Sowing period has a considerable effect on cotton productivity by regulating the use of thermal resources, but its effect on various soils depth layers has not been properly measured. The current study investigated the impacts of date sowing on cotton yield, cumulative growing degree days (GDD) and soil temperatures using a multi-depth monitoring system. A field study (2022–2024) was chosen to take place in Anyang, Henan Province, China, on the North China Plain, where the climate is that of a temperate continental monsoon. In a randomized block trial, the cotton cultivar CCRI 134 was sown on six different dates; 12 April (S1), 12 May (S6). Mean daily temperature and rainfall during the growing season were 21.9 °C and 656.6 mm in 2022, 23.66 °C and 793.5 mm in 2023, and 23.2 °C and 315.9 mm in 2024. Seed cotton yield varied significantly with sowing date, but the optimum differed by year: S3 yielded 3025.9 kg ha⁻¹ in 2022, S5 yielded 3398.5 kg ha⁻¹ in 2023, and S2 yielded 2561.7 kg ha⁻¹ in 2024, compared with 2118.4, 2815.5, and 1506.8 kg ha⁻¹ under S6, respectively. Earlier sowing reduced thermal accumulation under cooler conditions. Soil thermal responses were depth-dependent, with stronger variation in the 0–30 and 30–70 cm layers, while 70–110 cm remained more stable. Multi-depth soil temperature monitoring showed that very late sowing should be avoided, and sowing between 18 April and 6 May is the most reliable window for balancing soil heat accumulation and yield.

Context Sustainable cropping systems in intensity managed regions must simultaneously maintain high productivity while reducing environmental burdens and improving resource use efficiency. Evaluating these trade-offs at the system level is essential for identifying viable pathways towards sustainable agricultural intensification in the North China Plain. Objective This study aimed to compare the sustainability performance of representative cropping system by integrating productivity, environmental impacts, ecosystem services, and energy efficiency into a unified multi-criteria assessment framework. Methods Based on three years of field experiments, nine typical cropping systems in the North China Plain were evaluated using a LCA framework. System performance was evaluated using an integrated set of cradle-to-farm-gate carbon and nitrogen footprints, productivity, ecosystem service, and energy efficiency, which were synthesized into an overall sustainability index. Results and discussions Marked trade-offs were observed among productivity, environmental performance, and resource use across cropping systems. Legume-based systems, particularly those including soybean and peanut, exhibited lower carbon and nitrogen footprints due to reduced fertilizer inputs and greater soil organic carbon accumulation. The soybean-maize intercropping system achieved the highest overall sustainability index by combining high ecosystem service provision and energy productivity, while the wheat-maize rotation generated the greatest economic returns. In contrast, cotton monoculture showed consistently low sustainability performance because of high labor inputs and relatively low yields. Significance The results demonstrate how cropping system design influences multiple dimensions of sustainability and highlight the potential of diversified legume-cereal systems to improve system-level performance in intensively managed agricultural regions under comparable agroecological and management conditions.

Abstract Background Due to decreasing production and high input costs, there has been an increased focus on optimizing planting density. Maintaining a consistent yield across multiple plant populations is a desirable way to cut costs. A two-year field experiment was conducted in Anyang, Henan, China (36°06′ N, 114°21′ E) at the Institute of Cotton Research, Chinese Academy of Agricultural Sciences, on clay loam soil, to evaluate the photosynthetic efficiency and light interception (bottom, middle, and upper canopy layers) response to different planting densities (D1, 15 000; D2, 33 000; D3, 51 000; D4, 69 000; D5, 87 000, and D6, 105 000 plants‧hm−2) for higher cotton yield. The study was conducted under the environmental conditions of average temperatures of 22.2 °C in 2019 and 21.5 °C in 2020, and rainfall of 318 mm and 392 mm in the respective years. Results The highest planting density (D6) produced the maximum leaf area index (LAI) and overall light interception in both years; however, excessive canopy density led to increased inter-plant competition and reduced yield efficiency. Although D6 had the highest LAI and light interception, the highest yield was observed at D5. This result is attributed to a more balanced canopy structure at D5, which optimized light distribution and photosynthetic efficiency, particularly in the middle and lower canopy layers, leading to better biomass accumulation and higher yield. Yield decreased at D6 due to reduced boll weight and lower reproductive efficiency. In contrast, D5 maintained a well-structured canopy with high and efficiently distributed light interception and improved photosynthetic performance, particularly in the middle and lower canopy layers. This optimized canopy architecture enhanced biomass accumulation in reproductive organs, resulting in 33% and 30% higher lint yield compared with the lowest density (D1) during the two experimental years, respectively. More seed cotton yield (4 449 kg‧hm−2) and lint yield (1 695 kg‧hm−2) resulted from D5, which is attributed to increased biomass accumulation in the reproductive organs. Conversely, the yield dropped as planting density (D6) continued to grow. Conclusion Therefore, the results showed that the cotton yield can be increased through high light interception rate and photosynthetic efficiency by adjusting planting density and the structure of the canopy.

