ENT-YOLO: An improved lightweight YOLO for cotton organ detection in mulched drip irrigation systems in southern Xinjiang

The development of a crop production strategy through the use of a crop model represents a crucial method for the assurance of a stable agricultural yield and the subsequent enhancement thereof. There are currently no studies evaluating the suitability of the AquaCrop model for the drip irrigation of Gossypium barbadense in Southern Xinjiang, which is the primary planting region for Gossypium barbadense in China. In order to investigate the performance of the AquaCrop model in simulating the growth of cotton under mulched drip irrigation, the model was locally calibrated and validated according to different irrigation thresholds during a key growth period of two years. The results of the simulation for total soil water (TSW), crop evapotranspiration (ETc), canopy coverage (CC), aboveground biomass (Bio), and seed cotton yield demonstrated a high degree of correlation with the observed data, with a root mean square error (RMSE) of <11.58%. The Bio and yield simulations demonstrated a high degree of concordance with the corresponding measured values, with root mean square error (RMSE) values of 1.23 t ha−1 and 0.15 t ha−1, respectively. However, the predicted yield declined in the verification year, though the prediction error remained below 15%. Furthermore, the estimated evapotranspiration (ETc) value demonstrated a slight degree of overestimation. Generally, the middle and late stages of cotton growth led to an overestimation of the TSW content. However, the prediction error was less than 13.99%. Through the calculation of each performance index of the AquaCrop model, it is found that they are in the acceptable range. In conclusion, the AquaCrop model can be employed as a viable tool for predicting the water response of cotton to drip irrigation under mulched film in Southern Xinjiang. Based on 64 years of historical meteorological data, three years were selected as scenarios for simulation. Principal component analysis (PCA) showed that, in a local wet year in Southern Xinjiang, the irrigation quota was 520 mm, and the irrigation cycle was 6 days/time. In normal years, the irrigation quota was 520 mm, with an irrigation cycle of 6 days/time. In dry years, the irrigation quota was 595 mm, with an irrigation cycle of 10 days/time. This allowed for higher seed cotton yields and irrigation water productivity, as well as the maximization of cotton yields and net revenue in the arid oasis area of Southern Xinjiang.

To optimize suitable water-saving and soil salt-controlling irrigation needed for the high yield and good quality of cotton in southern Xinjiang, a field experiment was carried out to study the effects of soil water lower limits on water consumption, water use efficiency (WUE), yield, cotton fiber quality and soil salt accumulation under mulched drip irrigation in Korla, Xinjiang. The field capacity (FC) was regarded as the upper limit of soil moisture, and five soil water lower limits (85% FC, 75% FC, 65% FC, 55% FC, 45% FC, referred as T1~T5, respectively) were designed during the cotton growth period. The results indicated that the irrigation frequency and irrigation quota of cotton were gradually increased with the increase in the soil water lower limit, while the water consumption modulus for T2 treatment during the critical period of water demand arrived at the maximum value. Moreover, with the decrease in the soil water lower limit, the WUE, fiber micronaire value and fiber maturity index of cotton increased, whereas the yield, nitrogen partial factor productivity (PFPN) and fiber breaking elongation of cotton decreased. However, when the soil water lower limit exceeded 75% FC, the increase had little effect on the cotton yield increase and PFPN improvement, and the yield and PFPN for T2 treatment were 7146.4 kg∙hm−2 and 23.82 kg∙kg−1, respectively, In addition, the decrease in the soil water lower limit was unfavorable for an increase in fiber length, but it was conducive to the enhancement of fiber strength. Furthermore, soil salt accumulated inside and outside the film for the designed soil water lower limits, and the amount of accumulated salt in 0~100 cm followed T3 > T5 > T1 > T2 > T4. Based on a comprehensive analysis with the entropy TOPSIS method, the findings of the present study suggested that the suitable soil water lower limit for cotton under mulched drip irrigation was 75% FC in southern Xinjiang, China.

