In order to formulate a reasonable water input model for cotton fields in southern Xinjiang for scientific and rational fertilization, to reduce soil carbon leaching, and to improve soil carbon sequestration capacity, an undisturbed soil column leaching test was used to simulate the current field management method in the study area. Two methods, drip irrigation and flood irrigation, were set up, and three irrigation and three nitrogen fertilizer levels were used to carry out the undisturbed soil column leaching test. The results showed that the amount and mode of water and nitrogen input affected the distribution and leaching loss of organic carbon and inorganic carbon in the soil. When the nitrogen application rate increased from 270 kg·hm⁻² to 450 kg·hm⁻², the leaching loss of soluble organic carbon and soluble inorganic carbon increased significantly. When the water input increased from 6000 m³·hm⁻² to 9000 m³·hm⁻², the leaching loss of DOC and DIC increased significantly. The carbon leaching loss under drip irrigation was higher than that under flood irrigation. The leaching rates of DOC and DIC were fastest under the conditions of high water (9000 m³·hm⁻²) and high fertilizer (450 kg·hm⁻²). This shows that water and nitrogen input and irrigation methods are important factors affecting soil carbon leaching. In the case of excessive water input, long-term high-frequency irrigation is the main factor affecting carbon leaching.

Considering the pear in the arid region as the research object, single-factor testing and water–fertilizer coupling testing were conducted. The response of pear tree growth to water, nitrogen, and phosphorus was explored and provided a theoretical basis for efficient water and fertilizer management. Among them, the single-factor test set water, nitrogen, and phosphorus as the three factors, and five levels were set. Screening out W3, W4, N3, N4, P3, and P4 promoted plant nutrient uptake and fruit quality. Eight treatments were set up in the water and fertilizer coupling test: Treatment 1 (T1, W3N3P3), Treatment 2 (T2, W3N3P4), Treatment 3 (T3, W3N4P3), Treatment 4 (T4, W3N4P4), Treatment 5 (T5, W4N3P3), Treatment 6 (T6, W4N3P4), Treatment 7 (T7, W4N4P3), and Treatment 8 (T8, W4N4P4). The results showed that the leaf area index of the T1, T2, T3, and T4 treatments was significantly higher than that of the other treatments at maturity. The yield, single fruit weight, and primary fruit rate were the highest under T3 treatment. The gray correlation degree analysis of fruit quality showed that the T3 treatment had the highest degree of correlation and ranking of each fruit quality index, indicating that the T3 treatment had the highest fruit quality. The yield model showed that irrigation with 6510.06 m³ hm⁻², nitrogen fertilizer with 337.5 kg N hm⁻², and phosphate fertilizer with 262.5 kg P hm⁻² had the best yield. A detailed investigation of pear tree growth and fruit quality showed that the T3 treatment had the best fruit growth and development performance, and the pear fruit quality was the best.

