Agricultural Water Management最新文献_第7页

Effects of deficit irrigation at different phenological periods on yield, quality, and water productivity of Xanthoceras sorbifolium Bunge in the Horqin Sandy Land of China 不同物候期亏缺灌溉对科尔沁沙地文冠果产量、品质和水分生产力的影响

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-16 DOI: 10.1016/j.agwat.2026.110130

Shuqin Song , Benye Xi , Yang Liu , Jing Zhong , Xiaoxia Wang , Liming Jia

In arid and semi-arid regions, improving water efficiency is imperative for the sustainable advancement of both forestry and agriculture. The water demand period for Xanthoceras sorbifolium Bunge, an economically and ecologically important tree grown in the Horqin Sandy Land of China, is unknown. This knowledge gap has hindered the development of optimized deficit irrigation (DI) schemes aimed at conserving water while maintaining yield and quality. To identify this key period and better understand the water-use dynamics of this species, we conducted a two-year field experiment (2021–2022). Eight irrigation treatments were applied across three key phenological stages: flowering (F), fruit setting to expansion (S), and fruit-expansion to maturity (M). The irrigation treatments included full-stage (FSM), two-stage (FS, FM, and SM), single-stage (F, S, and M), and no irrigation (NI). The application of DI decreased fruit yields by 8.36–58.01 % (p < 0.05), while two-stage irrigation significantly reduced water consumption and evapotranspiration compared with full-stage irrigation (p < 0.01). FS significantly improved water productivity (

WP

), irrigation water productivity (

{WP}_{I}

), and fruit quality. All two-stage irrigation treatments demonstrated yield response factors (

k_{y}

) < 1. The FS treatment reduced irrigation volume by 43.7 %, while the yield decreased by only 8.36 %, suggesting that the irrigation savings did not significantly compromise yield. In summary, the FS treatment is recommended as the most optimal irrigation schedule, followed by SM and FM, for the production of X. sorbifolium in drylands. This approach conserves water while minimally impacting productivity, thus representing a sustainable water management strategy.

在干旱和半干旱地区，提高用水效率对于林业和农业的可持续发展至关重要。中国科尔沁沙地上生长着一种具有重要经济和生态意义的乔木——文冠（Xanthoceras sorbifolium Bunge），其需水周期尚不清楚。这种知识差距阻碍了旨在节水同时保持产量和质量的优化亏缺灌溉（DI）方案的发展。为了确定这一关键时期，更好地了解该物种的水分利用动态，我们进行了为期两年的野外试验（2021-2022）。8个灌溉处理跨越3个物候阶段：开花(F)、坐果至膨大(S)和果实膨大至成熟(M)。灌溉处理包括全阶段（FSM）、两阶段（FS、FM和SM）、单阶段（F、S和M）和不灌溉（NI）。直接灌水使果实产量降低了8.36 ~ 58.01 % (p <； 0.05)，两期灌溉较全期灌溉显著降低了耗水量和蒸散量（p <； 0.01）。FS显著提高了水分生产力（WP）、灌溉水分生产力（WPI）和果实品质。所有两期灌溉处理均表现出产量响应因子（ky） < 1。FS处理减少了43.7%的灌水量，而产量仅下降了8.36%，表明节水对产量没有显著影响。综上所述，旱地枸杞生产的最佳灌溉方式为FS处理，其次为SM和FM处理。这种方法节约用水，同时对生产力的影响最小，因此是一种可持续的水管理战略。

{"title":"Effects of deficit irrigation at different phenological periods on yield, quality, and water productivity of Xanthoceras sorbifolium Bunge in the Horqin Sandy Land of China","authors":"Shuqin Song , Benye Xi , Yang Liu , Jing Zhong , Xiaoxia Wang , Liming Jia","doi":"10.1016/j.agwat.2026.110130","DOIUrl":"10.1016/j.agwat.2026.110130","url":null,"abstract":"<div><div>In arid and semi-arid regions, improving water efficiency is imperative for the sustainable advancement of both forestry and agriculture. The water demand period for <em>Xanthoceras sorbifolium</em> Bunge, an economically and ecologically important tree grown in the Horqin Sandy Land of China, is unknown. This knowledge gap has hindered the development of optimized deficit irrigation (DI) schemes aimed at conserving water while maintaining yield and quality. To identify this key period and better understand the water-use dynamics of this species, we conducted a two-year field experiment (2021–2022). Eight irrigation treatments were applied across three key phenological stages: flowering (F), fruit setting to expansion (S), and fruit-expansion to maturity (M). The irrigation treatments included full-stage (FSM), two-stage (FS, FM, and SM), single-stage (F, S, and M), and no irrigation (NI). The application of DI decreased fruit yields by 8.36–58.01 % (p < 0.05), while two-stage irrigation significantly reduced water consumption and evapotranspiration compared with full-stage irrigation (p < 0.01). FS significantly improved water productivity (<span><math><mi>WP</mi></math></span>), irrigation water productivity (<span><math><msub><mrow><mi>WP</mi></mrow><mrow><mi>I</mi></mrow></msub></math></span>), and fruit quality. All two-stage irrigation treatments demonstrated yield response factors (<span><math><msub><mrow><mi>k</mi></mrow><mrow><mi>y</mi></mrow></msub></math></span>) < 1. The FS treatment reduced irrigation volume by 43.7 %, while the yield decreased by only 8.36 %, suggesting that the irrigation savings did not significantly compromise yield. In summary, the FS treatment is recommended as the most optimal irrigation schedule, followed by SM and FM, for the production of <em>X. sorbifolium</em> in drylands. This approach conserves water while minimally impacting productivity, thus representing a sustainable water management strategy.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"325 ","pages":"Article 110130"},"PeriodicalIF":6.5,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

From past to future: Spatiotemporal insights into drought stress on Belgian pome fruit production for 1991–2090 从过去到未来：1991-2090年干旱胁迫对比利时梨果生产的时空影响

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-16 DOI: 10.1016/j.agwat.2026.110151

Brecht Bamps , Anne Gobin , Ben Somers , Jos Van Orshoven

Recurring drought episodes cause significant economic losses and uncertainty in the Belgian pome fruit sector, requiring effective risk management strategies. Considering the expected increase in frequency, intensity and duration of drought episodes, our study focuses on quantifying the associated location-specific hazard for apple and pear orchards in Belgium. A spatially distributed implementation of the growth model AquaCrop was used to calculate a daily soil water balance at a 12.5 × 12.5 km spatial resolution based on nine bias corrected regional climate model projections from the EURO-CORDEX ensemble for emission scenarios RCP4.5 and RCP8.5. The spatiotemporal evolution of the precipitation deficit (mm), transpiration deficit (mm) and net irrigation requirement (mm) were used to assess current and future meteorological and agricultural drought hazards. We find a spatially averaged increase of the transpiration deficit of 61.4% (39.0%) and the net irrigation requirement of 32.3 % (18.5%) comparing a period in the far future (2061–2090) to a baseline period (1991–2020) under scenario RCP8.5 (RCP4.5). In addition, corresponding interquartile ranges of 204.2% (77.6%) and 30.8% (10.3%) show a large inter-climate model variability in the simulation results. The projected increase of the precipitation deficit during the summer months causes agricultural drought in a spatially heterogeneous manner, relating to the soil total available water content and depth to the shallow groundwater table. Despite the overall increase in drought hazard, the suitability of the current production regions is not projected to change relative to other agricultural regions in Belgium regarding agricultural drought.