Coordinating the efficient use of multiple natural resources is critical for ensuring sustainable crop production, particularly under increasingly variable climate conditions. The light interception rate (LIR) and soil water consumption (SWC) are fundamental to increasing cotton productivity, yet studies on their coupled dynamics across the soil-canopy continuum remain scarce. We established a spatial-grid, multisensor network that can be used to continuously monitor the LIR and SWC along with full soil-canopy profiles (-110-110 cm) in a staggered-sowing cotton experiment in China. Geostatistics and nonlinear models were employed to quantify the spatiotemporal changes in LIR and SWC and their links to biomass and yield. Sowing date substantially affected SWC patterns, which exhibited clear seasonal and vertical stratification characteristics. Compared with the early-sown treatments, the late-sown treatments consistently demonstrated lower consumption of shallow water (10-40 cm) but higher consumption of deep water (70-110 cm). Early sowing advanced the upper-canopy LIR peaks by 10-12 days, whereas late sowing sustained high interception (>90 %) for a longer period during peak flowering. SWC at each position along the soil profile was significantly positively correlated with the canopy LIR, highlighting the spatial coordination between above-and belowground resource capture. In normal years, the SWC at shallow positions was positively correlated with seed cotton yield, with correlations turning negative below 40 cm and being greatest at 90 cm in cotton rows, indicating an optimal zone (10-40 cm) for water extraction that supports productivity. However, this relationship may be disturbed by extreme weather. This study revealed critical spatiotemporal interactions across the soil profile and crop canopy, offering a new technical paradigm for spatiotemporally coupled soil-crop monitoring and providing a functional basis for optimizing cotton production in data-driven agricultural systems to guide precision management, yield optimization, and climate-adaptive decision-making in future smart agriculture.

Real-time monitoring of cotton above-ground biomass (AGB) is crucial for monitoring crop growth and optimizing management practices. This study evaluated UAV-based RGB and multispectral (MS) imagery for cotton AGB estimation across multiple growth stages under different planting densities and sowing dates in Anyang, China. Spectral features, vegetation indices (VIs), and Gray Level Co-occurrence Matrix (GLCM) texture metrics were extracted and organized into three scenarios: RGB + MS, RGB-only, and MS-only. Recursive feature elimination with cross-validation (RFECV) was applied for feature selection, and six machine learning models were evaluated using both baseline and selected feature sets. Results showed that model performance was strongly influenced by growth stage, sensor configuration, and feature composition. Accuracy was highest at the seedling and squaring stages and decreased at flowering due to canopy complexity and spectral saturation. MS-only and fused features generally performed best at the seedling stage, while RGB-only features were competitive or superior at the squaring stage, highlighting the importance of high-resolution structural information. At flowering, fused RGB–MS features provided the most stable performance, although improvements were limited. RFECV exhibited stage-dependent behavior, improving performance mainly at early growth stages but showing inconsistent benefits later. SHAP analysis revealed a shift from texture-dominated predictors at the seedling stage to balanced feature contributions at squaring and vegetation index (VIs) dominance at flowering. Overall, cotton AGB estimation is a stage-dependent process requiring adaptive sensor and feature selection strategies.