Maize is one of the top three crops globally, ranking only behind rice and wheat, making it an important crop of interest. Aboveground biomass is a key indicator for assessing maize growth and its yield potential. This study developed an efficient and stable biomass prediction model to estimate the aboveground biomass (AGB) of spring maize (Zea mays L.) under subsurface drip irrigation in arid regions, based on UAV multispectral remote sensing and machine learning techniques. Focusing on typical subsurface drip-irrigated spring maize in arid Xinjiang, multispectral images and field-measured AGB data were collected from 96 sample points (selected via stratified random sampling across 24 plots) over four key phenological stages in 2024 and 2025. Sixteen vegetation indices were calculated and 40 texture features were extracted using the gray-level co-occurrence matrix method, while an integrated feature-selection strategy combining Elastic Net and Random Forest was employed to effectively screen key predictor variables. Based on the selected features, six machine learning models were constructed, including Elastic Net Regression (ENR), Gradient Boosting Decision Trees (GBDT), Gaussian Process Regression (GPR), Partial Least Squares Regression (PLSR), Random Forest (RF), and Extreme Gradient Boosting (XGB). Results showed that the fused feature set comprised four vegetation indices (GRDVI, RERVI, GRVI, NDVI) and five texture features (R_Corr, NIR_Mean, NIR_Vari, B_Mean, B_Corr), thereby retaining red-edge and visible-light texture information highly sensitive to AGB. The GPR model based on the fused features exhibited the best performance (test set R² = 0.852, RMSE = 2890.74 kg ha⁻¹, MAE = 1676.70 kg ha⁻¹), demonstrating high fitting accuracy and stable predictive ability across both the training and test sets. Spatial inversions over the two growing seasons of 2024 and 2025, derived from the fused-feature GPR optimal model at four key phenological stages, revealed pronounced spatiotemporal heterogeneity and stage-dependent dynamics of spring maize AGB: the biomass accumulates rapidly from jointing to grain filling, slows thereafter, and peaks at maturity. At a constant planting density, AGB increased markedly with nitrogen inputs from N0 to N3 (420 kg N ha⁻¹), with the high-nitrogen N3 treatment producing the greatest biomass; this successfully captured the regulatory effect of the nitrogen gradient on maize growth, provided reliable data for variable-rate fertilization, and is highly relevant for optimizing water–fertilizer coordination in subsurface drip irrigation systems. Future research may extend this integrated feature selection and modeling framework to monitor the growth and estimate the yield of other crops, such as rice and cotton, thereby validating its generalizability and robustness in diverse agricultural scenarios.

In order to formulate a reasonable water input model for cotton fields in southern Xinjiang, scientific and rational fertilization, reduce soil carbon leaching, and improve soil carbon sequestration capacity, the undisturbed soil column leaching test was used to simulate the current field management method in the study area. Two methods of drip irrigation and flood irrigation were set up, and three irrigation and three nitrogen fertilizer levels were used to carry out the undisturbed soil column leaching test. The results showed that the amount and mode of water and nitrogen input affected the distribution and leaching loss of organic carbon and inorganic carbon in soil. When the nitrogen application rate increased from 270 kg·hm-2 to 450 kg·hm-2, the leaching loss of soluble organic carbon and soluble inorganic carbon increased significantly. When the water input increased from 6000 m3·hm-2 to 9000 m3·hm-2, the leaching loss of DOC and DIC increased significantly. The carbon leaching loss under drip irrigation was higher than that under flood irrigation. The leaching rates of DOC and DIC were the fastest under the conditions of high water ( 9000 m3·hm-2 ) and high fertilizer ( 450 kg·hm-2 ). It shows that water and nitrogen input and irrigation methods are important factors affecting soil carbon leaching. In the case of excessive water input, long-term high-frequency irrigation is the main factor affecting carbon leaching.

