Application of the AquaCrop model for cotton production under water scarce arid conditions

Cotton is an important fiber crop, yet traditional cultivation practices are heavily water intensive. As limited water resources increasingly constrain cotton productivity, it is important to investigate alternative irrigation strategies that can maintain stable yields under severe water scarcity conditions. Tools like the AquaCrop model play a key role in this process, allowing to simulate intricate relationships among management techniques, water use, crop development, and yield production. In this study, the model underwent calibration and validation based on three irrigation regimes (1-2-1; 1-3-1; 1-4-1) in three field capacity (F_C) datasets (F_C 65-65-60%, F_C 70-70-65%, and F_C 75-75-70%) and its effectiveness was assessed through simulations of canopy cover (CC), biomass accumulation, and water productivity (WP). The results showed that the highest cotton yield was obtained under F_C 70-70-65% irrigation regime (3.553 Mg ha⁻¹ for 2022; 3.325 Mg ha⁻¹ for 2023; 3.441 Mg ha⁻¹ for 2024), while the simulated and measured cotton yields were consistent with a deviation < 5%. WP values were also higher with the F_C 70-70-65% irrigation regime, exhibiting 0.66 kg m³ in 2024; 0.64 kg m³ in 2023 and 0.64 kg m³ in 2022. The deviations were in the range of <2.77 to 4.89% for cotton yield, <1.6 to 4.89% for WP, <3.48 to 5.01% for biomass, <1.17 to 5.17% for HI. Soil moisture content between measured and simulated values were observed with a coefficient of determination (R²) of 0.814, 0.799 and 0.604 for Fc 65-65-60%, Fc 70-70-65% and Fc 75-7570%, respectively. Under the optimal mode (70-70-65% of Fc) of irrigation schedule, WP increased by 13.15 to 24.13% compared other irrigation regimes. It was concluded that the digitalization of agricultural irrigation using the AquaCrop model will advance water-saving priorities, contribute to climate resilience and promote more sustainable crop production practices under the challenging water scarcity conditions of arid ecosystems.

Introduction

Agricultural production in Uzbekistan is heavily dependent on irrigation, producing over 95% of the total agricultural output, particularly through traditional furrow irrigation systems (Nurbekov et al., 2025). However, this reliance is increasingly challenged by the anticipated climate crisis and rapid population growth. As a result, water scarcity intensifies risks to agricultural output and complicates efforts to meet rising food demand sustainably. Furthermore, soil salinity and drought have become major concerns for crop producers, leading to a serious challenge to the efficient functioning of agricultural systems (Khaitov et al., 2025).

Agriculture consumes approximately 92% of the total water withdrawals in Uzbekistan, drawing primarily from the Amudarya and Syrdarya rivers. However, water levels in these rivers have significantly declined in recent years due to the impacts of climate change including reduced glacial melt and prolonged droughts (Abdullaev and Rakhmatullaev, 2025). Furthermore, the overall irrigation efficiency in the region remains critically low at just 26% due to inefficient and unsustainable use of water resources (Abdullaev et al., 2025). This inefficiency leads to negative consequences, contributing to the expansion of saline and abandoned lands. According to the Ministry of Agriculture of Uzbekistan (2024), saline irrigated lands cover approximately 2,270,700 hectares, constituting more than 50% of total irrigated lands (Sultonov et al., 2025). Therefore, optimizing water use in agriculture has become a national priority, while this goal can be advanced through the adoption of innovative digital irrigation methods and regenerative crop technologies.

Uzbekistan is one of the main cotton producers in the world with more than 3.6 million ton raw cotton produced yearly. The S-6524 cotton variety belonging to the Gossipium hirsutum family exhibits strong adaptive traits to drought and salt environment. Cotton production in this arid continental climate is accompanied by an acute shortage of irrigation water, especially in last years. Therefore, enhancing WP plays a crucial role in wisely managing irrigation practices within agricultural systems, particularly in regions reliant on furrow irrigation. Given the scarcity of water resources, implementing water-saving strategies is essential and this requires accurate assessment of crop-water interactions (Allanov et al., 2020; Wu et al., 2024).

Future climatic conditions in this region are likely to become warmer and drier, which could significantly impact crop evapotranspiration rates and increase irrigation demands (Hamidov et al., 2022). To anticipate how future climate-induced changes in water availability may affect crop yields, water production functions can be integrated with AquaCrop—crop simulation models developed by Food and Agriculture Organization (FAO) (Hsiao et al., 2023). Developed from empirical observations of crop yield responses to water availability, the model has been refined to include stabilized water productivity and has proven effective across a range of agricultural scenarios (Boulange et al., 2025).

AquaCrop serves as an effective tool for evaluating crop responses to varying water availability and optimizing WP. Previous experiments have consistently shown that this model is capable of simulating crop performance under rainfed, deficit, and full irrigation conditions in cotton-growing regions, providing estimates of attainable yields (Mumtaz et al., 2021; Zhang et al., 2022). Utilizing a water-driven algorithm, the model translates transpiration losses into biomass production based on crop-specific parameters (Steduto et al., 2009). Simulations conducted with AquaCrop using the default conservative settings for cotton yielded the most accurate results (Tsakmakis et al., 2019). Also, this model has demonstrated higher irrigation water productivity (Babel et al., 2019), reliably simulating aboveground biomass and canopy development under both optimal and limited irrigation conditions (Jin et al., 2014).