This study aimed to elucidate how canopy architecture influences the distribution and utilization of soil moisture and light, with the ultimate goal of improving cotton yield performance under different environmental conditions. This study, conducted over 2 years (2020-2021) at the Institute of Cotton Research, Chinese Academy of Agricultural Sciences, Anyang, China, evaluated six Gossypium hirsutum L. (Upland cotton) varieties (T-0, Ji228, SCRC28, TQ-1, CCRI50, and CCRI60) using a randomized complete block design. The investigation focused on patterns of soil moisture and light distribution across different canopy architectures to identify traits associated with high yield performance. Soil moisture was monitored using sensor grids, while light interception (LI) and leaf area index (LAI) were measured throughout the growing season across six cotton cultivars. Loose-type varieties such as Ji228 and SCRC28 exhibited higher LAI values (up to 4.23), greater vegetative biomass (up to 11,063.98 kg hm-2), and higher LI (ranging from 0.54 to 0.85). Soil moisture was mainly utilized in the 20- to 60-cm depth, with losses in the 0-20 cm and 60-80 cm layers. Ji228 and SCRC28 achieved the highest seed cotton yields in both years. Canopy structure significantly affects water and light distribution, influencing biomass accumulation and yield. Loose-type varieties, particularly Ji228 and SCRC28, demonstrated superior performance, indicating their adaptability and potential for higher yield under diverse environmental conditions.

In arid and semi-arid regions, improving water use irrigation efficiency under limited seasonal water supply is critical for sustainable cotton production. While the effects of seasonal irrigation amount have been widely studied, the independent and interactive roles of irrigation interval under a fixed seasonal irrigation constraint remain insufficiently quantified. This study aimed to evaluate how irrigation amount and interval jointly regulate soil water dynamics, evapotranspiration partitioning, yield formation, and water use efficiency (WUE) in cotton. A two-year, controlled soil-column experiment was conducted using a full-factorial design with two seasonal irrigation amounts (350 and 200 mm) and four irrigation intervals (every 3, 6, 9, or 12 days). The AquaCrop model was locally calibrated with 2024 data and validated with independent 2025 observations. The validated model was then used to conduct scenario simulations across 16 irrigation combinations to analyze process-level responses. The model reproduced canopy cover and soil water storage (SWS) dynamics with good accuracy (R2 > 0.89; NRMSE < 16%). The results showed that irrigation amount primarily controlled overall water availability, whereas irrigation interval reshaped the temporal fluctuation pattern of soil water content (SWC) in the shallow root zone. Under moderate irrigation levels, an intermediate interval (every 6 days) improved WUE by stabilizing SWC and maintaining high transpiration proportions. Under severe deficit conditions, prolonged intervals intensified periodic water stress and reduced yield. Simulated transpiration accounted for 95–97% of seasonal evapotranspiration in the controlled system, reflecting limited soil evaporation under column conditions. These findings highlight that irrigation interval, beyond total irrigation amount, is an important management variable for optimizing cotton irrigation scheduling under water-limited conditions. The combined experimental and modeling framework provides practical guidance for irrigation design in arid regions.

Radiation use efficiency (RUE) is closely associated with cotton biomass and yield, yet the synergistic regulation of phenotypic structure and physiological potential remains unclear. A field experiment (2024-2025) in Anyang, China, utilized three independent trials: six sowing dates (from 12 April to 12 May at 6-day intervals, S1-S6), six planting densities (1.5, 3.3, 5.1, 6.9, 8.7, and 10.5 & times; 10(4) plants & centerdot;ha(-1), D1-D6), and ten cultivars with distinct architectures (V1-V10). Feature importance and structural relationships were quantified via random forest (RF) and partial least squares structural equation modeling (PLS-SEM). Results indicated that delaying sowing reduced true leaf number (TLN) and plant height (PH), with the April 24 sowing (S3) optimizing leaf area index (LAI, 2.57) and light interception rate (iPAR, 0.61). Increasing density significantly enhanced population-level LAI, above-ground biomass, and RUE, despite a progressive decline in TLN. Among cultivars, CCRI 60 (V6) exhibited superior structural traits (PH: 72.94 cm; iPAR: 0.61), while CCRI 113 (V8) exhibited the highest maximum carboxylation rate (V-cmax, 88.9 mu mol & centerdot;m(-2)& centerdot;s(-1)) and RUE (4.88 g & centerdot;MJ(-1)). Across the comprehensive dataset (integrating the density, sowing date, and cultivar trials), iPAR exhibited the highest relative importance (42.01%) for RUE variation, while associated structural traits (PH, LAI, TLN) yielded a cumulative relative importance of 41.69%. RUE was strongly associated with biomass accumulation (path coefficient > 0.97), which subsequently optimized yield components. Conversely, within the cultivar-comparison subset, the relative importance of iPAR decreased to 17.95%, while V-cmax rose significantly to 19.20%. PLS-SEM indicated that canopy structure exerted a significant negative association with photosynthetic potential (V-cmax, J(max)) within this cultivar subset (path coefficient approximate to -0.51), whereas enhanced physiological potential was positively associated with resource allocation to yield components (path coefficient approximate to 0.57). Consequently, mitigating the inherent trade-off between canopy structure and leaf photosynthetic capacity is critical for further improving RUE and cotton yield under similar production environments.