The drip irrigation under mulch has become one of significant supporting technologies for cotton industry development in Xinjiang, and has shown the good economic and ecological benefits. With the rapid development of society and economy in Southern Xinjiang, the conventional mode of large-quota winter and spring irrigation, salt leaching and alkali decreasing is difficult to support sustainable development of land and water resources in Southern Xinjiang. This study tries to adjust soil moisture and salt content regulation mode of massive water salt leaching and drip irrigation under mulch in the non-growing period of cotton field in Southern Xinjiang, explores interannual soil salinity change features of drip irrigation cotton field without winter and spring irrigation, and provides experimental basis for drip irrigation technology under mulch which can reduce and exempt cotton irrigation in winter and spring. According to ET0, the dual-factor complete combination experiment involving 3 irrigating water quotas (I1, I2, I3) and 2 irrigation times (T12, T16) was designed, and 6 treatments were involved in total(I1T12,I2T12,I3T12,I1T16,I2T16 and I3T16). The investigation results of four-year (2012–2015) field positioning experiment showed that, under the condition of “germination under drip irrigation” without winter and spring irrigation, increasing irrigation quota and irrigation times could lower 0-100cm soil salinity accumulation, but the soil salinity accumulation degree was 40-100cm, and less than 0-30cm. In the seedling stage, bud stage, blossom and boll-forming stage, and boll opening stage, the average salinity of 0-100cm soil increased by 39.81%, 31.91%, 26.85% and 29.47%, respectively. Increasing irrigation quota and irrigation times could ease interannual soil salinity accumulation degree of cotton field with drip irrigation under mulch, without winter and spring irrigation. 0-100cm soil salinity before sowing was related to the irrigation quota of cotton in the growing stage of the last year. The larger the irrigation quota was, the smaller the soil salinity before sowing would be. The accumulation amount of soil salinity at the end of growing stage under different treatments was lower than that before sowing. The drip irrigation of cotton under mulch in the growing stage could effectively regulate soil salinity distribution and space-time migration process in the growing stage of cotton. Compared with the beginning of 2012, 0-100cm average soil salinity under 3 irrigation quotas (I1, I2, I3) was 33.66%, 5.60% and 1.24%, respectively. Salt accumulating rates under 12 irrigations and 16 irrigations were 20.66% and 6.33%, respectively. The soil had the risk of salinization when the “germination under drip irrigation” without winter and spring irrigation was used. Such results can provide the reference for prevention and treatment of soil moisture and salt content of cotton field with drip irrigation under mulch in the arid region.

A field experiment was carried out in southern Xinjiang, China, to reveal soil-water flow pattern beneath a combined plastic-mulch (film) and drip-irrigation system using brackish water. The soil-water flow system (SWFS) was characterized from soil surface to the water table based on observed spatio-temporal distribution of total soil-water potential, water content and electric conductivity. Root suction provided a strong inner sink. The results indicated that SWFS determined the soil salinity and moisture distribution. Drip-irrigation events could leach excess salts from the root zone and provide soil conditions with a tolerable salinity level that supports the growth of cotton. High-salinity strips were formed along the wetting front and at the bare soil surface. Hydrogeology conditions, irrigation regime, climate, plant growth and use of mulch would affect potential sources and sinks, boundary conditions and the size of the SWFS. At depth 0–60 cm, the soil salinity at the end of the irrigation season was 1.9 times that at the beginning. Beneath the mulch cover, the soil-water content in the ‘wide rows’ zone (55 cm between the two rows with no drip line) was higher than that in the ‘narrow rows’ zone (15 cm between the two rows with a drip line) due to the strong root-water uptake. The downward water flow below the divergent curved surface of zero flux before irrigation, and the water-table fluctuation with irrigation events, indicated that excessive irrigation occurred.

There are significant regional disparities between the northern and southern sides of the Tianshan Mountains, which the cultivated land soil quality shows different characteristics. In order to further understand the soil quality difference of the northern and southern sides of the Tianshan Mountains, we analyzed the influences of long-term drip irrigation (DI) in northern Xinjiang (NX) and drip irrigation during the growth season combined with flood irrigation in the non-growing season (DF) in southern Xinjiang (SX) on soil physical and chemical properties. The results from 1985 to 2020 indicated that DI increased the soil organic matter (SOM) (7.39%/year), total nitrogen (TN) (7.09%), available phosphorus (AP) (4.71%) and available potassium (AK) (1.12%) contents in NX. The results of field sampling in 2022 presented that the average values of SOM, AK content and pH in NX were 10.24%, 76.86% and 5.24% higher than those of SX, respectively, which validated our analysis of 1985 to 2020. Additionally, the integrated soil fertility index (IFI) trends confirmed that soil fertility in NX was higher than SX: NX showed sustained improvement (1985–2022), whereas SX fertility declined post-2009. In conclusion, NX has demonstrated superior soil fertility and nutrient retention compared to SX over the past thirty-five years under the influence of irrigation patterns, drip irrigation may represent a more sustainable long-term agricultural strategy than traditional flood irrigation in this region.

