Predicting Cotton Growth, Pod Development, and Lint Yield Responses to Irrigation and Cultivar Differences in Southwest Texas

Predicting Cotton Growth, Pod Development, and Lint Yield Responses to Irrigation and Cultivar Differences in Southwest Texas | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Predicting Cotton Growth, Pod Development, and Lint Yield Responses to Irrigation and Cultivar Differences in Southwest Texas Uzair Ahmad, Xuejun Dong, Thiago F. Duarte, Noemi Tortorici, Dale A. Mott, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9500734/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract In semi-arid regions, water scarcity limit root-zone water availability, restricting cotton growth and yield. Although prior studies evaluated drought responses and cultivar variations, interactions between irrigation and cultivar-specific dynamics—validated through process-based models—remain underexplored in southwest Texas. This study addresses this gap by quantifying the effects of four irrigation treatments on two cultivars, NG 4190 B3XF and ST 4990 B3XF, using field observations and DSSAT (Decision Support System for Agro-technology Transfer) CROPGRO-Cotton model. Full (100%) and moderate (75%) irrigations were based on field data, whereas half (50%) and no (0%) treatments were simulated. Key variables include phenology, pod numbers (m -2 ), pod, stem and leaf weights (all kg ha -1 ), AGB (aboveground biomass, kg ha -1 ), height (m), LAI (leaf area index), and lint yield (kg ha -1 ). Results showed that phenology varied, with differences of 5 days in anthesis, 4 in boll formation, 2 in seed set, and 5 in maturity. Lint yield varied across treatments, with MAE (mean absolute error) values of 11% and 13% for full and moderate treatments, respectively. Full irrigation maximized growth, with NG 4190 showing higher pod number (MAE 5%), pod weight (MAE 10%), and AGB (MAE 13%) than ST 4990. Half irrigation reduced pod numbers, pod weight, and AGB in both cultivars. No irrigation caused large declines in both cultivars, with MAE of 16–19% across pod number, pod weight, and AGB. Sensitivity analysis showed strong irrigation effects on boll development and AGB. Moderate irrigation sustains ~ 90% yield by preserving boll development and optimizing cotton production. Agronomy dssat cotton irrigation GCM climate yield Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Cotton ( Gossypium hirsutum L.) is one of the most valuable fiber crops globally and a cornerstone of the agricultural economy in Texas (Himanshu et al., 2023). In the United States, cotton is a key field crop that drives on-farm revenue, supports large-scale production systems, and provides significant value to national and international agricultural markets (Loka et al., 2011). However, cotton cultivation is water-intensive and sensitive to water supply timing, especially in semi-arid environments. As such, optimizing cotton water use while maintaining yield is a critical agronomic challenge, particularly in regions where groundwater and surface water resources are constrained (Singh et al., 2010). The southwest region in Texas typifies this challenge. This area experiences limited and highly variable rainfall, high atmospheric evaporative demand, and long growing seasons, which together place intense stress on water resources (Mishra et al., 2011). Many growers rely heavily on irrigation, often withdrawing water from underground sources such as the Edwards Aquifer (Ko et al., 2009; Piccinni et al., 2007). Climate variability, declining aquifers, and increasing regulatory restrictions make it essential to develop strategies that maximize cotton yield per unit of applied water. In this context, the use of crop simulation models offers a powerful tool to explore and optimize cotton growth under different irrigation treatments (Adhikari et al., 2016). Process-based models like DSSAT-CROPGRO-Cotton (Hoogenboom et al., 2019) simulate critical physiological processes—phenology, canopy expansion, carbon assimilation, biomass partitioning, and final yield—all in response to weather and soil conditions and management practices (Cammarano et al., 2012; Hoogenboom et al., 2024). When locally calibrated, these models can predict cultivar-specific responses to water stress, support decision-making around irrigation timing and rate, and identify cultivar traits that confer resilience. Despite the utility of these models, successful applications require careful calibration using field data, particularly in semi-arid zones where crop response to water varies strongly with cultivar and management (Guerra et al., 2007; Jones et al., 2003). In Texas, cotton cultivars such as NG 4190 and ST 4990 are widely used, yet their responses under reduced irrigation (e.g., 75%, 50%, and 0%) remain poorly quantified (Morgan et al., 2024). Without site-specific calibration, model predictions may misrepresent phenological shifts, leaf area dynamics, biomass allocation, or yield under water deficit. For instance, deficit irrigation may induce earlier flowering, accelerate leaf senescence, or alter biomass partitioning that reduces lint yield, but these trade-offs are not always obvious from field measurements alone (Ko et al., 2009). By calibrating CROPGRO-Cotton with field data from Texas, we evaluated hidden responses and generated realistic simulations to guide water management under water-limited production conditions. Our study therefore addresses two critical needs: first , to calibrate the CROPGRO-Cotton model for NG 4190 and ST 4990 using phenology, LAI, biomass (pods, stems, leaves), and yield data collected under four simulated irrigation treatments (100%, 75%, 50%, 0%); second , to use the calibrated model to simulate how these cultivars allocate biomass, adjust their growth, and set yield under water deficit. By combining field measurements and model simulations, we aim to generate insights that are difficult to observe in standard agronomic trials, such as how LAI declines under stress, how biomass partitioning shifts, and how reproductive sinks respond. The importance of this work lies in several interconnected factors. Phenology—particularly the timing of flowering and maturity—is a key determinant of yield under water stress (Bednarz & Nichols, 2005). Earlier or compressed reproductive windows can lead to more efficient use of water but may reduce the number of bolls or their fill (Herritt et al., 2022). Reductions in LAI under stress may lower the crop’s source capacity, limiting assimilate supply to the developing bolls (Bista et al., 2024). Biomass partitioning between pods, leaves, and stems underlies the sink–source balance: water deficits may shift allocation away from reproductive tissues, reducing lint yield (Tariq et al., 2024). Moreover, cultivar differences in all these traits determine their drought resilience: one genotype may sustain LAI and biomass better under stress, while another may prematurely abort fruit (Herritt et al., 2022). Using a calibrated model, we can quantify these physiological responses and provide actionable information for irrigation strategy, cultivar selection, and agronomic optimization in semi-arid cotton production. The objective of our study is to quantify the effects of four irrigation treatments on phenology, growth, and yield of two cotton cultivars (NG 4190 B3XF and ST 4990 B3XF) and calibrate the DSSAT CROPGRO-Cotton model against field-level observations under semi-arid conditions. Our hypothesis is that under semi-arid conditions, full irrigation maximizes LAI, biomass, and lint yield in both NG 4190 and ST 4990, while deficit irrigation triggers cultivar-specific changes in growth, biomass partitioning, and yield stability. To test this hypothesis, we posed three research questions, ( 1 ) How do different irrigation treatments (100%, 75%, 50%, 0%) influence the phenological development (emergence, flowering, maturity) of the two tested cultivars in a semi-arid region? ( 2 ) In what way does LAI respond to water deficit for the two tested cultivars? and ( 3 ) How does biomass partitioning (pods, stems, leaves) change under differing irrigation treatments and impact lint yield? Materials and Methods 2.1. Study site and crop management The field experiment was conducted from April to September 2025 at the Texas A&M AgriLife Research and Extension Center at Uvalde, Texas (29.21°N, 99.76°W; elevation 282 m). The study occupied approximately 0.404 hectares of levelled land previously planted with cover crops. The site has a semi-arid climate and clay soil of the Uvalde series (fine-silty, mixed, active, hyperthermic Aridic Calciustoll; USDA, 1976). Further information can be found in Ahmad et al. (2026). A 2×2 factorial design was used with two cotton cultivars (NG 4190 and ST 4990), two irrigation treatments (100% and 75%) and with three replications, totaling 12 plots. Each plot contained four rows, 60-m long with 76.2-cm row spacing. Three weeks before planting, the field received 50 mm of irrigation via a linear sprinkler system to ensure uniform soil moisture. Cotton seeds were sown on April 8 using a six-row vacuum planter. Post-emergence, irrigation was applied through surface drip tapes placed along individual rows. Pre-plant soil testing indicated adequate nitrogen (N) and phosphorus (P) in the top 15 cm of soil; 100 kg ha − 1 N (urea, 46-0-0) and 26 kg ha − 1 P 2 O 5 (diammonium phosphate, 18-46-0) were applied to maintain sufficient nutrient availability. Potassium (K) was not added due to adequate soil K levels. Thrips damage was observed on May 7 in young seedlings and controlled with AgriMek SC (0.13 L ha − 1 ) on May 12 and Acephate (0.29 L ha − 1 ) on May 16. Plants recovered within 10 days, and no further pest or disease issues occurred. 2.2. Field measurements 2.2.1. Pod numbers (m -2 ) Pod numbers (m − 2 ) were measured at the reproductive stage by sampling a defined ground area within each plot. A 1-m 2 area was delineated in the two central rows of each plot using a measuring tape to avoid border effects. All plants within the marked area were harvested, and the pods were manually counted. The total number of pods collected from the sampled area was recorded directly as pods m − 2 and averaged across sampling locations to represent plot-level pod density. 2.2.2. Pod, stem, and leaf weights (kg ha -1 ) Pod, stem, and leaf weights (kg ha − 1 ) were determined through destructive plant sampling. Five representative plants were collected from the central rows of each plot and immediately separated into pods, stems, and leaves in the field. Each component was oven-dried at 65°C until constant weight and weighed using a precision analytical balance. The measured dry masses were converted to kilograms per hectare (kg ha − 1 ) using the sampled area and plant population. 2.2.3. AGB (kg ha -1 ) AGB (kg ha − 1 ) was calculated to quantify total biomass accumulation above the soil surface. It was obtained by summing the dry weights of pods, stems, and leaves from the destructively sampled plants. Each component weight was first expressed on a per-hectare basis using plant density and sampling area. The resulting sum represented total AGB (kg ha − 1 ) for each plot and sampling event. 2.2.4. Height (m) Plant height was measured to quantify vertical plant growth during the season. Five representative plants were collected from the two central rows of each plot to avoid border effects and transported to the laboratory. Each plant was laid straight, and height was measured from the base of the stem to the terminal bud of the main stem using a measuring tape. Measurements were recorded in meters (m), and the mean value of the sampled plants represented the plot-level plant height. 2.2.5. LAI LAI was quantified from all 12 plots using a LI-2200C Plant Canopy Analyzer (LI-COR Biosciences, Lincoln, NE, USA) for in-field indirect measurements. Measurements were conducted once every two weeks on mostly clear days, avoiding periods with patchy clouds. Five measurements along a 15-m transect near the plot center were taken to obtain five LAI values per plot. For each measurement, one above-canopy and four below-canopy photosynthetically active radiation (PAR) readings were collected along a 2-m span at varying canopy positions between adjacent rows to capture within-plot light variability. The mean of the five measurements represented the plot-level LAI for each sampling date. 2.2.6. Yield (kg ha -1 ) Cotton was hand-harvested on September 3–4 from two middle row, 5-m long segments per plot. The collected seed cotton was ginned on a 20 saw Centennial Gin and weighed to determine seed cotton yield. Subsamples of ginned lint were then sent to the Fiber and Biopolymer Research Institute, Lubbock, Texas, for fiber quality analysis. Measurements were conducted using high volume instrument (HVI) system to assess key fiber quality parameters. 2.3. DSSAT workflow Cotton growth in this study was simulated using the CROPGRO-Cotton module within DSSAT v4.8.2 (Hoogenboom et al., 2019). This process-based model predicts cotton growth, development, and yield responses under varying irrigation treatments by integrating cultivar-specific genetic coefficients, daily weather, soil characteristics, and crop management practices. CROPGRO-Cotton simulates key physiological processes including phenology, LAI, and AGB partitioning into pods, stems and leaves, and lint yield. It has been widely validated for cotton under diverse environmental conditions and stress scenarios. Model inputs included detailed soil profiles with hydraulic properties, daily weather data (maximum and minimum temperatures, solar radiation, and precipitation), cultivar-specific parameters, and management practices such as sowing dates, irrigation schedules, fertilization, and row spacing. Calibration was performed using observed field data for cotton cultivar responses to ensure accurate simulation of phenology, pod numbers, pod, stem and leaf weights, AGB, height, LAI, and lint yield. DSSAT inputs were prepared with XBuild , weather (. WTH ), soil (. SOL ), and experimental treatment (. X ) files. Yield outputs were stored in average (. A ) files, and time-series pod numbers, pod, stem and leaf weights, AGB, height, LAI growth in time (. T ) files. FileX verified consistency among weather, soil, and management inputs, linking each treatment to site-specific soil and climate files. Time-series outputs were assigned to irrigation treatments for cultivar calibration. Cultivar and ecotype parameters were calibrated by manual trial-and-error in the CNGRO048.CUL and CNGRO048.ECO files. Daily weather data were formatted using Weathergen . For calibration, key cotton phenological and growth variables were implemented and iteratively adjusted in both files to closely match observed data, enabling robust simulation of phenology, growth, and lint yield under full, moderate, half, and no irrigation. 2.4. DSSAT inputs 2.4.1. Soil data The soil at the study site has 57% clay, 31% silt, and 12% sand. Bulk density measured 1.38 g cm − 3 (0–20 cm), 1.43 g cm − 3 (20–80 cm), and 1.58 g cm − 3 (80–120 cm). The field capacity and permanent wilting point were 0.35 and 0.18 cm 3 cm − 3 , respectively. Soil pH averaged 8.3 (0-120 cm), and organic matter in the top 0–20 cm was 3.6%. Laboratory analyses quantified nitrate-N (NO 3 − N), P, K, calcium (Ca), and magnesium (Mg). These physical and chemical properties were incorporated into DSSAT v4.8.2 by creating a site-specific soil input file to accurately simulate water dynamics and nutrient availability for cotton growth and yield prediction (Refer to Ahmad et al. (2026) for detailed soil chemical and physical properties across soil depths). 2.4.2. Crop data Cotton was planted on April 8 at a uniform seeding rate of 130,000 ha − 1 for the two cultivars, NG 4190 and ST 4990, across 12 plots, with two factors: cultivar and irrigation treatment (full vs. deficit). Field and crop management practices, including planting date, fertilization, irrigation, and pest control, were applied uniformly across plots. Key cotton phenological dates were recorded: emergence on April 15, anthesis on June 15, physiological maturity on July 31, and harvest on September 3–4. Crop management details were incorporated into DSSAT via crop management . X files. Cultivar-specific genetic coefficients were defined in the . CUL file. Planting density, row spacing, and fertilizer applications were recorded in the management and fertilizer files. 2.4.2. Weather data The study area has a semi-arid climate, with mean annual rainfall of 663 mm and average yearly evapotranspiration of 1506 mm. Monthly temperatures range from 24.7°C in May to 12.2°C in December. Weather data, including precipitation (mm), maximum and minimum temperatures (°C), and solar radiation (MJ m − 2 day − 1 ), were processed and imported into DSSAT to simulate cotton growth and yield ( Fig. 1 ). 2.5. DSSAT calibration The CROPGRO-Cotton model was calibrated using the 2025 field data for NG 4190 and ST 4990, including key observations for phenology, pod numbers, pod, stem and leaf weights, AGB, height, and LAI (Table 1 ). Calibration followed an iterative manual trial-and-error approach, which allowed direct control over parameter adjustments and rapid refinement compared to automated optimization routines that require large numbers of runs and longer processing time. This method made it possible to tune cultivar behavior efficiently while matching observed field performance and requires an experienced modeler, as expert judgment is essential for interpreting parameter interactions and selecting biologically realistic values. Calibration targeted cultivar-specific parameters governing phenology, canopy development, and boll formation. EM-FL (emergence to first flower) was adjusted first to align flowering onset. FL-SH (flower to first boll set) and FL-SD (flower to seed development) were refined to capture boll initiation and early reproductive dynamics. SD-PM (seed to physiological maturity) was tuned to synchronize simulated and observed crop maturity. Canopy growth was improved by adjusting FL-LF (time between first flower and end of leaf expansion), SLAVR (specific leaf area) , and SIZLF (size of full leaf), enhancing LAI and leaf expansion accuracy. Reproductive growth was calibrated using SFDUR (seed filling duration) and SDPDV (seed per pod division factor) to match observed boll growth rates and yield partitioning. Early phenological parameters were prioritized due to their cascading effects on later development. EM-FL and FL-SH were iteratively revisited to maintain thermal-time consistency, followed by refinement of SD-PM and SDPDV to resolve cultivar-specific differences in biomass allocation, boll retention, and lint yield. Calibration performance was accepted when simulated phenology, LAI, AGB, and lint yield deviated by ≤ 10% from observations and root mean square error (RMSE) for LAI and yield remained below 15% of observed means (Seidel et al., 2018). 2.6. Sensitivity analysis The Experiment option was selected from the DSSAT shell to establish a new experimental setup. Next, the Genetics menu was accessed, and Cultivars was chosen, followed by Genetic Coefficients . The cultivar-specific genetic coefficients EM-FL , CSDL (critical short-day length), PODUR (pod development duration), SFDUR (seed filling duration), and WTPSD (weight per seed) were then selected. For each genetic coefficient, the Starting Value , Increments , and Iterations were defined. This allowed each genetic coefficient to vary systematically within a biologically plausible range while holding other genetic coefficients constant, thereby isolating its individual effect on model outputs. After configuring all coefficients, the Run button was executed to generate the experiment, and the setup was saved for reproducibility. Subsequently, the Run Model function was activated to simulate crop growth responses across the defined parameter space. Upon completion of the simulations, the Analysis option was selected to retrieve model outputs. The file PlantGro.Out was extracted, as it provides detailed information on crop phenology, pod numbers, pod, stem and leaf weights, AGB, height, and LAI. Finally, the Plot function was used to visualize simulation results. 2.7. Adaptation strategies The Experiment option was selected to create a setup for adaptation analysis. Only two adaptation strategies were evaluated: crop management adaptations and irrigation-treatments adaptations. All other inputs were held constant to isolate the effects of each strategy. Crop management adaptations were implemented through the Management menu. The Planting Details option was used to modify planting strategies, including adjustments in planting dates to represent early, optimum, and delayed sowing scenarios. Row spacing and plant population were altered. Harvesting strategies were adjusted by modifying harvest timing to align with changes in crop phenology induced by altered planting configurations. Irrigation-treatments adaptations were implemented using the Irrigation module within the Management menu. Irrigation was applied at specific cotton growth stages to represent stage-specific water availability scenarios. Separate treatments included irrigation only during the vegetative stage, only during the flowering stage, only during the grain-filling stage, and full-season irrigation as a control. Irrigation amounts per event were kept constant, allowing evaluation of crop growth and yield responses and sensitive growth stages. After all, the Run button was executed to generate the experiments, and each configuration was saved. The Run Model function was then used to simulate crop growth responses under each adaptation strategy. Then, results were analyzed using the Analysis option, and outputs were extracted from PlantGro.Out , which provided detailed information on phenology, pod, stem and leaf weights, AGB, and LAI. The Plot function was used to visualize and compare crop responses across adaptation strategies. Table 1 Final calibration parameters for cotton cultivars NG 4190 and ST 4990 using manual trial-and-error in 2025 at Uvalde, Texas. Crop Crop cultivars Calibration methodology Evaluation metrics Parameters names Parameters threshold Parameters variability Final values Cotton NG 4190 Manual trial-and-error MAE / RMSE EM-FL 34.00–44.00 ± 5.0–10.0 39.5 FL-SH 8.00–12.00 ± 1.0–3.0 9.0 FL-SD 15.00–18.00 ± 2.0–3.0 17.5 FL-LF 65.00–75.00 ± 4.00–10.00 69.00 SLAVR 170.00-175.00 ± 3.0–5.0 173 SIZLF 250.00-300.00 ± 10.0–50.0 299.0 SFDUR 24.00–35.00 ± 4.0–8.0 29.0 SDPDV 20.00–27.00 ± 5.00–10.00 27.00 ST 4990 Manual trial-and-error MAE / RMSE EM-FL 34.00–44.00 + 5.0–10.0 39.0 FL-SH 8.00–12.00 + 1.0–3.0 11.0 FL-SD 15.00–18.00 + 2.0–3.0 17.0 FL-LF 65.00–75.00 + 4.00–10.00 72.00 SLAVR 170.00-175.00 + 3.0–5.0 175 SIZLF 250.00-300.00 + 10.0–50.0 263.0 SFDUR 24.00–35.00 + 4.0–8.0 32.0 SDPDV 20.00–27.00 + 5.00–10.00 25.00 Note : Calibration was performed using the manual trial-and-error method. Alternative methods, including GLUE, PEST, or TSE, can be applied depending on study objectives and user expertise. 2.8. Model performance evaluation Simulated and observed phenology, LAI, AGB partitioning into pods, stems, leaves, and lint yield were compared with the observed values to quantify CROPGRO model accuracy under site-specific inputs. Model performance was assessed using mean absolute error (MAE, Eq. 1 ) and root mean square error (RMSE, Eq. 2 ) (Jamshidi et al., 2024). All analyses were conducted in R (RStudio) using custom scripts with ggplot2, dplyr, readxl, and tidyr for data cleaning, analysis, and visualization. Model performance was considered acceptable when both MAE and RMSE were ≤ 20% of the observed mean. All MAE and RMSE values are reported as percentages relative to the mean of the observed values (Willmott & Matsuura, 2005): \(\:\begin{array}{cc}MAE=\frac{{\sum\:}_{\text{i}=1}^{\text{n}}{S}_{i}-{O}_{i}}{\text{n}}&\:\end{array}\) 1 where: O i corresponds to the observed data s i denotes the simulated data n represents the total number of samples \(\:RMSE=\frac{1}{n}\sqrt{\sum\:_{i-1}^{n}{({y}_{i}-{\widehat{y}}_{i})}^{2}}\) 2 where: y i represents the simulated data \(\:{\widehat{y}}_{i}\:\) corresponds to field-observed data employed n denotes the total number of samples 3. RESULTS 3.1. Model calibration 3.1.1. Phenological stages Under full irrigation (100%), the simulated anthesis of NG 4190 occurred on June 10 (63 days after planting [DAP]), closely matching the observed anthesis on June 15 (68 DAP), with an observed difference of 5 days (Table 2 ). For ST 4990, simulated anthesis was on June 9 (62 DAP), and the observed date was June 15. The simulated onset of boll formation was July 4–5 (87–88 DAP) under full irrigation for both cultivars, aligning closely with observed first-boll dates around 92 DAP. Under moderate water stress (75% irrigation), observed boll initiation occurred 1–2 days later than the full irrigation treatment, while simulated boll initiation under 50% irrigation was delayed by 2–3 days. Under no irrigation (0%), simulated boll initiation was delayed, indicating restricted reproductive development under severe water stress. Simulated first seed set occurred around June 27 (80 DAP) for NG 4190 and June 25 (78 DAP) for ST 4990 under full irrigation. Under the 75% irrigation treatment, this stage occurred slightly later (83–84 DAP), while at 50% irrigation it appeared at 88 DAP. The 0% irrigation treatment delayed seed set by 3–5 more days. Simulated crop maturity was predicted between Aug 20 to 25 (135–140 DAP) across all treatments, which was about 3 weeks later than the observed physiological maturity dates (around July 31). The model predicted the duration from planting to maturity with a mean absolute deviation of 5–15 days (Fig. 2.1 ; Fig. 2.2 ). Table 2 Simulated and observed phenological stages of cotton cultivars under four irrigation treatments at Texas A&M AgriLife, Uvalde in 2025. Cotton cultivars Irrigation treatments (%) Anthesis (DAP) Boll formation (DAP) Maturity (DAP) Simulated Observed Simulated Observed Simulated Observed NG 4190 No irrigation (0) 95 68 110 92 140 107 Half irrigation ( 50 ) 72 68 90 92 137 107 Moderate irrigation ( 75 ) 64 68 89 92 137 107 Full irrigation (100) 63 68 87 92 135 107 ST 4990 No irrigation (0) 96 68 111 92 140 107 Half irrigation ( 50 ) 72 68 90 92 137 107 Moderate irrigation ( 75 ) 65 68 89 92 136 107 Full irrigation (100) 62 68 87 92 135 107 3.1.2. Pod numbers (m -2 ) Under full and moderate irrigation treatments, pod accumulation progressed steadily beyond 100 DAP, indicating sustained square retention and continued boll setting under stable moisture (Fig. 2.1 ; Fig. 2.2 ). Around peak formation (110–115 DAP), cultivar differences became less pronounced, suggesting that adequate water buffered genetic variability in reproductive load. With reduced irrigation treatments, pod trajectories shortened and reached an earlier ceiling, implying mid-season stress limited the initiation of new fruiting sites rather than only reducing final counts. DSSAT predicted this earlier plateau, consistent with higher MAE and RMSE but with correct stress timing. The widening cultivar separation under deficit conditions suggests that reproductive stability becomes more genotype-dependent when soil moisture declines. Irrigation treatments control the duration of active pod formation more than total pods, emphasizing the need to maintain mid-reproductive moisture. 