The relationship between planting density and resource utilization, including water and light, has been extensively studied concerning cotton (Gossypium hirsutum L.) biomass and yield. However, understanding the effect of planting density on soil heat resource utilization remains limited. This study investigates the impact of planting density on soil heat utilization across different soil layers and growth stages, focusing on cotton biomass, yields and soil heat conversion efficiency (PEsoil). Conducted over 3 years (2022-2024), the experiment utilized various planting densities (D1:15,000, D2:33,000, D3:51,000, D4:69,000, D5:87,000 and D6: 105,000 plants ha-1) in a randomized complete block design. Our findings indicated that biomass accumulation showed a significant positive correlation with the effective accumulated temperature (EAT) in 2022, but this relationship turned negative in 2023 and 2024. PEsoil correlated positively with yield across treatments. Under warmer temperature conditions, D5 achieved the highest yield, exceeding D3 and D4 by 5% and 7%, respectively. In contrast, under moderate temperature conditions with well-distributed rainfall, D3 and D4 outperformed D5 by 4% and 3%, respectively. These findings suggest that excessively high planting densities did not further improve yield or PEsoil. Instead, optimizing planting density according to climatic conditions is critical for maximizing cotton productivity and soil heat conversion efficiency. Moderate planting densities (D3-D4) are recommended under normal climatic conditions, whereas slightly higher density (D5) may improve yield under warmer conditions. This climate-adaptive density management strategy provides a practical pathway for improving sustainable cotton production in the North China Plain.

Non-mulched crop planting can address residual film pollution in cotton fields in arid regions. However, little is known about the efficiency of soil hydrothermal resource utilization in different canopy layers under various drip irrigation regimes in non-mulched cotton fields. Therefore, field experiments with three irrigation regimes were conducted from 2020 to 2021, including deficit (W4, 3660 m3 ha-1;12-day interval), conventional (W6, 5040 m3 ha-1; 8-day interval), and excessive (W8, 6252 m3 ha-1; 6-day interval) irrigation. The drip irrigation regimes for the treatments were the same as those in film-mulched cotton fields at the seedling and budding stages. The volume for each irrigation for each treatment during the flowering and boll-setting stages was 690 m3 ha-1. The effects of different drip irrigation regimes on the growth and hydrothermal resource utilization across different canopy layers of non-mulched cotton were comprehensively evaluated. The results showed that W8 resulted in the highest water consumption (WC) during the flowering and boll-setting stages, which was 5.55%-112.43% higher than that of W4 and W6, but with lower effective accumulated temperature in soil (SEAT) during the boll-opening stage. In 2020 and 2021, at the boll-opening stage, the daily average effective soil temperature of W6 was 0.36 and 0.92 degrees C higher than that of W8, respectively, and the SEAT of W6 was 4.17% and 11.11% higher than that of W8, respectively. The W8 treatment stimulated horizontal and vertical growth and yield formation in the upper canopy. The W6 treatment stimulated yield formation in the middle and lower canopies, with yield comparable to that of W8. The accumulated heat productive efficiency in soil (PE) in the lower and middle canopies of W6 was not significantly different from that of W8. However, the water use efficiency (WUE) in the lower and middle canopies was 24.37%-78.43% higher than that of W8. In summary, the irrigation regime W6 (5040 m3 ha-1, 8-day interval) maintained a favorable soil hydrothermal environment, achieved total cotton yield comparable to that of W8, and improved hydrothermal resource utilization efficiency in the middle and lower canopies. This study provides an important reference for precision irrigation in non-mulched drip-irrigated cotton fields in arid regions.