Taking Korla fragrant pear as the research object, single factor testing and water and fertilizer coupling testing were conducted. Under the premise of the drip irrigation mode, the effects of water, nitrogen, and phosphorus on the growth, fruit quality, and yield of Korla Fragrant Pear were studied to provide a theoretical basis for efficient water and fertilizer management. Among them, the single factor test set water, nitrogen and phosphorus as the three factors, and each factor was set at five levels, namely, water (W1, 5460 m3·hm−2;W2, 5880 m3·hm−2;W3,6300 m3·hm−2; W4, 6720 m3·hm−2; W5, 7140 m3·hm−2), nitrogen (N1, 150 kg N·hm−2;N2, 225 kg N·hm−2; N3, 300 kg N·hm−2; N4, 375 kg N·hm−2; N5, 450 kg N·hm−2), phosphorus (P1,75 kg P2O5·hm−2; P2, 150 kg P2O5·hm−2; P3, 225 kg P2O5·hm−2; P4, 300 kg P2O5·hm−2; P5, 375 kg P2O5·hm−2). The results showed that the total nitrogen and total phosphorus contents in leaves under the W3, W4, N3, N4, P3, and P4 treatments were significantly higher than those under other treatments, and the fruit quality was better. Therefore, the W3, W4, N3, N4, P3, and P4 treatments were selected as the water and fertilizer inputs for the water–fertilizer coupling test for Korla fragrant pear. Eight treatments were set up in the water and fertilizer coupling test: Treatment 1 (T1, W3N3P3), Treatment 2 (T2, W3N3P4), Treatment 3 (T3, W3N4P3), Treatment 4 (T4,W3N4P4), Treatment 5 (T5, W4N3P3), Treatment 6 (T6, W4N3P4), Treatment 7 (T7, W4N4) P3), and Treatment 8 (T8, W4N4P4). The results showed that the leaf area index of the T1, T2, T3, and T4 treatments was significantly higher than that of the other treatments at maturity. The T3 treatment resulted in the highest fruit yield, single fruit weight, and primary fruit rate. The results of the grey correlation degree analysis of fruit quality showed that T3 treatment had the highest degree of correlation and ranking of each fruit quality index, indicating that T3 treatment had the highest fruit quality. Detailed investigation of pear tree growth and fruit quality showed that the T3 treatment (375 kg·hm−2 nitrogen application, 225 kg·hm−2 phosphorus application and 6300 m3·hm−2 irrigation) had the best fruit growth and development performance, and the sweet pear fruit quality was the best.

Taking Korla fragrant pear as the research object, single factor testing and water and fertilizer coupling testing were conducted. Under the premise of the drip irrigation mode, the effects of water, nitrogen, and phosphorus on the growth, fruit quality, and yield of Korla Fragrant Pear were studied to provide a theoretical basis for efficient water and fertilizer management. Among them, the single factor test set water, nitrogen and phosphorus as the three factors, and each factor was set at five levels, namely, water (W1, 5460 m3·hm−2;W2, 5880 m3·hm−2;W3,6300 m3·hm−2; W4, 6720 m3·hm−2; W5, 7140 m3·hm−2), nitrogen (N1, 150 kg N·hm−2;N2, 225 kg N·hm−2; N3, 300 kg N·hm−2; N4, 375 kg N·hm−2; N5, 450 kg N·hm−2), phosphorus (P1,75 kg P2O5·hm−2; P2, 150 kg P2O5·hm−2; P3, 225 kg P2O5·hm−2; P4, 300 kg P2O5·hm−2; P5, 375 kg P2O5·hm−2). The results showed that the total nitrogen and total phosphorus contents in leaves under the W3, W4, N3, N4, P3, and P4 treatments were significantly higher than those under other treatments, and the fruit quality was better. Therefore, the W3, W4, N3, N4, P3, and P4 treatments were selected as the water and fertilizer inputs for the water–fertilizer coupling test for Korla fragrant pear. Eight treatments were set up in the water and fertilizer coupling test: Treatment 1 (T1, W3N3P3), Treatment 2 (T2, W3N3P4), Treatment 3 (T3, W3N4P3), Treatment 4 (T4,W3N4P4), Treatment 5 (T5, W4N3P3), Treatment 6 (T6, W4N3P4), Treatment 7 (T7, W4N4) P3), and Treatment 8 (T8, W4N4P4). The results showed that the leaf area index of the T1, T2, T3, and T4 treatments was significantly higher than that of the other treatments at maturity. The T3 treatment resulted in the highest fruit yield, single fruit weight, and primary fruit rate. The results of the grey correlation degree analysis of fruit quality showed that T3 treatment had the highest degree of correlation and ranking of each fruit quality index, indicating that T3 treatment had the highest fruit quality. Detailed investigation of pear tree growth and fruit quality showed that the T3 treatment (375 kg·hm−2 nitrogen application, 225 kg·hm−2 phosphorus application and 6300 m3·hm−2 irrigation) had the best fruit growth and development performance, and the sweet pear fruit quality was the best.