经常性的干旱给比利时的梨类水果部门造成重大的经济损失和不确定性，需要有效的风险管理战略。考虑到预计干旱事件的频率、强度和持续时间会增加，我们的研究重点是量化比利时苹果和梨果园相关的特定地点危害。利用生长模式AquaCrop的空间分布式实现，基于EURO-CORDEX系统对RCP4.5和RCP8.5排放情景的9个校正偏差的区域气候模式预估，计算了12.5 × 12.5 km空间分辨率下的日土壤水分平衡。利用降水亏缺量（mm）、蒸腾亏缺量（mm）和净灌溉需要量（mm）的时空演变特征来评估当前和未来的气象和农业干旱灾害。研究发现，在RCP8.5 （RCP4.5）情景下，远未来时期（2061-2090）与基准期（1991-2020）相比，蒸腾亏缺的空间平均增幅为61.4%(39.0%)，净灌溉需水量为32.3% %（18.5%）。另外，对应的四分位数区间为204.2%（77.6%）和30.8%(10.3%)，在模拟结果中表现出较大的气候间模式变率。预估夏季降水亏缺的增加导致农业干旱的空间异质性，与土壤总有效含水量和浅层地下水位深度有关。尽管干旱灾害总体上有所增加，但预计当前生产区在农业干旱方面相对于比利时其他农业区的适宜性不会发生变化。

{"title":"From past to future: Spatiotemporal insights into drought stress on Belgian pome fruit production for 1991–2090","authors":"Brecht Bamps , Anne Gobin , Ben Somers , Jos Van Orshoven","doi":"10.1016/j.agwat.2026.110151","DOIUrl":"10.1016/j.agwat.2026.110151","url":null,"abstract":"<div><div>Recurring drought episodes cause significant economic losses and uncertainty in the Belgian pome fruit sector, requiring effective risk management strategies. Considering the expected increase in frequency, intensity and duration of drought episodes, our study focuses on quantifying the associated location-specific hazard for apple and pear orchards in Belgium. A spatially distributed implementation of the growth model AquaCrop was used to calculate a daily soil water balance at a 12.5 × 12.5 km spatial resolution based on nine bias corrected regional climate model projections from the EURO-CORDEX ensemble for emission scenarios RCP4.5 and RCP8.5. The spatiotemporal evolution of the precipitation deficit (mm), transpiration deficit (mm) and net irrigation requirement (mm) were used to assess current and future meteorological and agricultural drought hazards. We find a spatially averaged increase of the transpiration deficit of 61.4% (39.0%) and the net irrigation requirement of 32.3 % (18.5%) comparing a period in the far future (2061–2090) to a baseline period (1991–2020) under scenario RCP8.5 (RCP4.5). In addition, corresponding interquartile ranges of 204.2% (77.6%) and 30.8% (10.3%) show a large inter-climate model variability in the simulation results. The projected increase of the precipitation deficit during the summer months causes agricultural drought in a spatially heterogeneous manner, relating to the soil total available water content and depth to the shallow groundwater table. Despite the overall increase in drought hazard, the suitability of the current production regions is not projected to change relative to other agricultural regions in Belgium regarding agricultural drought.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"325 ","pages":"Article 110151"},"PeriodicalIF":6.5,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145969403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimizing maize canopy structure to enhance resource use efficiency in maize-soybean intercropping systems 优化玉米冠层结构提高玉米-大豆间作系统资源利用效率

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-16 DOI: 10.1016/j.agwat.2026.110161

Shiming Duan , Bin Du , Wencong Chen , Shichao Chen , Qi Liao , Manoj Shukla , Taisheng Du

In water-limited regions, optimizing irrigation is fundamental to the sustainability of maize-soybean intercropping. However, the synergistic mechanisms through which irrigation modulates canopy structure to alleviate interspecific light competition and enhance water productivity remain poorly understood. This study investigated the ecophysiological responses of the intercropping system to the interplay between maize canopy architectures and deficit irrigation strategies. A two-year field experiment was conducted in northwest China, a representative arid region. The sole cropping treatments included compact maize (M1), spreading maize (M2), and shade-tolerant soybean (S), while the intercropping treatments comprised M1S and M2S. Based on the traditional irrigation (full irrigation for both maize and soybean), single-crop deficit irrigation and dual-crop deficit irrigation strategies were implemented. Results showed that the compact maize architecture inherently improved light transmittance (Tau, +30.3 %), boosting photosynthetically active radiation (PAR, +20.8 %) and net photosynthetic rate (Pn, +13.9 %) for intercropped soybean. More significantly, the "maize-full, soybean-deficit" irrigation strategy synergistically modulated maize morphology, further increasing its canopy Tau by 19.0 %. This architectural modification amplified PAR for soybean by 40.9 % and its Pn by 46.4 %, effectively mitigating light competition and balancing interspecific aggressivity (Ams ≈ 0). Critically, this integrated strategy achieved substantial water savings (9.4–16.6 % less irrigation) without compromising system yield, while significantly increasing irrigation water productivity (IWP, +10.2 %) and water equivalent ratio (WER, +13.7 %). Our findings demonstrate that synergistic regulation of canopy architecture via precision irrigation is a potent strategy for achieving water-saving and high-yield goals in intercropping systems, offering a valuable paradigm for sustainable agriculture in arid zones globally.