In light of the aforementioned findings, this system represents a critical step toward integrating digital technologies into local agricultural practices, aligning with broader goals of climate resilience and resource efficiency (Wu et al., 2024). Therefore, this study aimed to evaluate the reliability of AquaCrop estimates in optimizing water use and improving cotton yields under semiarid conditions of Uzbekistan.

Materials and methods

Climate and soil conditions

The field experiment was conducted at the Experiment Station of the Cotton Growing Research Institute (41^°22’N, 60^°54′E; 572.2 m above sea level) at Tashkent, Uzbekistan, during the 2022–2024 vegetation seasons. Categorized by distinct agroecological zones, this area is predominantly characterized by aridity and a continental climate. Weather temperatures fluctuate rapidly despite this region recognized as an ideal location for cotton cultivation (Figure 1). Summers are hot, with July means of +23–28 °C and highs up to +43 °C. In 2023, a total of 17 days recorded maximum air temperatures exceeding 40 °C in 2023, with the highest reaching 43.1 °C in July. In 2024, this site experienced only 5 days with maximum air temperatures above 40 °C. Winters are unstable, with January means of −0.9 to −2.5 °C and lows reaching −25 °C; warm spells occur, but snow cover is brief. Spring arrives early, with fruit flowering in March, though frosts are possible through early spring. Annual precipitation is low and irregular, averaging about 200 mm (occasionally reaching 250–300 mm).

Figure 1

Figure 1. Climate data for 2022–2024 years.

The soil is an old irrigated typical sierozem soil (Calcic Xerosols), silt loam; and the water table is more than 15-m deep. The dominant soil texture at the experimental site is classified as silt loam, according to the United States Department of Agriculture (USDA) soil taxonomy. Over the 1.2-meter soil profile, the average texture composition was 8.2 ± 1.2% sand, 75.4 ± 0.7% silt, and 16.5 ± 0.7% clay. Bulk density measurements for the 0–30 cm and 30–50 cm layers were nearly identical, recorded at 1.29 and 1.31 g cm⁻³, respectively.

Key hydraulic properties including saturated soil water content (θS), field capacity (Fc), permanent wilting point (PWP), and infiltration rate (INF) were determined in situ using field measurements averaged over three replicates (Table 1). Notably, the measured parameters closely align with those estimated using the Saxton et al. (1986) pedotransfer functions, which are commonly applied in AquaCrop modeling (Table 1).

Table 1

Table 1. Soil physical characteristics.

The reported properties were measured directly in the field (upper rows) or estimated using the Saxton et al. (1986) pedotransfer method (lower rows). Values represent the mean and standard deviation calculated from three replicate samples.

As a starting point for investigations of irrigation scheduling, we adopted the Fc index, which was 0.298 m³ m⁻³ in this soil. Irrigations were scheduled when soil moisture in the root zone was depleted by the crop to specific fractions of Fc (for instance, irrigation at 70% of Fc) for each of the three main plant growth periods defined later. Soil moisture was determined using a neutron moisture meter “Hydroprobe-503 DR” according to the development periods of cotton, namely 0–70 cm before budding, 0–100 cm during flowering and fruiting, and 0–70 cm during ripening.

Experiment design

In this study, the cotton variety C-6524 belonging to Gossipium hirsutum L. species was cultivated during the 2022 and 2024 growing seasons, spanning April to October. C-6524 is a medium-fiber cotton cultivar, introduced in 1988, developed to improve upon the slow maturation characteristics of its predecessors (Khaitov et al., 2021).

The experiment used three replicates of three irrigation scheduling treatments, each applied to surface (conventional) irrigation. Each treatment consisted of scheduled irrigations exhibiting specific percentages of Fc (e.g., 65-65-60%) during plant growth periods, these numbers present F_C values from germination to pod forming, flowering-fruiting, and maturation stages of cotton, respectively (Table 2).

Table 2

Table 2. Irrigation regimes.

Each replicated plot was 72 m² (4.8 m by 15 m). Irrigation water volume applied through furrow irrigation was measured by an in-line propeller type flow meter (BCKM-5/20). The BCKM-5/20 turbine flowmeter used in this study has a manufacturer-reported accuracy of ±2.5% under standard operating conditions. The flowmeter was calibrated at the beginning of each growing season using volumetric tank testing and verified against manual bucket-and-stopwatch measurements. Calibration drift was monitored monthly, and no significant deviation (>1%) was observed during the study period. Fertilizer was applied at rates of 200 kg ha⁻¹ N, 140 kg ha⁻¹ P, and 100 kg ha⁻¹ K. Before plowing in autumn, 50% of P and 70% of K fertilizers were broadcasted, the rest was applied along with N during two to three true leaf, budding, and flowering stages.

