance of groundwater recharge and discharge. Our research results can provide a reference for water resource allocation and land-use optimization regulation in the Sanjiang Plain.

Understanding groundwater–surface water interaction is essential for water resource management and watershed ecological protection. However, the existing studies often emphasize the tracer role of hydrogeochemical methods (including hydrochemistry and isotopes) while underestimating the importance of analyzing watershed hydraulic characteristics, thus neglecting the indications of the driving mechanisms (hydraulic head difference) for the water exchange. Taking the Songhua River in the Sanjiang Plain as an example, this study combines hydraulic, hydrochemical, and isotopic methods to clarify the groundwater–surface water interactions from both a driving mechanism perspective and a hydrogeochemical characterization perspective within the water cycle. The results indicate that human exploitation has caused river water to infiltrate into groundwater, converting the section into a losing river, where surface water consistently exhibits a hydraulic tendency to recharge the aquifer. The influence zone of the river extends up to 3.5 km from the riverbank, with an average recharge rate from the river reaching 78.04% within this area. This recharge mixes and dilutes the adjacent groundwater, impacting its hydrogeochemical characteristics. This study enhances the understanding of combined methods for groundwater–surface water interaction and provides a scientific basis for water resource management and pollution control strategies in the local agricultural regions.

Sustaining rice productivity under the dual constraints of freshwater scarcity and low-temperature stress represents a pressing challenge for high-latitude japonica rice systems worldwide. There is an urgent need to develop coupled irrigation–agronomic management strategies that jointly safeguard yield stability and water use efficiency (WUE) in cold-region rice production. In this study, a two-year field experiment was conducted in 2024–2025 on albic soil (Albic Luvisols, WRB; θfc 38.2% v/v, pH 5.80, clayey texture with poor permeability and a propensity for subsurface waterlogging) in the Sanjiang Plain, Heilongjiang Province, China (47°15′ N, 133°28′ E), with nine coupled “irrigation regime × auxiliary practice” treatments, comprising conventional continuous flooding, four-level controlled irrigation (CI) at lower thresholds of 60%, 70%, 75%, and 80% θfc, and their combinations with film mulching (FM) or a humic-acid-based soil amendment (SA). An interpretable machine-learning diagnostic framework was developed, with elastic net (EN) as the primary analytical model and random forest (RF) as a nonlinear control, to simultaneously identify core yield predictors and outlier treatments. The principal findings were: (i) The soil-amendment-coupled 75% θfc CI treatment (SACI) increased grain yield by 12.3% and reduced water input by 17.0% relative to conventional continuous flooding, with WUE reaching 1.801 kg m⁻³, a 35.3% gain over the control (p < 0.05); these improvements were consistent across both individual years (year × treatment interaction: p = 0.601; inter-year rank correlation ρ = 0.967). Lowering the CI threshold below 75% θfc significantly reduced grain yield through diminished effective-panicle retention. (ii) Multi-method consensus analysis (Kendall’s W = 0.871, p < 0.01) identified root volume at the milk stage as the most strongly and consistently associated statistical predictor of yield formation, with convergent mechanistic support from independent rhizosphere evidence (Eh, TTC reductive activity). Definitive causal validation awaits isotope-tracing experiments. (iii) The film-mulching × continuous-flooding treatment (FMCG) was diagnosed as a yield-response outlier (permutation test p = 0.003), three in situ rhizosphere measurements (redox potential, root TTC-reducing activity, and rhizosphere temperature) supported the proposed mechanism of hot–anoxic rhizospheric inhibition. Methodologically, this study develops a four-level evidence convergence framework that integrates intra-model self-consistency, cross-model (EN vs. RF) consensus, independent rhizosphere evidence, and distribution-free permutation testing, with Jackknife+ conformal prediction and companion Monte Carlo simulations (1000 replicates) used to quantify the reliability boundaries under small-sample conditions (n = 27). These findings provide an evidence-based irrigation–soil co-management strategy for cold-region rice production in Northeast China, and the proposed diagnostic paradigm offers a generalizable, reliability-quantified methodological template for interpretable small-sample modeling in multifactorial coupled field experiments.