在缺水地区，优化灌溉对玉米-大豆间作的可持续性至关重要。然而，灌溉调节冠层结构以缓解种间光竞争和提高水分生产力的协同机制仍不清楚。研究了间作制度对玉米冠层结构与亏缺灌溉策略相互作用的生理生态响应。在具有代表性的西北干旱区进行了为期两年的田间试验。单作处理包括密实玉米（M1）、铺布玉米（M2）和耐荫大豆(S)，间作处理包括M1S和M2S。在传统灌溉方式（玉米和大豆全灌）的基础上，实施单亏灌和双亏灌策略。结果表明，紧凑的玉米结构内在地提高了间作大豆的透光率（Tau, +30.3 %），提高了光合有效辐射（PAR, +20.8 %）和净光合速率（Pn, +13.9 %）。更显著的是，“满玉米亏大豆”灌溉策略协同调节了玉米形态，进一步增加了19.0 %的冠层Tau。这种结构修饰使大豆的PAR和Pn分别提高了40.9 %和46.4 %，有效地缓解了光竞争，平衡了种间攻击性（Ams≈0）。关键是，这种综合策略在不影响系统产量的情况下实现了大量节水（灌溉减少9.4-16.6 %），同时显著提高了灌溉水生产率（IWP, +10.2 %）和水当量比（WER, +13.7 %）。我们的研究结果表明，通过精确灌溉对冠层结构进行协同调节是实现间作系统节水高产目标的有效策略，为全球干旱地区的可持续农业提供了有价值的范例。

{"title":"Optimizing maize canopy structure to enhance resource use efficiency in maize-soybean intercropping systems","authors":"Shiming Duan , Bin Du , Wencong Chen , Shichao Chen , Qi Liao , Manoj Shukla , Taisheng Du","doi":"10.1016/j.agwat.2026.110161","DOIUrl":"10.1016/j.agwat.2026.110161","url":null,"abstract":"<div><div>In water-limited regions, optimizing irrigation is fundamental to the sustainability of maize-soybean intercropping. However, the synergistic mechanisms through which irrigation modulates canopy structure to alleviate interspecific light competition and enhance water productivity remain poorly understood. This study investigated the ecophysiological responses of the intercropping system to the interplay between maize canopy architectures and deficit irrigation strategies. A two-year field experiment was conducted in northwest China, a representative arid region. The sole cropping treatments included compact maize (M1), spreading maize (M2), and shade-tolerant soybean (S), while the intercropping treatments comprised M1S and M2S. Based on the traditional irrigation (full irrigation for both maize and soybean), single-crop deficit irrigation and dual-crop deficit irrigation strategies were implemented. Results showed that the compact maize architecture inherently improved light transmittance (Tau, +30.3 %), boosting photosynthetically active radiation (PAR, +20.8 %) and net photosynthetic rate (Pn, +13.9 %) for intercropped soybean. More significantly, the \"maize-full, soybean-deficit\" irrigation strategy synergistically modulated maize morphology, further increasing its canopy Tau by 19.0 %. This architectural modification amplified PAR for soybean by 40.9 % and its Pn by 46.4 %, effectively mitigating light competition and balancing interspecific aggressivity (Ams ≈ 0). Critically, this integrated strategy achieved substantial water savings (9.4–16.6 % less irrigation) without compromising system yield, while significantly increasing irrigation water productivity (IWP, +10.2 %) and water equivalent ratio (WER, +13.7 %). Our findings demonstrate that synergistic regulation of canopy architecture via precision irrigation is a potent strategy for achieving water-saving and high-yield goals in intercropping systems, offering a valuable paradigm for sustainable agriculture in arid zones globally.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"325 ","pages":"Article 110161"},"PeriodicalIF":6.5,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Stomatal conductance modeling for drip-irrigated kiwifruit in seasonal drought regions of South China: Evaluation of improved empirical models and interpretable machine learning approaches 华南季节性旱区滴灌猕猴桃气孔导度建模：改进的经验模型和可解释性机器学习方法的评价

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-16 DOI: 10.1016/j.agwat.2026.110153

Shunsheng Zheng , Ningbo Cui , Quanshan Liu , Shouzheng Jiang , Daozhi Gong , Xiaoxian Zhang

Accurate modeling of stomatal conductance (g_s) enhances understanding of plant water relations and supports advancements in eco-physiological modeling and adaptive irrigation practices. This study provides a comprehensive evaluation of g_s modeling for drip-irrigated kiwifruit through parallel development of three Jarvis-type empirical models (JV, JV1, JV2) and five machine learning algorithms (XGBoost, LightGBM, CatBoost, SVR, LR) based on three years of field measurements comprising synchronized records of g_s and key environmental drivers. Models were assessed via year-wise grouped cross-validation, with performance measured by R², RMSE, and MAE, and interpretability analyzed using SHapley Additive exPlanations (SHAP) and Partial Dependence Plots (PDPs). Results showed that deficit irrigation significantly reduced g_s, with sensitivity being most pronounced during stage II. The incorporation of soil water content (SWC) substantially improved the accuracy of both empirical and machine learning models. Among empirical models, JV2, featuring a stage-specific nonlinear SWC response function, demonstrated the highest accuracy (R² ranging from 0.736 to 0.814) and minimized bias under extreme SWC conditions. Using vapor pressure deficit (VPD), air temperature (Ta), photosynthetically active radiation (PAR), and SWC as input variables, CatBoost outperformed both empirical models and other machine learning algorithms across all growth stages (R² = 0.815–0.839; RMSE = 0.065–0.076 mol m⁻² s⁻¹; MAE = 0.054–0.064 mol m⁻² s⁻¹). SHAP analysis and PDPs identified VPD as the dominant driver of g_s variation, followed by SWC. Overall, the improved JV2 model offers a structurally transparent framework for g_s estimation with acceptable accuracy, while CatBoost combined with SHAP analysis and PDPs provides superior predictive performance and robust interpretability under complex environmental conditions. These findings support the reliable modeling and regulation of kiwifruit g_s under varying SWC scenarios in drip-irrigated orchards.