Water deficiency, together with soil salinization, has been seriously restricting sustainable agriculture around the globe for a long time. Optimal soil moisture regulation contributes to the amelioration of soil water and salinity for crops, which is favorable for plant production. A field experiment with five soil water lower limit levels (T1: 85% FC, T2: 75% FC, T3: 65% FC, T4: 55% FC, and T5: 45% FC, where FC is the field capacity) was conducted in southern Xinjiang in 2018 to investigate the responses of soil water-salt dynamics and cotton performance to soil moisture regulation strategies. The results indicated that in the horizontal direction, the farther away the drip irrigation belt, the lower the soil moisture content and the greater the soil salinity. In the vertical direction, the soil moisture and soil salinity increased first and then decreased with an increase in soil depth after irrigation, and the distribution was similar to an ellipse. Moreover, the humid perimeter of soil water and the leaching range of soil salt increased with a decrease in the soil moisture lower limit. Though more soil salt was leached out for the T5 treatment at the flowering stage due to the higher single irrigation amount, soil salinity increased again at the boll setting stage owing to the long irrigation interval. After the cotton was harvested, soil salt accumulated in the 0-100 cm layer and the accumulation amount followed T3 > T5 > T1 > T2 > T4. Moreover, with a decline of soil moisture lower limit, both plant height and nitrogen uptake decreased significantly while the shoot-root ratio increased. Compared with the yield (7233.2 kg·hm-2) and water use efficiency (WUE, 1.27 kg·m-3) of the T1 treatment, the yield for the T2 treatment only decreased by 1.21%, while the WUE increased by 10.24%. Synthetically, considering the cotton yield, water-nitrogen use efficiency, and soil salt accumulation, the soil moisture lower limit of 75% FC is recommended for cotton cultivation in southern Xinjiang, China.

The root system plays a crucial role in water and nutrient absorption, making it a significant factor affected by nitrogen (N) availability in the soil. However, the intricate dynamics and distribution patterns of cotton (Gossypium hirsutum L.) root density and N nutrient under varying N supplies in Southern Xinjiang, China, have not been thoroughly understood. A two-year experiment (2021 and 2022) was conducted to determine the effects of five N rates (0, 150, 225, 300, and 450 kg N ha−1) on the root system, shoot growth, N uptake and distribution, and cotton yield. Compared to the N0 treatment (0 kg N ha−1), the application of N fertilizer at a rate of 300 kg N ha−1 resulted in consistent and higher seed cotton yields of 5875 kg ha−1 and 6815 kg ha−1 in 2021 and 2022, respectively. This N fertilization also led to a significant improvement in dry matter weight and N uptake by 32.4% and 53.7%, respectively. Furthermore, applying N fertilizer at a rate of 225 kg N ha−1 significantly increased root length density (RLD), root surface density (RSD), and root volume density (RVD) by 49.6–113.3%, 29.1–95.1%, and 42.2–64.4%, respectively, compared to the treatment without N fertilization (0 kg N ha−1). Notably, the roots in the 0–20 cm soil layers exhibited a stronger response to N fertilization compared to the roots distributed in the 20–40 cm soil layers. The root morphology parameters (RLD, RSD, and RVD) at specific soil depths (0–10 cm in the seedling stage, 10–25 cm in the bud stage, and 20–40 cm in the peak boll stage) were significantly associated with N uptake and seed cotton yield. Optimizing nitrogen fertilizer supply within the range of 225–300 kg N ha−1 can enhance root foraging, thereby promoting the interaction between roots and shoots and ultimately improving cotton production in arid areas.