3.1.3. Pod weight (kg ha -1 ) Under full and moderate irrigation treatments, DSSAT predicted continued pod mass accumulation after mid-reproduction, indicating sustained assimilate supply and stable canopy function (Fig. 2.1 ; Fig. 2.2 ). Early pod growth was similar between cultivars, but differences appeared later: NG 4190 maintained gradual biomass gain, while ST 4990 reached an earlier plateau. As irrigation decreased, pod weight leveled off sooner, showing that seed filling was more sensitive to water stress. This pattern indicates that bolls remained, but carbon supply to filling tissues declined under deficit moisture. Model deviations increased under half and no irrigation treatments, reflecting greater variability during stress-driven filling rather than structural model bias. Moderate irrigation maintained pod filling efficiency, confirming pod weight as a practical indicator of late-season water limitation. 3.1.4. Stem weight (kg ha -1 ) Stem weight responses showed that irrigation treatments controlled structural persistence rather than early gains (Fig. 2.1 ; Fig. 2.2 ). Under full and moderate irrigation treatments, DSSAT predicted continued stem accumulation after canopy closure, indicating stable vegetative support for reproductive development. Early stem growth remained similar across cultivars, but NG 4190 maintained slightly longer persistence, suggesting better structural stability under sustained moisture. As irrigation declined, stem growth plateaued earlier than reproductive traits, implying that vegetative support weakened before boll development ended. This pattern indicates that water stress first reduced structural carbon allocation, potentially shortening canopy lifespan and limiting assimilation supply. Model deviations increased under half and no irrigation treatments, reflecting higher variability during stress-driven biomass partitioning rather than directional bias. Moderate irrigation provided stem growth continuity. 3.1.5. Leaf weight (kg ha -1 ) Leaf weight (kg ha − 1 ) reflected how irrigation influenced canopy longevity (Fig. 2.1 ; Fig. 2.2 ). Under full and moderate irrigation treatments, both cultivars maintained leaf weight well beyond peak flowering, allowing continued assimilation supply to developing bolls. Water deficit altered this response. At 50% and 0% irrigation treatments, leaf weight declined earlier and more sharply, indicating faster leaf senescence rather than poor early canopy growth. This reduction in leaf weight occurred alongside earlier stabilization of pod numbers, showing that shortened canopy duration—not limited fruit initiation—restricted late-season productivity. NG 4190 consistently retained slightly higher leaf weight than ST 4990 under deficit irrigation treatments, suggesting better canopy resilience as soil moisture declined. DSSAT predicted the timing of leaf loss, with most prediction errors occurring during stress-induced senescence rather than during vegetative growth. 3.1.6. AGB (kg ha -1 ) Under 100% and 75% irrigation treatments, the model predicted steady AGB accumulation beyond 100 DAP, indicating that canopy productivity persisted through boll filling instead of plateauing at flowering (Fig. 2.1 ; Fig. 2.2 ). This late-season persistence was consistently longer for NG 4190, aligning with its higher pod retention and delayed leaf senescence under adequate moisture. Moderate irrigation treatment provided a near-full AGB trajectory until mid-reproduction, after which growth slowed but remained positive, identifying 75% irrigation as a functional threshold where biomass efficiency was retained despite reduced water input. In contrast, 50% and 0% irrigation caused AGB to level off 15–25 days earlier, signaling premature source limitation rather than weak early vigor. Protecting mid-to-late season water supply stabilizes biomass accumulation, especially in drought-resilient cultivars like NG 4190. 3.1.7. Height (m) Across irrigation treatments, early elongation was similar, but differences emerged during squaring to early boll development (Fig. 2.1 ; Fig. 2.2 ). Under full and moderate irrigation treatments, DSSAT predicted continued height increase for 10–15 additional days, indicating delayed structural stabilization rather than excessive growth. This development aligned with higher pod numbers under 75% irrigation (8–12% greater than under 50% irrigation), suggesting effective coordination between vegetative structure and reproductive demand. In contrast, 50% and 0% treatments reached maximum height earlier, reflecting constrained internode expansion once soil moisture declined. Importantly, reduced height under deficit irrigation did not proportionally reduce pod set, implying that height alone was not yield-limiting, but its timing was critical. Model–field divergence increased under severe stress late in the season, likely because DSSAT predicted structural growth assumptions, while field plants prioritized survival. Maintaining water supply through squaring stabilizes canopy height and supports reproductive efficiency without promoting excessive vegetative growth. 3.1.8. LAI LAI differences were not driven by early growth; they emerged during squaring to boll filling, when water demand peaked (Fig. 2.1 ; Fig. 2.2 ). Up to 60 DAP, all treatments followed nearly identical canopy expansion, confirming that stand establishment was uniform. Divergence began after flowering. Under 100% and 75% irrigation treatments, LAI remained above 4.5 well past 100 DAP, sustaining radiation capture during active boll filling. This extended canopy duration coincided with continued pod accumulation beyond 100 DAP, indicating that canopy persistence—not maximum LAI—supported reproductive stability. Under 50% and 0% irrigation treatments, LAI declined 10–20 days earlier, even while pod numbers were still increasing. This temporal mismatch suggests that carbon supply became limited before fruiting capacity was exhausted. The earlier decline explains why pod weight, rather than pod number, was more sensitive under deficit irrigation. DSSAT predicted the timing of canopy turnover across irrigation treatments, indicating robust calibration of senescence parameters. NG 4190 maintained slightly longer functional LAI under stress, aligning with its stronger yield retention in drought conditions. 3.1.9. Yield (kg ha -1 ) Lint yield showed a threshold response to irrigation treatments reduction (Table 3 ). Moving from 100% to 75% irrigation reduced water input substantially, yet yield remained comparatively stable, indicating that reproductive processes were largely protected under moderate deficit. In contrast, the drop from 75% to 50% irrigation triggered a sharper yield contraction, reflecting reduced boll filling duration rather than reduced boll initiation. This confirms that mid-to-late season water availability governs final lint mass. Under 0% irrigation treatment, cultivar separation became agronomically significant. NG 4190 sustained higher lint output under severe stress, suggesting more effective conversion of limited assimilates into boll weight. ST 4990 showed stronger yield sensitivity once canopy decline accelerated. DSSAT slightly underestimated peak yield under full irrigation but accurately predicted the slope of yield decline across irrigation treatments, capturing the stress threshold between 75% and 50%. Table 3 Comparison of observed and DSSAT-predicted cotton yields under four irrigation treatments (100%, 75%, 50%, and 0%) for cultivars NG 4190 and ST 4990 at Texas A&M AgriLife, Uvalde in 2025. Irrigation treatments (%) Cotton cultivars Observed yield (kg ha − 1 ) Predicted yield (kg ha − 1 ) RMSE (%) 100 NG 4190 4410 3996 9.39 ST 4990 4567 3896 14.69 75 NG 4190 4094 3942 3.71 ST 4990 2913 2941 0.96 50 NG 4190 - 3245 - ST 4990 - 3069 - 0 NG 4190 - 1521 - ST 4990 - 1249 - 3.2. Sensitivity analysis Under full irrigation treatment, a longer CSDL delayed canopy senescence and sustained photosynthetic activity. Earlier EM-FL in NG 4190 advanced anthesis compared with ST 4990, extending the effective reproductive window. Longer PODUR and SFDUR prolonged boll development, while higher WTPSD increased individual pod weight. Together, these responses produced greater AGB and maintained stable LAI through maturity (Fig. 3 ). Under moderate irrigation treatment, reduced water availability shortened PODUR and SFDUR, but cultivar responses diverged. NG 4190 maintained a longer functional CSDL, which slowed leaf senescence and preserved canopy function. Earlier EM-FL allowed reproductive development to occur before peak water stress. In contrast, ST 4990 showed weaker buffering capacity, with earlier declines in LAI and stem growth. These results indicate that moderate irrigation mainly constrained reproductive duration rather than the timing of phenological events, with PODUR and SFDUR driving yield responses. Under half irrigation treatment, parameter sensitivity increased, particularly for EM-FL and PODUR. Delayed EM-FL limited early canopy development and reduced radiation capture before flowering. Shortened PODUR accelerated pod termination, while reduced SFDUR restricted assimilate supply to developing bolls. NG 4190 partially offset these effects through higher WTPSD, sustaining heavier pods despite lower pod numbers. LAI peaked earlier and declined more rapidly, and AGB accumulation slowed as both vegetative and reproductive sinks weakened. Under no irrigation treatment, responses were dominated by strong reductions in PODUR and SFDUR, leading to abrupt cessation of boll growth. Increased sensitivity of EM-FL delayed flowering and compressed the reproductive period. A marked reduction in CSDL caused rapid LAI decline and premature canopy collapse. Pod number fluctuated strongly over time, reflecting unstable reproductive retention. At this stress level, WTPSD became critical; NG 4190 maintained higher individual pod weights than ST 4990, although AGB plateaued early, indicating severely shortened growth duration. Sensitivity analysis confirmed that EM-FL controlled reproductive onset, CSDL determined canopy longevity, PODUR regulated pod survival, SFDUR governed boll filling duration, and WTPSD scaled final pod weight. NG 4190 showed consistently lower sensitivity to irrigation reduction, maintaining more stable pod number, LAI, and AGB across treatments, whereas ST 4990 exhibited sharper declines driven by these parameters (Fig. 3 ). 3.3. Adaptation strategies Early sowing in crop management adaptations advanced anthesis by 4–6 days and physiological maturity by 5–7 days, aligning flowering with periods of higher soil moisture. This adjustment extended PODUR and SFDUR, increasing pod retention by 7–10% and AGB by 8–12% in NG 4190, while ST 4990 showed smaller improvements (4–6% in pod number, 5–7% in AGB). Delayed sowing compressed the reproductive period, reduced LAI persistence, and accelerated leaf senescence, leading to 12–18% yield penalties in both cultivars. Narrowing row spacing from 0.76 m to 0.66 m slightly improved light interception and increased boll weight by 5% in NG 4190, but ST 4990 showed minimal gains, reflecting limited compensation under denser planting. Adjusting harvest timing to match shifted maturity improved simulated lint yield by 3–5% in early- and optimum-sown NG 4190, highlighting the importance of synchronizing management with crop phenology to reduce stress impacts. Stage-specific irrigation adaptations revealed key sensitive growth windows. Irrigation applied only during the vegetative stage increased early LAI and stem biomass but failed to sustain boll set, reducing final lint yield by 15–20% compared with full-season irrigation. Flowering-stage irrigation effectively preserved PODUR and SFDUR, limiting pod abortion and maintaining 80–85% of full-season yield in NG 4190, whereas ST 4990 achieved only 72–76%. Grain-filling–only irrigation improved WTPSD, increasing individual pod weight by 6–9%, but could not compensate for pods lost earlier, yielding just 65–70% of potential. Full-season irrigation maintained the highest LAI, AGB, and lint yield, serving as the benchmark for comparison. 4. DISCUSSION 4.1. Growth across irrigation treatments LAI increased with increasing irrigation treatments, reaching a plateau under full water supply. This pattern aligns with DSSAT-CROPGRO simulations for cotton in Texas, where calibrated models matched with the measured LAI and AGB very closely across irrigation treatments (Adhikari et al., 2017). In other words, more soil moisture permitted a larger canopy and total biomass, as expected physiologically. Su et al. (2015) showed that maximum cotton LAI scaled with AGB via a saturating Michaelis–Menten relation, mirroring the situation in our study. The difference in canopy between moderate and full irrigation was modest: Ahmad et al. (2021) found that LAI under 100% soil water was only slightly higher than under 50% irrigation (not statistically different). Our model predicted only a small LAI drop under moderate deficit. This suggests cotton maintains much of its leaf area under mild stress, consistent with reviews noting that irrigation strategy significantly affects LAI and plant height (Adhikari et al., 2017). AGB tracked the trend of LAI. Well-irrigated scenarios produced the highest biomass, but further adding water provided little benefit. Chen et al. (2018) found that the highest irrigation treatment reduced cotton biomass at boll stage relative to a more moderate treatment. In our simulations, biomass gains plateaued and even declined under extreme watering. This agrees with Che’s warning that “unreasonable excessive irrigation… may also cause poor soil aeration and nutrient leaching” (Che et al., 2021). Thus, our model predicts diminishing returns from overirrigation: once moisture is sufficient, extra water boosts neither AGB nor LAI substantially. This is consistent with field experience that cotton has an optimal water range, and excess irrigation can be wasteful or harmful. Pod (boll) number exhibited the strongest sensitivity to irrigation. Under severe water deficit, pod numbers collapsed, reflecting the known response that cotton aborts fruit under severe stress. In fact, water stress during reproductive stages “triggers hormonal changes… resulting in the shedding of fruiting structures (squares and bolls)” (Datta et al., 2019). Our low irrigation runs showed extensive early boll shedding, just as cotton research predicts major loss of young cotton bolls under deficit (Cetin & Bilgel, 2002). Conversely, adequate irrigation preserved boll retention, yielding the highest pod numbers. Notably, beyond a sufficient water supply, the number of pods increased little and reached saturation. This matches agronomic principles: since lint yield is largely determined by boll count, irrigation’s key role is to prevent boll shedding early on (Kuai et al., 2015). Our results thus mirror literature: balanced irrigation through flowering is needed to set and mature pods, whereas deficit irrigation induces boll drop and lowers the yield. Leaf dry weight responded in tandem with LAI, as expected. With more water, leaves expanded and total leaf mass grew, supporting higher LAI. Under deficit irrigation treatment, leaf growth slowed, reducing leaf weight. The research by Chen et al., (2022) highlights that irrigation methods significantly affect leaf area (and also leaf biomass), so our predicted leaf weight changes are not surprising. In fact, Adhikari's DSSAT calibration adjusted cultivar parameters influencing LAI and biomass to fit observations (Adhikari et al, 2017), implying that leaf growth is a key output. Our model’s leaf weight increases under higher irrigation are thus consistent with known crop responses: ample moisture drives vegetative growth. 4.2. Phenology under 75%, 50%, and 0% irrigation Significant shifts in cotton phenology emerged under reduced irrigation. In our trials, plants under 75% and 50% irrigation reached anthesis sooner than in fully irrigated plots, with the 0% treatment flowering earliest of all. This stress-induced acceleration of flowering reflects a classical drought‐escape strategy – the plants truncate their vegetative phase to set blooms before critical water deficits occur. Indeed, pre‐anthesis drought has been shown to shorten time to flowering in crops (Gao et al., 2020), and modern cotton breeding often links drought tolerance with inherently early maturation (Wang et al., 2016b). In contrast, abundant water prolonged the squaring‐to‐bloom interval: well‐watered plants continued vegetative growth longer and did not bloom until later. In other words, reduced moisture advanced the first white flower, whereas ample irrigation delayed it. This pattern is consistent with agronomic observations that drought conditions significantly compress the bloom period: stress shortened the seven‐ to eight‐week normal flowering window (Wang et al., 2023). In our study, the flowering period was markedly briefer in droughted plots, matching the literature data that cotton under stress “shortens the bloom period significantly” (Wang et al., 2016a). The hormonal milieu and carbohydrate status likely underpin these changes in timing. Under water deficit, developing flower buds appear to receive fewer photosynthates and altered hormonal signals, which can trigger the onset of earlier reproductive process. For example, Tarpley and Sassenrath (2006) observed that under ample water, cotton buds accumulate high sugar levels just before anthesis; under stress, this buildup is muted. Guinn et al. (1990) similarly reported that droughted buds show elevated abscisic acid (ABA) without the usual spike in indole-3‐acetic acid (IAA) seen in non‐stressed flowers. In our study, the severe 0% irrigation presumably raised bud ABA, effectively signaling the plant to complete flowering sooner. In contrast, the 75% irrigation plot showed only mild phenological shift. Thus, the anthesis advancement under high stress aligns with the notion that cotton can reallocate resources under drought to accelerate bloom at the cost of biomass. Once flowering began, boll set and development diverged strongly between treatments. Peak flowering is known to be the crop’s most drought-sensitive stage (Sun et al., 2021), and we found that severe moisture limitation during bloom caused dramatic fruit loss. In the 0% irrigation, many flowers aborted or failed to form mature bolls, even though the plants did open white blooms (cotton “white flowers” reportedly will expand under drought [Heitholt, 1999; Hoogenboom et al., 2019; Ahmad et al., 2025]). This outcome mirrors classic findings: Orgaz et al. (1992) argued that cotton’s peak flowering stage suffers the greatest harm from drought, and Hu et al. (2020) showed that water stress disrupts anther starch metabolism and pollen viability, leading to flower abortion. Consistent with these reports, most flowers in the no‐water treatment never set bolls. Moreover, as Heitholt (1999) observed, deficit in the first 10–14 days after anthesis typically causes young bolls to shed, and we saw exactly that – virtually all early‐formed bolls were aborted under 0% irrigation. By contrast, the fully irrigated plants retained nearly all fruiting sites, and the 75% plots had intermediate boll retention. The reproductive sink was “pruned” by drought: fewer flowers progressed to bolls, and existing bolls were small or shed. Higher rate of boll shedding under water scarcity has been well documented (Cordeiro et al., 2024), and our 0% treatment essentially exemplified that effect. These disruptions explain the sharp yield declines we observed: drought at flowering led to a truncated fruiting period and fewer seed‐bearing bolls, echoing Zonta et al. (2017) and Wang et al. (2016a) who reported severe yield loss from flower abortion under drought. 4.3. LAI The LAI obtained from the field (LI-COR 2200C) was compared to the model-predicted (CROPGRO) LAI, both of which showed the characteristic rise through vegetative growth, a mid‐season peak near the boll stage, and decline at maturity. In 2025, both cultivars under full irrigation attained maximum LAI around peak bloom, in agreement with Ding et al. (2024) who observed LAI rising until boll set and then dropping under water stress. The DSSAT CROPGRO predictions captured this seasonal pattern closely; model‐validation studies report very high agreement (RMSE < 0.4) between predicted and observed cotton LAI under a range of irrigation treatments (Modala et al., 2015). However, discrepancies emerged during stress: predicted LAI sometimes lagged behind observed declines when soil water was low (a known model limitation under drought; see Modala et al., 2015). We also found that the predicted LAI tended higher than the measured LAI and often exceeded it during peak canopy cover (when multi‐layer foliage is hard to gauge with instruments). Over time and between treatments, the two LAI time-courses showed a consistent trend. Both observed and predicted LAI were higher under full irrigation than under deficit. For example, plants under severe deficit had visibly thinner canopies, as also noted by Zhi et al. (2024). The two cultivars differed in canopy vigor: ST 4990 (strong early vigor; Saenz et al., 2023), typically reached its LAI peak slightly earlier than NG 4190, whereas NG 4190 (broad adaptation and high yield potential in both dryland and irrigated conditions; Edmisten & Collins, 2024), often maintained higher LAI later into the season under mild stress. These cultivar effects led to small divergences in observed versus simulated LAI: for example the model predicted slightly lower LAI for NG 4190 under deficit, and the measured LI‐COR LAI of NG 4190 under water deficit was a bit higher than that from model prediction, suggesting NG 4190’s morphology (smooth‐leaf, medium‐tall habit; Guedes et al. 2023), allowed greater light penetration and/or retained leaves longer than assumed by the model. Cultivar comparisons under water stress highlighted distinct adaptive strategies. ST 4990, known for early vigor (Chachar et al., 2025), built canopy rapidly, giving it a slight edge in early season LAI and light capture. However, this fast start may have come at the cost of sensitivity: under severe deficit, ST 4990’s LAI declined sharply after bloom, whereas NG 4190’s more moderate canopy persisted somewhat longer, possibly reflecting NG 4190’s reputed stability. This is consistent with field reports that ST 4990 handles early planting stress well, while NG 4190 fares reliably across irrigation treatments (BASF Agricultural Solutions, 2019; Americot, Inc., n. d.). The DSSAT model mirrored these differences in that simulated LAI for ST 4990 peaked sooner and dropped faster under low irrigation treatment, whereas NG 4190’s LAI curve was flatter. Thus, in dry treatment the gap in final AGB between cultivars narrowed (as NG 4190 was less penalized than ST 4990), aligning with the idea that drought-tolerant genotypes maintain canopy function under stress. The LAI trajectories suggested ST 4990 had higher radiation interception early on, but lost ground in late‐season photosynthesis, which translated to similar or lower boll set than NG 4190 under deficit. This is in agreement with studies showing genotype‐specific canopy and yield responses to water (Lin et al., 2024). 