气孔导度（gs）的精确建模增强了对植物水分关系的理解，并支持生态生理建模和适应性灌溉实践的进步。本研究基于三年的田间测量数据，包括同步记录的gs和关键环境驱动因素，通过并行开发三种jarvis型经验模型（JV、JV1、JV2）和五种机器学习算法（XGBoost、LightGBM、CatBoost、SVR、LR），对滴灌猕猴桃gs模型进行了全面评估。通过年度分组交叉验证对模型进行评估，使用R2、RMSE和MAE测量模型的性能，并使用SHapley加性解释（SHAP）和部分依赖图（pdp）分析模型的可解释性。结果表明，亏缺灌溉显著降低了gs，在II期敏感性最为明显。土壤含水量（SWC）的结合大大提高了经验模型和机器学习模型的准确性。在经验模型中，JV2具有特定阶段的非线性SWC响应函数，在极端SWC条件下具有最高的精度（R2范围为0.736 ~ 0.814）和最小的偏差。使用蒸汽压差（VPD）、空气温度（Ta）、光合有效辐射（PAR）和SWC作为输入变量，CatBoost在所有生长阶段都优于经验模型和其他机器学习算法（R2 = 0.815-0.839; RMSE = 0.065-0.076 mol m−2 s−1;MAE = 0.054-0.064 mol m−2 s−1）。SHAP和pdp分析表明VPD是gs变化的主要驱动因素，其次是SWC。总体而言，改进的JV2模型为gs估计提供了一个结构透明的框架，具有可接受的精度，而CatBoost结合SHAP分析和pdp提供了优越的预测性能和复杂环境条件下的鲁棒可解释性。这些发现支持了滴灌果园中猕猴桃在不同SWC情景下的可靠建模和调控。

{"title":"Stomatal conductance modeling for drip-irrigated kiwifruit in seasonal drought regions of South China: Evaluation of improved empirical models and interpretable machine learning approaches","authors":"Shunsheng Zheng , Ningbo Cui , Quanshan Liu , Shouzheng Jiang , Daozhi Gong , Xiaoxian Zhang","doi":"10.1016/j.agwat.2026.110153","DOIUrl":"10.1016/j.agwat.2026.110153","url":null,"abstract":"<div><div>Accurate modeling of stomatal conductance (g<sub>s</sub>) enhances understanding of plant water relations and supports advancements in eco-physiological modeling and adaptive irrigation practices. This study provides a comprehensive evaluation of g<sub>s</sub> modeling for drip-irrigated kiwifruit through parallel development of three Jarvis-type empirical models (JV, JV1, JV2) and five machine learning algorithms (XGBoost, LightGBM, CatBoost, SVR, LR) based on three years of field measurements comprising synchronized records of g<sub>s</sub> and key environmental drivers. Models were assessed via year-wise grouped cross-validation, with performance measured by R<sup>2</sup>, RMSE, and MAE, and interpretability analyzed using SHapley Additive exPlanations (SHAP) and Partial Dependence Plots (PDPs). Results showed that deficit irrigation significantly reduced g<sub>s</sub>, with sensitivity being most pronounced during stage II. The incorporation of soil water content (SWC) substantially improved the accuracy of both empirical and machine learning models. Among empirical models, JV2, featuring a stage-specific nonlinear SWC response function, demonstrated the highest accuracy (R<sup>2</sup> ranging from 0.736 to 0.814) and minimized bias under extreme SWC conditions. Using vapor pressure deficit (VPD), air temperature (Ta), photosynthetically active radiation (PAR), and SWC as input variables, CatBoost outperformed both empirical models and other machine learning algorithms across all growth stages (R<sup>2</sup> = 0.815–0.839; RMSE = 0.065–0.076 mol m<sup>−2</sup> s<sup>−1</sup>; MAE = 0.054–0.064 mol m<sup>−2</sup> s<sup>−1</sup>). SHAP analysis and PDPs identified VPD as the dominant driver of g<sub>s</sub> variation, followed by SWC. Overall, the improved JV2 model offers a structurally transparent framework for g<sub>s</sub> estimation with acceptable accuracy, while CatBoost combined with SHAP analysis and PDPs provides superior predictive performance and robust interpretability under complex environmental conditions. These findings support the reliable modeling and regulation of kiwifruit g<sub>s</sub> under varying SWC scenarios in drip-irrigated orchards.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"325 ","pages":"Article 110153"},"PeriodicalIF":6.5,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145969402","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Toward sustainable Brassica napus production: Optimizing fertilization regimes for yield, water, and nutrient efficiency 朝着可持续的甘蓝型油菜生产：优化施肥制度的产量，水分和养分效率

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-16 DOI: 10.1016/j.agwat.2026.110145

Qiang Zheng , Peng Song , Xin Wang , Siqi Han , Kai Zhang , Peng Hou , Peiling Yang

Enhancing both crop productivity and resource-use efficiency is essential for sustainable intensification, particularly in high-value vegetable systems. Conventional surface irrigation combined with one-time application of compound fertilizer often leads to poor synchronization between nutrient supply and crop demand, resulting in inefficient resource use and elevated environmental risks. This study investigated the effects of split N P K application via precision drip fertigation on soil conditions, crop performance, and water productivity in Brassica napus cultivated for edible shoots. Four fertilization treatments were compared: (T1) conventional surface irrigation with a single top-dressing of compound fertilizer, and three drip-fertigation regimes using organic water-soluble fertilizer-(T2) fertilizer applied in two equal splits, (T3) fertilizer applied in three equal splits, and (T4) fertilizer applied in four equal splits-synchronized with successive shoot-harvest stages. Results demonstrated that drip fertigation significantly improved subsoil moisture and reduced soil electrical conductivity. Notably, treatments T3 and T4 enhanced nitrate nitrogen availability, stem diameter, plant height, and biomass accumulation. Compared with T1, T3 and T4 increased shoot yield by 17.1 % and 9.31 %, irrigation water productivity (WP_I) by 17.1 % and 9.17 %, and partial factor productivity of N fertilizer (PFP_N) by 1.62 % and 7.63 %, respectively. Structural equation modeling identified stem diameter, dry weight, and inflorescence number as key yield drivers, while PFP_N was affected by both morphological and physiological traits. A combined AHP-EWM evaluation framework identified T3 as the optimal fertilization regime. The fertigation strategy and evaluation framework developed here offer a practical and scalable pathway for enhancing water-nutrient efficiency in multi-cut vegetable systems, supporting sustainable intensification beyond the study region.