In order to study the effects of different irrigation amounts on the root characteristics and yield of cotton without film mulching (abbreviated as filmless cotton) under drip irrigation in Southern Xinjiang, five irrigation amounts of filmless cotton (300, 375, 450, 525 and 600 mm, represented by W1, W2, W3, W4 and W5) and one irrigation amount of cotton with film mulching (abbreviated as filmed cotton) (450 mm, represented by WCK) were set. The effects of irrigation amount on root length density (RLD), root surface area (RSA), root average diameter (RAD), root volume (RV), root weight density (RED) and yield of filmless cotton were analyzed. The results of the two-year experiment showed the following: (1) The indexes of cotton root growth reached the maximum at the flowering and bolling stage, and the growth of soil root in the periphery (30 cm from the main root) and the lower layer (40–60 cm soil layer) reached the peak at the flowering and bolling stages, respectively; (2) The average value of root growth index of film-free cotton in each treatment was W2 > W3 > W4 > W5 and W1. The RLD of W2 and W3 increased by 19.41–106.67% and 13.66–84.22% in the peripheral and lower soil layer, and the proportion of RSA in the peripheral and middle soil layer (20–40 cm soil layer) increased by 1.64–3.41% and 0.49–4.09% compared with other treatments. The RAD, RV and RWD after W2 treatment were relatively large at various distribution points in the soil, followed by W3 treatment; (3) The average root indexes of WCK were not significantly different from those of W3, but the indexes of the lower soil layer were the smallest, at only 29.18–66.84% of the average value of the non-film mulching treatment, while the root indexes of the surface layer (0–20 cm soil layer) and the surrounding soil were larger, with an increase of 11.43–102.17% and 29.60–111.57%, respectively, compared with the non-film mulching treatment; (4) The seed cotton yield of W3 was the highest in the non-film mulching treatment, reaching 4833.25 kg·hm−2, but was still lower than that of WCK by 27.79%. Conclusion: An appropriate water deficit is conducive to root growth and increases the uniformity of its distribution in the soil layer. The irrigation amount of 375–450 mm for filmless cotton in Southern Xinjiang can promote root growth, prevent senescence and increase yield, which can be used as a reference in production.

The stability and adaptability of the critical nitrogen concentration in drip-irrigated cotton fields in Southern Xinjiang were investigated by measuring the nitrogen uptake (Nuptake), nitrogen nutrition index (NNI), and accumulated nitrogen deficit (Nand). A two-year field study was conducted using five nitrogen levels (0, 75, 150, 300, and 450 kg·hm-2), with cotton cultivars 'Xinluzhong55' and 'Xinluzhong78' as the test materials. We analyzed the effects of nitrogen addition on plant dry matter (PDM) and critical nitrogen concentration dilution curves (PNCc). The Nuptake, NNI, and Nand models as well as a model of the relationships between NNI, Nand, and the relative seed cotton yield (RY) were established based on the PNCc for both cultivars. The results showed that the PNCc, maximum nitrogen concentration (PNCmax), and minimum nitrogen concentration (PNCmin) dilution models in drip-irrigated cotton showed power function relationships with plant dry matter (PDM), with determination coefficients (R2) of 0.905 and 0.960, 0.919 and 0.950, 0.934 and 0.969 for Xinluzhong55 and Xinluzhong78, respectively. Model validation with independent experimental data showed that the critical nitrogen concentrations in cotton grown under drip-irrigated conditions were simulated satisfactorily. The Xinluzhong55 and Xinluzhong78 1:1 line R2, root mean square error (RMSE) and relative estimation error (RE) of the simulated and observed critical nitrogen concentrations were 0.987, 0.051 g·kg-1, 1.95% and 0.984, 0.044 g·kg-1, 1.91%, respectively, showing good stability and adaptability between the two years. According to the Nuptake, NNI, and Nand model, we concluded that 300 kg·hm-2 was the optimum level of nitrogen application in drip-irrigated cotton in Southern Xinjiang. Based on the fitting results of nitrogen level and seed cotton yield, the theoretical appropriate nitrogen applications for Xinluzhong55 and Xinluzhong78 were 322.7 and 336.4 kg·hm-2, respectively. The R2 of NNI, Nand and RY in different growth stages were greater than 0.900, the NNI, Nand and RY estimation model of seed cotton yield based on 1:1 lines of R2, RMSE, and RE were 0.899-0.989, 0.05-0.13, and 4.1%-9.9% and 0.902-0.981, 0.04-0.12, and 5.1%-9.5%, all of which were highly significant. These results could provide new directions for the evaluation and diagnosis of optimum nitrogen application levels and nitrogen status, as well as the estimation of seed cotton yield in drip-irrigated cotton in Southern Xinjiang.