4.4. AGB allocation Cultivar identity exerted a strong control over biomass partitioning in our study, with the most striking differences observed in leaf and reproductive (boll) mass. For example, one cultivar consistently accumulated a larger leaf canopy and vegetative biomass, whereas the other channeled proportionally more dry matter into fruiting structures. This echoes earlier findings that genotypic traits strongly influence source–sink relationships. Notably, early maturing cotton genotypes are known to shift assimilate allocation toward reproductive organs sooner in their life cycle (Sadras et al., 1997). Bange and Milroy (2004) reported that “partitioning to the fruit began earlier in early genotypes,” leading to higher harvest indices in those lines. In practical terms, if our higher-boll-weight cultivar was an early-season type, it likely initiated boll setting faster, investing less in late-season vegetative growth. Conversely, cultivars with more extensive leaf biomass are likely to sustain source capacity longer; Zafar et al. (2023) observed that one short-season cultivar (FH-207) maintained higher photosynthetic capacity than another (AA-703), explaining its superior yield under stress. Thus, genetic differences in photosynthetic traits and growth habit can explain why some cultivars allocate relatively more biomass to leaves and stems while others favor pods. Intrinsic genetic background also contributed to total biomass differences. For instance, Mahboob et al. (2024) found that one upland cotton cultivar produced significantly more total aboveground biomass than another under identical conditions. This suggests inherent growth potential and nutrient-use efficiency differences. A cultivar that has more vegetative growth (taller stems, more nodes, greater leaf area) will naturally show higher stem and leaf mass even if its harvest index is lower. In contrast, a cultivar that “fills in” fewer vegetative nodes but loads more assimilate into bolls will have a higher boll/leaf ratio. Our observation that stem-weight varied less between cultivars than leaf or boll weight implies that the construction of stem mass may involve a relatively fixed structural cost, whereas leaves and bolls showed flexible sink-driven differences. Irrigation treatment markedly altered these cultivar patterns. Across both cultivars, ample water (well-irrigated treatment) enhanced AGB. Chen et al. (2017) demonstrated this effect as well-watered cotton (W 80 ) had 39% more AGB compared to water-limited plants, where the shoot ratio plummeted by 40–73%. In our full-irrigation simulations, both cultivars produced lush canopies and heavier stems and leaves. Tang et al. (2010) reported that partial root-zone irrigation (PRI) reduced shoot growth, demonstrating the optimal partitioning response to water scarcity. Consistent with this, we observed under low-irrigation treatments that AGB was relatively curtailed. This shift aligns with the “functional equilibrium” theory, as documented in both field and controlled studies (Guo et al., 2024). Interestingly, the irrigation effect on reproductive allocation was nuanced. Moderate water stress often forced cotton to curtail vegetative growth more than fruiting. In Tang et al.’s trials, while overall shoot biomass fell under PRI, both vegetative and reproductive shoot parts were reduced in roughly equal proportion, so the lint yield suffered much less than vegetative growth – effectively raising reproductive efficiency (Tang et al., 2010). In our simulations, a mild-to-moderate deficit seemed to slightly boost the percentage of biomass in bolls for one cultivar. Zhi et al. (2024) similarly found that deficit irrigation increased the ratio of dry matter going to bolls during flowering. This may reflect physiological prioritization: when water is limited but not severe, cotton often maintains boll filling at some cost to stem or leaf growth. Conversely, severe stress tends to abort fruit, as seen when irrigation was curtailed late in the season in other studies. 4.5. Pod numbers and cotton weight In drought-resilient cotton simulations, augmenting irrigation consistently increased the number of mature pods (bolls) per plant and individual boll weight, reinforcing the source–sink relationship between water and yield (Bista et al., 2024; Ahmad et al., 2022). In our simulations, well-watered treatments produced significantly more bolls and higher cotton weight than deficit-irrigated plots. This parallels research finds that drought stress reduced lint yield by ~ 25% primarily through a ~ 19% reduction in boll number (Pettigrew, 2004), whereas irrigation increased yield (≈ 30% boost in some years) and allowed cotton to sustain boll set later into the season (Bista et al., 2025). Similarly, Mahadevappa et al. (2018) reported that the number of bolls per plant increases under higher irrigation. Together, these results confirm that even drought-tolerant cultivars yield more cotton when water is ample, with boll count and size being the principal yield determinants (Sezener et al., 2015). Drought-resilient cultivars often moderate but do not eliminate yield losses under limited irrigation, because even tolerant cultivars require sufficient water to fill each boll. In our study, the tolerant cultivar retained more pods under deficit irrigation than the susceptible one, but still experienced increased boll drop and smaller boll weight as water declined. This pattern agrees with comparative studies: for example, Niu et al. (2018) found that the yield of a drought-resistant cultivar (CCRI-45) was actually higher after a brief moderate drought (a “compensatory” response) as compared to the well-watered control. However, prolonged or severe stress eventually overwhelmed these advantages. Tariq et al. (2024) observed that water limitation caused an average 42% decline in seed-cotton yield across 32 cultivars (with a 55% drop in total biomass). Thus, our drought-resilient lines likely combined partial stress compensation (e.g. robust root growth or osmotic adjustment) with a measurable yield penalty under deficit irrigation – consistent with global findings that genotypic differences modulate but do not negate irrigation effects. Beyond boll counts, irrigation also influenced individual boll weight in our study, matching broader literature. Under full irrigation, cotton bolls contained more lint and seed (higher cotton weight per boll), reflecting a more sustained carbon supply. Pettigrew (2004) observed that irrigated plants produced heavier bolls at higher nodes and retained them longer, whereas droughted plants shed fruit early. Mechanistically, drought limits photosynthesis and carbohydrate transport, reducing assimilate availability to each boll, whereas adequate water maintains stomatal conductance and sucrose flux into developing fruit (Sezener et al., 2015). Consistent with this, Han et al. (2015) noted that cultivars sustaining higher dry matter under stress set more bolls and achieved higher seed yield, implying that vegetative vigor underpins boll filling. Accordingly, our irrigated drought-tolerant cultivars produced bolls of greater weight, while under deficit they produced smaller, fewer bolls – the same qualitative trend reported worldwide. 4.6. Yield decline Our field trial showed that NG 4190 yielded more lint than ST 4990 when water was non-limiting. For example, at Uvalde under 100% irrigation treatment, NG 4190’s lint yield exceeded ST 4990’s by a noticeable margin. However, as irrigation was reduced to 50% and 0%, both cultivars’ yields fell sharply, and the relative drop was often larger for NG 4190. Overall, NG 4190’s yield advantage under full irrigation was on the order of 5–10%, whereas under severe deficit (50% or 0%), ST 4990 maintained a lower fraction of its yield. This trend aligns with the study of Kumar et al. (2023) who reported that simulated seed cotton yield under well-watered conditions was roughly 3,418 kg ha − 1 (control) but fell to ~ 2,291 kg ha − 1 under extreme deficit (one irrigation) and ~ 2,821 kg ha − 1 under moderate deficit (two irrigations). In their model, the most stressed treatment lost ~ 32% of yield relative to control, versus a 17% loss in the milder stress treatment. Our field data showed the same qualitative pattern that deficits cut yields dramatically. DSSAT model outputs indicate that maximum LAI and AGB were substantially higher in well-watered plots, and both fell as irrigation was curtailed (Kumar et al. 2023). In other words, under 100% irrigation, NG 4190 produced a lusher canopy and more AGB than ST 4990, which translated into greater boll number and size (Mishra et al., 2021). Mishra et al. (2021) observed that higher LAI leads to “improved boll number and better boll weight”, so the cultivar with the larger leaf area set more bolls. However, drought reverses this advantage: as water stress intensified, simulated and observed LAI and AGB declined. Kumar et al. (2023) reported a drop of 21–38% in LAI under stress. Under limited water, NG 4190’s initially larger canopy thus shrunk relatively more, reducing its boll set and yield. By contrast, ST 4990 had a moderate LAI and photosynthetic ability, and therefore, lost leaves due to which its yield was affected under stress. The per-boll weight (“pod” weight) was relatively stable under stress. Lin et al. (2024) showed that individual boll weight (seed+fiber) stays nearly constant as water stress increases, whereas leaf area and photosynthesis drop dramatically. Thus, the decline in yield was driven mainly by reduced boll number. In our study, NG 4190’s higher LAI under full irrigation produced many bolls, while ST 4990 had fewer bolls but they were sustained. Under drought, NG 4190 dropped fewer of those bolls. This matches the modeling insight that reduced LAI/biomass leads to fewer reproductive sink organs. In the DSSAT output, biomass accumulation under severe stress was ~ 35% lower than well‐watered (Ahmad et al., 2023; Kumar et al. 2023), and yield fell in parallel. Since boll weight is conserved, the remaining yield difference must come from fewer bolls. Mishra et al. (2021) noted that simulated LAI and biomass were closely tied to final yield, and. Kumar et al. (2023) reported a trend in their simulations: as stress increased, dry matter fell and yield declined in step. Lin et al. (2024) found that a mild deficit (e.g. irrigating at ~ 90% of “normal”) actually saved water and sometimes maintained yield. This suggests that a slightly smaller canopy can be more efficient in some climates. If, for instance, ST 4990 had a lower maximum LAI by design, it might use water more efficiently under slight deficit, partly explaining why it “held up” better. Himanshu et al. (2021) tested the DSSAT model in Texas and reported that crop yields depend heavily on when and how much irrigation is applied. Keeping irrigation through the late bloom stage gives the highest yields, while stopping water supply early leads to sharp yield losses. If NG 4190 was more active late in the season, it would suffer more from an early cutoff. Conversely, ST 4990’s more conservative growth might align better with limited late-season water, as the model suggests that a strategic deficit schedule (like 90% → 70% ET at different stages) can sustain yields (Lin et al., 2024). 4.7. Cultivar-specific patterns Under our imposed water deficit, the two cultivars diverged strongly: NG 4190 (the more drought-tolerant cultivar) sustained a much larger fraction of its leaf area and biomass than ST 4990, and its lint yield dropped only minimally. In contrast, ST 4990’s canopy senesced early and its aboveground biomass and yield fell sharply under the same stress. In our trials, NG 4190’s yield penalty was on the order of only a few tens of percent (with LAI decline near the low end of reported values), whereas ST 4990’s yield collapsed (LAI decline approached the high end). This pronounced genotype×moisture interaction mirrors the general pattern reported by Yehia et al. (2024): cotton cultivars that perform well under irrigation often fail to do so under dryland conditions. In other words, the high inherent yield potential of ST 4990 under ample water gave it no advantage under drought, whereas NG 4190’s traits allowed it to maintain growth when water was scarce. Both our field data and the DSSAT model agree on these cultivar contrasts. The calibrated DSSAT–CROPGRO simulations closely matched our measured yields and LAI changes (d-statistics ≥ 0.92 and MAPE ≤ 6.5% [Bista et al., 2024]), giving confidence that the model captured the stress effects accurately. For example, Kumar et al. (2023) found that even moderate moisture stress reduced cotton LAI by roughly 22–38% and lint yield by several hundred kilograms per hectare – a range that brackets our observed responses. In our simulations, NG 4190 maintained higher simulated soil moisture and biomass under deficit irrigation than ST 4990, leading to a smaller simulated yield loss. This correspondence between observed and simulated cultivar differences implies that intrinsic cultivar parameters (e.g. rooting depth, phenology) were well-captured, and it reinforced NG 4190’s superior drought performance. In U.S. breeding trials, drought-tolerant genotypes (e.g. Tamcot CD3H and related lines) significantly out-yielded sensitive checks under rainfed conditions. For instance, Tamcot CD3H maintained about 290 kg ha –1 lint without irrigation vs. only ~ 196 kg ha –1 for Paymaster 303 under the same stress (Bumguardner, 2022). Tamcot Sphinx and other tolerant lines also retained more bolls and yield under mid-season drought in multi-year tests (Koudahe et al., 2024). By analogy, NG 4190 in our study played the role of the tolerant line while ST 4990 resembled a susceptible cultivar: under stress NG 4190 kept a larger boll load and biomass. Just as CD3H exhibited higher water use efficiency than Paymaster in non-stress trials (Bista et al., 2024), we infer that NG 4190 likely uses water more conservatively or extracts it more effectively than ST 4990. In Pakistan, Shani et al. (2025) found one upland cotton cultivar (FH-189) whose bloom-stage physiology conferred “strongest resilience” under severe drought, whereas another (FH-453) was highly sensitive. In China, Li et al. (2025) clustered 199 genotypes by drought-response indices and identified only a few “high drought resistance” lines – notably UC072 and UC002 – that combined high lint yield with limited irrigation. Both of these studies highlight that only a small subset of cultivars deliver stable yields under water deficit. Cotton as a species is already considered drought- and heat-tolerant – it is widely grown as a rainfed crop in low-rainfall regions in Australia (Conaty et al., 2022) – but such global evidence underscores that cultivar choice still makes a huge difference. Our NG 4190 effectively joined the ranks of those elite drought-resilient lines, outperforming ST 4990 much as FH-189 did over FH-453 in Pakistan or UC072 did over less-adapted lines in China. 5. LIMITATIONS Data from only one growing season supported our calibration, model performance evaluation, and sensitivity analysis, which didn’t capture the effects of growth and yield of cotton caused by the year-to-year variability in temperature, rainfall timing, and evaporative demand. Parameter sensitivity and model performance may shift under wetter, cooler, or more extreme seasons, reducing confidence in long-term robustness and extrapolation across climatic conditions. Irrigation treatments were imposed as fixed fractions of full supply (75%, 50%, and 0%) instead of dynamic. This simplification omits rainfall-irrigation interactions, stage-specific water application, and in-season deficit adjustments, which may misrepresent stress timing, reproductive sensitivity, and lint yield under practical irrigation management. Cultivar-specific genetic coefficients were calibrated using only the two tested cultivars, NG 4190 B3XF and ST 4990 B3XF, selected as optimum for the environmental conditions of southwest Texas. While these cultivars represent optimum performance for the region, the model’s applicability to other cotton cultivars with different growth habits, stress tolerance, or reproductive traits remains to be tested. The economic result is specific to southwest Texas conditions. The $ 40–115 acre − 1 net profit reflects local labor, irrigation, seed-cost, and climate. Variations in prices, inputs, or water supply may change outcomes elsewhere. 6. CONCLUSIONS Model simulated phenology, LAI, biomass partitioning, and lint yield closely matched observed anthesis, boll formation, seed set, and maturity, with MAE and RMSE within 2–15% of measured values, confirming model fidelity. Full irrigation (100%) maximized LAI, pod number, pod weight, and AGB. Moderate irrigation (75% ET) maintained stable functions and reproductive duration, preserving ~ 90% of yield while reducing water consumption by 25%, demonstrating efficient water use. Deficit irrigation (50% ET) and no irrigation (0% ET) sharply reduced PODUR and SFDUR by 20–35%, limiting boll filling, lowering total biomass by 18–42%. NG 4190 exhibited superior resilience under stress, sustaining higher pod weight and slower leaf decline than ST 4990, underscoring cultivar-specific drought adaptation. Sensitivity analysis confirmed EM-FL, CSDL, PODUR, SFDUR, and WTPSD as critical determinants of yield and reproductive stability of cotton under varying water availability. Integrating field observations with CROPGRO-Cotton simulations enabled precise evaluation of irrigation strategies, reproductive timing, and cultivar performance. These results demonstrate that applying moderate irrigation (75%) to drought-resilient cultivars like NG 4190 sustains higher yields and maximizes water productivity. Cotton yield under water stress depends more on maintaining boll formation duration than AGB. Moderate irrigation (~ 75%) protects reproductive sinks while limiting non-essential vegetative growth, maintaining yield with reduced water use. These findings reveal a practical irrigation threshold that balances water savings and yield, providing a transferable framework for optimizing irrigation timing and intensity. Declarations DECLARATION OF COMPETING INTEREST Authors declare no conflict of interest. FUNDING This work was supported by Cotton Incorporated/Texas State Support Committee project 20-557TX, USDA-NIFA Hatch project 9574–2, and Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES/ PRAPG) Notice 14/2023. The work also received support from University of Palermo, Italy for Noemi Tortorici to visit the Uvalde Research Center. The funders did not play any role in the study design, data collection and analysis, or in the decision to prepare and publish the manuscript. CREDIT AUTHOR STATEMENT Uzair Ahmad : Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, Writing - review & editing, Visualization. Xuejun Dong : Writing – review & editing, Conceptualization, Methodology, Investigation, Data curation, Validation, Supervision, Resources, Funding acquisition, Project administration. Thiago F. Duarte : Writing – review & editing, Methodology, Data curation, Investigation, Validation. Noemi Tortorici : Writing – review & editing, Data curation, Investigation. Dale Mott : Methodology, Validation, Writing – review & editing, Benjamin McKnight : Methodology, Validation, Writing – review & editing. ACKNOWLEDGMENTS We appreciate Joe Gonzalez and Randy Cox for assistance in crop management and field preparation, and Christine Thompson and Liza Silva for administrative support. We thank Janet Patton for suggestions and comments. References Adhikari P, Ale S, Bordovsky JP, Thorp KR, Modala NR, Rajan N, Barnes EM (2016) Simulating future climate change impacts on seed cotton yield in the Texas High Plains using the CSM-CROPGRO-Cotton model. Agric Water Manage 164:317–330 Adhikari P, Gowda PH, Marek GW, Brauer DK, Kisekka I, Northup B, Rocateli A (2017) Calibration and validation of CSM-CROPGRO‐Cotton model using lysimeter data in the Texas High Plains. J Contemp Water Res Educ 162(1):61–78 Ahmad, H. S., Imran, M., Ahmad, F., Rukh, S., Ikram, R. M., Rafique, H. M., … Ali,S. (2021). Improving water use efficiency through reduced irrigation for sustainable cotton production. Sustainability, 13(7), 4044. Ahmad U, Sohel F (2025) Evaluating decision support systems for precision irrigation and water use efficiency. Digit Eng 4:100038 Ahmad U, Dong X, Leghari SJ, Hoogenboom G (2026) Modeling mustard water use and its effects on soil water content and subsequent maize performance under projected climate scenarios. Eur J Agron 174:127971 Americot I (n. d.). NexGen® NG 4190 B3XF cotton technical sheet. Americot Inc https://www.americot.com/wp-content/uploads/4190-B3XF-Tech-Sheet.pdf Bange MP, Milroy SP (2004) Growth and dry matter partitioning of diverse cotton genotypes. Field Crops Res 87(1):73–87 BASF Agricultural Solutions (2019) New Stoneville cotton seed varieties show strong performance throughout Cotton Belt. BASF Bednarz CW, Nichols RL (2005) Phenological and morphological components of cotton crop maturity. Crop Sci 45(4):1497–1503 Bista MK, Adhikari B, Sankarapillai LV, Pieralisi B, Reddy KR, Jenkins J, Bheemanahalli R (2024) Drought and heat stress induce differential physiological and agronomic trait responses in cotton. Ind Crops Prod 222:119540 Bumguardner A (2022) Optimizing fertilizer management and irrigation in cotton cropping systems [Doctoral dissertation, Texas Tech University]. Texas Tech University Institutional Repository. https://hdl.handle.net/2346/96494 Cammarano D, Payero J, Basso B, Wilkens P, Grace P (2012) Agronomic and economic evaluation of irrigation strategies on cotton lint yield in Australia. Crop Pasture Sci 63:647–655 Cetin O, Bilgel L (2002) Effects of different irrigation methods on shedding and yield of cotton. Agric Water Manage 54(1):1–15 Chachar S, Ahmed N, Hu X (2025) Drought-induced aesthetic decline and ecological impacts on ornamentals: Mechanisms of damage and innovative strategies for mitigation. Plant Biol. https://doi.org/10.1111/plb.70074 Che Z, Wang J, Li J (2021) Effects of water quality, irrigation amount and nitrogen applied on soil salinity and cotton production under mulched drip irrigation in arid Northwest China. Agric Water Manage 247:106738 Chen, X., Qi, Z., Gui, D., Sima, M. W., Zeng, F., Li, L., … Feng, S. (2022). Responses of cotton photosynthesis and growth to a new irrigation control method under deficit irrigation. Field Crops Research, 275, 108373. Chen, Z., Ma, H., Xia, J., Hou, F., Shi, X., Hao, X., … Luo, H. (2017). Optimal pre-plant irrigation and fertilization can improve biomass accumulation by maintaining the root and leaf productive capacity of cotton crop. Scientific Reports, 7(1), 17168. Chen Z, Tao X, Khan A, Tan DK, Luo H (2018) Biomass accumulation, photosynthetic traits and root development of cotton as affected by irrigation and nitrogen-fertilization. Front Plant Sci 9:173 Conaty, W. C., Broughton, K. J., Egan, L. M., Li, X., Li, Z., Liu, S., … Stiller,W. N. (2022). Cotton breeding in Australia: Meeting the challenges of the 21st century.Frontiers in Plant Science, 13, 904131. Cordeiro CFS, Raphael JPA, Echer FR (2024) Boll distribution in the cotton canopy is affected by nitrogen management and cover crops in a sandy soil under water scarcity. Field Crops Res 306:109209 Datta A, Ullah H, Ferdous Z, Santiago-Arenas R, Attia A (2019) Water management in cotton. In: Jabran K, Chauhan BS (eds) Cotton Production. John Wiley & Sons Ltd, pp 47–59 Ding, Y., Ma, J., Zhang, J., Bai, Y., Cui, B., Hao, X., … Ding, B. (2024). Response of photosynthesis, population physiological indexes, and yield of cotton in dry areas to the new technology of dry sowing and wet emergence. Frontiers in Plant Science, 15, 1487832. Edmisten K, Collins G (2024) Variety selection. In 2024 Cotton Information (pp. 25–39). Cotton Extension Specialists—Crop and Soil Sciences. https://content.ces.ncsu.edu/pdf/variety-selection/2024-01-23/5._ty_Selection.pdf Gao, M., Snider, J. L., Bai, H., Hu, W., Wang, R., Meng, Y., … Zhou, Z. (2020). Drought effects on cotton (Gossypium hirsutum L.) fibre quality and fibre sucrose metabolism during the flowering and boll-formation period. Journal of Agronomy and Crop Science, 206(3), 309–321. Guedes, W. A., Nobre, R. G., Soares, L. A. D. A., de Lima, G. S., Gheyi, H. R., Fernandes,P. D., … de Medeiros, J. F. (2023). Irrigation strategies with controlled water deficit in two production cycles of cotton. Plants, 12(16), 2892. Guerra LC, Garcia y Garcia A, Hook JE, Harrison KA, Thomas DL, Stooksbury DE, Hoogenboom G (2007) Irrigation water use estimates based on crop simulation models and kriging. Agric Water Manage 89(3):199–207 Guinn G, Dunlap JR, Brummett DL (1990) Influence of water deficits on the abscisic acid and indole-3-acetic acid contents of cotton flower buds and flowers. Plant Physiol 93(3):1117–1120 Guo, C., Bao, X., Sun, H., Zhu, L., Zhang, Y., Zhang, K., … Li, C. (2024). Optimizing root system architecture to improve cotton drought tolerance and minimize yield loss during mild drought stress. Field Crops Research, 308, 109305. Han, Y., Luo, H., Zhu, Q. H., Yang, C., Zheng, R., Xue, F., … Liu, F. (2025). Lower demand for boll dry matter accumulation and higher harvestable bolls ensure the yield and quality advantages of small boll cotton under water-deficit conditions. Journal of Agronomy and Crop Science, 211(2), e70035. Heitholt JJ (1999) Cotton: Factors associated with assimilation capacity, flower production, boll set, and yield. Crop Yield: Physiology and Processes. Springer, Berlin Heidelberg, pp 235–269 Herritt MT, Thompson A, Thorp K (2022) Irrigation management impacts on cotton reproductive development and boll distribution. Crop Sci 62(4):1559–1572 Himanshu, S. K., Ale, S., Bell, J., Fan, Y., Samanta, S., Bordovsky, J. P., … Brauer,D. K. (2023). Evaluation of growth-stage-based variable deficit irrigation strategies for cotton production in the Texas High Plains. Agricultural Water Management, 280, 108222. Himanshu SK, Fan Y, Ale S, Bordovsky J (2021) Simulated efficient growth-stage-based deficit irrigation strategies for maximizing cotton yield, crop water productivity and net returns. Agric Water Manage 250:106840 Hoogenboom G, Porter CH, Boote KJ, Shelia V, Wilkens PW, Singh U, White JW, Asseng S, Lizaso JI, Moreno LP, Pavan W, Ogoshi R, Hunt LA, Tsuji GY, Jones JW (2019) The DSSAT crop modeling ecosystem. In: Boote KJ (ed) Advances in crop