提高作物生产力和资源利用效率对于可持续集约化至关重要，特别是在高价值蔬菜系统中。传统的地表灌溉与一次性施用复合肥相结合，往往导致养分供给与作物需求不同步，导致资源利用效率低下，环境风险加大。研究了精确滴灌分步施氮磷钾对甘蓝型可食苗土壤条件、作物生产性能和水分生产力的影响。比较了4种施肥处理：（T1）常规地表灌溉单次追肥，和3种有机水溶性肥料滴灌施肥方案——（T2）两次等距施肥，（T3）三等距施肥，（T4）四等距施肥，与连续的芽收期同步。结果表明，滴灌施肥显著改善了土壤水分，降低了土壤电导率。T3和T4处理显著提高了硝态氮有效性、茎粗、株高和生物量积累。与T1相比，T3和T4分别提高了地上部产量17.1 %和9.31 %，灌溉水生产力（WPI）分别提高了17.1 %和9.17 %，氮肥部分要素生产力（PFPN）分别提高了1.62 %和7.63 %。结构方程模型表明茎粗、干重和花序数是主要的产量驱动因素，而PFPN受形态和生理性状的共同影响。AHP-EWM综合评价框架确定T3为最佳施肥方案。这里制定的施肥策略和评价框架为提高多切蔬菜系统的水-养分效率提供了一条实用且可扩展的途径，支持研究区域以外的可持续集约化。

{"title":"Toward sustainable Brassica napus production: Optimizing fertilization regimes for yield, water, and nutrient efficiency","authors":"Qiang Zheng , Peng Song , Xin Wang , Siqi Han , Kai Zhang , Peng Hou , Peiling Yang","doi":"10.1016/j.agwat.2026.110145","DOIUrl":"10.1016/j.agwat.2026.110145","url":null,"abstract":"<div><div>Enhancing both crop productivity and resource-use efficiency is essential for sustainable intensification, particularly in high-value vegetable systems. Conventional surface irrigation combined with one-time application of compound fertilizer often leads to poor synchronization between nutrient supply and crop demand, resulting in inefficient resource use and elevated environmental risks. This study investigated the effects of split N P K application via precision drip fertigation on soil conditions, crop performance, and water productivity in <em>Brassica napus</em> cultivated for edible shoots. Four fertilization treatments were compared: (T1) conventional surface irrigation with a single top-dressing of compound fertilizer, and three drip-fertigation regimes using organic water-soluble fertilizer-(T2) fertilizer applied in two equal splits, (T3) fertilizer applied in three equal splits, and (T4) fertilizer applied in four equal splits-synchronized with successive shoot-harvest stages. Results demonstrated that drip fertigation significantly improved subsoil moisture and reduced soil electrical conductivity. Notably, treatments T3 and T4 enhanced nitrate nitrogen availability, stem diameter, plant height, and biomass accumulation. Compared with T1, T3 and T4 increased shoot yield by 17.1 % and 9.31 %, irrigation water productivity (WP<sub>I</sub>) by 17.1 % and 9.17 %, and partial factor productivity of N fertilizer (PFP<sub>N</sub>) by 1.62 % and 7.63 %, respectively. Structural equation modeling identified stem diameter, dry weight, and inflorescence number as key yield drivers, while PFP<sub>N</sub> was affected by both morphological and physiological traits. A combined AHP-EWM evaluation framework identified T3 as the optimal fertilization regime. The fertigation strategy and evaluation framework developed here offer a practical and scalable pathway for enhancing water-nutrient efficiency in multi-cut vegetable systems, supporting sustainable intensification beyond the study region.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"325 ","pages":"Article 110145"},"PeriodicalIF":6.5,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145969401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Opposite responses of water use efficiency and carbon use efficiency to agricultural drought during winter wheat growth period in the core area of grain production in the ancient course of Yellow River 黄河古河道粮食生产核心区冬小麦生育期水分利用效率和碳利用效率对农业干旱的相反响应

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-15 DOI: 10.1016/j.agwat.2026.110144

Xiaojuan Ren , Guodong Li , Xuejian Sun , Hui He , Qingtao Zhao , Longsheng Wang , Yunfei Gong , Bingqian Han , Chenxi Cao

In the context of global climate change, agricultural drought threatens food security in the ancient course of Yellow River, a key grain production region in China. Based on the temperature vegetation dryness index (TVDI), carbon use efficiency (CUE), and water use efficiency (WUE), this study employed methods such as random forest model and partial dependence plots to systematically evaluate the spatiotemporal variations in winter wheat carbon-water use efficiency and their threshold responses to agricultural drought and environmental factors. Results indicated that the winter wheat growing period was dominated by mild drought, with TVDI exhibiting a fluctuating upward trend. In drought-intensified regions, 63.9 % of the area showed a decline in CUE, while 98 % experienced an increase in WUE. CUE and WUE exhibited distinct inverse responses at TVDI = 0.5, and this threshold can serve as a key decision point for irrigation initiation. Multi-factor synergistic analysis revealed that mild drought combined with lower hydrothermal conditions or high vapor pressure deficit tended to form low-CUE zones, whereas mild drought combined with suitable hydrothermal conditions optimized WUE. Conversely, severe drought led to carbon-water metabolic damage. These findings provide a quantitative basis for differentiated irrigation management. This study elucidates the threshold responses of winter wheat carbon-water cycling to drought stress and multi-factor regulatory mechanisms, providing a quantitative basis for precision irrigation and optimal agricultural water resource allocation in core grain production regions.

在全球气候变化背景下，黄河古河道作为中国重要的粮食产区，农业干旱对其粮食安全构成威胁。基于温度植被干燥指数（TVDI）、碳利用效率（CUE）和水分利用效率（WUE），采用随机森林模型和部分依赖图等方法，系统评价了冬小麦碳水利用效率的时空变化及其对农业干旱和环境因子的阈值响应。结果表明：冬小麦生育期以轻度干旱为主，TVDI呈波动上升趋势；在干旱加剧区，63.9 %的面积出现CUE下降，而98 %的面积出现WUE增加。在TVDI = 0.5时，CUE和WUE表现出明显的反向响应，该阈值可以作为灌溉启动的关键决策点。多因素协同分析表明，轻度干旱与低热液条件或高蒸汽压亏缺相结合会形成低cue区，而轻度干旱与适宜的热液条件相结合会优化WUE。相反，严重干旱导致碳水代谢损伤。这些发现为差别化灌溉管理提供了定量依据。本研究阐明了冬小麦碳水循环对干旱胁迫的阈值响应及其多因素调控机制，为粮食主产区精准灌溉和农业水资源优化配置提供定量依据。