Field observations

The performance of crop models is typically assessed based on their ability to accurately simulate CC dynamics, biomass accumulation, and final yield. In this study, cotton CC and above-ground biomass were monitored at regular intervals, approximately every 10 days. CC was determined by capturing overhead images from a height of 2.0–2.5 meters directly above the crop. These images were promptly analyzed using the Canopeo smartphone application (Patrignani and Ochsner, 2015), which calculates fractional CC based on color indices, including red-to-green, blue-to-green ratios, and the excess green index.

Above-ground biomass was quantified by harvesting cotton stems clipped at ground level within a 0.25 m² sampling area. The samples were oven-dried at 65 °C for 48 h and then weighed. Both CC and biomass measurements were conducted in triplicate, with results reported as mean values and standard deviations across four replicates.

Cotton yield was evaluated within a 10 m² area designated in each of the three treatment plots. These areas were carefully selected to minimize edge effects and ensure representative sampling.

AquaCrop builds upon the foundational yield response to water model proposed by Doorenbos et al. (1979), enhancing it with a daily time-step and a process-oriented approach to simulate crop growth. Conceptually, the root zone functions like a reservoir, reacting to incoming water sources such as rainfall, irrigation, and capillary rise and outgoing losses through runoff, evaporation, transpiration, and deep percolation (Vanuytrecht et al., 2014).

In this model, crop growth is tracked daily through canopy expansion and biomass production, both of which directly affect yield. Canopy development is quantified as CC, indicating the proportion of soil surface shaded by the crop. Initially, CC is zero before emergence and gradually increases, reaching a peak value (CCx) under optimal conditions. As the crop enters senescence later in the season, CC begins to decline.

The simulation of canopy dynamics relies on first-order kinetic equations and requires four key parameters: the initial canopy cover after emergence (CCo), the maximum canopy cover (CCx), the canopy growth coefficient (CGC), the canopy decline coefficient (CDC), and the onset of senescence (Tsen). Notably, CGC is temperature-dependent, meaning that canopy expansion is influenced by thermal conditions. Equations (1a) and (1b) govern canopy growth, while Equation (1c) describes its decline post-senescence.

$\begin{array}{ll}\mathit{CC}=\mathit{CCo}\times e^{t\times \mathit{CGC}}\text{for}\mathit{CC}\le \frac{\mathit{CCx}}{2}& (1)\end{array}$

$\begin{array}{ll}\mathit{CC}=\mathit{CCx}–0.25\times \frac{\mathit{CCx}}{\mathit{CCo}}{e}^{--t\times \mathit{CGC}}\text{for}\mathit{CC}>\frac{\mathit{CCx}}{2}& (1)\end{array}$

$\begin{array}{ll}\mathit{CC}=\mathit{CCx}[1–0.05(e\frac{3.33\times \mathit{CDC}}{\mathit{CCx}+2.29}\times t–1\left)\right]\text{if}t\ge \mathit{\text{Tsen}}& (1)\end{array}$

where t is the number of days after seeding and all other parameters were previously defined.

Crop transpiration (Tr, mm) largely depends on CC. As shown in Equation (2), Tr is a function of ETo, the evaporative power of the atmosphere typically calculated using the Penman-Monteith method, a crop coefficient (KcTr), which is proportional to CC, and by considering the influence of both water stress (Ks) and cold stress (KsTr):

$\begin{array}{ll}\mathit{Tr}=(\mathit{Ks}\times \mathit{\text{KsTr}})\mathit{\text{KcTr}}\times \mathit{ETo}& (2)\end{array}$

Evapotranspiration, ETo (mm), is a forcing input provided to the AquaCrop model, Ks is a stress coefficient representing both water logging and water scarcity. In turn, Tr is used in conjunction with the normalized water productivity coefficient (WP*) to determine biomass production [Equation (3)].

$\begin{array}{ll}B=\mathit{Ksb}•\mathit{WP}\ast \Sigma \left(\mathit{Tr}\mathit{ETo}\right)& (3)\end{array}$

where WP* is adjusted by Ksb, a low temperature stress coefficient (Vanuytrecht et al., 2014). Water stress does not affect WP*, but directly impacts biomass accumulation since its effect is captured via Tr, as described in Equation (2).

Water productivity was determined with Equation (4) Fernandez (2023):

$\begin{array}{ll}\text{Water}\mathrm{WP}(\mathrm{kg}/{\mathrm{m}}^3)=\text{Yield}(\mathrm{kg}/\mathrm{ha})/\text{water used}({\mathrm{m}}^3/\mathrm{ha})& (4)\end{array}$

The crop yield (Y) was determined as the product of accumulated biomass (Bacc) and a harvest index (HI). The value of HI is the product of a reference harvest index (HIo) with a factor accounting for stress effects (fHI). Y is calculated with Equation (5):

$\begin{array}{ll}Y=\mathit{fHI}•\mathit{HIo}•\mathit{\text{Bacc}}& (5)\end{array}$

Crop water use (ET) was calculated from the soil water balance, where P is precipitation, I is irrigation, R is the sum of runoff and run-on, F is flux across the lower boundary of the soil profile (control volume), and ΔS is change in the soil water stored in the profile as shown in Equation (6):

$\begin{array}{ll}\mathit{ET}+\mathit{\Delta S}+R–P–I–F=0& (6)\end{array}$

where the sign conventions are as given in Evett (2002).

