ORIGINAL RESEARCH article
Front. Plant Sci.
Sec. Crop and Product Physiology
Volume 16 - 2025 | doi: 10.3389/fpls.2025.1678277
This article is part of the Research TopicPhysiological Growth of Crops in Saline-Alkali Land and Its New Quality Productive Control MethodsView all 5 articles
Optimization of deficit irrigation system for drip-irrigated corn in northern Xinjiang using dynamic reconstruction and dual physics-informed neural networks to drive AquaCrop
Provisionally accepted
- Xinjiang Agricultural University, Ürümqi, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Abstract: 【Introduction】To optimize the irrigation schedule for corn in northern Xinjiang and save water resources while maintaining stable production.【Methods】Based on the actual water shortage in northern Xinjiang during summer 2024, this study set up different deficit irrigation gradient treatments according to the crop water requirement (ETc) of each growth stage of corn. Combined with the corn growth and yield data of farmers from 2022 to 2024, the model parameters were calibrated and validated through global sensitivity analysis using AquaCrop-OS MATLAB. Then, the Dynamic Reconstruction and Dual Physics-Informed Neural Networks (DR-DPINNs) were integrated with water balance constraints during the corn growth period to optimize the deficit irrigation system for corn in northern Xinjiang.【Results】The results showed that in the global sensitivity analysis of the AquaCrop model, the water productivity (wp) and canopy growth coefficient (cgc) parameters had a significant impact on biomass accumulation (STi>0.10), and the canopy senescence parameter (psen) had a marked effect on yield (Si>0.05). The model parameters obtained through sensitivity analysis could meet the application requirements for simulating biomass, canopy cover, soil water content, and yield in the AquaCrop model.After optimization with DR-DPINNs, when the total irrigation amount was 472 mm, the yield increased by 10.8% and the water use efficiency rose by 11.15% compared with the conventional scheme. The DR-DPINNs method, by combining physical mechanisms with dynamic feature extraction, could significantly enhance the solving capability for high-dimensional nonlinear irrigation optimization problems. The optimized spatial and temporal irrigation distribution under a total water volume of 472 mm could achieve a simultaneous increase in yield and water use efficiency. 【Discussion】This study can provide theoretical methods with both mechanistic interpretability and decision-making accuracy for the dynamic optimal systems of drip-irrigated corn under water resource constraints in arid regions, and offer theoretical support and technical reference for agricultural water management in arid regions.
Keywords: corn, deficit irrigation, AquaCrop model, Optimization method, DPINNs
Received: 02 Aug 2025; Accepted: 29 Aug 2025.
Copyright: © 2025 Zhang, Zhao, Hong and Ma. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Jinghua Zhao, Xinjiang Agricultural University, Ürümqi, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.