Conventional integrated-structure nozzles suffer from fixed parameters and high maintenance costs, whereas modular nozzles allow flexible adjustment of structural parameters through interchangeable nozzle inserts, with atomization performance determined directly by the nozzle insert. This study uses a hollow-cone nozzle as the base structure and investigates the effects of insert material and structural parameters on atomization quality by varying one factor at a time. The experimental results indicate that, in terms of material, the YG6 carbide nozzle insert outperforms the 304 stainless steel and TiN-coated nozzle inserts; structurally, the spray angle increases as the outlet straight-section length decreases and decreases as the inlet cone angle increases. When the outlet straight-section length is 0.5 mm or smaller and the pressure ranges from 0.2 to 0.5 MPa, the droplet volume median diameter remains between 105 and 150 μm, with a relative span factor of less than 1.1, indicating fine atomization and uniform droplet distribution. It is also found that the nozzle inserts with a 60° inlet cone angle achieved the maximum spray angle (64.28°) at 0.5 MPa, while demonstrating favorable droplet size distribution and uniformity. Accordingly, the YG6 carbide nozzle insert combined with a 0.5 mm outlet straight-section length and a 60° inlet cone angle maximizes the spray angle, refines the droplet size, and ensures uniform distribution, offering a direct selection basis for the refined design of agricultural sprayers and precision pesticide application.

Afforestation is an important approach to prevent soil erosion in dry areas, reduce water and soil loss, and solve ecological and environmental problems. However, soil drought, uneven distribution of precipitation, and low rainwater utilization efficiency in dryland seriously restrict the survival rate of seedlings. To address these issues, a Rainwater Harvesting Tray (RHT) with a multi-ribbed fold design was developed. It can effectively increase the rainwater collection area. When placed under the forest canopy, it achieves in-situ rainwater collection and infiltration to replenish soil moisture. After digitizing the structure of the rainwater harvesting tray product, the collection performance of RHT was estimated using an evaluation model based on Grasshopper-SWMM. The relative error between the calculated and measured rainwater collection amount was within 2.5%. The simulation results show that when the number of ribs on the surface of the rainwater collection tray is 26, the height of the ribs is 3 cm, and the inclination angle of the collection groove is 13°, the collection performance of the rainwater collection tray is optimal.

Accurate diagnosis and prediction of crop water deficit are essential for enhancing water resource utilization efficiency and advancing precision agriculture, serving as critical prerequisites for the intelligent management of irrigation water resources. The traditional methods for acquiring crop water information in irrigation districts are primarily at the point scale, which are time-consuming, labor-intensive, and have certain spatial limitations. Satellite remote sensing technology, with its wide coverage and continuous monitoring capabilities, provides extensive and long-term land surface information, making it an effective solution for assessing crop water deficit over large regions. This paper begins by summarizing the common methods used to acquire soil and crop water information in modern agriculture, analyzing their strengths and weaknesses. Then, based on the different responses of crops to water stress, the research progress of satellite remote sensing in crop water deficit diagnosis is described from the aspects of spectral reflection characteristics, microwave remote sensing technology, infrared thermometry, and multi-modal remote sensing data coupling. Finally, this study reviews the progress in applying crop growth models, machine learning algorithms, and hybrid-driven models for predicting crop water deficit and drought early warnings, while also addressing the challenges in crop water deficit diagnosis research. The aim is to clarify the application prospects and development directions of satellite remote sensing technology in crop water deficit diagnosis and drought prediction, offering new insights and references for future research in crop water deficit diagnosis using satellite remote sensing methods.