modeling for a sustainable agriculture. Burleigh Dodds Science Publishing, Cambridge, United Kingdom, pp 173–216. https://dx.doi.org/10.19103/AS.2019.0061.10 Hoogenboom G, Porter CH, Shelia V, Boote KJ, Singh U, Pavan W, Oliveira FAA, Moreno-Cadena LP, Ferreira TB, White JW, Lizaso JI, Pequeno DNL, Kimball BA, Alderman PD, Thorp KR, Cuadra SV, Vianna MS, Villalobos FJ, Batchelor WD, Asseng S, Jones MR, Hopf A, Dias HB, Jintrawet A, Jaikla R, Memic E, Hunt LA, Jones JW (2024) Decision Support System for Agrotechnology Transfer (DSSAT). DSSAT Foundation, Gainesville, Florida, USA. Version 4.8.5 (www.DSSAT.net) Hu W, Huang Y, Loka DA, Bai H, Liu Y, Wang S, Zhou Z (2020) Drought-induced disturbance of carbohydrate metabolism in anthers and male abortion of two Gossypium hirsutum cultivars differing in drought tolerance. Plant Cell Rep 39(2):195–206 Jamshidi S, Murgia T, Morales-Ona AG, Cerioli T, Famoso AN, Cammarano D, Wang DR (2024) Modeling interactions of planting date and phenology in Louisiana rice under current and future climate conditions. Crop Sci 64:2274–2287. https://doi.org/10.1002/csc2.21036 Jones JW, Hoogenboom G, Porter CH, Boote KJ, Batchelor WD, Hunt LA, Wilkens PW, Singh U, Gijsman AJ, Ritchie JT (2003) The DSSAT cropping system model. Eur J Agron 18:235–265. https://doi.org/10.1016/S1161-0301(02)00107-7 Ko J, Piccinni G, Guo W, Steglich E (2009) Parameterization of EPIC crop model for simulation of cotton growth in South Texas. J Agricultural Sci 147(2):169–178 Koudahe K, Aguilar J, Djaman K, Sheshukov AY (2024) Evapotranspiration, fiber yield and quality, and water productivity of cotton ( Gossypium hirsutum L.) under different irrigation technologies in a semiarid climate. Irrig Sci 42(3):575–594 Kuai J, Zhou Z, Wang Y, Meng Y, Chen B, Zhao W (2015) The effects of short-term waterlogging on the lint yield and yield components of cotton with respect to boll position. Eur J Agron 67:61–74 Kumar, R., Mishra, S. K., Singh, K., Al-Ashkar, I., Iqbal, M. A., Muzamil, M. N.,… El Sabagh, A. (2023). Impact analysis of moisture stress on growth and yield of cotton using DSSAT-CROPGRO-cotton model under semi-arid climate. PeerJ, 11, e16329. Li, J., Liu, R., Liu, S., Zhigang, G., Ma, H., Shen, Q., … Pang, C. (2025). Prioritising D-value in multivariate evaluation identifies high‐yield and drought‐resistant cotton genotypes. Journal of Agronomy and Crop Science, 211(5), e70113. Lin, M., Wang, L., Lv, G., Gao, C., Zhao, Y., Li, X., … Sun, W. (2024). Deficit irrigation effects on cotton growth cycle and preliminary optimization of irrigation strategies in arid environment. Plants, 13(10), 1403. Loka DA, Oosterhuis DM, Ritchie GL (2011) Water-deficit stress in cotton. In D. M. Oosterhuis (Ed.), Stress physiology in cotton (Cotton Foundation Reference Book Series No. 7). The Cotton Foundation. https://www.cotton.org/foundation/upload/Stress-Physiology-in-Cotton.pdf Mahadevappa SG, Sreenivas G, Reddy DR, Madhavi A, Rao SS (2018) Effect of different levels of irrigation and nitrogen on growth and yield of Bt cotton. Int J Curr Microbiol Appl Sci 7(8):4599–4604 Mahboob W, Sarwar N, Abdul Hafeez B, Arif O, Akhtar MAR, M., Yang G (2024) Exploring genotypic variations in cotton associated with growth and nitrogen use efficiency. J Plant Nutr 47(19):3231–3250 Mishra AK, Özger M, Singh VP (2011) Association between uncertainties in meteorological variables and water-resources planning for the state of Texas. J Hydrol Eng 16(12):984–999 Mishra SK, Kaur V, Singh K (2021) Evaluation of DSSAT-CROPGRO‐cotton model to simulate phenology, growth, and seed cotton yield in northwestern India. Agron J 113(5):3975–3990 Modala, N. R., Ale, S., Rajan, N., Munster, C. L., DeLaune, P. B., Thorp, K. R., …Barnes, E. M. (2015). Evaluation of the CSM-CROPGRO-Cotton model for the Texas rolling plains region and simulation of deficit irrigation strategies for increasing water use efficiency. Transactions of the ASABE, 58(3), 685–696. Morgan G, McGinty J, Mott D (2024) Replicated Agronomic Cotton Evaluation (RACE) South, East and Central Regions of Texas. Texas A&M AgriLife Extension Service, CS-2025-01 Niu, J., Zhang, S., Liu, S., Ma, H., Chen, J., Shen, Q., … Zhao, X. (2018). The compensation effects of physiology and yield in cotton after drought stress. Journal of Plant Physiology, 224, 30–48. Orgaz F, Mateos L, Fereres E (1992) Agroclimatology and modeling: Season length and cultivar determine the optimum evapotranspiration deficit in cotton. Agron J 84:700–706 Pettigrew WT (2004) Moisture deficit effects on cotton lint yield, yield components, and boll distribution. Agron J 96(2):377–383 Piccinni G, Ko J, Wentz A, Leskovar D, Marek T, Howell T (2007) Determination of crop coefficients (Kc) for irrigation management of crops. 28th Annual International Irrigation Show , San Diego, CA, December, 9–12 Sadras VO, Bange MP, Milroy SP (1997) Reproductive allocation of cotton in response to plant and environmental factors. Ann Botany 80(1):75–81 Saenz V, Sekula D, McGinty J, Ramirez J, Livingston C, Alaniz R (2023) Hidalgo County RACE Trial, 2023 [Data table]. Texas A&M AgriLife Extension. https://southtexas.tamu.edu/files/2023/10/2023-Hidalgo-Co.-Cotton-Variety-Trials.pdf Seidel SJ, Palosuo T, Thorburn P, Wallach D (2018) Towards improved calibration of crop models–Where are we now and where should we go? Eur J Agron 94:25–35 Sezener V, Basal H, Peynırcıoglu C, Gurbuz T, Kızılkaya K (2015) Screening of cotton cultivars for drought tolerance under field conditions. Turkish J Field Crops 20(2):223–232 Shani MY, Ditta A, Khan MKR (2025) Characterizing drought-resilient cotton genotypes through morpho-physiological and biochemical traits at flowering stage. BMC Plant Biol 25(1):1–22 Singh Y, Rao SS, Regar PL (2010) Deficit irrigation and nitrogen effects on seed cotton yield, water productivity and yield response factor in shallow soils of semi-arid environment. Agric Water Manage 97(7):965–970 Su L, Wang Q, Wang C, Shan Y (2015) Simulation models of leaf area index and yield for cotton grown with different soil conditioners. PLoS ONE 10(11):e0141835 Sun F, Chen Q, Chen Q, Jiang M, Gao W, Qu Y (2021) Screening of key drought tolerance indices for cotton at the flowering and boll setting stage using the dimension reduction method. Front Plant Sci 12:619926 Tang LS, Li Y, Zhang J (2010) Biomass allocation and yield formation of cotton under partial rootzone irrigation in arid zone. Plant Soil 337(1):413–423 Tariq, M. S., Imran, M., Ud Din, S., Murtaza, B., Naeem, M. A., Amjad, M., … AbdElgawad,H. (2024). Magnetic nanocomposite of maize offal biomass for effective sequestration of Congo red and methyl orange dyes from contaminated water: Modeling, kinetics and reusability. International Journal of Phytoremediation, 26(6), 975–992. Ahmad U, Sharma L (2023) A review of best management practices for potato crop using precision agricultural technologies. Smart Agricultural Technol 4:100220 Ahmad U, Alvino A, Marino S (2022) Solar fertigation: A sustainable and smart IoT-based irrigation and fertilization system for efficient water and nutrient management. Agronomy 12(5):1012 Tarpley L, Sassenrath GF (2006) Carbohydrate profiles during cotton floral bud (square) development. J Agron Crop Sci 192(5):363–372 USDA (1976) Soil survey of Uvalde County, Texas. USDA Soil Conservation Service, Washington, DC Wang L, Lin M, Han Z, Han L, He L, Sun W (2023) Simulating the effects of drought stress timing and the amount irrigation on cotton yield using the CSM-CROPGRO-cotton model. Agronomy 14(1):14 Wang R, Gao M, Ji S, Wang S, Meng Y, Zhou Z (2016a) Carbon allocation, osmotic adjustment, antioxidant capacity and growth in cotton under long-term soil drought during flowering and boll-forming period. Plant Physiol Biochem 107:137–146 Wang R, Ji S, Zhang P, Meng Y, Wang Y, Chen B, Zhou Z (2016b) Drought effects on cotton yield and fiber quality on different fruiting branches. Crop Sci 56(3):1265–1276 Willmott CJ, Matsuura K (2005) Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Res 30(1):79–82 Yehia WMB, Zaazaa EEDI, El-Hashash EF, El-Enin A, M. M., Shaaban A (2024) Genotype-by-environment interaction analysis for cotton seed yield using various biometrical methods under irrigation levels in a semi-arid region. Arch Agron Soil Sci 70(1):1–23 Zafar, M. M., Chattha, W. S., Khan, A. I., Zafar, S., Subhan, M., Saleem, H., … Xuefei,J. (2023). Drought and heat stress on cotton genotypes suggested agro-physiological and biochemical features for climate resilience. Frontiers in Plant Science, 14, 1265700. Zhi, X., George-Jaeggli, B., Han, Y., Chen, Q., Zhang, S., Feng, L., … Yang, B. (2024).Comprehensive analysis on investigating water-saving potentials of irrigated cotton in semi-arid area in China. Agricultural Water Management, 301, 108960. Zonta JH, Brandao ZN, Rodrigues JIDS, Sofiatti V (2017) Cotton response to water deficits at different growth stages. Revista Caatinga 30(4):980–990 Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9500734","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":628028322,"identity":"c5147cf6-c4b9-4cc1-96a9-986f76a52022","order_by":0,"name":"Uzair Ahmad","email":"","orcid":"https://orcid.org/0000-0002-7384-9449","institution":"Texas A\u0026M AgriLife Research \u0026 Extension Center, Uvalde - 78801, Texas, USA","correspondingAuthor":false,"prefix":"","firstName":"Uzair","middleName":"","lastName":"Ahmad","suffix":""},{"id":628028323,"identity":"e88ced2e-3dcd-46af-85e4-832da900a209","order_by":1,"name":"Xuejun Dong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYJACAyBmbADiBwgxNuK0MIMYEkRpYYBqYZMgSovBjeQDxbw7GGT7pduvVfz4xVDHP+2MAcOHssM4tUjOSEsw5j3DYDxzzpmym719DBISt3MMGGecw62FXyLHwJi3jSFxw42ctBu8PUCHAbUw87bh1sImkf8BrqXwL1CLPEjLXzxagLYwQLWkH2Pm+cEgYQDSwohHi2TPMwPDuW0SxjNn5DBLyzZISG68nVZwsOdcOk4tBseTnxm8bbOR7ZdIf/jxzR8bfrnbyRsf/CizxqkF5B0DSGzwGDAwtkGi5QA+9UDA/ABCswPpPwTUjoJRMApGwYgEALlSU+wAQk5oAAAAAElFTkSuQmCC","orcid":"","institution":"Texas A\u0026M AgriLife Research \u0026 Extension Center, Uvalde - 78801, Texas, USA","correspondingAuthor":true,"prefix":"","firstName":"Xuejun","middleName":"","lastName":"Dong","suffix":""},{"id":628028324,"identity":"cc6124cd-6975-478b-8e55-4338e8b31269","order_by":2,"name":"Thiago F. Duarte","email":"","orcid":"","institution":"Universidade Federal de Rondonópolis, Rondonópolis - 78736, Brazil","correspondingAuthor":false,"prefix":"","firstName":"Thiago","middleName":"F.","lastName":"Duarte","suffix":""},{"id":628028325,"identity":"0dc4b37e-3cac-4d2a-85ed-4490b2bdd069","order_by":3,"name":"Noemi Tortorici","email":"","orcid":"","institution":"Università degli Studi di Palermo, Palermo - 90128, Italy","correspondingAuthor":false,"prefix":"","firstName":"Noemi","middleName":"","lastName":"Tortorici","suffix":""},{"id":628028326,"identity":"856d521d-b764-404b-8089-49b31f8ad228","order_by":4,"name":"Dale A. Mott","email":"","orcid":"","institution":"Texas A\u0026M AgriLife Extension Service, College Station - 77843, USA","correspondingAuthor":false,"prefix":"","firstName":"Dale","middleName":"A.","lastName":"Mott","suffix":""},{"id":628028327,"identity":"de86686f-8a84-4093-82c3-21b6fe2ad866","order_by":5,"name":"Benjamin M. McKnight","email":"","orcid":"","institution":"Texas A\u0026M AgriLife Extension Service, College Station - 77843, USA","correspondingAuthor":false,"prefix":"","firstName":"Benjamin","middleName":"M.","lastName":"McKnight","suffix":""}],"badges":[],"createdAt":"2026-04-22 23:40:34","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9500734/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9500734/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107878584,"identity":"2a26befc-ff3e-4c4f-b422-21739a222981","added_by":"auto","created_at":"2026-04-27 08:21:43","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":359951,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure 1.\u003c/strong\u003e Daily maximum temperature (Tmax, °C, red), minimum temperature (Tmin, °C, green), precipitation (mm, blue bars), and solar radiation (MJ m\u003csup\u003e-2\u003c/sup\u003e day\u003csup\u003e-1\u003c/sup\u003e, yellow) recorded at Uvalde, Texas, during the 2025 cotton growing season. Data were obtained from the NASA POWER Data Access Viewer (https://power.larc.nasa.gov/data-access-viewer/).\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9500734/v1/0f9379cc64b34a04b089e20c.jpg"},{"id":107878630,"identity":"19c8ad60-8752-4375-9f8c-bf4db4f48e26","added_by":"auto","created_at":"2026-04-27 08:21:44","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":614036,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure 2.1.\u003c/strong\u003e Simulated (lines) vs. observed (shapes) \u003cu\u003e\u003cstrong\u003epod numbers (m\u003c/strong\u003e\u003c/u\u003e\u003csup\u003e\u003cu\u003e\u003cstrong\u003e-2\u003c/strong\u003e\u003c/u\u003e\u003c/sup\u003e\u003cu\u003e\u003cstrong\u003e), pod, stem, and leaf weights (kg ha\u003c/strong\u003e\u003c/u\u003e\u003csup\u003e\u003cu\u003e\u003cstrong\u003e-1\u003c/strong\u003e\u003c/u\u003e\u003c/sup\u003e\u003cu\u003e\u003cstrong\u003e), AGB (kg ha\u003c/strong\u003e\u003c/u\u003e\u003csup\u003e\u003cu\u003e\u003cstrong\u003e-1\u003c/strong\u003e\u003c/u\u003e\u003c/sup\u003e\u003cu\u003e\u003cstrong\u003e), height (m), and LAI\u003c/strong\u003e\u003c/u\u003e\u003cstrong\u003e \u003c/strong\u003eunder full (100%) and moderate (75%) irrigation treatments in two cotton cultivars NG 4190 B3XF and ST 4990 B3XF during the 2025 cotton growing season at Uvalde, Texas.\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9500734/v1/105c099e2b9073621a44a8b3.jpg"},{"id":107878629,"identity":"0018ebcb-ba61-4894-a45f-9bf69234b1f3","added_by":"auto","created_at":"2026-04-27 08:21:44","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":352323,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure 2.2.\u003c/strong\u003e Simulated \u003cu\u003e\u003cstrong\u003epod numbers (m\u003c/strong\u003e\u003c/u\u003e\u003csup\u003e\u003cu\u003e\u003cstrong\u003e-2\u003c/strong\u003e\u003c/u\u003e\u003c/sup\u003e\u003cu\u003e\u003cstrong\u003e), pod, stem, and leaf weights (kg ha\u003c/strong\u003e\u003c/u\u003e\u003csup\u003e\u003cu\u003e\u003cstrong\u003e-1\u003c/strong\u003e\u003c/u\u003e\u003c/sup\u003e\u003cu\u003e\u003cstrong\u003e), AGB (kg ha\u003c/strong\u003e\u003c/u\u003e\u003csup\u003e\u003cu\u003e\u003cstrong\u003e-1\u003c/strong\u003e\u003c/u\u003e\u003c/sup\u003e\u003cu\u003e\u003cstrong\u003e), height (m), and LAI \u003c/strong\u003e\u003c/u\u003eunder half (50%), and no irrigation (0%) treatments in two cotton cultivars NG 4190 B3XF and ST 4990 B3XF during the 2025 cotton growing season at Uvalde, Texas.\u003c/p\u003e","description":"","filename":"Picture3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9500734/v1/6eda10ca8d8d124ce0d47846.jpg"},{"id":107878702,"identity":"92f3a8a4-7abc-4c50-b8ce-ab0f4f37af6f","added_by":"auto","created_at":"2026-04-27 08:21:56","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":357953,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFigure 3.\u003c/strong\u003e Sensitivity analysis of simulated cotton pod numbers (m\u003csup\u003e-2\u003c/sup\u003e), pod, and leaf weights (all kg ha\u003csup\u003e-1\u003c/sup\u003e), AGB (kg ha\u003csup\u003e-1\u003c/sup\u003e), and LAI under full, moderate, half, and no irrigation treatments during the 2025 growing season at Uvalde, Texas. \u003cstrong\u003eNote:\u003c/strong\u003e Each line indicates baseline starting values (16, 24) with increments (1, 2): colors denote combinations of yellow (16, +1), orange (16, +2), green (24, +1) and blue (24, +2) for each parameter. These values show sensitivity to the specific cultivar parameter whose baseline starting values were set at 16 and 24.\u003c/p\u003e","description":"","filename":"Picture4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9500734/v1/a383a54ae809926d95b17566.jpg"},{"id":107878761,"identity":"682abfac-ddc2-4e43-8555-237ee203961d","added_by":"auto","created_at":"2026-04-27 08:22:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2241398,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9500734/v1/2bef3085-2582-471f-8a2c-6e3d0ec16980.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003ePredicting Cotton Growth, Pod Development, and Lint Yield Responses to Irrigation and Cultivar Differences in Southwest Texas\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCotton (\u003cem\u003eGossypium hirsutum\u003c/em\u003e L.) is one of the most valuable fiber crops globally and a cornerstone of the agricultural economy in Texas (Himanshu et al., 2023). In the United States, cotton is a key field crop that drives on-farm revenue, supports large-scale production systems, and provides significant value to national and international agricultural markets (Loka et al., 2011). However, cotton cultivation is water-intensive and sensitive to water supply timing, especially in semi-arid environments. As such, optimizing cotton water use while maintaining yield is a critical agronomic challenge, particularly in regions where groundwater and surface water resources are constrained (Singh et al., 2010).\u003c/p\u003e \u003cp\u003eThe southwest region in Texas typifies this challenge. This area experiences limited and highly variable rainfall, high atmospheric evaporative demand, and long growing seasons, which together place intense stress on water resources (Mishra et al., 2011). Many growers rely heavily on irrigation, often withdrawing water from underground sources such as the Edwards Aquifer (Ko et al., 2009; Piccinni et al., 2007). Climate variability, declining aquifers, and increasing regulatory restrictions make it essential to develop strategies that maximize cotton yield per unit of applied water. In this context, the use of crop simulation models offers a powerful tool to explore and optimize cotton growth under different irrigation treatments (Adhikari et al., 2016).\u003c/p\u003e \u003cp\u003eProcess-based models like DSSAT-CROPGRO-Cotton (Hoogenboom et al., 2019) simulate critical physiological processes\u0026mdash;phenology, canopy expansion, carbon assimilation, biomass partitioning, and final yield\u0026mdash;all in response to weather and soil conditions and management practices (Cammarano et al., 2012; Hoogenboom et al., 2024). When locally calibrated, these models can predict cultivar-specific responses to water stress, support decision-making around irrigation timing and rate, and identify cultivar traits that confer resilience. Despite the utility of these models, successful applications require careful calibration using field data, particularly in semi-arid zones where crop response to water varies strongly with cultivar and management (Guerra et al., 2007; Jones et al., 2003).\u003c/p\u003e \u003cp\u003eIn Texas, cotton cultivars such as NG 4190 and ST 4990 are widely used, yet their responses under reduced irrigation (e.g., 75%, 50%, and 0%) remain poorly quantified (Morgan et al., 2024). Without site-specific calibration, model predictions may misrepresent phenological shifts, leaf area dynamics, biomass allocation, or yield under water deficit. For instance, deficit irrigation may induce earlier flowering, accelerate leaf senescence, or alter biomass partitioning that reduces lint yield, but these trade-offs are not always obvious from field measurements alone (Ko et al., 2009). By calibrating CROPGRO-Cotton with field data from Texas, we evaluated hidden responses and generated realistic simulations to guide water management under water-limited production conditions.\u003c/p\u003e \u003cp\u003eOur study therefore addresses two critical needs: \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003efirst\u003c/span\u003e, to calibrate the CROPGRO-Cotton model for NG 4190 and ST 4990 using phenology, LAI, biomass (pods, stems, leaves), and yield data collected under four simulated irrigation treatments (100%, 75%, 50%, 0%); \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003esecond\u003c/span\u003e, to use the calibrated model to simulate how these cultivars allocate biomass, adjust their growth, and set yield under water deficit. By combining field measurements and model simulations, we aim to generate insights that are difficult to observe in standard agronomic trials, such as how LAI declines under stress, how biomass partitioning shifts, and how reproductive sinks respond.\u003c/p\u003e \u003cp\u003eThe importance of this work lies in several interconnected factors. Phenology\u0026mdash;particularly the timing of flowering and maturity\u0026mdash;is a key determinant of yield under water stress (Bednarz \u0026amp; Nichols, 2005). Earlier or compressed reproductive windows can lead to more efficient use of water but may reduce the number of bolls or their fill (Herritt et al., 2022). Reductions in LAI under stress may lower the crop\u0026rsquo;s source capacity, limiting assimilate supply to the developing bolls (Bista et al., 2024). Biomass partitioning between pods, leaves, and stems underlies the sink\u0026ndash;source balance: water deficits may shift allocation away from reproductive tissues, reducing lint yield (Tariq et al., 2024). Moreover, cultivar differences in all these traits determine their drought resilience: one genotype may sustain LAI and biomass better under stress, while another may prematurely abort fruit (Herritt et al., 2022). Using a calibrated model, we can quantify these physiological responses and provide actionable information for irrigation strategy, cultivar selection, and agronomic optimization in semi-arid cotton production.\u003c/p\u003e \u003cp\u003eThe \u003cb\u003eobjective\u003c/b\u003e of our study is to quantify the effects of four irrigation treatments on phenology, growth, and yield of two cotton cultivars (NG 4190 B3XF and ST 4990 B3XF) and calibrate the DSSAT CROPGRO-Cotton model against field-level observations under semi-arid conditions. Our \u003cb\u003ehypothesis\u003c/b\u003e is that under semi-arid conditions, full irrigation maximizes LAI, biomass, and lint yield in both NG 4190 and ST 4990, while deficit irrigation triggers cultivar-specific changes in growth, biomass partitioning, and yield stability. To test this hypothesis, we posed three research questions, (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) How do different irrigation treatments (100%, 75%, 50%, 0%) influence the phenological development (emergence, flowering, maturity) of the two tested cultivars in a semi-arid region? (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) In what way does LAI respond to water deficit for the two tested cultivars? and (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) How does biomass partitioning (pods, stems, leaves) change under differing irrigation treatments and impact lint yield?\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study site and crop management\u003c/h2\u003e \u003cp\u003eThe field experiment was conducted from April to September 2025 at the Texas A\u0026amp;M AgriLife Research and Extension Center at Uvalde, Texas (29.21\u0026deg;N, 99.76\u0026deg;W; elevation 282 m). The study occupied approximately 0.404 hectares of levelled land previously planted with cover crops. The site has a semi-arid climate and clay soil of the Uvalde series (fine-silty, mixed, active, hyperthermic Aridic Calciustoll; USDA, 1976). Further information can be found in Ahmad et al. (2026). A 2\u0026times;2 factorial design was used with two cotton cultivars (NG 4190 and ST 4990), two irrigation treatments (100% and 75%) and with three replications, totaling 12 plots. Each plot contained four rows, 60-m long with 76.2-cm row spacing. Three weeks before planting, the field received 50 mm of irrigation via a linear sprinkler system to ensure uniform soil moisture. Cotton seeds were sown on April 8 using a six-row vacuum planter. Post-emergence, irrigation was applied through surface drip tapes placed along individual rows. Pre-plant soil testing indicated adequate nitrogen (N) and phosphorus (P) in the top 15 cm of soil; 100 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e N (urea, 46-0-0) and 26 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e P\u003csub\u003e2\u003c/sub\u003eO\u003csub\u003e5\u003c/sub\u003e (diammonium phosphate, 18-46-0) were applied to maintain sufficient nutrient availability. Potassium (K) was not added due to adequate soil K levels. Thrips damage was observed on May 7 in young seedlings and controlled with AgriMek SC (0.13 L ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) on May 12 and Acephate (0.29 L ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) on May 16. Plants recovered within 10 days, and no further pest or disease issues occurred.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Field measurements\u003c/h2\u003e \u003cdiv id=\"Sec5\" class=\"Section3\"\u003e \u003ch2\u003e2.2.1. Pod numbers (m\u003csup\u003e-2\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003ePod numbers (m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e) were measured at the reproductive stage by sampling a defined ground area within each plot. A 1-m\u003csup\u003e2\u003c/sup\u003e area was delineated in the two central rows of each plot using a measuring tape to avoid border effects. All plants within the marked area were harvested, and the pods were manually counted. The total number of pods collected from the sampled area was recorded directly as pods m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e and averaged across sampling locations to represent plot-level pod density.