Low water use efficiency and soil salinization are two main factors limiting cotton production in southern Xinjiang. A field experiment was conducted to investigate the effects of brackish water irrigation levels on cotton growth, yield and soil water–salt dynamics in southern Xinjiang, so as to provide a theoretical and experimental basis for the development and utilization of brackish water. There were three irrigation levels: W1 (75 mm + 80%ETc), W2 (150 mm + 80%ETc) and W3 (240 mm + 80%ETc) at the seeding stage (S1), seeding stage + budding stage (S2) and seeding stage + budding stage + flowering stage (S3), with an irrigation amount of 450 mm during spring as the CK (the local reference level) (10 treatments in total). The salt of the local brackish water used was 3 g·L−1. Film-mulched drip irrigation experiments were conducted to observe cotton growth, aboveground dry matter, cotton yield, soil water and salt distribution, as well as other indicators. The results showed that the irrigation applications of S3 can improve the soil moisture and salt distribution of the root zone. The salt accumulation at the harvest stage of W3S3 was reduced by 39.5% and 2.8% compared with W3S1 and W3S2, respectively. More frequent irrigation applications can reduce a soil’s total dissolved solids (TDS), avoid exceeding the salt tolerance threshold of 4.8 g kg−1 and lead to higher aboveground dry matter and cotton yields. W3S3 obtained the highest yield of 5685 kg ha−1, which was increased by 39.59%, 7.85% and 11.25% compared with W3S1, W3S2 and CK, respectively. The higher the irrigation amount, the less water use efficiency (WUE), following the order of S3 > S2 > S1 > CK at various growth stages. W3S1 obtained the lowest WUE of 0.64 kg·m−3. Comprehensively considering the effects of soil moisture retention and salt suppression, cotton growth, yield and water use efficiency, an irrigation amount of 240 mm brackish water at three growth stages, with 80%ETc for irrigation, is recommended for the sustainable production of cotton in southern Xinjiang.

In order to provide a scientific technology reference for cotton planting and management in southern Xinjiang that is based on practical investigations and existing achievements, cotton planting and irrigation methods were divided into four groups. In addition, the applications of these methods are discussed. The four types of cotton‐planting methods are as follows: (i) one film + no pipes + four rows; (ii) one film + single pipe + four rows; (iii) one film + double pipes + four rows; and (iv) double films + double pipes + four rows. The four types of cotton field irrigation methods are as follows: (i) winter irrigation (spring irrigation) + routine flooding irrigation; (ii) winter irrigation (spring irrigation) + drip irrigation under mulch; (iii) seeding after drip irrigation + drip irrigation under mulch; and (iv) dry seeding before drip irrigation + drip irrigation under mulch. The planting and irrigation methods for cotton supplement each other and are coupled based on local climate, soil, surface water and groundwater conditions. In addition, the most scientifically reasonable, cost‐effective, water‐saving and productive methods must be selected. Copyright © 2016 John Wiley & Sons, Ltd.

Intensive and large-scale drip irrigation plays a crucial role in ensuring cotton production in Northwest China. However, significant differences in cotton production have occurred at times within large-scale irrigation systems, and quantitative information on the importance and interactions of factors related to cotton growth and constraints is scarce. In 2018–2019, we monitored six possible constraints (irrigation depth, soil texture, soil salt, soil moisture, soil inorganic nitrogen and soil organic matter) associated with drip irrigation management and seed cotton yields in a collective drip irrigation system (CDIS, composed of several drip irrigation subsystems (DISs)) in southern Xinjiang to assess the importance of different factors and identify the main constraints. In 2023, other more refined field trials were conducted to further evaluate the influencing mechanism of the main constraints on crop growth in one typical DIS within the selected CDIS. The results revealed large yield differences within the CDIS; although the average seed cotton yield was good (2018: 8051 kg ha−1, 2019: 6617 kg ha−1). Excessive irrigation depths (>500 mm) and coarse soil texture (sand content > 70%) were identified as the main constraints, affecting more than 45% of the plant area in the CDIS based on boundary line analysis (a typical analysis method to study the responses between variables) The results from the DISs revealed that the two constraints directly affected the soil moisture and soil inorganic nitrogen in the root zone, which reduced the effectiveness of irrigation and fertilization under drip irrigation. The Structural Equation Model (used to evaluate the causal relationships among multiple variables) revealed that both irrigation depth and soil texture indirectly affect yield by affecting soil inorganic nitrogen and plant N uptake and that soil nitrogen management is critical in resisting yield decline caused by constraints. An optimized irrigation schedule, improved uniformity of the drip irrigation network and adjusted drip fertilization strategies could be used for site-specific management to address the yield decline due to the main constraints and improve water and fertilizer use efficiency under drip irrigation management.