In order to achieve rapid and precise monitoring of soil moisture content in flue-cured tobacco fields, this study was conducted in nine flue-cured tobacco experimental plots in Tuokeng Village, Menggu Town, Qiaojia County, Zhaotong City, Yunnan Province as the study area. The research integrated multispectral and thermal infrared UAV imagery with machine learning models to retrieve soil moisture content. Data acquisition involved multispectral imagery from a DJI Mavic 3 UAV (four bands, spatial resolution of 0.003 m) and thermal infrared imagery from a DJI Matrice 4T UAV (VOx sensor, spatial resolution of 0.005 m), along with in-situ measurements of soil volumetric water content. The images were preprocessed using ENVI and Pix4D software, followed by the extraction of nine key multispectral features. These were combined with thermal infrared DN values to construct the feature set. A series of machine learning models—including Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Convolutional Neural Network (CNN)—were applied to develop both single-source and multi-source soil moisture retrieval models. Model performance was evaluated based on root mean square error (RMSE) and the coefficient of determination (R2). The results demonstrated that multi-source models significantly outperformed their single-source counterparts. The RF model, incorporating multispectral features and thermal infrared DN values, yielded the best performance (RMSE=1.30%, R2= 0.79), with XGBoost providing comparable results (RMSE=1.31%, R2=0.78). SHAP analysis further revealed that the Normalized Difference Red Edge Index (NREI), NDVI_RVI, and thermal infrared DN values were the key features. This study highlights the potential of combining multispectral and thermal UAV imagery with machine learning for efficient soil moisture retrieval in flue-cured tobacco fields, providing technical support for precision irrigation and sustainable water resource management in agricultural regions.

Drip irrigation under plastic mulch is widely practiced in cotton fields of northern Xinjiang, following the empirical mode of "frequent and light irrigation". However, the lack of quantitative criteria for irrigation frequency and volume has led to shallow root systems, low water use efficiency (WUE), and large interannual yield fluctuations. To address this issue, seven irrigation treatments were established in a typical oasis cotton field in Changji City, and the response thresholds of five fine-root diameter classes (0~4.5 mm) to soil moisture lower limits were systematically quantified using WinRHIZO Pro. The results showed that: ① During the bud stage, the 70% field capacity (FC) treatment (F2B) significantly increased the length, surface area, and volume of 0~2.5 mm fine roots by 53.81%, 54.80%, and 60.63%, respectively, compared with conventional farmers’ practice (CK), while the total irrigation amount decreased by 34.27%. ② The seed cotton yield of F2B3 (70% FC at both bud and flowering and boll stages) reached 7 154.7 kg/hm2, a significant increase of 32.45% over CK, while WUE increased from 1.56 kg/m3 to 2.78 kg/m3 (a 78% improvement). The upper half mean length and breaking strength of fibers increased by 6.92% and 4.93%, respectively. ③ Soil moisture content in the 20 cm layer was significantly positively correlated with yield (R=0.85, p<0.05) and boll number per plant (R=0.73, p<0.05), indicating the critical role of surface soil moisture in boll formation, while root indices indirectly influenced yield. A threshold-based irrigation strategy (70% FC at bud and flowering and boll stages + 21 high-frequency irrigations) was proposed, providing replicable technical parameters for precision drip irrigation of cotton in arid regions.

Understanding the relationship between precipitation and rice irrigation requirements in the cold regions of Inner Mongolia provides crucial evidence for optimizing irrigation schedule and promoting efficient water resource utilization. Based on daily meteorological data from 1959 to 2017 at four representative rice stations in the cold regions of Inner Mongolia, this study constructed a water balance model. The Mann–Kendall test was applied to analyze the trends of water balance components, while the Pearson correlation coefficient and maximal information coefficient were used to reveal the relationships among these components. In addition, precipitation concentration degree and concentration period were employed to characterize the temporal distribution of rainfall. The results show that during the rice growing season in Inner Mongolia's cold regions, rainfall is relatively low (<500 mm) while crop water demand is high (>600 mm), resulting in significant overall irrigation demands: Kailu (921.1 mm) > Bairin Left Banner (799.7 mm) > Jalaid Banner (777.6 mm) > Zhalantun (669.5 mm). All sites exhibited high rainfall utilization rates (>80%) but low rainfall contribution rates (<40%). Although limited rainfall was efficiently utilized, it remained insufficient to meet irrigation requirements. In recent years, rainfall has shown a decreasing trend, while ET_c has increased, leading to a growing dependence on irrigation. Under future climate change scenarios characterized by reduced rainfall and enhanced ET_c, Inner Mongolia's cold regions can optimize irrigation patterns by integrating rainfall forecasts to enhance rainfall contribution and alleviate water resource pressure.