{"title":"Opposite responses of water use efficiency and carbon use efficiency to agricultural drought during winter wheat growth period in the core area of grain production in the ancient course of Yellow River","authors":"Xiaojuan Ren , Guodong Li , Xuejian Sun , Hui He , Qingtao Zhao , Longsheng Wang , Yunfei Gong , Bingqian Han , Chenxi Cao","doi":"10.1016/j.agwat.2026.110144","DOIUrl":"10.1016/j.agwat.2026.110144","url":null,"abstract":"<div><div>In the context of global climate change, agricultural drought threatens food security in the ancient course of Yellow River, a key grain production region in China. Based on the temperature vegetation dryness index (TVDI), carbon use efficiency (CUE), and water use efficiency (WUE), this study employed methods such as random forest model and partial dependence plots to systematically evaluate the spatiotemporal variations in winter wheat carbon-water use efficiency and their threshold responses to agricultural drought and environmental factors. Results indicated that the winter wheat growing period was dominated by mild drought, with TVDI exhibiting a fluctuating upward trend. In drought-intensified regions, 63.9 % of the area showed a decline in CUE, while 98 % experienced an increase in WUE. CUE and WUE exhibited distinct inverse responses at TVDI = 0.5, and this threshold can serve as a key decision point for irrigation initiation. Multi-factor synergistic analysis revealed that mild drought combined with lower hydrothermal conditions or high vapor pressure deficit tended to form low-CUE zones, whereas mild drought combined with suitable hydrothermal conditions optimized WUE. Conversely, severe drought led to carbon-water metabolic damage. These findings provide a quantitative basis for differentiated irrigation management. This study elucidates the threshold responses of winter wheat carbon-water cycling to drought stress and multi-factor regulatory mechanisms, providing a quantitative basis for precision irrigation and optimal agricultural water resource allocation in core grain production regions.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"324 ","pages":"Article 110144"},"PeriodicalIF":6.5,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Soil water and salt migration characteristics under freezing saline water irrigation: An isotope tracing and phase analysis perspective 冰冻盐水灌溉条件下土壤水盐迁移特征：同位素示踪和物相分析

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-15 DOI: 10.1016/j.agwat.2026.110150

Junjie Li , Zhongyi Qu , Wei Yang , Gerile Hasi , Liping Wang , Ruxin Zhang , Dongliang Zhang

In arid cold regions, optimizing saline water use for soil improvement presents significant challenges in water-salt regulation. This study investigates freezing saline water irrigation (FSWI) as a promising strategy, examining its effects on soil water-heat-salt dynamics through a three-year field experiment with varying irrigation amounts (0, 90, 180, and 270 mm, designated FSWI₀–FSWI₂₇₀). Results demonstrate that FSWI considerably moderates deep soil temperature variation during freeze-thaw cycles, shortening the freezing period by up to 8 days. Meltwater from surface ice effectively leaches shallow soil (0–40 cm), evidenced by decreased soil water δ¹ ⁸O. Although irrigation introduced salt (peak salt load: 3.15 kg m⁻²), freezing-induced phase changes promoted salt migration to deeper layers, reducing shallow soil salinity by 40.47 % under FSWI₁₈₀ after thawing. Conversely, excessive irrigation (FSWI₂₇₀) increased surface salinity by 4.67 %, indicating salinization risk. The study identifies 180 mm as the optimal irrigation volume. These findings elucidate the mechanisms of water-salt transport under FSWI and support its potential as a viable strategy for saline soil amelioration and water conservation in cold regions.

在干旱寒冷地区，优化盐碱水利用以改善土壤是水盐调控面临的重大挑战。本研究研究了冷冻盐水灌溉（FSWI）作为一种有前景的策略，通过为期三年的不同灌溉量（0、90、180和270 mm，指定FSWI0-FSWI270）的田间试验，研究了其对土壤水-热-盐动态的影响。结果表明，FSWI显著调节了冻融循环中深层土壤温度的变化，使冻结期缩短了8天。表层冰的融水有效地淋滤浅层土壤（0-40 cm），土壤水分δ¹ ⁸O降低。虽然灌溉引入了盐（峰值盐负荷：3.15 kg m⁻²），但冻结引起的相变促进了盐向更深层的迁移，在FSWI180下，解冻后浅层土壤的盐度降低了40.47 %。相反，过量灌溉（FSWI270）使地表盐度增加4.67 %，表明存在盐碱化风险。研究确定180 mm为最佳灌溉水量。这些发现阐明了FSWI下水盐运移的机制，并支持其作为寒地盐渍土改良和保水的可行策略的潜力。

{"title":"Soil water and salt migration characteristics under freezing saline water irrigation: An isotope tracing and phase analysis perspective","authors":"Junjie Li , Zhongyi Qu , Wei Yang , Gerile Hasi , Liping Wang , Ruxin Zhang , Dongliang Zhang","doi":"10.1016/j.agwat.2026.110150","DOIUrl":"10.1016/j.agwat.2026.110150","url":null,"abstract":"<div><div>In arid cold regions, optimizing saline water use for soil improvement presents significant challenges in water-salt regulation. This study investigates freezing saline water irrigation (FSWI) as a promising strategy, examining its effects on soil water-heat-salt dynamics through a three-year field experiment with varying irrigation amounts (0, 90, 180, and 270 mm, designated FSWI<sub>0</sub>–FSWI<sub>270</sub>). Results demonstrate that FSWI considerably moderates deep soil temperature variation during freeze-thaw cycles, shortening the freezing period by up to 8 days. Meltwater from surface ice effectively leaches shallow soil (0–40 cm), evidenced by decreased soil water δ¹ ⁸O. Although irrigation introduced salt (peak salt load: 3.15 kg m⁻²), freezing-induced phase changes promoted salt migration to deeper layers, reducing shallow soil salinity by 40.47 % under FSWI<sub>180</sub> after thawing. Conversely, excessive irrigation (FSWI<sub>270</sub>) increased surface salinity by 4.67 %, indicating salinization risk. The study identifies 180 mm as the optimal irrigation volume. These findings elucidate the mechanisms of water-salt transport under FSWI and support its potential as a viable strategy for saline soil amelioration and water conservation in cold regions.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"324 ","pages":"Article 110150"},"PeriodicalIF":6.5,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Water governance matters: Self-supplied versus collective irrigation services and their impact on technical efficiency and production in olive oil farming 水治理问题：自供与集体灌溉服务及其对橄榄油农业技术效率和产量的影响

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-15 DOI: 10.1016/j.agwat.2026.110148

S. Russo , L. Mirra , F. Caracciolo , G. Giannoccaro

This study examines the impact of irrigation water services in shaping the economic performance of olive farms, with particular attention to their effects on production variability and technical efficiency. Employing a stochastic frontier analysis, we evaluate how different modes of irrigation service provision (self-supplied versus collective) affect farm-level outcomes. The empirical analysis is based on 954 observations from the Farm Accountancy Data Network, covering olive farms in Apulia, Italy’s principal olive-producing region, over the 2016–2019 period. The findings point out that self-supplied irrigation is associated with lower variability in both output and technical efficiency, reflecting greater reliability of water service. Conversely, farms relying on collective irrigation services experience higher variability in technical efficiency. These results highlight the importance of targeted policy interventions aimed at enhancing the reliability of collective irrigation services, also integrating current water allocation mechanisms with pricing differentiation policy.