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section3\"\u003e \u003ch2\u003e2.2.2. Pod, stem, and leaf weights (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003ePod, stem, and leaf weights (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) were determined through destructive plant sampling. Five representative plants were collected from the central rows of each plot and immediately separated into pods, stems, and leaves in the field. Each component was oven-dried at 65\u0026deg;C until constant weight and weighed using a precision analytical balance. The measured dry masses were converted to kilograms per hectare (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) using the sampled area and plant population.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.2.3. AGB (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eAGB (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) was calculated to quantify total biomass accumulation above the soil surface. It was obtained by summing the dry weights of pods, stems, and leaves from the destructively sampled plants. Each component weight was first expressed on a per-hectare basis using plant density and sampling area. The resulting sum represented total AGB (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) for each plot and sampling event.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.2.4. Height (m)\u003c/h2\u003e \u003cp\u003ePlant height was measured to quantify vertical plant growth during the season. Five representative plants were collected from the two central rows of each plot to avoid border effects and transported to the laboratory. Each plant was laid straight, and height was measured from the base of the stem to the terminal bud of the main stem using a measuring tape. Measurements were recorded in meters (m), and the mean value of the sampled plants represented the plot-level plant height.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.2.5. LAI\u003c/h2\u003e \u003cp\u003eLAI was quantified from all 12 plots using a LI-2200C Plant Canopy Analyzer (LI-COR Biosciences, Lincoln, NE, USA) for in-field indirect measurements. Measurements were conducted once every two weeks on mostly clear days, avoiding periods with patchy clouds. Five measurements along a 15-m transect near the plot center were taken to obtain five LAI values per plot. For each measurement, one above-canopy and four below-canopy photosynthetically active radiation (PAR) readings were collected along a 2-m span at varying canopy positions between adjacent rows to capture within-plot light variability. The mean of the five measurements represented the plot-level LAI for each sampling date.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e \u003ch2\u003e2.2.6. Yield (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eCotton was hand-harvested on September 3\u0026ndash;4 from two middle row, 5-m long segments per plot. The collected seed cotton was ginned on a 20 saw Centennial Gin and weighed to determine seed cotton yield. Subsamples of ginned lint were then sent to the Fiber and Biopolymer Research Institute, Lubbock, Texas, for fiber quality analysis. Measurements were conducted using high volume instrument (HVI) system to assess key fiber quality parameters.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.3. DSSAT workflow\u003c/h2\u003e \u003cp\u003eCotton growth in this study was simulated using the CROPGRO-Cotton module within DSSAT v4.8.2 (Hoogenboom et al., 2019). This process-based model predicts cotton growth, development, and yield responses under varying irrigation treatments by integrating cultivar-specific genetic coefficients, daily weather, soil characteristics, and crop management practices. CROPGRO-Cotton simulates key physiological processes including phenology, LAI, and AGB partitioning into pods, stems and leaves, and lint yield. It has been widely validated for cotton under diverse environmental conditions and stress scenarios. Model inputs included detailed soil profiles with hydraulic properties, daily weather data (maximum and minimum temperatures, solar radiation, and precipitation), cultivar-specific parameters, and management practices such as sowing dates, irrigation schedules, fertilization, and row spacing. Calibration was performed using observed field data for cotton cultivar responses to ensure accurate simulation of phenology, pod numbers, pod, stem and leaf weights, AGB, height, LAI, and lint yield.\u003c/p\u003e \u003cp\u003eDSSAT inputs were prepared with \u003cem\u003eXBuild\u003c/em\u003e, weather (.\u003cem\u003eWTH\u003c/em\u003e), soil (.\u003cem\u003eSOL\u003c/em\u003e), and experimental treatment (.\u003cem\u003eX\u003c/em\u003e) files. Yield outputs were stored in average (.\u003cem\u003eA\u003c/em\u003e) files, and time-series pod numbers, pod, stem and leaf weights, AGB, height, LAI growth in time (.\u003cem\u003eT\u003c/em\u003e) files. FileX verified consistency among weather, soil, and management inputs, linking each treatment to site-specific soil and climate files. Time-series outputs were assigned to irrigation treatments for cultivar calibration. Cultivar and ecotype parameters were calibrated by manual trial-and-error in the \u003cem\u003eCNGRO048.CUL\u003c/em\u003e and \u003cem\u003eCNGRO048.ECO\u003c/em\u003e files. Daily weather data were formatted using \u003cem\u003eWeathergen\u003c/em\u003e. For calibration, key cotton phenological and growth variables were implemented and iteratively adjusted in both files to closely match observed data, enabling robust simulation of phenology, growth, and lint yield under full, moderate, half, and no irrigation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.4. DSSAT inputs\u003c/h2\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003ch2\u003e2.4.1. Soil data\u003c/h2\u003e \u003cp\u003eThe soil at the study site has 57% clay, 31% silt, and 12% sand. Bulk density measured 1.38 g cm\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e (0\u0026ndash;20 cm), 1.43 g cm\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e (20\u0026ndash;80 cm), and 1.58 g cm\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e (80\u0026ndash;120 cm). The field capacity and permanent wilting point were 0.35 and 0.18 cm\u003csup\u003e3\u003c/sup\u003e cm\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e, respectively. Soil pH averaged 8.3 (0-120 cm), and organic matter in the top 0\u0026ndash;20 cm was 3.6%. Laboratory analyses quantified nitrate-N (NO\u003csub\u003e3\u003c/sub\u003e\u003csup\u003e\u0026minus;\u003c/sup\u003eN), P, K, calcium (Ca), and magnesium (Mg). These physical and chemical properties were incorporated into DSSAT v4.8.2 by creating a site-specific soil input file to accurately simulate water dynamics and nutrient availability for cotton growth and yield prediction (Refer to Ahmad et al. (2026) for detailed soil chemical and physical properties across soil depths).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e \u003ch2\u003e2.4.2. Crop data\u003c/h2\u003e \u003cp\u003eCotton was planted on April 8 at a uniform seeding rate of 130,000 ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for the two cultivars, NG 4190 and ST 4990, across 12 plots, with two factors: cultivar and irrigation treatment (full vs. deficit). Field and crop management practices, including planting date, fertilization, irrigation, and pest control, were applied uniformly across plots. Key cotton phenological dates were recorded: emergence on April 15, anthesis on June 15, physiological maturity on July 31, and harvest on September 3\u0026ndash;4. Crop management details were incorporated into DSSAT via crop management .\u003cem\u003eX\u003c/em\u003e files. Cultivar-specific genetic coefficients were defined in the .\u003cem\u003eCUL\u003c/em\u003e file. Planting density, row spacing, and fertilizer applications were recorded in the management and fertilizer files.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e \u003ch2\u003e2.4.2. Weather data\u003c/h2\u003e \u003cp\u003eThe study area has a semi-arid climate, with mean annual rainfall of 663 mm and average yearly evapotranspiration of 1506 mm. Monthly temperatures range from 24.7\u0026deg;C in May to 12.2\u0026deg;C in December. Weather data, including precipitation (mm), maximum and minimum temperatures (\u0026deg;C), and solar radiation (MJ m\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e day\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), were processed and imported into DSSAT to simulate cotton growth and yield \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e2.5. DSSAT calibration\u003c/h2\u003e \u003cp\u003eThe CROPGRO-Cotton model was calibrated using the 2025 field data for NG 4190 and ST 4990, including key observations for phenology, pod numbers, pod, stem and leaf weights, AGB, height, and LAI (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Calibration followed an iterative manual trial-and-error approach, which allowed direct control over parameter adjustments and rapid refinement compared to automated optimization routines that require large numbers of runs and longer processing time. This method made it possible to tune cultivar behavior efficiently while matching observed field performance and requires an experienced modeler, as expert judgment is essential for interpreting parameter interactions and selecting biologically realistic values.\u003c/p\u003e \u003cp\u003eCalibration targeted cultivar-specific parameters governing phenology, canopy development, and boll formation. \u003cem\u003eEM-FL\u003c/em\u003e (emergence to first flower) was adjusted first to align flowering onset. \u003cem\u003eFL-SH\u003c/em\u003e (flower to first boll set) and \u003cem\u003eFL-SD\u003c/em\u003e (flower to seed development) were refined to capture boll initiation and early reproductive dynamics. \u003cem\u003eSD-PM\u003c/em\u003e (seed to physiological maturity) was tuned to synchronize simulated and observed crop maturity. Canopy growth was improved by adjusting \u003cem\u003eFL-LF\u003c/em\u003e (time between first flower and end of leaf expansion), \u003cem\u003eSLAVR (specific leaf area)\u003c/em\u003e, and \u003cem\u003eSIZLF\u003c/em\u003e (size of full leaf), enhancing LAI and leaf expansion accuracy. Reproductive growth was calibrated using \u003cem\u003eSFDUR\u003c/em\u003e (seed filling duration) and \u003cem\u003eSDPDV\u003c/em\u003e (seed per pod division factor) to match observed boll growth rates and yield partitioning. Early phenological parameters were prioritized due to their cascading effects on later development. \u003cem\u003eEM-FL\u003c/em\u003e and \u003cem\u003eFL-SH\u003c/em\u003e were iteratively revisited to maintain thermal-time consistency, followed by refinement of \u003cem\u003eSD-PM\u003c/em\u003e and \u003cem\u003eSDPDV\u003c/em\u003e to resolve cultivar-specific differences in biomass allocation, boll retention, and lint yield. Calibration performance was accepted when simulated phenology, LAI, AGB, and lint yield deviated by \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026le;\u003c/span\u003e\u0026thinsp;10% from observations and root mean square error (RMSE) for LAI and yield remained below 15% of observed means (Seidel et al., 2018).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e2.6. Sensitivity analysis\u003c/h2\u003e \u003cp\u003eThe \u003cem\u003eExperiment\u003c/em\u003e option was selected from the DSSAT shell to establish a new experimental setup. Next, the \u003cem\u003eGenetics\u003c/em\u003e menu was accessed, and \u003cem\u003eCultivars\u003c/em\u003e was chosen, followed by \u003cem\u003eGenetic Coefficients\u003c/em\u003e. The cultivar-specific genetic coefficients \u003cem\u003eEM-FL\u003c/em\u003e, \u003cem\u003eCSDL\u003c/em\u003e (critical short-day length), \u003cem\u003ePODUR\u003c/em\u003e (pod development duration), \u003cem\u003eSFDUR\u003c/em\u003e (seed filling duration), and \u003cem\u003eWTPSD\u003c/em\u003e (weight per seed) were then selected. For each genetic coefficient, the \u003cem\u003eStarting Value\u003c/em\u003e, \u003cem\u003eIncrements\u003c/em\u003e, and \u003cem\u003eIterations\u003c/em\u003e were defined. This allowed each genetic coefficient to vary systematically within a biologically plausible range while holding other genetic coefficients constant, thereby isolating its individual effect on model outputs. After configuring all coefficients, the \u003cem\u003eRun\u003c/em\u003e button was executed to generate the experiment, and the setup was saved for reproducibility. Subsequently, the \u003cem\u003eRun Model\u003c/em\u003e function was activated to simulate crop growth responses across the defined parameter space. Upon completion of the simulations, the \u003cem\u003eAnalysis\u003c/em\u003e option was selected to retrieve model outputs. The file \u003cem\u003ePlantGro.Out\u003c/em\u003e was extracted, as it provides detailed information on crop phenology, pod numbers, pod, stem and leaf weights, AGB, height, and LAI. Finally, the \u003cem\u003ePlot\u003c/em\u003e function was used to visualize simulation results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e2.7. Adaptation strategies\u003c/h2\u003e \u003cp\u003eThe \u003cem\u003eExperiment\u003c/em\u003e option was selected to create a setup for adaptation analysis. Only two adaptation strategies were evaluated: crop management adaptations and irrigation-treatments adaptations. All other inputs were held constant to isolate the effects of each strategy.\u003c/p\u003e \u003cp\u003eCrop management adaptations were implemented through the \u003cem\u003eManagement\u003c/em\u003e menu. The \u003cem\u003ePlanting Details\u003c/em\u003e option was used to modify planting strategies, including adjustments in planting dates to represent early, optimum, and delayed sowing scenarios. Row spacing and plant population were altered. Harvesting strategies were adjusted by modifying harvest timing to align with changes in crop phenology induced by altered planting configurations.\u003c/p\u003e \u003cp\u003eIrrigation-treatments adaptations were implemented using the \u003cem\u003eIrrigation\u003c/em\u003e module within the \u003cem\u003eManagement\u003c/em\u003e menu. Irrigation was applied at specific cotton growth stages to represent stage-specific water availability scenarios. Separate treatments included irrigation only during the vegetative stage, only during the flowering stage, only during the grain-filling stage, and full-season irrigation as a control. Irrigation amounts per event were kept constant, allowing evaluation of crop growth and yield responses and sensitive growth stages.\u003c/p\u003e \u003cp\u003eAfter all, the \u003cem\u003eRun\u003c/em\u003e button was executed to generate the experiments, and each configuration was saved. The \u003cem\u003eRun Model\u003c/em\u003e function was then used to simulate crop growth responses under each adaptation strategy. Then, results were analyzed using the \u003cem\u003eAnalysis\u003c/em\u003e option, and outputs were extracted from \u003cem\u003ePlantGro.Out\u003c/em\u003e, which provided detailed information on phenology, pod, stem and leaf weights, AGB, and LAI. The \u003cem\u003ePlot\u003c/em\u003e function was used to visualize and compare crop responses across adaptation strategies.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eFinal calibration parameters for cotton cultivars NG 4190 and ST 4990 using manual trial-and-error in 2025 at Uvalde, Texas.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCrop\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrop cultivars\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCalibration methodology\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eEvaluation metrics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eParameters names\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eParameters threshold\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eParameters variability\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eFinal values\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"15\" rowspan=\"16\"\u003e \u003cp\u003eCotton\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eNG 4190\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eManual trial-and-error\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eMAE / RMSE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEM-FL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e34.00\u0026ndash;44.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;5.0\u0026ndash;10.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e39.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFL-SH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e8.00\u0026ndash;12.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1.0\u0026ndash;3.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e9.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFL-SD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e15.00\u0026ndash;18.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;2.0\u0026ndash;3.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e17.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFL-LF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e65.00\u0026ndash;75.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;4.00\u0026ndash;10.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e69.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSLAVR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e170.00-175.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;3.0\u0026ndash;5.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e173\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSIZLF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e250.00-300.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;10.0\u0026ndash;50.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e299.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSFDUR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e24.00\u0026ndash;35.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;4.0\u0026ndash;8.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e29.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSDPDV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e20.00\u0026ndash;27.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;5.00\u0026ndash;10.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e27.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eST 4990\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eManual trial-and-error\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\" morerows=\"7\" rowspan=\"8\"\u003e \u003cp\u003eMAE / RMSE\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEM-FL\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e34.00\u0026ndash;44.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;5.0\u0026ndash;10.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e39.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFL-SH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e8.00\u0026ndash;12.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;1.0\u0026ndash;3.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e11.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFL-SD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e15.00\u0026ndash;18.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;2.0\u0026ndash;3.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e17.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eFL-LF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e65.00\u0026ndash;75.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;4.00\u0026ndash;10.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e72.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSLAVR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e170.00-175.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;3.0\u0026ndash;5.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e175\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSIZLF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e250.00-300.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;10.0\u0026ndash;50.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e263.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSFDUR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e24.00\u0026ndash;35.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;4.0\u0026ndash;8.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e32.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSDPDV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e20.00\u0026ndash;27.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e+\u0026thinsp;5.00\u0026ndash;10.00\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e25.00\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"8\"\u003e\u003cb\u003eNote\u003c/b\u003e: \u003cem\u003eCalibration was performed using the manual trial-and-error method. Alternative methods, including GLUE, PEST, or TSE, can be applied depending on study objectives and user expertise.\u003c/em\u003e\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e2.8. Model performance evaluation\u003c/h2\u003e \u003cp\u003eSimulated and observed phenology, LAI, AGB partitioning into pods, stems, leaves, and lint yield were compared with the observed values to quantify CROPGRO model accuracy under site-specific inputs. Model performance was assessed using mean absolute error (MAE, \u003cb\u003eEq.\u0026nbsp;1\u003c/b\u003e) and root mean square error (RMSE, \u003cb\u003eEq.\u0026nbsp;2\u003c/b\u003e) (Jamshidi et al., 2024). All analyses were conducted in R (RStudio) using custom scripts with ggplot2, dplyr, readxl, and tidyr for data cleaning, analysis, and visualization. Model performance was considered acceptable when both MAE and RMSE were \u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026le;\u003c/span\u003e\u0026thinsp;20% of the observed mean. All MAE and RMSE values are reported as percentages relative to the mean of the observed values (Willmott \u0026amp; Matsuura, 2005):\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:\\begin{array}{cc}MAE=\\frac{{\\sum\\:}_{\\text{i}=1}^{\\text{n}}{S}_{i}-{O}_{i}}{\\text{n}}\u0026amp;\\:\\end{array}\\)\u003c/span\u003e \u003c/span\u003e \u003cem\u003e1\u003c/em\u003e\u003c/p\u003e \u003cp\u003ewhere:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003eO\u003c/em\u003e \u003csub\u003e \u003cem\u003ei\u003c/em\u003e \u003c/sub\u003e corresponds to the observed data\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003es\u003c/em\u003e \u003csub\u003e \u003cem\u003ei\u003c/em\u003e \u003c/sub\u003e denotes the simulated data\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003en\u003c/em\u003e represents the total number of samples\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:RMSE=\\frac{1}{n}\\sqrt{\\sum\\:_{i-1}^{n}{({y}_{i}-{\\widehat{y}}_{i})}^{2}}\\)\u003c/span\u003e \u003c/span\u003e \u003cem\u003e2\u003c/em\u003e \u003c/p\u003e \u003cp\u003ewhere:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003ey\u003c/em\u003e \u003csub\u003e \u003cem\u003ei\u003c/em\u003e \u003c/sub\u003e represents the simulated data\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{\\widehat{y}}_{i}\\:\\)\u003c/span\u003e \u003c/span\u003ecorresponds to field-observed data employed\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003e \u003cem\u003en\u003c/em\u003e denotes the total number of samples\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. RESULTS","content":"\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Model calibration\u003c/h2\u003e \u003cdiv id=\"Sec22\" class=\"Section3\"\u003e \u003ch2\u003e3.1.1. Phenological stages\u003c/h2\u003e \u003cp\u003eUnder full irrigation (100%), the simulated anthesis of NG 4190 occurred on June 10 (63 days after planting [DAP]), closely matching the observed anthesis on June 15 (68 DAP), with an observed difference of 5 days (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). For ST 4990, simulated anthesis was on June 9 (62 DAP), and the observed date was June 15.\u003c/p\u003e \u003cp\u003eThe simulated onset of boll formation was July 4\u0026ndash;5 (87\u0026ndash;88 DAP) under full irrigation for both cultivars, aligning closely with observed first-boll dates around 92 DAP. Under moderate water stress (75% irrigation), observed boll initiation occurred 1\u0026ndash;2 days later than the full irrigation treatment, while simulated boll initiation under 50% irrigation was delayed by 2\u0026ndash;3 days. Under no irrigation (0%), simulated boll initiation was delayed, indicating restricted reproductive development under severe water stress.\u003c/p\u003e \u003cp\u003eSimulated first seed set occurred around June 27 (80 DAP) for NG 4190 and June 25 (78 DAP) for ST 4990 under full irrigation. Under the 75% irrigation treatment, this stage occurred slightly later (83\u0026ndash;84 DAP), while at 50% irrigation it appeared at 88 DAP. The 0% irrigation treatment delayed seed set by 3\u0026ndash;5 more days.\u003c/p\u003e \u003cp\u003eSimulated crop maturity was predicted between Aug 20 to 25 (135\u0026ndash;140 DAP) across all treatments, which was about 3 weeks later than the observed physiological maturity dates (around July 31). The model predicted the duration from planting to maturity with a mean absolute deviation of 5\u0026ndash;15 days (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSimulated and observed phenological stages of cotton cultivars under four irrigation treatments at Texas A\u0026amp;M AgriLife, Uvalde in 2025.