In recent years, the frequency of extreme rainfall events has increased, making it particularly urgent to improve the accuracy of precipitation prediction to support crop growth. Leveraging the advantages of Variational Mode Decomposition (VMD) in time series decomposition, and the complementary strengths of Long Short-Term Memory (LSTM) networks—which excel in handling local temporal information, and FEDformer, which is adept at processing global dependencies and possesses frequency domain characteristics, this paper proposes a precipitation prediction method based on a combined VMD-LSTM-FEDformer model. Five meteorological stations with different geographical features in Henan Province were selected for prediction analysis. The results indicated that the prediction errors of the combined model at all meteorological stations were within 5 mm, demonstrating the strong robustness of the model. In comparative experiments, the Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R²) of the VMD-LSTM-FEDformer combined model were 9.339 8 mm, 12.703 5 mm, and 0.964 4, respectively. These metrics outperformed those of other models, proving that the proposed model possesses excellent predictive capability and practical application value.

Soil evaporation is an important part of the water cycle in farmland. Accurately measuring inter-plant soil evaporation is particularly important for precise farmland water management. To explore effective measures to improve the accuracy of soil evaporation measurement, this study took loam soil in the North China Plain as the research object. The effects of different specifications of micro-lysimeters such as material, height, inner diameter, and bottom sealing materials on soil evaporation were compared, and the observation accuracy of different burial positions and soil replacement cycles was analyzed. The results showed that the evaporation amount of PVC material was lower than that of galvanized iron sheet, and the difference in evaporation amount between the two materials was not significant; the measured evaporation of the micro-lysimeter decreased with the increase of the diameter, and the highest accuracy was achieved when the diameter was 10 cm; the measured evaporation of the micro-lysimeter increased with the height of the micro-lysimeter, and the accuracy was highest at a height of 20 cm; the relative error of observation results for gauze sealing was generally lower than that for plastic cloth sealing; when the soil replacement cycle was 1~2 d, the observation accuracy was the highest; and inter-row evaporation was greater than that between plants and near plants. Therefore, in the North China Plain, in order to ensure the stability and operability of sampling, it is recommended that the micro-lysimeter should be made of galvanized iron sheet, the bottom sealing material should be gauze, with a diameter of 10 cm and a height of 10 cm respectively. The suitable burial location of the micro-lysimeter is inter-row, and the suitable soil replacement cycle is 1~2 d.

This paper proposes a TOPMODEL-LSTM coupled model to address the difficulty in integrating physical characteristics of watersheds into Long Short-Term Memory (LSTM) neural network models. Based on hydrological data in the Andun River Basin from 2008 to 2018, the performance of runoff simulation over the 1~6 hour forecast periods under three input schemes was compared and analyzed. The results indicate that: ① the coupled model achieved better forecast accuracy than the LSTM model in all forecasting periods, with average reductions of 15% in RMSE values and 13% in MAE values, and an increase in R2 from 0.78 to 0.83, which improved the accuracy of flood simulation to some extent; ② in the simulation of hydrological processes in a watershed, the scheme of using hourly precipitation data and simultaneous runoff observations as the basic input data for the model was validated through multiple comparative experiments, and its simulation accuracy was significantly superior to that of a single runoff input or other combined input schemes; ③ as the forecast period extended, the simulation accuracy of all models and their schemes decreased, but the coupled model effectively alleviated the memory decay issue of LSTM and exhibited more stable simulation performance. The research results can provide a technical reference for the coupling of traditional flood simulation models and artificial intelligence models.

The crop waterlogging degree index (WI) has been primarily validated for assessing waterlogging stress in wheat, yet its applicability for characterizing waterlogging stress in maize remains unclear. Based on published experimental data of maize waterlogging stress, this study systematically analyzed the relationship between the WI index and the relative yield of maize under single waterlogging, single submergence, and sequential submergence-waterlogging treatments, based on published experimental data of maize waterlogging stress. The results showed that under different waterlogging stress treatments, the WI index exhibited a highly significant linear negative correlation with the relative yield of maize. The multiple correlation coefficients for all fitted samples were above 0.74. Like the traditional SDI index, the WI index was also effective in quantifying the comprehensive waterlogging stress degree throughout the entire maize growth period. Therefore, the WI index can be extended as a universal indicator for crop waterlogging stress, providing a theoretical basis for large-scale agricultural waterlogging disaster monitoring.