Legume-based intercropping enhances asymbiotic biological nitrogen fixation (BNF); however, the underlying mechanisms remain unclear, including the roles of soil keystone diazotroph taxa with varying niche breadths. A field experiment was conducted to evaluate soil BNF variations between rhizosphere and bulk soils in peanut/cotton intercropping systems and monocultures. BNF activities were measured by nitrogen fixation rates, nitrogenase activity, and nifH gene abundance. Phylogenetic null models, co-occurrence networks, and niche breadth analysis were applied to investigate the roles of diazotrophic keystone taxa and their ecological niches. Rhizosphere soils exhibited 7.8-125.5% higher BNF potentials than bulk soils, whereas intercropping systems showed 11.6-323.0% increases over monocultures for nitrogen fixation rate, nitrogenase activity, and nifH gene abundance (all P<0.05). Diazotrophic community composition and diversity differed significantly, with Proteobacteria (excluding Alphaproteobacteria) enriched in intercropping and rhizosphere soils, while Cyanobacteria and Firmicutes were less abundant. Deterministic processes, particularly heterogeneous selection, dominated community assembly in the rhizosphere (91.9%) and intercropping soils (86.3%). The co-occurrence networks consistently revealed more complex and interconnected communities in intercropping and rhizosphere soils that were dominated by opportunistic diazotrophs (78.8-85.9%), followed by specialists (10.2-18.5%) and generalists (1.38-3.80%). Keystone taxa, including opportunists such as Azoarcus, Azohydromonas, and Steroidobacter, and generalists like Pseudomonas and Azotobacter, correlated positively with microbial biomass carbon and nitrate nitrogen, contributing to enhanced BNF. Peanut/cotton intercropping enhances BNF by selectively enriching the keystone diazotrophic taxa with varying ecological roles, particularly opportunists and generalists. Such targeted intercropping strategies can optimize BNF, improve soil fertility, and promote sustainable agricultural production.

Drought threatens to destroy almost 70 % of the world's cotton supply. Optimizing sowing dates is an agricultural strategy that may help synchronize ecology and productivity. Field data on the coupling impact of various environmental resources on cotton and its response to climate change under sowing date control is still lacking, though. This study examined how resource use efficiencies like water use efficiency (WUE), water consumption, water productivity and heat production efficiency (PEsoil) changed during six sowing dates (S1-S6) over two years (2023 and 2024), characterized by distinct temperature and rainfall. Results revealed that in 2023, optimal climatic conditions and well-timed rainfall events led to a maximum seed cotton yield under S4 (+178 % increase), whereas late sowing (S6) led to a -10 % decrease compared to S1. However, in 2024, delayed sowing had a more adverse impact, with yield declined up to -39 %, likely due to irregular rainfall and suboptimal temperature distribution during critical reproductive stages. The highest water use amounted to the flowering and boll development stages, exceeding 700 mm in late sowing treatments. However, WUE and WPc in delayed sowing were substantially lower than in early sowing, indicating inefficient resource conversion. Furthermore, statistical analysis of year-to-year specific positive correlations with resource use metrics were found to be significant with seed cotton yield. In 2023, WUE (R2 = 0.8350), WPc (R2 = 0.7189), and PEsoil (R2 = 0.8586) were correlated (strongly) with early sowing dates (S1 and S2) due to optimal timing of growth stages with respect to temperature and rainfall regimes. Though the overall R2 values were slightly reduced with changed rainfall pattern and cooler peak temperatures, early sowing still had a positive correlation with WUE (R2 = 0.81), WPc (R2 = 0.69), and PEsoil (R2 = 0.78) during 2024, implying stable performance under variable climatic conditions. Similarly, these early sowing treatments also had more stable aboveground biomass, had higher LAI and demonstrated the ability to synchronize phenological state with hydrothermal availability. Principal component analysis (PCA) also confirmed that early sowing increased resource use coupling and yield resilience under the two climatic years. This study introduces a novel integration of temporal sowing optimization, multisensor environmental monitoring, and resource coupling analysis. Future studies should focus on integrating climate forecasting models with sowing date recommendations to enable dynamic, site-specific cotton management.