Re-arranging this equation gives the crop water use or ET as shown in Equation (7):

$\begin{array}{ll}\mathit{ET}=-\mathit{\Delta S}+P+I–R+F& (7)\end{array}$

Soil profile water content was determined twice weekly with the soil moisture neutron probe (SMNP) (Campbell Pacific Nuclear International, model Hydroprobe-503DR1.5), which was calibrated using methods described in Evett and Steiner (1995).

WP was computed using measured ETa derived from the water balance method. Simulated ETa values were used solely for model evaluation purposes. Calibration equations were established for the important soil layers (Evett, 2002). Since model parameterization depends on the specific site, the relevance of key calibrated parameters should be reassessed when applied to different environmental or management conditions.

Calibration and validation

Maximum Canopy Cover (CCx = 0.86): Measured and calibrated based on field image analysis using the Canopeo app (Table 3). This value aligns with literature-reported CCx ranges for Gossypium hirsutum under optimal conditions (Tsakmakis et al., 2019; Jin et al., 2014).

Table 3

Table 3. Cotton input parameters (calibrated and measured) used in the AquaCrop model.

Reference Harvest Index (HI₀ = 0.40): Adopted from AquaCrop default settings and validated against field measurements. Comparable values are reported in Ran et al. (2017) and Hussein et al. (2011).

Maximum Rooting Depth (Rmax = 0.98 m): Measured in situ using soil coring and root density profiling. This value is consistent with previous studies on cotton root architecture in arid soils (Khaitov et al., 2021; Vanuytrecht et al., 2014).

Statistical analysis

Statistical analysis using ANOVA (CropStat 2015 software) was conducted to assess the impact of three irrigation regimes (Fc 65-65-60; Fc 70-70-65; Fc 75-75-70) on cotton yield, biomass accumulation, and water productivity. Differences among treatment means were evaluated using the least significant difference (LSD) test at a 5% significance level (p = 0.05), based on data from three experimental replicates.

Results and discussion

Cotton yield and water productivity parameters with AquaCrop simulations

The calibrated model was validated using data sets from the 2022 and 2023 growing seasons. The results across all three experimental seasons showed minimal variation, indicating consistency in the observed data and robustness of the model. The validation runs with the calibrated AquaCrop showed good results for the simulated CC as indicated by deviation values in a range of <2.77 to 4.89% for cotton yield, <1.6 to 4.89% for WP, <3.48 to 5.01% for biomass, and <1.17 to 5.17% for HI (Tables 4, 5).

Table 4

Table 4. Measured versus simulated parameters for cotton yield and water productivity under three irrigation management practices.

Table 5

Table 5. Measured versus simulated values for biomass (Mg ha⁻¹) and harvest index (HI) under three irrigation management practices.

Seed cotton yields under varying irrigation regimes were also simulated with high precision, as indicated by deviation indices below 5% (Table 4). Cotton yield was higher in the 2022 season than those in 2023 and 2024. The largest error was around 5% underprediction of seed cotton yield in the Fc 65-65-60 were detected in the 2024 growing season. The year effect was significant for most parameters, likely because weather variability impacted on crop development at a different rate. The highest cotton yield was recorded under the Fc 70-70-65% regime, exceeding the Fc 75-75-70% regime by 6.46% in 2022, 8.33% in 2023 and 7.03% in 2024, while yield differences for these irrigation regimes were statistically significant (p ≤ 0.05) across seasons. Notably, the difference was substantially higher between the irrigation schedules of 70-70-65% of Fc and 65-65-60% of Fc, showing increases of 19.94% in 2022, 17.75% in 2023 and 20.6% in 2024. This is likely due to the water stress experienced by plants under the Fc 65-65-60% irrigation regime. On the contrary, abundant moisture availability under the Fc 75-75-70 regime resulted in a slight decline in cotton yield compared to the Fc 70-70-65 regime.

WP followed a similar trend with increases ranging from 0.01 to 0.13 kg m³ across the respective irrigation modes. The Fc 70-70-65% irrigation regime was highly productive over Fc 65-65-60%, enhancing WP by 14.51% in 2022, 13.15% in 2023, and 14.04% in 2024. The advantages of applying the Fc 70-70-65% regime over the Fc 75-75-70% regime were even higher, presenting an increase by 19.89% in 2022, 15.97% in 2023 and 24.13% in 2024. WP was generally higher for 2022 than 2023 and 2024 seasons, likely attributable to favorable temperature during summer months (Table 4). During the 2023 growing season, simulated cotton yields consistently exceeded field-measured values, probably due to high temperature stress that adversely affected the generative development phase of the crop. The application of the Fc 70-70-65% irrigation schedule resulted in a notable improvement in cotton yield, while simultaneously demonstrating strong potential for water savings in this semi-arid environment.