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eCotton cultivars\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eIrrigation treatments (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eAnthesis (DAP)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eBoll formation (DAP)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e \u003cp\u003eMaturity (DAP)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSimulated\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eObserved\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSimulated\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eObserved\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eSimulated\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eObserved\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNG 4190\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo irrigation (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e140\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHalf irrigation (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModerate irrigation (\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFull irrigation (100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e135\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eST 4990\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo irrigation (0)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e111\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e140\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHalf irrigation (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e90\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e137\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eModerate irrigation (\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e65\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e136\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFull irrigation (100)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e87\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e135\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003e3.1.2. Pod numbers (m\u003csup\u003e-2\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eUnder full and moderate irrigation treatments, pod accumulation progressed steadily beyond 100 DAP, indicating sustained square retention and continued boll setting under stable moisture (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Around peak formation (110\u0026ndash;115 DAP), cultivar differences became less pronounced, suggesting that adequate water buffered genetic variability in reproductive load. With reduced irrigation treatments, pod trajectories shortened and reached an earlier ceiling, implying mid-season stress limited the initiation of new fruiting sites rather than only reducing final counts. DSSAT predicted this earlier plateau, consistent with higher MAE and RMSE but with correct stress timing. The widening cultivar separation under deficit conditions suggests that reproductive stability becomes more genotype-dependent when soil moisture declines. Irrigation treatments control the duration of active pod formation more than total pods, emphasizing the need to maintain mid-reproductive moisture.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section3\"\u003e \u003ch2\u003e3.1.3. Pod weight (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eUnder full and moderate irrigation treatments, DSSAT predicted continued pod mass accumulation after mid-reproduction, indicating sustained assimilate supply and stable canopy function (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Early pod growth was similar between cultivars, but differences appeared later: NG 4190 maintained gradual biomass gain, while ST 4990 reached an earlier plateau. As irrigation decreased, pod weight leveled off sooner, showing that seed filling was more sensitive to water stress. This pattern indicates that bolls remained, but carbon supply to filling tissues declined under deficit moisture. Model deviations increased under half and no irrigation treatments, reflecting greater variability during stress-driven filling rather than structural model bias. Moderate irrigation maintained pod filling efficiency, confirming pod weight as a practical indicator of late-season water limitation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003e3.1.4. Stem weight (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eStem weight responses showed that irrigation treatments controlled structural persistence rather than early gains (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Under full and moderate irrigation treatments, DSSAT predicted continued stem accumulation after canopy closure, indicating stable vegetative support for reproductive development. Early stem growth remained similar across cultivars, but NG 4190 maintained slightly longer persistence, suggesting better structural stability under sustained moisture. As irrigation declined, stem growth plateaued earlier than reproductive traits, implying that vegetative support weakened before boll development ended. This pattern indicates that water stress first reduced structural carbon allocation, potentially shortening canopy lifespan and limiting assimilation supply. Model deviations increased under half and no irrigation treatments, reflecting higher variability during stress-driven biomass partitioning rather than directional bias. Moderate irrigation provided stem growth continuity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003e3.1.5. Leaf weight (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eLeaf weight (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) reflected how irrigation influenced canopy longevity (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Under full and moderate irrigation treatments, both cultivars maintained leaf weight well beyond peak flowering, allowing continued assimilation supply to developing bolls. Water deficit altered this response. At 50% and 0% irrigation treatments, leaf weight declined earlier and more sharply, indicating faster leaf senescence rather than poor early canopy growth. This reduction in leaf weight occurred alongside earlier stabilization of pod numbers, showing that shortened canopy duration\u0026mdash;not limited fruit initiation\u0026mdash;restricted late-season productivity. NG 4190 consistently retained slightly higher leaf weight than ST 4990 under deficit irrigation treatments, suggesting better canopy resilience as soil moisture declined. DSSAT predicted the timing of leaf loss, with most prediction errors occurring during stress-induced senescence rather than during vegetative growth.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003e3.1.6. AGB (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eUnder 100% and 75% irrigation treatments, the model predicted steady AGB accumulation beyond 100 DAP, indicating that canopy productivity persisted through boll filling instead of plateauing at flowering (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). This late-season persistence was consistently longer for NG 4190, aligning with its higher pod retention and delayed leaf senescence under adequate moisture. Moderate irrigation treatment provided a near-full AGB trajectory until mid-reproduction, after which growth slowed but remained positive, identifying 75% irrigation as a functional threshold where biomass efficiency was retained despite reduced water input. In contrast, 50% and 0% irrigation caused AGB to level off 15\u0026ndash;25 days earlier, signaling premature source limitation rather than weak early vigor. Protecting mid-to-late season water supply stabilizes biomass accumulation, especially in drought-resilient cultivars like NG 4190.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section3\"\u003e \u003ch2\u003e3.1.7. Height (m)\u003c/h2\u003e \u003cp\u003eAcross irrigation treatments, early elongation was similar, but differences emerged during squaring to early boll development (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Under full and moderate irrigation treatments, DSSAT predicted continued height increase for 10\u0026ndash;15 additional days, indicating delayed structural stabilization rather than excessive growth. This development aligned with higher pod numbers under 75% irrigation (8\u0026ndash;12% greater than under 50% irrigation), suggesting effective coordination between vegetative structure and reproductive demand. In contrast, 50% and 0% treatments reached maximum height earlier, reflecting constrained internode expansion once soil moisture declined. Importantly, reduced height under deficit irrigation did not proportionally reduce pod set, implying that height alone was not yield-limiting, but its timing was critical. Model\u0026ndash;field divergence increased under severe stress late in the season, likely because DSSAT predicted structural growth assumptions, while field plants prioritized survival. Maintaining water supply through squaring stabilizes canopy height and supports reproductive efficiency without promoting excessive vegetative growth.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section3\"\u003e \u003ch2\u003e3.1.8. LAI\u003c/h2\u003e \u003cp\u003eLAI differences were not driven by early growth; they emerged during squaring to boll filling, when water demand peaked (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e2.2\u003c/span\u003e). Up to 60 DAP, all treatments followed nearly identical canopy expansion, confirming that stand establishment was uniform. Divergence began after flowering. Under 100% and 75% irrigation treatments, LAI remained above 4.5 well past 100 DAP, sustaining radiation capture during active boll filling. This extended canopy duration coincided with continued pod accumulation beyond 100 DAP, indicating that canopy persistence\u0026mdash;not maximum LAI\u0026mdash;supported reproductive stability. Under 50% and 0% irrigation treatments, LAI declined 10\u0026ndash;20 days earlier, even while pod numbers were still increasing. This temporal mismatch suggests that carbon supply became limited before fruiting capacity was exhausted. The earlier decline explains why pod weight, rather than pod number, was more sensitive under deficit irrigation. DSSAT predicted the timing of canopy turnover across irrigation treatments, indicating robust calibration of senescence parameters. NG 4190 maintained slightly longer functional LAI under stress, aligning with its stronger yield retention in drought conditions.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec30\" class=\"Section3\"\u003e \u003ch2\u003e3.1.9. Yield (kg ha\u003csup\u003e-1\u003c/sup\u003e)\u003c/h2\u003e \u003cp\u003eLint yield showed a threshold response to irrigation treatments reduction (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Moving from 100% to 75% irrigation reduced water input substantially, yet yield remained comparatively stable, indicating that reproductive processes were largely protected under moderate deficit. In contrast, the drop from 75% to 50% irrigation triggered a sharper yield contraction, reflecting reduced boll filling duration rather than reduced boll initiation. This confirms that mid-to-late season water availability governs final lint mass. Under 0% irrigation treatment, cultivar separation became agronomically significant. NG 4190 sustained higher lint output under severe stress, suggesting more effective conversion of limited assimilates into boll weight. ST 4990 showed stronger yield sensitivity once canopy decline accelerated. DSSAT slightly underestimated peak yield under full irrigation but accurately predicted the slope of yield decline across irrigation treatments, capturing the stress threshold between 75% and 50%.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of observed and DSSAT-predicted cotton yields under four irrigation treatments (100%, 75%, 50%, and 0%) for cultivars NG 4190 and ST 4990 at Texas A\u0026amp;M AgriLife, Uvalde in 2025.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIrrigation treatments (%)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCotton cultivars\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eObserved yield (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePredicted yield (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRMSE (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNG 4190\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4410\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3996\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9.39\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eST 4990\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4567\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3896\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14.69\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e75\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNG 4190\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4094\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3942\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3.71\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eST 4990\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2913\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2941\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.96\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNG 4190\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3245\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eST 4990\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3069\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNG 4190\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1521\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eST 4990\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1249\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003e3.2. Sensitivity analysis\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eUnder full irrigation treatment, a longer CSDL delayed canopy senescence and sustained photosynthetic activity. Earlier EM-FL in NG 4190 advanced anthesis compared with ST 4990, extending the effective reproductive window. Longer PODUR and SFDUR prolonged boll development, while higher WTPSD increased individual pod weight. Together, these responses produced greater AGB and maintained stable LAI through maturity (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eUnder moderate irrigation treatment, reduced water availability shortened PODUR and SFDUR, but cultivar responses diverged. NG 4190 maintained a longer functional CSDL, which slowed leaf senescence and preserved canopy function. Earlier EM-FL allowed reproductive development to occur before peak water stress. In contrast, ST 4990 showed weaker buffering capacity, with earlier declines in LAI and stem growth. These results indicate that moderate irrigation mainly constrained reproductive duration rather than the timing of phenological events, with PODUR and SFDUR driving yield responses.\u003c/p\u003e \u003cp\u003eUnder half irrigation treatment, parameter sensitivity increased, particularly for EM-FL and PODUR. Delayed EM-FL limited early canopy development and reduced radiation capture before flowering. Shortened PODUR accelerated pod termination, while reduced SFDUR restricted assimilate supply to developing bolls. NG 4190 partially offset these effects through higher WTPSD, sustaining heavier pods despite lower pod numbers. LAI peaked earlier and declined more rapidly, and AGB accumulation slowed as both vegetative and reproductive sinks weakened.\u003c/p\u003e \u003cp\u003eUnder no irrigation treatment, responses were dominated by strong reductions in PODUR and SFDUR, leading to abrupt cessation of boll growth. Increased sensitivity of EM-FL delayed flowering and compressed the reproductive period. A marked reduction in CSDL caused rapid LAI decline and premature canopy collapse. Pod number fluctuated strongly over time, reflecting unstable reproductive retention. At this stress level, WTPSD became critical; NG 4190 maintained higher individual pod weights than ST 4990, although AGB plateaued early, indicating severely shortened growth duration.\u003c/p\u003e \u003cp\u003eSensitivity analysis confirmed that EM-FL controlled reproductive onset, CSDL determined canopy longevity, PODUR regulated pod survival, SFDUR governed boll filling duration, and WTPSD scaled final pod weight. NG 4190 showed consistently lower sensitivity to irrigation reduction, maintaining more stable pod number, LAI, and AGB across treatments, whereas ST 4990 exhibited sharper declines driven by these parameters (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003e3.3. Adaptation strategies\u003c/h2\u003e \u003cp\u003eEarly sowing in crop management adaptations advanced anthesis by 4\u0026ndash;6 days and physiological maturity by 5\u0026ndash;7 days, aligning flowering with periods of higher soil moisture. This adjustment extended PODUR and SFDUR, increasing pod retention by 7\u0026ndash;10% and AGB by 8\u0026ndash;12% in NG 4190, while ST 4990 showed smaller improvements (4\u0026ndash;6% in pod number, 5\u0026ndash;7% in AGB). Delayed sowing compressed the reproductive period, reduced LAI persistence, and accelerated leaf senescence, leading to 12\u0026ndash;18% yield penalties in both cultivars. Narrowing row spacing from 0.76 m to 0.66 m slightly improved light interception and increased boll weight by 5% in NG 4190, but ST 4990 showed minimal gains, reflecting limited compensation under denser planting. Adjusting harvest timing to match shifted maturity improved simulated lint yield by 3\u0026ndash;5% in early- and optimum-sown NG 4190, highlighting the importance of synchronizing management with crop phenology to reduce stress impacts.\u003c/p\u003e \u003cp\u003eStage-specific irrigation adaptations revealed key sensitive growth windows. Irrigation applied only during the vegetative stage increased early LAI and stem biomass but failed to sustain boll set, reducing final lint yield by 15\u0026ndash;20% compared with full-season irrigation. Flowering-stage irrigation effectively preserved PODUR and SFDUR, limiting pod abortion and maintaining 80\u0026ndash;85% of full-season yield in NG 4190, whereas ST 4990 achieved only 72\u0026ndash;76%. Grain-filling\u0026ndash;only irrigation improved WTPSD, increasing individual pod weight by 6\u0026ndash;9%, but could not compensate for pods lost earlier, yielding just 65\u0026ndash;70% of potential. Full-season irrigation maintained the highest LAI, AGB, and lint yield, serving as the benchmark for comparison.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. DISCUSSION","content":"\u003cdiv id=\"Sec34\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Growth across irrigation treatments\u003c/h2\u003e \u003cp\u003eLAI increased with increasing irrigation treatments, reaching a plateau under full water supply. This pattern aligns with DSSAT-CROPGRO simulations for cotton in Texas, where calibrated models matched with the measured LAI and AGB very closely across irrigation treatments (Adhikari et al., 2017). In other words, more soil moisture permitted a larger canopy and total biomass, as expected physiologically. Su et al. (2015) showed that maximum cotton LAI scaled with AGB via a saturating Michaelis\u0026ndash;Menten relation, mirroring the situation in our study. The difference in canopy between moderate and full irrigation was modest: Ahmad et al. (2021) found that LAI under 100% soil water was only slightly higher than under 50% irrigation (not statistically different). Our model predicted only a small LAI drop under moderate deficit. This suggests cotton maintains much of its leaf area under mild stress, consistent with reviews noting that irrigation strategy significantly affects LAI and plant height (Adhikari et al., 2017).\u003c/p\u003e \u003cp\u003eAGB tracked the trend of LAI. Well-irrigated scenarios produced the highest biomass, but further adding water provided little benefit. Chen et al. (2018) found that the highest irrigation treatment reduced cotton biomass at boll stage relative to a more moderate treatment. In our simulations, biomass gains plateaued and even declined under extreme watering. This agrees with Che\u0026rsquo;s warning that \u0026ldquo;unreasonable excessive irrigation\u0026hellip; may also cause poor soil aeration and nutrient leaching\u0026rdquo; (Che et al., 2021). Thus, our model predicts diminishing returns from overirrigation: once moisture is sufficient, extra water boosts neither AGB nor LAI substantially. This is consistent with field experience that cotton has an optimal water range, and excess irrigation can be wasteful or harmful.\u003c/p\u003e \u003cp\u003ePod (boll) number exhibited the strongest sensitivity to irrigation. Under severe water deficit, pod numbers collapsed, reflecting the known response that cotton aborts fruit under severe stress. In fact, water stress during reproductive stages \u0026ldquo;triggers hormonal changes\u0026hellip; resulting in the shedding of fruiting structures (squares and bolls)\u0026rdquo; (Datta et al., 2019). Our low irrigation runs showed extensive early boll shedding, just as cotton research predicts major loss of young cotton bolls under deficit (Cetin \u0026amp; Bilgel, 2002). Conversely, adequate irrigation preserved boll retention, yielding the highest pod numbers. Notably, beyond a sufficient water supply, the number of pods increased little and reached saturation. This matches agronomic principles: since lint yield is largely determined by boll count, irrigation\u0026rsquo;s key role is to prevent boll shedding early on (Kuai et al., 2015). Our results thus mirror literature: balanced irrigation through flowering is needed to set and mature pods, whereas deficit irrigation induces boll drop and lowers the yield.\u003c/p\u003e \u003cp\u003eLeaf dry weight responded in tandem with LAI, as expected. With more water, leaves expanded and total leaf mass grew, supporting higher LAI. Under deficit irrigation treatment, leaf growth slowed, reducing leaf weight. The research by Chen et al., (2022) highlights that irrigation methods significantly affect leaf area (and also leaf biomass), so our predicted leaf weight changes are not surprising. In fact, Adhikari's DSSAT calibration adjusted cultivar parameters influencing LAI and biomass to fit observations (Adhikari et al, 2017), implying that leaf growth is a key output. Our model\u0026rsquo;s leaf weight increases under higher irrigation are thus consistent with known crop responses: ample moisture drives vegetative growth.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec35\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Phenology under 75%, 50%, and 0% irrigation\u003c/h2\u003e \u003cp\u003eSignificant shifts in cotton phenology emerged under reduced irrigation. In our trials, plants under 75% and 50% irrigation reached anthesis sooner than in fully irrigated plots, with the 0% treatment flowering earliest of all. This stress-induced acceleration of flowering reflects a classical drought‐escape strategy \u0026ndash; the plants truncate their vegetative phase to set blooms before critical water deficits occur. Indeed, pre‐anthesis drought has been shown to shorten time to flowering in crops (Gao et al., 2020), and modern cotton breeding often links drought tolerance with inherently early maturation (Wang et al., 2016b). In contrast, abundant water prolonged the squaring‐to‐bloom interval: well‐watered plants continued vegetative growth longer and did not bloom until later. In other words, reduced moisture advanced the first white flower, whereas ample irrigation delayed it. This pattern is consistent with agronomic observations that drought conditions significantly compress the bloom period: stress shortened the seven‐ to eight‐week normal flowering window (Wang et al., 2023). In our study, the flowering period was markedly briefer in droughted plots, matching the literature data that cotton under stress \u0026ldquo;shortens the bloom period significantly\u0026rdquo; (Wang et al., 2016a).\u003c/p\u003e \u003cp\u003eThe hormonal milieu and carbohydrate status likely underpin these changes in timing. Under water deficit, developing flower buds appear to receive fewer photosynthates and altered hormonal signals, which can trigger the onset of earlier reproductive process. For example, Tarpley and Sassenrath (2006) observed that under ample water, cotton buds accumulate high sugar levels just before anthesis; under stress, this buildup is muted. Guinn et al. (1990) similarly reported that droughted buds show elevated abscisic acid (ABA) without the usual spike in indole-3‐acetic acid (IAA) seen in non‐stressed flowers. In our study, the severe 0% irrigation presumably raised bud ABA, effectively signaling the plant to complete flowering sooner. In contrast, the 75% irrigation plot showed only mild phenological shift. Thus, the anthesis advancement under high stress aligns with the notion that cotton can reallocate resources under drought to accelerate bloom at the cost of biomass.\u003c/p\u003e \u003cp\u003eOnce flowering began, boll set and development diverged strongly between treatments. Peak flowering is known to be the crop\u0026rsquo;s most drought-sensitive stage (Sun et al., 2021), and we found that severe moisture limitation during bloom caused dramatic fruit loss. In the 0% irrigation, many flowers aborted or failed to form mature bolls, even though the plants did open white blooms (cotton \u0026ldquo;white flowers\u0026rdquo; reportedly will expand under drought [Heitholt, 1999; Hoogenboom et al., 2019; Ahmad et al., 2025]). This outcome mirrors classic findings: Orgaz et al. (1992) argued that cotton\u0026rsquo;s peak flowering stage suffers the greatest harm from drought, and Hu et al. (2020) showed that water stress disrupts anther starch metabolism and pollen viability, leading to flower abortion. Consistent with these reports, most flowers in the no‐water treatment never set bolls. Moreover, as Heitholt (1999) observed, deficit in the first 10\u0026ndash;14 days after anthesis typically causes young bolls to shed, and we saw exactly that \u0026ndash; virtually all early‐formed bolls were aborted under 0% irrigation. By contrast, the fully irrigated plants retained nearly all fruiting sites, and the 75% plots had intermediate boll retention. The reproductive sink was \u0026ldquo;pruned\u0026rdquo; by drought: fewer flowers progressed to bolls, and existing bolls were small or shed. Higher rate of boll shedding under water scarcity has been well documented (Cordeiro et al., 2024), and our 0% treatment essentially exemplified that effect. These disruptions explain the sharp yield declines we observed: drought at flowering led to a truncated fruiting period and fewer seed‐bearing bolls, echoing Zonta et al. (2017) and Wang et al. (2016a) who reported severe yield loss from flower abortion under drought.