Integrating green manure with reduced nitrogen (N) fertilization is a promising strategy to mitigate N emissions in intensive cotton cultivation, however, the underlying mechanisms remain poorly understood. This study investigated the effects of three green manure incorporation patterns-no green manure (NG), Orychophragmus violaceus (OVG), and Vicia villosa (VVG)-combined with four N reduction levels (100, 50, 25%, and conventional) on gaseous N emissions (NH3 and N2O), soil physicochemical properties, and bacterial community characteristics using a cotton field experiment in the Yellow River Basin. Results showed that OVG incorporation with 25% N reduction (N2 treatment) significantly reduced total gaseous N emissions by 36.07% on average during the cotton growth period, reducing NH3 and N2O emissions by 13.31-54.11% and 32.25-68.77%, respectively, compared with N2 application without OVG. OVG application also increased the relative abundance of Proteobacteria (28.10%), enhanced heterogeneous selection in bacterial community assembly (200%), and increased the complexity of co-occurrence networks, compared with NG. Compared with conventional N fertilization (N3 treatment), ≥50% N reduction significantly lowered NH3 (>25.51%) and N2O (>32.76%) emissions, reduced the relative abundance of Acidobacteria (-20.23%), simplified co-occurrence networks, and increased homogeneous selection in bacterial assembly (50.00%). Integrating green manure with 25% N reduction substantially reduced gaseous N emissions, which was associated with the enhanced microbial biomass carbon (MBC) and facilitated recruitment of key bacterial taxa (e.g., Sphingosinicella, Azohydromonas, Phototrophicus) within the microbial co-occurrence network. These findings provide insight into how green manure application coupled with N reduction can mitigate gaseous N losses and reshape soil microbial ecology, offering a theoretical basis for sustainable nutrient management during cotton production.

Cost-effective remote sensing solutions are critically needed to democratize precision agriculture technologies. While hyperspectral and LiDAR systems deliver high accuracy, their prohibitive costs limit widespread adoption. This study demonstrates that systematic multi-modal feature integration transforms standard UAV-based RGB imagery into a powerful phenotyping instrument, achieving crop trait prediction accuracy comparable to systems costing 10-50 times more. We developed a comprehensive framework integrating spectral indices, geometric parameters, and texture metrics from commodity RGB sensors to predict five critical cotton traits: leaf area index (LAI), intercepted photosynthetically active radiation (IPAR), above-ground biomass, lint yield, and seed cotton yield. The progressive integration approach employed Random Forest regression with four feature configurations: baseline color indices (CIbase), refined color indices (CIref), geometric parameters (CIref + GP), and texture metrics (CIref + GP + T). Field experiments across three trials over two growing seasons (2022-2023) with varying genotypes, planting densities, and sowing dates provided 2,126 ground truth measurements for model development and validation. The optimal multi-modal model achieved R2 = 0.97 for IPAR (rRMSE = 6 %), R2 = 0.91 for LAI (rRMSE = 15 %), and R2 = 0.85 for biomass (rRMSE = 32 %), with lint yield and seed cotton yield demonstrating R2 values of 0.92 and 0.77, respectively. Variance partitioning analysis revealed texture features as the dominant contributor (16.2 % +/- 7.1 %), followed by spectral indices (9.1 % +/- 4.2 %) and geometric parameters (8.0 % +/- 2.8 %), with substantial shared variance (45-65 %) indicating strong feature complementarity. Phenological analysis demonstrated that flowering-stage imagery outperformed boll opening stage measurements, while stage-general models showed superior robustness. Cross-temporal validation confirmed model generalizability, with trial-general models achieving R2 values of 0.91-0.97 for IPAR across diverse environmental conditions. The framework enables sub-meter spatial resolution trait mapping while maintaining operational simplicity and cost-effectiveness, demonstrating that systematic feature engineering can democratize high-precision phenotyping technologies for broader agricultural applications.

In the cotton-growing region of the Yellow River Basin in China, cotton sowing dates typically range from early May to early June. The yield of early-maturing cotton varies significantly with sowing dates and growing seasons. We hypothesize that these yield variations are primarily due to differences in resource utilization efficiency. A 3-year field experiment was conducted from 2021 to 2023 with early maturing cotton variety 'Chinese Cotton Research Institute 134' to monitor light and heat resources at different sowing dates. Results showed that cotton sown on May 6, 2023, achieved the highest yield (3455.26 kg ha(-1)), while April 12, 2021, had the lowest yield (1832.41 kg ha(-1)), with the highest yield exceeding the lowest by 28.0 %, 42.8 %, and 46.97 % over three years. LAI negatively correlated with yield, while true leaf number and biomass were positively correlated. PLS analysis showed that light energy utilization (Eu) influences yield through biomass and accumulated temperature utilization efficiency (TPE), while heat resource utilization (HUE) and TPE directly influence yield (r = 0.776, r = -0.971, P < 0.001). Sowing date significantly affected true leaf number, LAI, and yield, with climate factors indirectly impacting yield. Biomass, HUE, and TPE directly influenced yield. 2022 showed higher biomass and LAI peaks in the late sowing, and higher yield in the early sowing. Late sowing resulted in higher yield in 2021 and 2023. This study highlights the role of sowing date and growing season in regulating cotton yield through resource utilization, providing a basis for optimizing cotton production management.