The AquaCrop model demonstrated strong performance in simulating cotton biomass across all three experimental seasons, with measured and simulated values showing close agreement (Table 5). Harvest index (HI), calculated as the ratio of seed cotton yield to biomass, followed a typical developmental pattern: a slow increase from flowering (lag phase), followed by a linear rise until physiological maturity. Simulations of HI were reasonably accurate, with deviations not exceeding 5% (Table 5). Across all treatments, the values of 0.30 to 0.39 were found for HI described the impact of the applied irrigation regimes to cotton yield rate relative biomass accumulation. In this irrigation experiment, HI serves as an indicator of water stress during the yield formation stage. The simulated versus observed values for HI showed satisfactory agreement with a deviation less than 5% (Table 5). The highest HI values were found under the Fc 70-70-65 irrigation regime (0.38 for 2022 and 0.39 for 2023 and 2024 seasons), followed by the Fc 65-65-60 irrigation regime (0.36 for 2022 and 0.38 for 2023 and 0.37 for 2024). Whereas, the lowest values were observed under the Fc 75-75-70 irrigation regime with values 0.31 for 2022, 0.32 for 2023, and 0.30 for 2024.

Previous studies (Masasi et al., 2020) have noted that the AquaCrop model tends to overestimate WP under severe water stress conditions, primarily due to its reliance on a constant normalized WP value. Nevertheless, AquaCrop has demonstrated strong performance when compared to more complex models (Raes, 2023). Unlike other models that require extensive calibration and numerous input parameters (Ahmadi et al., 2015), AquaCrop demands minimal adjustment and focuses on a limited set of key variables.

In the present study, WP values initially increased as actual evapotranspiration (ETa) declined, peaked at moderate water stress levels, and then decreased again under more severe deficits. The highest WP was consistently observed under the Fc 70-70-65% irrigation treatment across all three growing seasons. While AquaCrop generally overestimated WP across treatments, this trend did not apply to the Fc 70-70-65% regime (Table 4). Despite these tendencies, the model accurately simulated seed cotton yield and WP. Therefore, in contexts with limited input data and for practical management applications, the use of a simplified model like AquaCrop remains highly reliable.

The reference harvest index (HIo) determined in this study was between 0.30 and 0.39, consistent with findings by Ran et al. (2017), although other sources have reported a value of 0.35. Adjustments to HI under water stress conditions depend on both the timing and severity of the stress during the growing season. Despite its drought-tolerant characteristics, cotton remains vulnerable to water stress during the reproductive stage, which may have contributed to the observed decline in yield and HI.

According to previous findings (Aziz et al., 2022), HI may be improved under pre-anthesis water stress, potentially due to a relative reduction in biomass accumulation before flowering. Although this effect is incorporated into the model, no increase in HI was observed in this study under the Fc 65-65-60 treatment, as the water stress initiated prior to the flowering stage (Datta et al., 2019). Recent research applying the AquaCrop model to simulate seed cotton yield and HI reported strong correlation with WP, highlighting the model’s effectiveness in evaluating WP under varying irrigation regimes (Hussein et al., 2011).

The accuracy of the simulated parameters slightly declined at the higher irrigation regime (Fc 75-75-70) in most cases, with the highest error values observed during the hot summer period. WP was overestimated in the Fc 75-75-70 treatment throughout the study period due to inflated yield predictions and underestimated evapotranspiration rates, resulting in simulated WP values that exceeded those observed in the field. These results are consistent with previous studies (Karimov et al., 2025; Wu et al., 2022), highlighting the heavy reliance on irrigation and water deficiency are the primary constraints on crop productivity in the region (Nurbekov et al., 2024).

These monitored results suggest that the AquaCrop model serves as a reliable and efficient tool for assessing crop WP, supporting yield potential, and reducing water loss, all of which contribute to enhanced climate resilience in water-scarce environments.

AquaCrop simulations on canopy cover, evapotranspiration and soil water content

Differences in vegetative components such as CC, evapotranspiration, soil water content (SWC) were statistically significant by irrigation mode. Temporal patterns of SWC under different irrigation treatments in 2023 and 2024 followed the same trend observed in 2022, confirming the model’s ability to replicate drying cycles and irrigation events. Although some deviations were noted in the absolute SWC values (Figure 2), these discrepancies had minimal impact on the estimation of actual evapotranspiration (ETa) across treatments.

Figure 2

Figure 2. Soil moisture content values (measured and simulated) under three irrigation regimes.

The model successfully captured temporal variations in SWC within the 0.60 m soil profile, with a maximum deviation of approximately 5% across all irrigation treatments (Figure 2). This indicates a satisfactory estimation of ETa within the soil water balance component of the model. Under deficit irrigation conditions, a tendency to overpredict SWC was observed; however, this had only a minor impact on ETa simulations, which were slightly underpredicted. Similar findings were reported by Sajid et al. (2024) where detailed analysis of the soil profile showed consistent overestimation of SWC in the surface layer and underestimation in deeper layers. The model did not simulate deep percolation under any irrigation treatment during the 2022 season, and the results aligned well with measured SWC values throughout the soil profile in the fully irrigated plots.