本研究考察了灌溉用水服务在塑造橄榄农场经济绩效方面的影响，特别关注它们对生产变异性和技术效率的影响。采用随机前沿分析，我们评估了不同的灌溉服务提供模式（自供与集体）如何影响农场层面的结果。实证分析基于农场会计数据网络（Farm accounting Data Network）在2016-2019年期间对意大利主要橄榄产区普利亚（Apulia）橄榄农场的954项观察结果。研究结果指出，自给灌溉与产量和技术效率的变异性较低有关，反映了供水服务的可靠性较高。相反，依靠集体灌溉服务的农场在技术效率方面的变化较大。这些结果强调了有针对性的政策干预的重要性，这些政策干预旨在提高集体灌溉服务的可靠性，并将当前的水资源分配机制与价格差异化政策结合起来。

{"title":"Water governance matters: Self-supplied versus collective irrigation services and their impact on technical efficiency and production in olive oil farming","authors":"S. Russo , L. Mirra , F. Caracciolo , G. Giannoccaro","doi":"10.1016/j.agwat.2026.110148","DOIUrl":"10.1016/j.agwat.2026.110148","url":null,"abstract":"<div><div>This study examines the impact of irrigation water services in shaping the economic performance of olive farms, with particular attention to their effects on production variability and technical efficiency. Employing a stochastic frontier analysis, we evaluate how different modes of irrigation service provision (self-supplied versus collective) affect farm-level outcomes. The empirical analysis is based on 954 observations from the Farm Accountancy Data Network, covering olive farms in Apulia, Italy’s principal olive-producing region, over the 2016–2019 period. The findings point out that self-supplied irrigation is associated with lower variability in both output and technical efficiency, reflecting greater reliability of water service. Conversely, farms relying on collective irrigation services experience higher variability in technical efficiency. These results highlight the importance of targeted policy interventions aimed at enhancing the reliability of collective irrigation services, also integrating current water allocation mechanisms with pricing differentiation policy.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"324 ","pages":"Article 110148"},"PeriodicalIF":6.5,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Interpretable machine learning workflow for estimating reference crop evapotranspiration in China's five major dry-wet regions using limited meteorological data 基于有限气象数据估算中国五大干湿区参考作物蒸散量的可解释机器学习工作流程

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-15 DOI: 10.1016/j.agwat.2026.110143

Ziyu Guan , Changhai Qin , Yong Zhao , Junlin Qu , Rong Liu , Yuan Liu , Wenxin Che , Tao Wang

Accurate estimation of reference crop evapotranspiration (

{ET}_{0}

) is crucial for improving water use efficiency and the design and operation of agricultural water management systems. Machine learning (ML) can accurately estimate

{ET}_{0}

across different climatic zones in China when meteorological data are limited, but its “black box” nature restricts interpretability. This study developed an interpretable machine learning workflow to enhance

{ET}_{0}

prediction transparency. It utilized four meta-heuristic algorithms and four machine learning algorithms based on meteorological data observed at 2382 stations across five climatic zones in China from 1960 to 2022. Model performance was evaluated using mean absolute error (MAE), root mean square error (RMSE), Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (

R^{2}

), and global performance index (GPI). Results indicate that the XGBoost model optimized by the Grey Wolf Optimization (GWO) algorithm (GWO-XGB) achieved the highest fitting accuracy. Its Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (

R^{2}

), and global prediction index (GPI) were 0.087, 0.116, 0.993, 0.993, and 1.783, respectively. Cross-validation across basins revealed that GWO-XGB maintained an

R^{2}

above 0.96 on the independent validation dataset, indicating robust stability and generalization of the interpretable machine learning framework in

{ET}_{0}

prediction. SHAP accurately captured underlying hydrological and climatic processes, identifying solar radiation and extreme temperatures as the primary predictors of

{ET}_{0}

, while humidity and wind speed exerted lesser influences. This study offers a promising approach for precise

{ET}_{0}

estimation in data-scarce regions, thereby supporting scientific water resource management and achieving water conservation and efficiency goals. The open-source prediction application is available at https://github.com/guangian.

准确估算参考作物蒸散量（ET0）对提高水分利用效率和农业水管理系统的设计和运行至关重要。在气象数据有限的情况下，机器学习（ML）可以准确估计中国不同气候带的ET0，但其“黑匣子”性质限制了其可解释性。本研究开发了一个可解释的机器学习工作流程，以提高ET0预测的透明度。基于1960 - 2022年中国5个气候带2382个站点的气象观测数据，采用了4种元启发式算法和4种机器学习算法。采用平均绝对误差（MAE）、均方根误差（RMSE）、Nash-Sutcliffe效率系数（NSE）、决定系数（R2）和全局性能指数（GPI）评价模型的性能。结果表明，采用灰狼优化（GWO）算法（GWO- xgb）优化的XGBoost模型拟合精度最高。平均绝对误差（MAE）、均方根误差（RMSE）、Nash-Sutcliffe效率系数（NSE）、决定系数（R2）和全局预测指数（GPI）分别为0.087、0.116、0.993、0.993和1.783。跨流域的交叉验证表明，独立验证数据集的GWO-XGB R2保持在0.96以上，表明可解释机器学习框架在ET0预测中的鲁棒稳定性和泛化性。SHAP准确地捕获了潜在的水文和气候过程，确定太阳辐射和极端温度是ET0的主要预测因子，而湿度和风速的影响较小。该研究为数据稀缺地区的ET0精确估算提供了一种有希望的方法，从而支持科学的水资源管理，实现节水和效率目标。开源预测应用程序可在https://github.com/guangian上获得。