In response to the practical difficulty of accurately quantifying the comprehensive benefits of intelligent sprinkler irrigation for field crops, this study introduces the Analytic Hierarchy Process (AHP) and entropy weight method of game theory to establish an improved TOPSIS model. Taking the water-saving and yield increasing experimental area of sprinkler irrigation for wheat in Dacaozhuang Farm, Ningjin County, Hebei Province as a case study, the effectiveness and stability of the model are verified, and the key factors affecting the comprehensive benefits of Intelligent sprinkler irrigation are identified and analyzed. The analysis indicated that the T2 treatment (25% water deficit during the green-up to jointing stage) achieved balanced performance in yield increase, water use efficiency, and soil health, ranking as the optimal model. Spearman correlation analysis further revealed that yield increase rate (ρ=0.988), water use efficiency (ρ=0.875), and soil microbial metabolic intensity (ρ=0.855) were key drivers of comprehensive system benefits, while excessive water-saving ( >30%) led to a significant efficiency decline. This study provides scientific decision-making basis and methods for the selection of Intelligent sprinkler irrigation technology model in the North China Plain, and has important reference value for promoting the accurate allocation and efficient utilization of agricultural water resources in the North China Plain.

Agricultural water metering and the allocation of water use quotas are important means of standardizing and strengthening water management. Aiming at the problems in water management, such as the incomplete planning and layout of agricultural water metering facilities, incomplete metering data, unclear division of water use units, and irregular allocation and application of agricultural water use quotas, this paper innovatively defined the basic concept and characteristics of the water use unit. Starting from the irrigation system as a whole, it systematically analyzed the principles and methods for rationally dividing water use units, setting water supply and use handover sections, and scientifically planning the layout of a complete agricultural water metering system. It also selected the metering modes, methods, and facility types for metering facilities, as well as the principles and methods for agricultural water use quota allocation oriented towards water use units. This provides a beneficial reference and guidance for coordinating irrigation project construction and water metering facility construction, deepening the comprehensive reform of agricultural water pricing, improving the water management mechanism, standardizing the construction of the agricultural water metering system and the allocation of agricultural water use quotas, and resolving the confusion in improving the water management mechanism.

Agriculture is a major source of greenhouse gas (GHG) emissions, and selecting an appropriate model for its precise quantification is a key prerequisite for research on farmland carbon sequestration and emission reduction. This study aims to provide a basis for model selection in various research scenarios by reviewing and comparing six mainstream process-based models: DNDC, DAYCENT, WNMM, SPACSYS, DRAINMOD-N II, and DLEM. Their research progress and simulation methodologies were systematically reviewed. The results indicated that model applications were predominantly concentrated in the United States and China. DLEM demonstrated better performance in simulating CO?, while DAYCENT was more effective for CH?. Both DNDC and DAYCENT showed superior performance in simulating N?O. For regional-scale simulations, DLEM is suitable for carbon simulation, and DNDC and WNMM for nitrogen simulation. DNDC and DAYCENT are more suitable for field-scale gas analysis. It is concluded that future model development should focus on enhancing the basis of physical mechanisms, improving the capability to integrate multi-source data, and increasing the spatiotemporal resolution of output results.

In order to explore the migration and distribution characteristics of Cd in typical soda saline-alkali soil in the Songyuan irrigation area under different irrigation modes, and to study the simulation applicability of the HYDRUS-1D model, based on an indoor dynamic soil column test, three irrigation modes of drip irrigation, flood irrigation, and dry-wet alternate irrigation were set up to analyze the basic law of Cd migration in soil under different irrigation modes. A numerical model of Cd migration was constructed using HYDRUS-1D, and the model was used to simulate and verify the accumulation and migration of Cd in the region. The results showed that under the drip irrigation mode, Cd was mainly enriched in the shallow topsoil, the adsorption rate of Cd was the fastest, and the Cd content gradually decreased with the increase of soil depth. Under the flood irrigation mode, Cd was relatively evenly distributed at all depths, and the Cd adsorption rate was the slowest. Under the dry-wet alternate irrigation mode, the distribution law of Cd was not obvious, and the adsorption rate was between that of drip irrigation and flood irrigation. Additionally, the Cd content of each soil layer under the three irrigation modes increased with the increase of irrigation amount, and the overall trend was increasing. The dynamic changes of Cd concentration in simulated water flow under drip irrigation and flood irrigation modes showed a trend of increasing first and then stabilizing with time. Under the dry-wet alternate irrigation mode, the Cd concentration of the water flow showed a fluctuating increase before tending towards stable fluctuation. The simulated values of the model and the measured values of the soil column test were verified. It was found that the drip irrigation test and the flood irrigation test had a good fit, while the dry-wet alternate irrigation test had a poor fit. In summary, the HYDRUS-1D model is suitable for the simulation of Cd migration in soda saline-alkali soil under drip irrigation and flood irrigation modes in the Songyuan irrigation area, which can provide a theoretical basis for the prevention and control of soil Cd pollution in this area.