Furthermore, the simulated versus measured results were well correlated, as shown in Figure 2, good fit is illustrated by slopes very close to one and small intercepts with high R² values (R² = 0.8147 for Fc 65-65-60%; R² = 0.7996 for Fc 70-70-65%, R² = 0.6043 for Fc 75-75-70%).

The regression lines showed slopes close to unity and minimal intercepts regarding ETa, with high coefficients of determination: R² = 0.93 for Fc 65-65-60%, R² = 0.92 for Fc 70-70-65%, and R² = 0.93 for Fc 75-75-70%, indicating a reliable model fit across all irrigation treatments (Figure 3).

Figure 3

Figure 3. Evapotranspiration values (measured and simulated) under three irrigation regimes.

Accurate simulation of CC is essential for the effective performance of AquaCrop, as it directly influences transpiration rates and, in turn, biomass accumulation. The AquaCrop model effectively simulated CC development during the 2022 growing season across various irrigation treatments, as shown in Figure 4. The low experimental error (<5%) calculated for all treatments indicates strong agreement between observed and simulated data. This is further supported by robust linear regression results between measured and simulated CC values: R² = 0.87 for Fc 65-65-60%, R² = 0.79 for Fc 70-70-65%, and R² = 0.85 for Fc 75-75-70%.

Figure 4

Figure 4. Canopy cover values (measured and simulated) under three irrigation regimes.

Simulations of CC showed high accuracy, with a maximum deviation of 5.4% across irrigation levels (Figure 4). The model fit was excellent, with R² = 0.9097 (n = 3), a slope near unity, and a small intercept. The largest deviation (5.6%) was recorded under the Fc 75-75-70% treatment, yet final aboveground biomass was simulated with high precision across all treatments.

As anticipated, climate variability characterized by rising temperatures and increased aridity is likely to intensify evaporation rates, thereby raising the risk of secondary salinization which could further compromise crop productivity (Wu et al., 2023). In these uncertain situations, this tool harnesses digital technologies may play a great role in strengthening agricultural water management and crop output.

It is well known that water is a key driver of many essential ecosystem functions, particularly in arid environments. Projected climate change impacts are likely to aggravate the current water scarcity issues facing Uzbekistan. As this study showed, the AquaCrop model developed by FAO offers a robust framework for simulating crop responses to varying irrigation strategies under limited water availability. By integrating climate data, soil characteristics, and crop parameters, AquaCrop enables precise decision-making that supports both yield optimization and environmental stewardship. Results from the AquaCrop simulation experiments suggest that these findings contribute to a better understanding of crop water management in harsh arid environments and highlight opportunities for more sustainable irrigation practices. Digital transformation in agriculture is increasingly recognized as a critical pathway toward sustainable resource management, particularly in water-scarce regions (Elsadek, et al., 2025). This approach not only enhances irrigation efficiency but also strengthens agricultural resilience and promotes long-term food security in regions vulnerable to climate stress.

Conclusion

In this study, the application of the AquaCrop model has contributed to a better understanding of crop water management by identifying strategies to minimize water loss, improve WP, and predict cotton yield. Cotton cultivated under the Fc 70-70-65% irrigation regime with the AquaCrop model had greater effect on yield and WP components, likely due to improved SWC, better nutrient distribution, and reduced soil erosion. This treatment resulted in higher cotton with increases ranging from 6.46 to 20.6% and WP improvements from 13.15 to 24.13% compared to other tested irrigation regimes (Fc 65-65-60% and Fc 75-75-70%). The identified optimal irrigation regime (Fc 70-70-65%) with the model effectively supported water use optimization and contributed sustainable cotton production.

The AquaCrop model proved to be a reliable and accurate tool for promoting water-saving measures, shedding light on the digitalization of crop production during periods of climate uncertainty.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

OS: Software, Writing – original draft, Conceptualization, Resources, Writing – review & editing. SabI: Software, Data curation, Investigation, Methodology, Writing – original draft, Supervision, Writing – review & editing. KK: Conceptualization, Writing – review & editing, Formal analysis, Writing – original draft, Data curation. PB: Writing – original draft, Writing – review & editing, Formal analysis, Data curation, Methodology. SR: Writing – review & editing, Validation, Formal analysis, Writing – original draft, Project administration. SS: Data curation, Conceptualization, Writing – review & editing, Formal analysis, Visualization, Writing – original draft. SZ: Supervision, Methodology, Writing – review & editing, Writing – original draft, Data curation. AK: Formal analysis, Writing – original draft, Visualization, Conceptualization, Writing – review & editing, Validation. FA: Writing – review & editing, Resources, Formal analysis, Writing – original draft, Methodology, Validation. SaiI: Writing – review & editing, Writing – original draft, Visualization, Formal analysis, Validation. UN: Data curation, Methodology, Supervision, Writing – original draft, Writing – review & editing. BKhali_ Supervision, Conceptualization, Software, Writing – review & editing, Writing – original draft, Methodology, Data curation. IT: Writing – review & editing, Supervision, Writing – original draft, Data curation, Methodology. BKhaki: Conceptualization, Project administration, Methodology, Writing – review & editing, Validation, Supervision, Investigation, Writing – original draft, Visualization, Software, Formal analysis, Resources, Funding acquisition. OK: Writing – original draft, Writing – review & editing, Methodology, Validation, Software. BKhai: Writing – original draft, Writing – review & editing, Methodology, Validation, Conceptualization.