{"title":"Interpretable machine learning workflow for estimating reference crop evapotranspiration in China's five major dry-wet regions using limited meteorological data","authors":"Ziyu Guan , Changhai Qin , Yong Zhao , Junlin Qu , Rong Liu , Yuan Liu , Wenxin Che , Tao Wang","doi":"10.1016/j.agwat.2026.110143","DOIUrl":"10.1016/j.agwat.2026.110143","url":null,"abstract":"<div><div>Accurate estimation of reference crop evapotranspiration (<span><math><msub><mrow><mi>ET</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span>) is crucial for improving water use efficiency and the design and operation of agricultural water management systems. Machine learning (ML) can accurately estimate <span><math><msub><mrow><mi>ET</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span> across different climatic zones in China when meteorological data are limited, but its “black box” nature restricts interpretability. This study developed an interpretable machine learning workflow to enhance <span><math><msub><mrow><mi>ET</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span> prediction transparency. It utilized four meta-heuristic algorithms and four machine learning algorithms based on meteorological data observed at 2382 stations across five climatic zones in China from 1960 to 2022. Model performance was evaluated using mean absolute error (MAE), root mean square error (RMSE), Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>), and global performance index (GPI). Results indicate that the XGBoost model optimized by the Grey Wolf Optimization (GWO) algorithm (GWO-XGB) achieved the highest fitting accuracy. Its Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>), and global prediction index (GPI) were 0.087, 0.116, 0.993, 0.993, and 1.783, respectively. Cross-validation across basins revealed that GWO-XGB maintained an <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> above 0.96 on the independent validation dataset, indicating robust stability and generalization of the interpretable machine learning framework in <span><math><msub><mrow><mi>ET</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span> prediction. SHAP accurately captured underlying hydrological and climatic processes, identifying solar radiation and extreme temperatures as the primary predictors of <span><math><msub><mrow><mi>ET</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span>, while humidity and wind speed exerted lesser influences. This study offers a promising approach for precise <span><math><msub><mrow><mi>ET</mi></mrow><mrow><mn>0</mn></mrow></msub></math></span> estimation in data-scarce regions, thereby supporting scientific water resource management and achieving water conservation and efficiency goals. The open-source prediction application is available at <span><span>https://github.com/guangian</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"324 ","pages":"Article 110143"},"PeriodicalIF":6.5,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Shifting climatic sensitivities of drought-related yield gaps signal potential increases in irrigation reliance in the Yellow River Basin 与干旱相关的产量差距的气候敏感性变化表明黄河流域对灌溉依赖的潜在增加

IF 6.5 1区农林科学 Q1 AGRONOMY

Agricultural Water Management

Pub Date : 2026-01-15 DOI: 10.1016/j.agwat.2026.110137

Linchao Li , Zhongshan Xu , Yajie Zhang , Ning Yao , Yi Li , Qiang Yu , Hao Feng , Guijun Yang , Qinsi He

Climate change is intensifying drought risk and threatening food production, yet how drought-driven yield losses evolve with warming remains poorly quantified. Here, we combine an ensemble of nine crop models with 38 global climate models to quantify shifting sensitivities of the irrigated–rainfed yield gap to key climate drivers across the Yellow River Basin under SSP126, SSP245, and SSP585. We find that yield gaps increase for maize, soybean, and rice under future climates, while wheat exhibits a slight decrease. The precipitation is negatively associated with the yield gap, but this negative effect weakens (shifts toward zero) across large areas of the basin. For maize, 69.8 %, 66.7 %, and 77.8 % of grid cells show increasing sensitivity under SSP126, SSP245, and SSP585, respectively. This indicates rainfall is becoming less effective at narrowing the gap as atmospheric demand rises, implying greater reliance on irrigation. Evapotranspiration (ET) generally shows the opposite spatial pattern, except in some extremely arid upstream areas where ET-yield gap coupling weakens or even decouples due to high atmospheric demand. Yield gap sensitivity to atmospheric CO₂ is negative in most regions, consistent with improved water-use efficiency that reduces the yield gap; this effect is more pronounced in the arid upstream. By identifying spatiotemporal hotspots of intensifying yield-gap sensitivities, this study informs targeted irrigation investment and drought-time water-allocation prioritization, supporting climate-smart water management to stabilize production and long-term sustainability.

气候变化正在加剧干旱风险并威胁粮食生产，然而，干旱导致的产量损失如何随着气候变暖而演变，仍然缺乏量化。本文将9种作物模型与38个全球气候模型相结合，量化了SSP126、SSP245和SSP585条件下黄河流域灌雨产量缺口对主要气候驱动因素的变化敏感性。我们发现，在未来的气候条件下，玉米、大豆和水稻的产量差距会增加，而小麦的产量差距会略有减少。降水与产量缺口呈负相关，但这种负影响在盆地的大部分地区减弱（向零转移）。对于玉米，69.8 %、66.7 %和77.8 %的网格细胞在SSP126、SSP245和SSP585下的敏感性分别提高。这表明，随着大气需求的增加，降雨在缩小缺口方面的效果越来越差，这意味着对灌溉的依赖越来越大。蒸散发（ET）总体上呈现相反的空间格局，但在一些极端干旱的上游地区，由于高大气需求，蒸散发-产差耦合减弱甚至解耦。在大多数地区，产量缺口对大气CO2的敏感性为负，这与提高水分利用效率减少产量缺口相一致；这种影响在干旱的上游地区更为明显。通过识别产量差距敏感性加剧的时空热点，本研究为有针对性的灌溉投资和旱期水资源分配优先级提供信息，支持气候智慧型水资源管理，以稳定生产和长期可持续性。

{"title":"Shifting climatic sensitivities of drought-related yield gaps signal potential increases in irrigation reliance in the Yellow River Basin","authors":"Linchao Li , Zhongshan Xu , Yajie Zhang , Ning Yao , Yi Li , Qiang Yu , Hao Feng , Guijun Yang , Qinsi He","doi":"10.1016/j.agwat.2026.110137","DOIUrl":"10.1016/j.agwat.2026.110137","url":null,"abstract":"<div><div>Climate change is intensifying drought risk and threatening food production, yet how drought-driven yield losses evolve with warming remains poorly quantified. Here, we combine an ensemble of nine crop models with 38 global climate models to quantify shifting sensitivities of the irrigated–rainfed yield gap to key climate drivers across the Yellow River Basin under SSP126, SSP245, and SSP585. We find that yield gaps increase for maize, soybean, and rice under future climates, while wheat exhibits a slight decrease. The precipitation is negatively associated with the yield gap, but this negative effect weakens (shifts toward zero) across large areas of the basin. For maize, 69.8 %, 66.7 %, and 77.8 % of grid cells show increasing sensitivity under SSP126, SSP245, and SSP585, respectively. This indicates rainfall is becoming less effective at narrowing the gap as atmospheric demand rises, implying greater reliance on irrigation. Evapotranspiration (ET) generally shows the opposite spatial pattern, except in some extremely arid upstream areas where ET-yield gap coupling weakens or even decouples due to high atmospheric demand. Yield gap sensitivity to atmospheric CO<sub>2</sub> is negative in most regions, consistent with improved water-use efficiency that reduces the yield gap; this effect is more pronounced in the arid upstream. By identifying spatiotemporal hotspots of intensifying yield-gap sensitivities, this study informs targeted irrigation investment and drought-time water-allocation prioritization, supporting climate-smart water management to stabilize production and long-term sustainability.</div></div>","PeriodicalId":7634,"journal":{"name":"Agricultural Water Management","volume":"324 ","pages":"Article 110137"},"PeriodicalIF":6.5,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974099","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0