Agriculture, particularly rice production, poses a substantial threat to global water security due to its high freshwater demand for irrigation and substantial wastewater discharge during drainage. Rice farming consumes approximately 40% of global freshwater resources and exhibits lower water use efficiency compared to upland crops, resulting in higher wastewater release. The intensification of irrigation and drainage systems exacerbates water stress, further aggravated by climate change and its associated extreme weather events. These events increase agricultural nutrient losses and negatively impact water quality. This growing water scarcity jeopardizes food production and global food security, highlighting the urgent need for sustainable and climate-resilient agricultural practices. Achieving the United Nations' Sustainable Development Goals related to ending hunger, promoting sustainable agriculture, improving water quality, and increasing water use efficiency requires both on-farm technological advancements and transformative changes to agricultural systems. While on-farm technologies such as water-saving irrigation and controlled drainage have been extensively studied, research on irrigation drainage system transformation remains limited. Past studies have focused on specific aspects like water resource regulation, water quality improvement, or system resilience. This research addresses the need for a more comprehensive analysis and solution to tackle escalating water stress challenges.

Existing literature highlights the potential of irrigation system transformation for cost-effective water resource regulation, particularly in regions with distinct monsoon and dry seasons. Studies demonstrate the effectiveness of drainage system redesign in mitigating nutrient losses, enhancing fertilizer use efficiency, and reducing greenhouse gas emissions. Modern water storage strategies advocate for integrated systems incorporating reservoirs, ponds, tanks, aquifers, and wetlands to enhance resilience. However, a comprehensive approach that addresses both sustainability and resilience in irrigation and drainage systems is lacking. This research draws inspiration from historical irrigation systems, such as the Beitang system in China, which demonstrated a successful balance between rice production and environmental impact for over 2500 years. This traditional system, characterized by decentralized small ponds used for both water storage and irrigation, offers a valuable model for addressing contemporary water stress issues.

This study uses China as a case study to explore pathways toward more sustainable and resilient rice irrigation and drainage systems. A systematic survey was conducted to assess the evolution and current status of rice irrigation and drainage systems in China, focusing on the utilization of small water bodies. The survey included statistical data on the number of Beitang systems, irrigated and drainage field areas, population, rice yields, and the number of aquaculture ponds from 1950 to 2020. Expert knowledge on four irrigation and drainage management styles was also gathered. Finally, remote sensing data were utilized to determine the area percentages of ditches and ponds in current rice irrigation drainage systems. Only small ponds with a surface area of less than 0.33 ha were included to avoid confounding effects from ponds used for aquaculture. The area percentages of ditches were determined using Google Earth imagery. Daily climate data, field agricultural management information (including fertilization and water level management), and nutrient retention parameters for ditches and ponds were collected. A water quantity and quality model for paddy irrigation and drainage units (WQQM-PIDU) was developed and used to simulate water consumption and wastewater release for different management styles. The model incorporates daily water quantity and quality variations within an irrigation drainage unit (IDU) including fields, ditches, and ponds (if present). To simulate freshwater irrigation from remote reservoirs, freshwater irrigation was assumed to be sufficient in normal years but limited in dry conditions. The model was calibrated by comparing simulated nutrient export with published literature values, and uncertainty analysis was performed using Monte Carlo sampling of nutrient retention parameters. The sustainability of rice production was evaluated using the water footprint (WF), comprising green (rainfall), blue (freshwater irrigation), and gray (wastewater) components. System resilience was assessed using metrics such as local storage volume, irrigation self-sufficiency, and potential yield loss under dry conditions. Three system redesign approaches—recycling irrigation, pond reconnection, and pond construction—were proposed and analyzed to determine their potential benefits, costs, and implementation feasibility.

The study reveals a fundamental shift in China's irrigation and drainage systems over recent decades, characterized by the decline of decentralized pond-based systems (Beitang) and the rise of centralized systems relying on large reservoirs. This shift has led to a substantial decrease in the number and area of small water bodies, primarily due to land conversion and poor maintenance. The disuse of ponds has resulted in a significant increase in the water footprint of rice production, particularly the blue water footprint (irrigation freshwater consumption). The water footprint of the quasi-decentralized system (utilizing small water bodies) is 33% less than that of the totally centralized system. The quasi-decentralized system reduces its blue WF by an average of 23% due to the recycled usage of gray water, leading to a reduction of 59% in gray WF entering surrounding surface waters. The difference in WF between the two systems was more pronounced in provinces with a higher area percentage of ditches and ponds. The study also demonstrates that the disuse of ponds has reduced the resilience of rice production systems to extreme weather events. Irrigation self-sufficiency is strongly correlated with local storage volume provided by ditches and ponds. Decentralized systems exhibit higher irrigation self-sufficiency, even in dry years, providing greater resilience. In 25% dry years, centralized systems showed 2.5% greater potential yield loss than decentralized systems with irrigation self-sufficiency of 15-40%. This difference increases to 3.1% under extreme dry conditions. Three system redesign approaches were proposed to integrate more ditches and ponds: recycling irrigation, pond reconnection, and pond construction. Recycling irrigation, the most cost-effective approach, involves prioritizing the use of retained ditch and pond water for irrigation. Pond reconnection involves reconnecting existing ponds to rice fields. Pond construction involves building new ponds. Analysis of observations in redesigned irrigation drainage units (IDUs) compared to control IDUs showed significant reductions in gray WF (65-85%), TN loads (21-85%), and TP loads (30-85%) and slight increases in rice yield (2-3%). The cost-benefit analysis shows that recycling irrigation is the most cost-effective approach, while pond construction is the most expensive, with trade-offs regarding land use and potential yield reduction. However, ongoing government policies in China may partially alleviate these costs and incentivize implementation.

The findings of this study directly address the research question of how to enhance rice production sustainability and resilience. The results demonstrate that reactivating small water bodies offers a significant opportunity to improve both sustainability and resilience. The substantial reduction in water footprint and alleviation of yield loss under dry conditions highlight the importance of integrating traditional practices with modern agricultural techniques. The proposed system redesign approaches provide practical and adaptable solutions within the existing infrastructure. The success of the recycling irrigation approach is particularly noteworthy due to its cost-effectiveness and ease of implementation. The study's findings are highly relevant to the broader field of sustainable agriculture and water resource management, emphasizing the potential of integrating ecological engineering principles to enhance agricultural resilience in the face of climate change. The findings are directly applicable to ongoing policies and projects aimed at improving water management in China's agricultural sector.

This study demonstrates the significant benefits of reactivating small water bodies in rice irrigation and drainage systems in China, leading to improved water use efficiency, reduced water footprint, and enhanced resilience to drought. The three proposed redesign approaches offer different levels of complexity and cost, with recycling irrigation presenting a particularly promising cost-effective solution. Future research could focus on further refining cost-benefit analyses considering various socio-economic factors and exploring the generalizability of these findings to other rice-producing regions.

While the study presents a comprehensive analysis, certain limitations need to be acknowledged. The WQQM-PIDU model, although calibrated, is a relatively new model, and further validation using data from more sites is needed to enhance model robustness. The analysis primarily focuses on drought resilience, while the benefits of additional water storage capacity in mitigating waterlogging are not fully quantified. The cost-benefit analysis assumes a certain level of government support for implementation. The generalizability of the findings beyond the specific context of China requires further research.