Funding

The author(s) declare that no financial support was received for the research and/or publication of this article.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The authors declare that no Gen AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abdullaev, I., Assubayeva, A., Bobojonov, I., Djanibekov, N., Dombrowsky, I., Gafurov, A., et al. (2025). Current challenges in central Asian water governance and their implications for research, higher education, and science-policy interaction. Cent. Asian J. Water Res. 11, 47–58. doi: 10.29258/CAJWR/2025-R1.v11-1/47-58.eng

Crossref Full Text | Google Scholar

Abdullaev, I., and Rakhmatullaev, S. (2025). “Water security in Central Asia: towards sustainable development and resilience” in Sustainable water management and irrigation systems: Tackling climate change (Cham: Springer Nature Switzerland), 47–70.

Google Scholar

Ahmadi, S. H., Mosallaeepour, E., Kamgar-Haghighi, A. A., and Sepaskhah, A. R. (2015). Modeling maize yield and soil water content with AquaCrop under full and deficit irrigation managements. Water Resour. Manag. 29, 2837–2853. doi: 10.1007/s11269-015-0973-3

Crossref Full Text | Google Scholar

Allanov, K., Shamsiev, A., Durdiev, N., Avliyakulov, M., Karimov, A., and Khaitov, B. (2020). Improving nutrition and water use efficiencies of pima cotton (Gossypium barbadense L.) varieties under arid conditions of Uzbekistan. J. Plant Nutr. 43, 2590–2600. doi: 10.1080/01904167.2020.1793186

Crossref Full Text | Google Scholar

Aziz, M., Rizvi, S. A., Sultan, M., Bazmi, M. S., Shamshiri, R. R., Ibrahim, S. M., et al. (2022). Simulating cotton growth and productivity using AquaCrop model under deficit irrigation in a semi-arid climate. Agri 12, 1234–1245. doi: 10.3390/agriculture12020242

Crossref Full Text | Google Scholar

Babel, M. S., Deb, P., and Soni, P. (2019). Performance evaluation of AquaCrop and DSSAT-CERES for maize under different irrigation and manure application rates in the Himalayan region of India. Agric. Res. 8, 207–217. doi: 10.1007/s40003-018-0366-y

Crossref Full Text | Google Scholar

Boulange, J., Nizamov, S., Nurbekov, A., Ziyatov, M., Kamilov, B., Nizamov, S., et al. (2025). Calibration and validation of the AquaCrop model for simulating cotton growth under a semi-arid climate in Uzbekistan. Agric. Water Manag. 310:109360. doi: 10.1016/j.agwat.2025.109360

Crossref Full Text | Google Scholar

Datta, A., Ullah, H., Ferdous, Z., Santiago-Arenas, R., and Attia, A. (2019). “Water management in cotton” in Cotton production, ed. B. S. Chauhan. 47–59.

Google Scholar

Doorenbos, J., Kassam, A. H., and Bentvelsen, C. I. M. (1979). Yield response to water. Rome: Food and Agriculture Organization of the United Nations.

Google Scholar

Elsadek, E. A., Attalah, S., Waller, P., Norton, R., Hunsaker, D. J., Williams, C., et al. (2025). Simulating water use and yield for full and deficit flood-irrigated cotton in Arizona, USA. Agronomy 15:2023. doi: 10.3390/agronomy15092023

Crossref Full Text | Google Scholar

Evett, S. R. (2002). “Water and energy balances at soil-plant-atmosphere interfaces” in The soil physics companion. ed. A. A. Warrick (Boca Raton, FL: CRC Press LLC), 127–188.

Google Scholar

Evett, S. R., and Steiner, J. L. (1995). Precision of neutron scattering and capacitance type moisture gages based on field calibration. Soil Sci. Soc. Am. J. 59, 961–968.

Google Scholar

Fernandez, E. (2023). Editorial note on terms for crop evapotranspiration, water use efficiency and water productivity. Agric. Water Manag. 289:108548. doi: 10.1016/j.agwat.2023.108548

Crossref Full Text | Google Scholar

Hamidov, A., Kasymov, U., Djumaboev, K., and Paul, C. (2022). Rebound effects in irrigated agriculture in Uzbekistan: a stakeholder-based assessment. Sustainability 14:8375. doi: 10.3390/su14148375

Crossref Full Text | Google Scholar

Hsiao, T., Fereres, E., Steduto, P., and Raes, D. (2023). “Chapter 4: calibration guidance” in AquaCrop version 7.1 reference manual (Rome: FAO).