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec36\" class=\"Section2\"\u003e \u003ch2\u003e4.3. LAI\u003c/h2\u003e \u003cp\u003eThe LAI obtained from the field (LI-COR 2200C) was compared to the model-predicted (CROPGRO) LAI, both of which showed the characteristic rise through vegetative growth, a mid‐season peak near the boll stage, and decline at maturity. In 2025, both cultivars under full irrigation attained maximum LAI around peak bloom, in agreement with Ding et al. (2024) who observed LAI rising until boll set and then dropping under water stress. The DSSAT CROPGRO predictions captured this seasonal pattern closely; model‐validation studies report very high agreement (RMSE\u0026thinsp;\u0026lt;\u0026thinsp;0.4) between predicted and observed cotton LAI under a range of irrigation treatments (Modala et al., 2015). However, discrepancies emerged during stress: predicted LAI sometimes lagged behind observed declines when soil water was low (a known model limitation under drought; see Modala et al., 2015). We also found that the predicted LAI tended higher than the measured LAI and often exceeded it during peak canopy cover (when multi‐layer foliage is hard to gauge with instruments).\u003c/p\u003e \u003cp\u003eOver time and between treatments, the two LAI time-courses showed a consistent trend. Both observed and predicted LAI were higher under full irrigation than under deficit. For example, plants under severe deficit had visibly thinner canopies, as also noted by Zhi et al. (2024). The two cultivars differed in canopy vigor: ST 4990 (strong early vigor; Saenz et al., 2023), typically reached its LAI peak slightly earlier than NG 4190, whereas NG 4190 (broad adaptation and high yield potential in both dryland and irrigated conditions; Edmisten \u0026amp; Collins, 2024), often maintained higher LAI later into the season under mild stress. These cultivar effects led to small divergences in observed versus simulated LAI: for example the model predicted slightly lower LAI for NG 4190 under deficit, and the measured LI‐COR LAI of NG 4190 under water deficit was a bit higher than that from model prediction, suggesting NG 4190\u0026rsquo;s morphology (smooth‐leaf, medium‐tall habit; Guedes et al. 2023), allowed greater light penetration and/or retained leaves longer than assumed by the model.\u003c/p\u003e \u003cp\u003eCultivar comparisons under water stress highlighted distinct adaptive strategies. ST 4990, known for early vigor (Chachar et al., 2025), built canopy rapidly, giving it a slight edge in early season LAI and light capture. However, this fast start may have come at the cost of sensitivity: under severe deficit, ST 4990\u0026rsquo;s LAI declined sharply after bloom, whereas NG 4190\u0026rsquo;s more moderate canopy persisted somewhat longer, possibly reflecting NG 4190\u0026rsquo;s reputed stability. This is consistent with field reports that ST 4990 handles early planting stress well, while NG 4190 fares reliably across irrigation treatments (BASF Agricultural Solutions, 2019; Americot, Inc., n. d.). The DSSAT model mirrored these differences in that simulated LAI for ST 4990 peaked sooner and dropped faster under low irrigation treatment, whereas NG 4190\u0026rsquo;s LAI curve was flatter. Thus, in dry treatment the gap in final AGB between cultivars narrowed (as NG 4190 was less penalized than ST 4990), aligning with the idea that drought-tolerant genotypes maintain canopy function under stress. The LAI trajectories suggested ST 4990 had higher radiation interception early on, but lost ground in late‐season photosynthesis, which translated to similar or lower boll set than NG 4190 under deficit. This is in agreement with studies showing genotype‐specific canopy and yield responses to water (Lin et al., 2024).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec37\" class=\"Section2\"\u003e \u003ch2\u003e4.4. AGB allocation\u003c/h2\u003e \u003cp\u003eCultivar identity exerted a strong control over biomass partitioning in our study, with the most striking differences observed in leaf and reproductive (boll) mass. For example, one cultivar consistently accumulated a larger leaf canopy and vegetative biomass, whereas the other channeled proportionally more dry matter into fruiting structures. This echoes earlier findings that genotypic traits strongly influence source\u0026ndash;sink relationships. Notably, early maturing cotton genotypes are known to shift assimilate allocation toward reproductive organs sooner in their life cycle (Sadras et al., 1997). Bange and Milroy (2004) reported that \u0026ldquo;partitioning to the fruit began earlier in early genotypes,\u0026rdquo; leading to higher harvest indices in those lines. In practical terms, if our higher-boll-weight cultivar was an early-season type, it likely initiated boll setting faster, investing less in late-season vegetative growth. Conversely, cultivars with more extensive leaf biomass are likely to sustain source capacity longer; Zafar et al. (2023) observed that one short-season cultivar (FH-207) maintained higher photosynthetic capacity than another (AA-703), explaining its superior yield under stress. Thus, genetic differences in photosynthetic traits and growth habit can explain why some cultivars allocate relatively more biomass to leaves and stems while others favor pods.\u003c/p\u003e \u003cp\u003eIntrinsic genetic background also contributed to total biomass differences. For instance, Mahboob et al. (2024) found that one upland cotton cultivar produced significantly more total aboveground biomass than another under identical conditions. This suggests inherent growth potential and nutrient-use efficiency differences. A cultivar that has more vegetative growth (taller stems, more nodes, greater leaf area) will naturally show higher stem and leaf mass even if its harvest index is lower. In contrast, a cultivar that \u0026ldquo;fills in\u0026rdquo; fewer vegetative nodes but loads more assimilate into bolls will have a higher boll/leaf ratio. Our observation that stem-weight varied less between cultivars than leaf or boll weight implies that the construction of stem mass may involve a relatively fixed structural cost, whereas leaves and bolls showed flexible sink-driven differences.\u003c/p\u003e \u003cp\u003eIrrigation treatment markedly altered these cultivar patterns. Across both cultivars, ample water (well-irrigated treatment) enhanced AGB. Chen et al. (2017) demonstrated this effect as well-watered cotton (W\u003csub\u003e80\u003c/sub\u003e) had 39% more AGB compared to water-limited plants, where the shoot ratio plummeted by 40\u0026ndash;73%. In our full-irrigation simulations, both cultivars produced lush canopies and heavier stems and leaves. Tang et al. (2010) reported that partial root-zone irrigation (PRI) reduced shoot growth, demonstrating the optimal partitioning response to water scarcity. Consistent with this, we observed under low-irrigation treatments that AGB was relatively curtailed. This shift aligns with the \u0026ldquo;functional equilibrium\u0026rdquo; theory, as documented in both field and controlled studies (Guo et al., 2024).\u003c/p\u003e \u003cp\u003eInterestingly, the irrigation effect on reproductive allocation was nuanced. Moderate water stress often forced cotton to curtail vegetative growth more than fruiting. In Tang et al.\u0026rsquo;s trials, while overall shoot biomass fell under PRI, both vegetative and reproductive shoot parts were reduced in roughly equal proportion, so the lint yield suffered much less than vegetative growth \u0026ndash; effectively raising reproductive efficiency (Tang et al., 2010). In our simulations, a mild-to-moderate deficit seemed to slightly boost the percentage of biomass in bolls for one cultivar. Zhi et al. (2024) similarly found that deficit irrigation increased the ratio of dry matter going to bolls during flowering. This may reflect physiological prioritization: when water is limited but not severe, cotton often maintains boll filling at some cost to stem or leaf growth. Conversely, severe stress tends to abort fruit, as seen when irrigation was curtailed late in the season in other studies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec38\" class=\"Section2\"\u003e \u003ch2\u003e4.5. Pod numbers and cotton weight\u003c/h2\u003e \u003cp\u003eIn drought-resilient cotton simulations, augmenting irrigation consistently increased the number of mature pods (bolls) per plant and individual boll weight, reinforcing the source\u0026ndash;sink relationship between water and yield (Bista et al., 2024; Ahmad et al., 2022). In our simulations, well-watered treatments produced significantly more bolls and higher cotton weight than deficit-irrigated plots. This parallels research finds that drought stress reduced lint yield by ~\u0026thinsp;25% primarily through a\u0026thinsp;~\u0026thinsp;19% reduction in boll number (Pettigrew, 2004), whereas irrigation increased yield (\u0026asymp;\u0026thinsp;30% boost in some years) and allowed cotton to sustain boll set later into the season (Bista et al., 2025). Similarly, Mahadevappa et al. (2018) reported that the number of bolls per plant increases under higher irrigation. Together, these results confirm that even drought-tolerant cultivars yield more cotton when water is ample, with boll count and size being the principal yield determinants (Sezener et al., 2015).\u003c/p\u003e \u003cp\u003eDrought-resilient cultivars often moderate but do not eliminate yield losses under limited irrigation, because even tolerant cultivars require sufficient water to fill each boll. In our study, the tolerant cultivar retained more pods under deficit irrigation than the susceptible one, but still experienced increased boll drop and smaller boll weight as water declined. This pattern agrees with comparative studies: for example, Niu et al. (2018) found that the yield of a drought-resistant cultivar (CCRI-45) was actually higher after a brief moderate drought (a \u0026ldquo;compensatory\u0026rdquo; response) as compared to the well-watered control. However, prolonged or severe stress eventually overwhelmed these advantages. Tariq et al. (2024) observed that water limitation caused an average 42% decline in seed-cotton yield across 32 cultivars (with a 55% drop in total biomass). Thus, our drought-resilient lines likely combined partial stress compensation (e.g. robust root growth or osmotic adjustment) with a measurable yield penalty under deficit irrigation \u0026ndash; consistent with global findings that genotypic differences modulate but do not negate irrigation effects.\u003c/p\u003e \u003cp\u003eBeyond boll counts, irrigation also influenced individual boll weight in our study, matching broader literature. Under full irrigation, cotton bolls contained more lint and seed (higher cotton weight per boll), reflecting a more sustained carbon supply. Pettigrew (2004) observed that irrigated plants produced heavier bolls at higher nodes and retained them longer, whereas droughted plants shed fruit early. Mechanistically, drought limits photosynthesis and carbohydrate transport, reducing assimilate availability to each boll, whereas adequate water maintains stomatal conductance and sucrose flux into developing fruit (Sezener et al., 2015). Consistent with this, Han et al. (2015) noted that cultivars sustaining higher dry matter under stress set more bolls and achieved higher seed yield, implying that vegetative vigor underpins boll filling. Accordingly, our irrigated drought-tolerant cultivars produced bolls of greater weight, while under deficit they produced smaller, fewer bolls \u0026ndash; the same qualitative trend reported worldwide.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec39\" class=\"Section2\"\u003e \u003ch2\u003e4.6. Yield decline\u003c/h2\u003e \u003cp\u003eOur field trial showed that NG 4190 yielded more lint than ST 4990 when water was non-limiting. For example, at Uvalde under 100% irrigation treatment, NG 4190\u0026rsquo;s lint yield exceeded ST 4990\u0026rsquo;s by a noticeable margin. However, as irrigation was reduced to 50% and 0%, both cultivars\u0026rsquo; yields fell sharply, and the relative drop was often larger for NG 4190. Overall, NG 4190\u0026rsquo;s yield advantage under full irrigation was on the order of 5\u0026ndash;10%, whereas under severe deficit (50% or 0%), ST 4990 maintained a lower fraction of its yield. This trend aligns with the study of Kumar et al. (2023) who reported that simulated seed cotton yield under well-watered conditions was roughly 3,418 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (control) but fell to ~\u0026thinsp;2,291 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e under extreme deficit (one irrigation) and ~\u0026thinsp;2,821 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e under moderate deficit (two irrigations). In their model, the most stressed treatment lost\u0026thinsp;~\u0026thinsp;32% of yield relative to control, versus a 17% loss in the milder stress treatment. Our field data showed the same qualitative pattern that deficits cut yields dramatically.\u003c/p\u003e \u003cp\u003eDSSAT model outputs indicate that maximum LAI and AGB were substantially higher in well-watered plots, and both fell as irrigation was curtailed (Kumar et al. 2023). In other words, under 100% irrigation, NG 4190 produced a lusher canopy and more AGB than ST 4990, which translated into greater boll number and size (Mishra et al., 2021). Mishra et al. (2021) observed that higher LAI leads to \u0026ldquo;improved boll number and better boll weight\u0026rdquo;, so the cultivar with the larger leaf area set more bolls. However, drought reverses this advantage: as water stress intensified, simulated and observed LAI and AGB declined. Kumar et al. (2023) reported a drop of 21\u0026ndash;38% in LAI under stress. Under limited water, NG 4190\u0026rsquo;s initially larger canopy thus shrunk relatively more, reducing its boll set and yield. By contrast, ST 4990 had a moderate LAI and photosynthetic ability, and therefore, lost leaves due to which its yield was affected under stress.\u003c/p\u003e \u003cp\u003eThe per-boll weight (\u0026ldquo;pod\u0026rdquo; weight) was relatively stable under stress. Lin et al. (2024) showed that individual boll weight (seed+fiber) stays nearly constant as water stress increases, whereas leaf area and photosynthesis drop dramatically. Thus, the decline in yield was driven mainly by reduced boll number. In our study, NG 4190\u0026rsquo;s higher LAI under full irrigation produced many bolls, while ST 4990 had fewer bolls but they were sustained. Under drought, NG 4190 dropped fewer of those bolls. This matches the modeling insight that reduced LAI/biomass leads to fewer reproductive sink organs. In the DSSAT output, biomass accumulation under severe stress was ~\u0026thinsp;35% lower than well‐watered (Ahmad et al., 2023; Kumar et al. 2023), and yield fell in parallel. Since boll weight is conserved, the remaining yield difference must come from fewer bolls. Mishra et al. (2021) noted that simulated LAI and biomass were closely tied to final yield, and. Kumar et al. (2023) reported a trend in their simulations: as stress increased, dry matter fell and yield declined in step.\u003c/p\u003e \u003cp\u003eLin et al. (2024) found that a mild deficit (e.g. irrigating at ~\u0026thinsp;90% of \u0026ldquo;normal\u0026rdquo;) actually saved water and sometimes maintained yield. This suggests that a slightly smaller canopy can be more efficient in some climates. If, for instance, ST 4990 had a lower maximum LAI by design, it might use water more efficiently under slight deficit, partly explaining why it \u0026ldquo;held up\u0026rdquo; better. Himanshu et al. (2021) tested the DSSAT model in Texas and reported that crop yields depend heavily on when and how much irrigation is applied. Keeping irrigation through the late bloom stage gives the highest yields, while stopping water supply early leads to sharp yield losses. If NG 4190 was more active late in the season, it would suffer more from an early cutoff. Conversely, ST 4990\u0026rsquo;s more conservative growth might align better with limited late-season water, as the model suggests that a strategic deficit schedule (like 90% \u0026rarr; 70% ET at different stages) can sustain yields (Lin et al., 2024).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec40\" class=\"Section2\"\u003e \u003ch2\u003e4.7. Cultivar-specific patterns\u003c/h2\u003e \u003cp\u003eUnder our imposed water deficit, the two cultivars diverged strongly: NG 4190 (the more drought-tolerant cultivar) sustained a much larger fraction of its leaf area and biomass than ST 4990, and its lint yield dropped only minimally. In contrast, ST 4990\u0026rsquo;s canopy senesced early and its aboveground biomass and yield fell sharply under the same stress. In our trials, NG 4190\u0026rsquo;s yield penalty was on the order of only a few tens of percent (with LAI decline near the low end of reported values), whereas ST 4990\u0026rsquo;s yield collapsed (LAI decline approached the high end). This pronounced genotype\u0026times;moisture interaction mirrors the general pattern reported by Yehia et al. (2024): cotton cultivars that perform well under irrigation often fail to do so under dryland conditions. In other words, the high inherent yield potential of ST 4990 under ample water gave it no advantage under drought, whereas NG 4190\u0026rsquo;s traits allowed it to maintain growth when water was scarce.\u003c/p\u003e \u003cp\u003eBoth our field data and the DSSAT model agree on these cultivar contrasts. The calibrated DSSAT\u0026ndash;CROPGRO simulations closely matched our measured yields and LAI changes (d-statistics\u0026thinsp;\u0026ge;\u0026thinsp;0.92 and MAPE\u0026thinsp;\u0026le;\u0026thinsp;6.5% [Bista et al., 2024]), giving confidence that the model captured the stress effects accurately. For example, Kumar et al. (2023) found that even moderate moisture stress reduced cotton LAI by roughly 22\u0026ndash;38% and lint yield by several hundred kilograms per hectare \u0026ndash; a range that brackets our observed responses. In our simulations, NG 4190 maintained higher simulated soil moisture and biomass under deficit irrigation than ST 4990, leading to a smaller simulated yield loss. This correspondence between observed and simulated cultivar differences implies that intrinsic cultivar parameters (e.g. rooting depth, phenology) were well-captured, and it reinforced NG 4190\u0026rsquo;s superior drought performance.\u003c/p\u003e \u003cp\u003eIn U.S. breeding trials, drought-tolerant genotypes (e.g. Tamcot CD3H and related lines) significantly out-yielded sensitive checks under rainfed conditions. For instance, Tamcot CD3H maintained about 290 kg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e lint without irrigation vs. only\u0026thinsp;~\u0026thinsp;196 kg ha\u003csup\u003e\u0026ndash;1\u003c/sup\u003e for Paymaster 303 under the same stress (Bumguardner, 2022). Tamcot Sphinx and other tolerant lines also retained more bolls and yield under mid-season drought in multi-year tests (Koudahe et al., 2024). By analogy, NG 4190 in our study played the role of the tolerant line while ST 4990 resembled a susceptible cultivar: under stress NG 4190 kept a larger boll load and biomass. Just as CD3H exhibited higher water use efficiency than Paymaster in non-stress trials (Bista et al., 2024), we infer that NG 4190 likely uses water more conservatively or extracts it more effectively than ST 4990.\u003c/p\u003e \u003cp\u003eIn Pakistan, Shani et al. (2025) found one upland cotton cultivar (FH-189) whose bloom-stage physiology conferred \u0026ldquo;strongest resilience\u0026rdquo; under severe drought, whereas another (FH-453) was highly sensitive. In China, Li et al. (2025) clustered 199 genotypes by drought-response indices and identified only a few \u0026ldquo;high drought resistance\u0026rdquo; lines \u0026ndash; notably UC072 and UC002 \u0026ndash; that combined high lint yield with limited irrigation. Both of these studies highlight that only a small subset of cultivars deliver stable yields under water deficit. Cotton as a species is already considered drought- and heat-tolerant \u0026ndash; it is widely grown as a rainfed crop in low-rainfall regions in Australia (Conaty et al., 2022) \u0026ndash; but such global evidence underscores that cultivar choice still makes a huge difference. Our NG 4190 effectively joined the ranks of those elite drought-resilient lines, outperforming ST 4990 much as FH-189 did over FH-453 in Pakistan or UC072 did over less-adapted lines in China.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. LIMITATIONS","content":"\u003cp\u003eData from only one growing season supported our calibration, model performance evaluation, and sensitivity analysis, which didn\u0026rsquo;t capture the effects of growth and yield of cotton caused by the year-to-year variability in temperature, rainfall timing, and evaporative demand. Parameter sensitivity and model performance may shift under wetter, cooler, or more extreme seasons, reducing confidence in long-term robustness and extrapolation across climatic conditions.\u003c/p\u003e \u003cp\u003eIrrigation treatments were imposed as fixed fractions of full supply (75%, 50%, and 0%) instead of dynamic. This simplification omits rainfall-irrigation interactions, stage-specific water application, and in-season deficit adjustments, which may misrepresent stress timing, reproductive sensitivity, and lint yield under practical irrigation management.\u003c/p\u003e \u003cp\u003eCultivar-specific genetic coefficients were calibrated using only the two tested cultivars, NG 4190 B3XF and ST 4990 B3XF, selected as optimum for the environmental conditions of southwest Texas. While these cultivars represent optimum performance for the region, the model\u0026rsquo;s applicability to other cotton cultivars with different growth habits, stress tolerance, or reproductive traits remains to be tested.\u003c/p\u003e \u003cp\u003eThe economic result is specific to southwest Texas conditions. The \u003cspan\u003e$\u003c/span\u003e40\u0026ndash;115 acre\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e net profit reflects local labor, irrigation, seed-cost, and climate. Variations in prices, inputs, or water supply may change outcomes elsewhere.\u003c/p\u003e"},{"header":"6. CONCLUSIONS","content":"\u003cp\u003eModel simulated phenology, LAI, biomass partitioning, and lint yield closely matched observed anthesis, boll formation, seed set, and maturity, with MAE and RMSE within 2\u0026ndash;15% of measured values, confirming model fidelity. Full irrigation (100%) maximized LAI, pod number, pod weight, and AGB. Moderate irrigation (75% ET) maintained stable functions and reproductive duration, preserving\u0026thinsp;~\u0026thinsp;90% of yield while reducing water consumption by 25%, demonstrating efficient water use. Deficit irrigation (50% ET) and no irrigation (0% ET) sharply reduced PODUR and SFDUR by 20\u0026ndash;35%, limiting boll filling, lowering total biomass by 18\u0026ndash;42%. NG 4190 exhibited superior resilience under stress, sustaining higher pod weight and slower leaf decline than ST 4990, underscoring cultivar-specific drought adaptation. Sensitivity analysis confirmed EM-FL, CSDL, PODUR, SFDUR, and WTPSD as critical determinants of yield and reproductive stability of cotton under varying water availability. Integrating field observations with CROPGRO-Cotton simulations enabled precise evaluation of irrigation strategies, reproductive timing, and cultivar performance. These results demonstrate that applying moderate irrigation (75%) to drought-resilient cultivars like NG 4190 sustains higher yields and maximizes water productivity.\u003c/p\u003e \u003cp\u003eCotton yield under water stress depends more on maintaining boll formation duration than AGB. Moderate irrigation (~\u0026thinsp;75%) protects reproductive sinks while limiting non-essential vegetative growth, maintaining yield with reduced water use. These findings reveal a practical irrigation threshold that balances water savings and yield, providing a transferable framework for optimizing irrigation timing and intensity.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eDECLARATION OF COMPETING INTEREST\u003c/h2\u003e \u003cp\u003eAuthors declare no conflict of interest.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFUNDING\u003c/h2\u003e \u003cp\u003eThis work was supported by Cotton Incorporated/Texas State Support Committee project 20-557TX, USDA-NIFA Hatch project 9574\u0026ndash;2, and Brazilian Federal Agency for Support and Evaluation of Graduate Education (CAPES/ PRAPG) Notice 14/2023. The work also received support from University of Palermo, Italy for Noemi Tortorici to visit the Uvalde Research Center. The funders did not play any role in the study design, data collection and analysis, or in the decision to prepare and publish the manuscript.