To investigate the spatio-temporal variation characteristics of agricultural drainage water quality in the polder of southern China and their influencing factors, a multi-point water quality monitoring experiment was conducted from July 2023 to June 2024 in Baihu Polder and Yangliu Polder in the Chaohu Lake Basin as the study area. The study focused on analyzing the patterns of water quality changes and key driving factors in paddy fields, ditches and external rivers. The results indicate that on the temporal scale, influenced by time-varying factors such as rainfall and agricultural management, the concentrations of total nitrogen, ammonia nitrogen, nitrate nitrogen, and total phosphorus in paddy field surface water within the polders showed an initial increase followed by a decrease after topdressing, associated with the irrigation and drainage process, while COD concentration exhibited a steady decline. On an annual scale, nitrogen, phosphorus, and COD concentrations in the ditches and external rivers exhibited a double-peak pattern during flood and non-flood seasons. Spatially, influenced by sediment resuspension and purification effects in ditches, nitrogen, phosphorus, and COD concentrations along the flow paths generally followed an upstream-high, downstream-low distribution pattern. Analysis of water quality influencing factors indicates that larger pump station control areas result in lower nitrogen, phosphorus, and COD concentrations in pumped effluent. Agricultural non-point source pollution and domestic sewage are the primary drivers of water quality deterioration in the polder. It is recommended to enhance water quality in the polder by reducing non-point source pollution from farmland and intercepting pollutants during transport and migration. Concurrently, attention should be paid to domestic sewage treatment and prevention in areas with a high risk of soil erosion.

Soil macropores are the main channels for the movement of soil moisture. The soil macropore structure under different vegetation types is conducive to increasing precipitation infiltration and thus reducing the degree of water and soil loss. In this study, three typical vegetation types, namely Populus simonii，Caragana korshinskii and grassland of the loess region of northwestern Shanxi Province，were taken as the research objects. Undisturbed soil columns of 0~33 cm were collected, and the soil porosity, pore number, and other pore characteristics were quantitatively analyzed using CT scanning and image processing technology. And combined soil physicochemical properties, the main factors affecting soil pore structure were investigated. The results showed that: ①The macroporosity, pore number, and connectivity of Caragana korshinskii and Populus simonii were significantly better than those of grassland (p<0.05). Among them, the number of pores in Populus simonii reached 4 437 ± 1 190 per cm³, and the average pore diameter was 5.29±0.95 mm, which was significantly greater than that of the grassland (p<0.05). ②The relationship between bulk density and p parameters was weak (p>0.05), and the pH value was significantly positively correlated with porosity (p<0.05), which may be related to the inhibition of aggregate cementation by acidic environment; the correlation with water content was low, which might be restricted by the regional soil dry layer. ③Populus simonii and Caragana korshinskii expanded the pore diameter distribution and increased the number of large pores through deep root penetration and soil animal activity (diameter > 4mm), while the grassland roots were shallow and fine (diameter < 4mm), resulting in poor pore connectivity. The study showed that due to their well-developed soil pore structure, Caragana korshinskii and Populus simonii are significantly superior to grasslands and can effectively improve the soil water infiltration rate. Although Caragana korshinskii and Populus simonii are often regarded as preferred vegetation for slope stabilization and water conservation, their strong transpiration can also lead to varying degrees of soil water deficit. Therefore, in the large-scale vegetation configuration of the loess region in northwest Shanxi, the water consumption characteristics of different vegetation types should be considered for a long term to realize the sustainable utilization of regional water resources.