The individual effects of cropping patterns and planting densities on cotton yield formation and resource utilization have been extensively studied in the arid regions of western China, but research on their combined impacts remains limited. This study hypothesized that optimizing cropping patterns and planting densities would enhance hydrothermal resource productivity and cotton yield in the region. To test this, a two-year field experiment (2022-2023) employed a split-plot design with two main planting patterns (four rows per film and six rows per film) and three planting densities (low, medium, and high) as subplots. Using internet of sensor technology, soil temperature and moisture were monitored to assess their spatial and temporal distributions. The effects of planting pattern, density, and their interactions on cotton yield, yield components, biomass accumulation, and water and heat utilization were evaluated. The interaction between pattern and density significantly influenced cotton yield, harvest index, and water productivity, with planting density exerting a stronger effect on water productivity than planting pattern. In 2023, the four-row pattern at low and medium densities produced higher yields than the high-density treatment. Over the two-year period, the four-row, low-density treatment achieved 8.77 % and 13.40 % greater water productivity than the medium- and high-density treatments, respectively, while the six-row, medium-density treatment outperformed low and high densities, increasing water productivity by 3.64 % and 8.74 %. Seed cotton yield was also higher, with a 2.88 % and 6.15 % increase in the four-row, low-density treatment and an 8.51 % and 4.79 % increase in the six-row, medium-density treatment compared to higher-density treatments. The study further analyzed spatial and temporal variations in soil moisture and temperature and their link to resource productivity and cotton yield. Soil water content differences ranged from 0.10 to 0.90 mm in the four-row pattern and from 0.20 to 0.70 mm in the six-row pattern between low- and high-density treatments. Planting density significantly affected soil temperature during flowering and boll-setting stages. Lint and seed cotton yields showed positive correlations with soil heat production efficiency (PEsoil) and negative correlations with water production efficiency (WPc), with optimal patterns observed in the four-row, low-density and six-row, medium-density treatments. These findings explain why these configurations led to a higher harvest index and enhanced hydrothermal resource productivity. This study provides valuable insights into the optimal configurations for maximizing cotton yield and resource efficiency in arid regions, supporting sustainable cotton production under resource-limited conditions.

With the advancement of agricultural information technology, sensors have become instrumental in monitoring soil water environments, opening new avenues for optimizing water management strategies. This study utilized high spatiotemporal resolution sensors and a grid sampling method to monitor soil moisture distribution during the squaring and flowering and boll developing stage of cotton (Gossypium hirsutum L.) under varying planting densities. Geostatistical methods were employed to calculate soil water consumption (SWC) distribution and dynamics, quantifying water competition patterns among cotton populations at different planting densities. The analysis integrated cotton growth dynamics and yield to examine the relationship between biomass, yield, and soil water utilization. Results showed significant interannual variability in cotton growth curves, with planting density notably affecting underground biomass, yield, and the distribution of SWC within the soil profile. A positive correlation was found between water consumption at depths of 30-50 cm and yield, even under low water consumption conditions. The three-dimensional efficiency map showed that a planting density of approximately 210,000 plants center dot hm-2 with SWC between 250.0 and 400.0 mm resulted in stable, high biomass and yield. The double Gaussian model indicated that with increasing SWC, a first yield peak was at the SWC of 255.9 mm, after which there was a decline in water use efficiency (WUE) (the slope of yield vs. SWC). The second yield peak was at a greater SWC of around 630.0 mm. This study also found that by controlling the soil water consumption of cotton at different densities, the biomass and yield of cotton can be quantitatively regulated, thereby reducing the yield differences caused by interannual effects and varying planting densities. These findings provide valuable insights into the spatial competition and efficient utilization of soil moisture in cotton populations, offering important guidance for achieving high, stable yields and precision water management in cotton production.