\u003c/p\u003e\u003ch2\u003eCREDIT AUTHOR STATEMENT\u003c/h2\u003e \u003cp\u003e \u003cb\u003eUzair Ahmad\u003c/b\u003e: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, Writing - review \u0026amp; editing, Visualization. \u003cb\u003eXuejun Dong\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Conceptualization, Methodology, Investigation, Data curation, Validation, Supervision, Resources, Funding acquisition, Project administration. \u003cb\u003eThiago F. Duarte\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Methodology, Data curation, Investigation, Validation. \u003cb\u003eNoemi Tortorici\u003c/b\u003e: Writing \u0026ndash; review \u0026amp; editing, Data curation, Investigation. \u003cb\u003eDale Mott\u003c/b\u003e: Methodology, Validation, Writing \u0026ndash; review \u0026amp; editing, \u003cb\u003eBenjamin McKnight\u003c/b\u003e: Methodology, Validation, Writing \u0026ndash; review \u0026amp; editing.\u003c/p\u003e\u003ch2\u003eACKNOWLEDGMENTS\u003c/h2\u003e \u003cp\u003eWe appreciate Joe Gonzalez and Randy Cox for assistance in crop management and field preparation, and Christine Thompson and Liza Silva for administrative support. We thank Janet Patton for suggestions and comments.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAdhikari P, Ale S, Bordovsky JP, Thorp KR, Modala NR, Rajan N, Barnes EM (2016) Simulating future climate change impacts on seed cotton yield in the Texas High Plains using the CSM-CROPGRO-Cotton model. Agric Water Manage 164:317\u0026ndash;330\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdhikari P, Gowda PH, Marek GW, Brauer DK, Kisekka I, Northup B, Rocateli A (2017) Calibration and validation of CSM-CROPGRO‐Cotton model using lysimeter data in the Texas High Plains. J Contemp Water Res Educ 162(1):61\u0026ndash;78\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhmad, H. S., Imran, M., Ahmad, F., Rukh, S., Ikram, R. M., Rafique, H. M., \u0026hellip; Ali,S. (2021). Improving water use efficiency through reduced irrigation for sustainable cotton production. Sustainability, 13(7), 4044.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhmad U, Sohel F (2025) Evaluating decision support systems for precision irrigation and water use efficiency. Digit Eng 4:100038\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhmad U, Dong X, Leghari SJ, Hoogenboom G (2026) Modeling mustard water use and its effects on soil water content and subsequent maize performance under projected climate scenarios. Eur J Agron 174:127971\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmericot I (n. d.). NexGen\u0026reg; NG 4190 B3XF cotton technical sheet. Americot Inc \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.americot.com/wp-content/uploads/4190-B3XF-Tech-Sheet.pdf\u003c/span\u003e\u003cspan address=\"https://www.americot.com/wp-content/uploads/4190-B3XF-Tech-Sheet.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBange MP, Milroy SP (2004) Growth and dry matter partitioning of diverse cotton genotypes. Field Crops Res 87(1):73\u0026ndash;87\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBASF Agricultural Solutions (2019) New Stoneville cotton seed varieties show strong performance throughout Cotton Belt. BASF\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBednarz CW, Nichols RL (2005) Phenological and morphological components of cotton crop maturity. Crop Sci 45(4):1497\u0026ndash;1503\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBista MK, Adhikari B, Sankarapillai LV, Pieralisi B, Reddy KR, Jenkins J, Bheemanahalli R (2024) Drought and heat stress induce differential physiological and agronomic trait responses in cotton. Ind Crops Prod 222:119540\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBumguardner A (2022) \u003cem\u003eOptimizing fertilizer management and irrigation in cotton cropping systems\u003c/em\u003e [Doctoral dissertation, Texas Tech University]. Texas Tech University Institutional Repository. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://hdl.handle.net/2346/96494\u003c/span\u003e\u003cspan address=\"https://hdl.handle.net/2346/96494\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCammarano D, Payero J, Basso B, Wilkens P, Grace P (2012) Agronomic and economic evaluation of irrigation strategies on cotton lint yield in Australia. Crop Pasture Sci 63:647\u0026ndash;655\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCetin O, Bilgel L (2002) Effects of different irrigation methods on shedding and yield of cotton. Agric Water Manage 54(1):1\u0026ndash;15\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChachar S, Ahmed N, Hu X (2025) Drought-induced aesthetic decline and ecological impacts on ornamentals: Mechanisms of damage and innovative strategies for mitigation. Plant Biol. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/plb.70074\u003c/span\u003e\u003cspan address=\"10.1111/plb.70074\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChe Z, Wang J, Li J (2021) Effects of water quality, irrigation amount and nitrogen applied on soil salinity and cotton production under mulched drip irrigation in arid Northwest China. Agric Water Manage 247:106738\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, X., Qi, Z., Gui, D., Sima, M. W., Zeng, F., Li, L., \u0026hellip; Feng, S. (2022). Responses of cotton photosynthesis and growth to a new irrigation control method under deficit irrigation. Field Crops Research, 275, 108373.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, Z., Ma, H., Xia, J., Hou, F., Shi, X., Hao, X., \u0026hellip; Luo, H. (2017). Optimal pre-plant irrigation and fertilization can improve biomass accumulation by maintaining the root and leaf productive capacity of cotton crop. Scientific Reports, 7(1), 17168.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen Z, Tao X, Khan A, Tan DK, Luo H (2018) Biomass accumulation, photosynthetic traits and root development of cotton as affected by irrigation and nitrogen-fertilization. Front Plant Sci 9:173\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConaty, W. C., Broughton, K. J., Egan, L. M., Li, X., Li, Z., Liu, S., \u0026hellip; Stiller,W. N. (2022). Cotton breeding in Australia: Meeting the challenges of the 21st century.Frontiers in Plant Science, 13, 904131.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCordeiro CFS, Raphael JPA, Echer FR (2024) Boll distribution in the cotton canopy is affected by nitrogen management and cover crops in a sandy soil under water scarcity. Field Crops Res 306:109209\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDatta A, Ullah H, Ferdous Z, Santiago-Arenas R, Attia A (2019) Water management in cotton. In: Jabran K, Chauhan BS (eds) Cotton Production. John Wiley \u0026amp; Sons Ltd, pp 47\u0026ndash;59\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDing, Y., Ma, J., Zhang, J., Bai, Y., Cui, B., Hao, X., \u0026hellip; Ding, B. (2024). Response of photosynthesis, population physiological indexes, and yield of cotton in dry areas to the new technology of dry sowing and wet emergence. Frontiers in Plant Science, 15, 1487832.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdmisten K, Collins G (2024) Variety selection. In 2024 \u003cem\u003eCotton Information\u003c/em\u003e (pp. 25\u0026ndash;39). Cotton Extension Specialists\u0026mdash;Crop and Soil Sciences. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://content.ces.ncsu.edu/pdf/variety-selection/2024-01-23/5._ty_Selection.pdf\u003c/span\u003e\u003cspan address=\"https://content.ces.ncsu.edu/pdf/variety-selection/2024-01-23/5._ty_Selection.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, M., Snider, J. L., Bai, H., Hu, W., Wang, R., Meng, Y., \u0026hellip; Zhou, Z. (2020). Drought effects on cotton (Gossypium hirsutum L.) fibre quality and fibre sucrose metabolism during the flowering and boll-formation period. Journal of Agronomy and Crop Science, 206(3), 309\u0026ndash;321.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuedes, W. A., Nobre, R. G., Soares, L. A. D. A., de Lima, G. S., Gheyi, H. R., Fernandes,P. D., \u0026hellip; de Medeiros, J. F. (2023). Irrigation strategies with controlled water deficit in two production cycles of cotton. Plants, 12(16), 2892.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuerra LC, Garcia y Garcia A, Hook JE, Harrison KA, Thomas DL, Stooksbury DE, Hoogenboom G (2007) Irrigation water use estimates based on crop simulation models and kriging. Agric Water Manage 89(3):199\u0026ndash;207\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuinn G, Dunlap JR, Brummett DL (1990) Influence of water deficits on the abscisic acid and indole-3-acetic acid contents of cotton flower buds and flowers. Plant Physiol 93(3):1117\u0026ndash;1120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo, C., Bao, X., Sun, H., Zhu, L., Zhang, Y., Zhang, K., \u0026hellip; Li, C. (2024). Optimizing root system architecture to improve cotton drought tolerance and minimize yield loss during mild drought stress. Field Crops Research, 308, 109305.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan, Y., Luo, H., Zhu, Q. H., Yang, C., Zheng, R., Xue, F., \u0026hellip; Liu, F. (2025). Lower demand for boll dry matter accumulation and higher harvestable bolls ensure the yield and quality advantages of small boll cotton under water-deficit conditions. Journal of Agronomy and Crop Science, 211(2), e70035.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeitholt JJ (1999) Cotton: Factors associated with assimilation capacity, flower production, boll set, and yield. Crop Yield: Physiology and Processes. Springer, Berlin Heidelberg, pp 235\u0026ndash;269\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHerritt MT, Thompson A, Thorp K (2022) Irrigation management impacts on cotton reproductive development and boll distribution. Crop Sci 62(4):1559\u0026ndash;1572\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHimanshu, S. K., Ale, S., Bell, J., Fan, Y., Samanta, S., Bordovsky, J. P., \u0026hellip; Brauer,D. K. (2023). Evaluation of growth-stage-based variable deficit irrigation strategies for cotton production in the Texas High Plains. Agricultural Water Management, 280, 108222.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHimanshu SK, Fan Y, Ale S, Bordovsky J (2021) Simulated efficient growth-stage-based deficit irrigation strategies for maximizing cotton yield, crop water productivity and net returns. Agric Water Manage 250:106840\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoogenboom G, Porter CH, Boote KJ, Shelia V, Wilkens PW, Singh U, White JW, Asseng S, Lizaso JI, Moreno LP, Pavan W, Ogoshi R, Hunt LA, Tsuji GY, Jones JW (2019) The DSSAT crop modeling ecosystem. In: Boote KJ (ed) Advances in crop modeling for a sustainable agriculture. Burleigh Dodds Science Publishing, Cambridge, United Kingdom, pp 173\u0026ndash;216. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dx.doi.org/10.19103/AS.2019.0061.10\u003c/span\u003e\u003cspan address=\"10.19103/AS.2019.0061.10\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHoogenboom G, Porter CH, Shelia V, Boote KJ, Singh U, Pavan W, Oliveira FAA, Moreno-Cadena LP, Ferreira TB, White JW, Lizaso JI, Pequeno DNL, Kimball BA, Alderman PD, Thorp KR, Cuadra SV, Vianna MS, Villalobos FJ, Batchelor WD, Asseng S, Jones MR, Hopf A, Dias HB, Jintrawet A, Jaikla R, Memic E, Hunt LA, Jones JW (2024) Decision Support System for Agrotechnology Transfer (DSSAT). DSSAT Foundation, Gainesville, Florida, USA. Version 4.8.5 (www.DSSAT.net)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu W, Huang Y, Loka DA, Bai H, Liu Y, Wang S, Zhou Z (2020) Drought-induced disturbance of carbohydrate metabolism in anthers and male abortion of two \u003cem\u003eGossypium hirsutum\u003c/em\u003e cultivars differing in drought tolerance. Plant Cell Rep 39(2):195\u0026ndash;206\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJamshidi S, Murgia T, Morales-Ona AG, Cerioli T, Famoso AN, Cammarano D, Wang DR (2024) Modeling interactions of planting date and phenology in Louisiana rice under current and future climate conditions. Crop Sci 64:2274\u0026ndash;2287. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/csc2.21036\u003c/span\u003e\u003cspan address=\"10.1002/csc2.21036\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones JW, Hoogenboom G, Porter CH, Boote KJ, Batchelor WD, Hunt LA, Wilkens PW, Singh U, Gijsman AJ, Ritchie JT (2003) The DSSAT cropping system model. Eur J Agron 18:235\u0026ndash;265. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S1161-0301(02)00107-7\u003c/span\u003e\u003cspan address=\"10.1016/S1161-0301(02)00107-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKo J, Piccinni G, Guo W, Steglich E (2009) Parameterization of EPIC crop model for simulation of cotton growth in South Texas. J Agricultural Sci 147(2):169\u0026ndash;178\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoudahe K, Aguilar J, Djaman K, Sheshukov AY (2024) Evapotranspiration, fiber yield and quality, and water productivity of cotton (\u003cem\u003eGossypium hirsutum\u003c/em\u003e L.) under different irrigation technologies in a semiarid climate. Irrig Sci 42(3):575\u0026ndash;594\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuai J, Zhou Z, Wang Y, Meng Y, Chen B, Zhao W (2015) The effects of short-term waterlogging on the lint yield and yield components of cotton with respect to boll position. Eur J Agron 67:61\u0026ndash;74\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar, R., Mishra, S. K., Singh, K., Al-Ashkar, I., Iqbal, M. A., Muzamil, M. N.,\u0026hellip; El Sabagh, A. (2023). Impact analysis of moisture stress on growth and yield of cotton using DSSAT-CROPGRO-cotton model under semi-arid climate. PeerJ, 11, e16329.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, J., Liu, R., Liu, S., Zhigang, G., Ma, H., Shen, Q., \u0026hellip; Pang, C. (2025). Prioritising D-value in multivariate evaluation identifies high‐yield and drought‐resistant cotton genotypes. Journal of Agronomy and Crop Science, 211(5), e70113.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLin, M., Wang, L., Lv, G., Gao, C., Zhao, Y., Li, X., \u0026hellip; Sun, W. (2024). Deficit irrigation effects on cotton growth cycle and preliminary optimization of irrigation strategies in arid environment. Plants, 13(10), 1403.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLoka DA, Oosterhuis DM, Ritchie GL (2011) Water-deficit stress in cotton. In D. M. Oosterhuis (Ed.), \u003cem\u003eStress physiology in cotton\u003c/em\u003e (Cotton Foundation Reference Book Series No. 7). The Cotton Foundation. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cotton.org/foundation/upload/Stress-Physiology-in-Cotton.pdf\u003c/span\u003e\u003cspan address=\"https://www.cotton.org/foundation/upload/Stress-Physiology-in-Cotton.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahadevappa SG, Sreenivas G, Reddy DR, Madhavi A, Rao SS (2018) Effect of different levels of irrigation and nitrogen on growth and yield of Bt cotton. Int J Curr Microbiol Appl Sci 7(8):4599\u0026ndash;4604\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMahboob W, Sarwar N, Abdul Hafeez B, Arif O, Akhtar MAR, M., Yang G (2024) Exploring genotypic variations in cotton associated with growth and nitrogen use efficiency. J Plant Nutr 47(19):3231\u0026ndash;3250\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMishra AK, \u0026Ouml;zger M, Singh VP (2011) Association between uncertainties in meteorological variables and water-resources planning for the state of Texas. J Hydrol Eng 16(12):984\u0026ndash;999\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMishra SK, Kaur V, Singh K (2021) Evaluation of DSSAT-CROPGRO‐cotton model to simulate phenology, growth, and seed cotton yield in northwestern India. Agron J 113(5):3975\u0026ndash;3990\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eModala, N. R., Ale, S., Rajan, N., Munster, C. L., DeLaune, P. B., Thorp, K. R., \u0026hellip;Barnes, E. M. (2015). Evaluation of the CSM-CROPGRO-Cotton model for the Texas rolling plains region and simulation of deficit irrigation strategies for increasing water use efficiency. Transactions of the ASABE, 58(3), 685\u0026ndash;696.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorgan G, McGinty J, Mott D (2024) Replicated Agronomic Cotton Evaluation (RACE) South, East and Central Regions of Texas. Texas A\u0026amp;M AgriLife Extension Service, CS-2025-01\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNiu, J., Zhang, S., Liu, S., Ma, H., Chen, J., Shen, Q., \u0026hellip; Zhao, X. (2018). The compensation effects of physiology and yield in cotton after drought stress. Journal of Plant Physiology, 224, 30\u0026ndash;48.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOrgaz F, Mateos L, Fereres E (1992) Agroclimatology and modeling: Season length and cultivar determine the optimum evapotranspiration deficit in cotton. Agron J 84:700\u0026ndash;706\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePettigrew WT (2004) Moisture deficit effects on cotton lint yield, yield components, and boll distribution. Agron J 96(2):377\u0026ndash;383\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiccinni G, Ko J, Wentz A, Leskovar D, Marek T, Howell T (2007) Determination of crop coefficients (Kc) for irrigation management of crops. \u003cem\u003e28th Annual International Irrigation Show\u003c/em\u003e, San Diego, CA, December, 9\u0026ndash;12\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSadras VO, Bange MP, Milroy SP (1997) Reproductive allocation of cotton in response to plant and environmental factors. Ann Botany 80(1):75\u0026ndash;81\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaenz V, Sekula D, McGinty J, Ramirez J, Livingston C, Alaniz R (2023) Hidalgo County RACE Trial, 2023 [Data table]. Texas A\u0026amp;M AgriLife Extension. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://southtexas.tamu.edu/files/2023/10/2023-Hidalgo-Co.-Cotton-Variety-Trials.pdf\u003c/span\u003e\u003cspan address=\"https://southtexas.tamu.edu/files/2023/10/2023-Hidalgo-Co.-Cotton-Variety-Trials.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeidel SJ, Palosuo T, Thorburn P, Wallach D (2018) Towards improved calibration of crop models\u0026ndash;Where are we now and where should we go? Eur J Agron 94:25\u0026ndash;35\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSezener V, Basal H, Peynırcıoglu C, Gurbuz T, Kızılkaya K (2015) Screening of cotton cultivars for drought tolerance under field conditions. Turkish J Field Crops 20(2):223\u0026ndash;232\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShani MY, Ditta A, Khan MKR (2025) Characterizing drought-resilient cotton genotypes through morpho-physiological and biochemical traits at flowering stage. BMC Plant Biol 25(1):1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSingh Y, Rao SS, Regar PL (2010) Deficit irrigation and nitrogen effects on seed cotton yield, water productivity and yield response factor in shallow soils of semi-arid environment. Agric Water Manage 97(7):965\u0026ndash;970\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu L, Wang Q, Wang C, Shan Y (2015) Simulation models of leaf area index and yield for cotton grown with different soil conditioners. PLoS ONE 10(11):e0141835\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun F, Chen Q, Chen Q, Jiang M, Gao W, Qu Y (2021) Screening of key drought tolerance indices for cotton at the flowering and boll setting stage using the dimension reduction method. Front Plant Sci 12:619926\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang LS, Li Y, Zhang J (2010) Biomass allocation and yield formation of cotton under partial rootzone irrigation in arid zone. Plant Soil 337(1):413\u0026ndash;423\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTariq, M. S., Imran, M., Ud Din, S., Murtaza, B., Naeem, M. A., Amjad, M., \u0026hellip; AbdElgawad,H. (2024). Magnetic nanocomposite of maize offal biomass for effective sequestration of Congo red and methyl orange dyes from contaminated water: Modeling, kinetics and reusability. International Journal of Phytoremediation, 26(6), 975\u0026ndash;992.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhmad U, Sharma L (2023) A review of best management practices for potato crop using precision agricultural technologies. Smart Agricultural Technol 4:100220\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhmad U, Alvino A, Marino S (2022) Solar fertigation: A sustainable and smart IoT-based irrigation and fertilization system for efficient water and nutrient management. Agronomy 12(5):1012\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTarpley L, Sassenrath GF (2006) Carbohydrate profiles during cotton floral bud (square) development. J Agron Crop Sci 192(5):363\u0026ndash;372\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUSDA (1976) Soil survey of Uvalde County, Texas. USDA Soil Conservation Service, Washington, DC\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang L, Lin M, Han Z, Han L, He L, Sun W (2023) Simulating the effects of drought stress timing and the amount irrigation on cotton yield using the CSM-CROPGRO-cotton model. Agronomy 14(1):14\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang R, Gao M, Ji S, Wang S, Meng Y, Zhou Z (2016a) Carbon allocation, osmotic adjustment, antioxidant capacity and growth in cotton under long-term soil drought during flowering and boll-forming period. Plant Physiol Biochem 107:137\u0026ndash;146\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang R, Ji S, Zhang P, Meng Y, Wang Y, Chen B, Zhou Z (2016b) Drought effects on cotton yield and fiber quality on different fruiting branches. Crop Sci 56(3):1265\u0026ndash;1276\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWillmott CJ, Matsuura K (2005) Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance. Climate Res 30(1):79\u0026ndash;82\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYehia WMB, Zaazaa EEDI, El-Hashash EF, El-Enin A, M. M., Shaaban A (2024) Genotype-by-environment interaction analysis for cotton seed yield using various biometrical methods under irrigation levels in a semi-arid region. Arch Agron Soil Sci 70(1):1\u0026ndash;23\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZafar, M. M., Chattha, W. S., Khan, A. I., Zafar, S., Subhan, M., Saleem, H., \u0026hellip; Xuefei,J. (2023). Drought and heat stress on cotton genotypes suggested agro-physiological and biochemical features for climate resilience. Frontiers in Plant Science, 14, 1265700.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhi, X., George-Jaeggli, B., Han, Y., Chen, Q., Zhang, S., Feng, L., \u0026hellip; Yang, B. (2024).Comprehensive analysis on investigating water-saving potentials of irrigated cotton in semi-arid area in China. Agricultural Water Management, 301, 108960.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZonta JH, Brandao ZN, Rodrigues JIDS, Sofiatti V (2017) Cotton response to water deficits at different growth stages. Revista Caatinga 30(4):980\u0026ndash;990\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Texas A\u0026M University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"dssat, cotton, irrigation, GCM, climate, yield","lastPublishedDoi":"10.21203/rs.3.rs-9500734/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9500734/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn semi-arid regions, water scarcity limit root-zone water availability, restricting cotton growth and yield. Although prior studies evaluated drought responses and cultivar variations, interactions between irrigation and cultivar-specific dynamics\u0026mdash;validated through process-based models\u0026mdash;remain underexplored in southwest Texas. This study addresses this gap by quantifying the effects of four irrigation treatments on two cultivars, NG 4190 B3XF and ST 4990 B3XF, using field observations and DSSAT (Decision Support System for Agro-technology Transfer) CROPGRO-Cotton model. Full (100%) and moderate (75%) irrigations were based on field data, whereas half (50%) and no (0%) treatments were simulated. Key variables include phenology, pod numbers (m\u003csup\u003e-2\u003c/sup\u003e), pod, stem and leaf weights (all kg ha\u003csup\u003e-1\u003c/sup\u003e), AGB (aboveground biomass, kg ha\u003csup\u003e-1\u003c/sup\u003e), height (m), LAI (leaf area index), and lint yield (kg ha\u003csup\u003e-1\u003c/sup\u003e). Results showed that phenology varied, with differences of 5 days in anthesis, 4 in boll formation, 2 in seed set, and 5 in maturity. Lint yield varied across treatments, with MAE (mean absolute error) values of 11% and 13% for full and moderate treatments, respectively. Full irrigation maximized growth, with NG 4190 showing higher pod number (MAE 5%), pod weight (MAE 10%), and AGB (MAE 13%) than ST 4990. Half irrigation reduced pod numbers, pod weight, and AGB in both cultivars. No irrigation caused large declines in both cultivars, with MAE of 16\u0026ndash;19% across pod number, pod weight, and AGB. Sensitivity analysis showed strong irrigation effects on boll development and AGB. Moderate irrigation sustains\u0026thinsp;~\u0026thinsp;90% yield by preserving boll development and optimizing cotton production.\u003c/p\u003e","manuscriptTitle":"Predicting Cotton Growth, Pod Development, and Lint Yield Responses to Irrigation and Cultivar Differences in Southwest Texas","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-27 08:20:07","doi":"10.21203/rs.3.rs-9500734/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4368e45a-8419-445c-bda6-e6c22d598c69","owner":[],"postedDate":"April 27th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":66846128,"name":"Agronomy"}],"tags":[],"updatedAt":"2026-04-27T08:20:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-27 08:20:07","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9500734","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9500734","identity":"rs-9500734","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}