Evaluation of evapotranspiration and deep percolation under mulched drip irrigation in an oasis of Tarim basin, China

This study investigates the impact of Aerated Subsurface Drip Irrigation (ASDI) on the growth and yield of mulched cotton, aiming to identify the optimal water-air combination pattern for ASDI in cotton cultivation. Conducted during 2021–2022, the experimental setup involved two aeration modes (aerated A1 and unaerated A0) and four irrigation quotas (W1, W2, W3, and W4), organized in a two-factor randomized block design resulting in eight distinct treatments. The findings revealed that ASDI significantly promoted soil moisture depletion from 0 to 40 cm during the cotton flowering and boll opening stages. Specifically, aerated A1 reduced soil water content by 5.84% to 7.83% during the flowering stage and 7.45% to 13.39% during the boll opening stage compared to unaerated A0. Additionally, both aerating and increasing irrigation quotas not only enhanced the cotton leaf area index (LAI) but also delayed leaf area decay, contributing to prolonged photosynthetic activity. Aerating also favorably influenced the distribution of above-ground biomass in cotton towards budding and boll stages, with the biomass share of buddings, flowers, and bolls averaging 62.98% under aerated conditions versus 62.27% under non-aerated conditions during the boll opening stage. Furthermore, aerating combined with increased irrigation quotas resulted in higher seed cotton yields, with aerated irrigation boosting yields by 1.79% in 2021 and 4.43% in 2022 compared to non-aerated irrigation. This approach also increased cotton’s water demand and average daily water consumption significantly (p < 0.01). Importantly, aerating improved IWUE, achieving 1.72 kg/m3 in 2021 and 1.62 kg/m3 in 2022 for ASDI, versus 1.69 kg/m3 and 1.57 kg/m3 for unaerated subsurface drip irrigation, respectively. In conclusion, from a water conservation and yield enhancement perspective, an irrigation quota of 337.4 mm during the reproductive stage under ASDI is recommended as an effective strategy for “one film three tubes and six rows” mulched cotton in Southern Xinjiang.

A field experiment was conducted on 16-year-old ‘Wen 185’ walnut trees in Aksu, Southern Xinjiang, to identify optimal water and fertilizer management under subsurface drip irrigation. Four irrigation levels were established: 75% ETc (W1), 100% ETc (W2), 125% ETc (W3), and 150% ETc (W4). These were combined with three fertilizer levels: N 270, P 240, K 300 kg ha-1 (F1), N 360, P 320, K 400 kg ha−1 (F2), and N 450, P 400, K 500 kg ha−1 (F3). This resulted in a total of 12 treatments. This study assessed the impact of different water and fertilizer treatments on walnut growth dynamics, yield, fruit quality, water and fertilizer use efficiency, and soil nitrate residue. Principal component analysis (PCA) was used to construct comprehensive growth and photosynthesis indices (CGI and CPI). Parameters significantly correlated with yield and quality were then screened via Pearson analysis, and a game theory-based combination weighting method was adopted to determine weights for integrating six categories of indicators: growth, photosynthesis, yield, quality, resource use efficiency, and environmental impact. A coupled TOPSIS-GRA model was developed for comprehensive evaluation. Furthermore, binary quadratic regression was employed to optimize the application ranges of water and fertilizer. The results showed that the W2F2 treatment achieved the highest rank by synergistically enhancing growth, photosynthetic performance, yield, and quality. This treatment also maintained high water use efficiency (WUE) and partial factor productivity of fertilizer (PFP) and effectively reduced nitrate accumulation in deep soil layers. The CGI and CPI, derived from PCA, effectively quantified phenological growth and photosynthetic characteristics. Correlation analysis identified seven core parameters, among which IV-CPI correlated most strongly with yield. In contrast, II-CPI was more closely associated with increased single-fruit weight and reduced tannin content. Within the comprehensive evaluation system that used game theory-based combination weighting, yield received the highest weight (0.215), while IV-CPI was assigned the lowest (0.011). The TOPSIS-GRA coupled model identified the W2F2 treatment as the highest-ranked. Furthermore, regression optimization determined the optimal total seasonal application ranges to be 5869.94–6519.81 m3 ha−1 for irrigation and 975.54–1107.49 kg ha−1 for fertilization. The coupled TOPSIS-GRA model enabled a balanced assessment of the objectives: high yield, superior quality, resource use efficiency, and environmental sustainability. Thus, it provides a theoretical foundation and practical guidance for enhancing the productivity and sustainability of subsurface drip-irrigated walnut orchards in Southern Xinjiang.