Legume-based intercropping, such as the peanut-cotton system, stands out as a promising strategy for enhancing soil ecosystem multifunctionality (EMF); however, the underlying microbial mechanisms driving these enhancements remain inadequately explored. In this study, after implementing peanut-cotton intercropping for six consecutive years, a data set of 13 ecosystem functional indicators including 41 soil variables, was obtained and used to quantify the average EMF index. We investigated changes in microbial keystone taxa in co-occurrence networks, community assembly processes, carbon (C) cycling profiles, and their collective impacts on soil EMF. Soil EMF increased by an average of 140.0 % in the peanut-cotton intercropping system, compared with monoculture systems of both peanut and cotton, driven by significant increases in C-cycling (159.9 %), nutrient provisioning (91.2 %), and microbial growth efficiency functions (53.9 %). The peanut-cotton intercropping system significantly increased the average well-color developments (AWCD), abundance of C-fixation and Cdegradation genes, and related pathways, resulting in a highly vigorous microbial C-cycling profile. The microbial community assembly processes shifted from a balance of stochastic and deterministic processes in monocultures to predominantly deterministic processes (>70 %) in the intercropping system. Additionally, the peanut-cotton intercropping system fostered a more efficient and stable bacterial-fungal cross-kingdom network than the monocultures, characterized by a higher average clustering coefficient, higher robustness, and shorter average path length. This intercropping system also recruited a group of keystone taxa affiliated with Proteobacteria, Actinobacteria, and Ascomycota phyla. The enhancement of EMF in the peanut-cotton intercropping system resulted from the positive impact of key microbial community members and their assembly, C/N ratios, AWCD, and C-fixation and C-degradation genes. Our study provides insights into the complex ecological linkages between microbial communities, C-cycling profiles, and soil ecosystem functions, providing valuable insights into the microbial mechanisms underlying the benefits of intercropping systems.

Climate change and market demands emphasize the importance of both the average quality and long-term stability of end-use fiber quality in cotton production. Intercropping, a sustainable agriculture practice, has demonstrated potential to enhance cotton fiber quality compared to monoculture. However, the impacts of intercropping on cotton fiber quality stability and key factors influencing quality variability remain poorly understood. This study analyzed five years of data to investigate the variability and controlling factors of cotton fiber quality in both monoculture and intercropping. The results indicated that intercropping maintained cotton fiber length within the long fiber range. While intercropping did not influence fiber quality temporal stability or the risk of quality decline, higher stability was associated with a reduced probability of quality deterioration. Trade-offs among fiber quality traits were observed, indicating that improving one trait often negatively impacted another. Intercropping had minimal impact on these trade-offs. Weather factors accounted for 17-33 % of fiber quality variability, with the effects varying by specific quality trait. Precipitation and photosynthetically active radiation positively influenced fiber length, strength, and uniformity index but negatively affected micronaire and elongation. In contrast, temperature positively influenced all measured fiber quality traits. Overall, our results demonstrate intercropping maintained relatively stable fiber quality over time. This study also reveals strong trade-offs among multiple cotton fiber quality traits, underscoring the need to close trade-offs and achieve synergistic improvements in future research.

Unmanned Aerial Vehicles (UAVs) offer enhanced spatial and temporal resolution for agricultural remote sensing, surpassing traditional satellite-based methods. Given the abundance of evolving machine-learning methods for crop recognition, this study evaluates and compares five machine learning algorithms (ML) and tests an Ensemble Learning method as a sixth approach, integrated with object-based image analysis (OBIA) for crop-type classification using UAV multispectral imagery, aiming to identify the most effective model and produce a classification map based on the best-performing method. Image segmentation was built using eCognition software, and spectral, index, and gray level co-occurrence matrix (GLCM) features were extracted from the segmented object. A machine learning model integrating multiple classification algorithms (SVM, ANN, RF, XGBoost, KNN, Ensemble Learning) with automated hyperparameter optimization was developed and executed in Google Colab using Python 3.10. All classifiers achieved accuracies exceeding 80% and Area Under the Curve (AUC) values above 0.9. SVM and ANN are the best classifiers, with the same value of accuracy (94%), followed by XGBoost (93%), RF (92%), and KNN (89%). The Ensemble Learning method (SVM + ANN) as a sixth approach outperformed all single models, with an accuracy value of 95%. Cotton, maize, peanut, and soybean were classified with the highest accuracy, with index and GLCM features contributing most significantly, followed by spectral features. The integration of high-resolution UAV imagery with ML and OBIA demonstrates strong potential for automated crop-type classification, offering valuable support for precision agriculture applications.