Google Scholar

Hussein, F., Janat, M., and Yakoub, A. (2011). Simulating cotton yield response to deficit irrigation with the FAO aquacrop model. Span. J. Agric. Res. 9, 1319–1330. doi: 10.5424/sjar/20110904-358-10

Crossref Full Text | Google Scholar

Jin, X., Feng, H., Zhu, X., Li, Z., Song, S., and Song, X. (2014). Assessment of the AquaCrop model for use in simulation of irrigated winter wheat canopy cover, biomass, and grain yield in the North China plain. PLoS One 9, 1–11. doi: 10.1371/journal.pone.0086938

PubMed Abstract | Crossref Full Text | Google Scholar

Karimov, A., Khalikov, B., Isaev, S., Sindarov, O., Khayitov, K., Avliyakulov, M., et al. (2025). Efficiency of drip irrigation in amaranth production using the HYDRUS-1D model. Front. Sustain. Food Syst. 9:1612679. doi: 10.3389/fsufs.2025.1612679

Crossref Full Text | Google Scholar

Khaitov, B., Rajabov, N., Murtazayeva, G., Durdiev, N., Norqulov, U., Abdalova, G., et al. (2025). The effectiveness of bio-treatment on licorice (Glycyrrhiza glabra) productivity and soil restoration in saline ecosystems. Eurasian J. Soil Sci. 14, 198–207. doi: 10.18393/ejss.1657415

Crossref Full Text | Google Scholar

Khaitov, B., Teshaev, S., Karimov, A., Mukhammadiev, A., Teshaev, F., Yuldasheva, R., et al. (2021). Agronomic performance of cotton (Gossypium hirsutum L.) under the influence of UVA radiation. Res. Crops 22, 986–993. doi: 10.31830/2348-7542.2021.160

Crossref Full Text | Google Scholar

Masasi, B., Taghvaeian, S., Gowda, P. H., Marek, G., and Boman, R. (2020). Validation and application of AquaCrop for irrigated cotton in the southern Great Plains of US. Irrig. Sci. 38, 593–607. doi: 10.1007/s00271-020-00665-4

Crossref Full Text | Google Scholar

Mumtaz, H., Farhan, M., Amjad, M., Riaz, F., Kazim, A. H., Sultan, M., et al. (2021). Biomass waste utilization for adsorbent preparation in CO2 capture and sustainable environment applications. Sustain Energy Technol Assess 46:101288. doi: 10.1016/j.seta.2021.101288

Crossref Full Text | Google Scholar

Nurbekov, A., Kosimov, M., Islamov, S., Khaitov, B., Qodirova, D., Yuldasheva, Z., et al. (2024). No-till, crop residue management and winter wheat-based crop rotation strategies under rainfed environment. Front. Agron. 6:1453976. doi: 10.3389/fagro.2024.1453976

Crossref Full Text | Google Scholar

Nurbekov, A., Umarov, S., Azizov, S., Umarov, M., Khaitov, B., Khalilova, L., et al. (2025). Arid irrigated winter wheat and soybean cropping under conservation tillage systems. Front. Sustain. Food Syst. 9:1512277. doi: 10.3389/fsufs.2025.1512277

Crossref Full Text | Google Scholar

Patrignani, A., and Ochsner, T. E. (2015). Canopeo: a powerful new tool for measuring fractional green canopy cover. Agron. J. 107, 2312–2320. doi: 10.2134/agronj15.0150

Crossref Full Text | Google Scholar

Raes, D. (2023). AquaCrop training handbook I. Understanding AquaCrop. Rome: Food and Agriculture Organization of the United Nations.

Google Scholar

Ran, H., Kang, S., Li, F., Tong, L., Ding, R., Du, T., et al. (2017). Performance of AquaCrop and SIMDualKc models in evapotranspiration partitioning on full and deficit irrigated maize for seed production under plastic film-mulch in an arid region of China. Agric. Syst. 151, 20–32. doi: 10.1016/j.agsy.2016.11.001

Crossref Full Text | Google Scholar

Sajid, I., Tischbein, B., Borgemeister, C., García-Vila, M., Bakhsh, A., and Florke, M. (2024). Assessing cotton irrigation scheduling strategies under rotational delivery schedules in Pakistan. Ecohydrol. Hydrobiol. 24, 710–724. doi: 10.1016/j.ecohyd.2024.04.005

Crossref Full Text | Google Scholar

Saxton, K. E., Rawls, W. J., Romberger, J. S., and Papendick, R. I. (1986). Estimating generalized soil-water characteristics from texture. Soil Sci. Soc. Am. J. 50, 1031–1036. doi: 10.2136/sssaj1986.03615995005000040039x

Crossref Full Text | Google Scholar

Steduto, P., Hsiao, T. C., Raes, D., and Fereres, E. (2009). Aquacrop—the FAO crop model to simulate yield response to water: i. concepts and underlying principles. Agron. J. 101, 426–437. doi: 10.2134/agronj2008.0139s

Crossref Full Text | Google Scholar

Sultonov, K., Kholmurodova, G., Eshonkulov, J., Allanov, K., Sindarov, O., Khayitov, K., et al. (2025). Effect of integrative NPK soil and foliar nutrition on winter wheat (Triticum aestivum L.) productivity in irrigated arid lands. Front. Sustain. Food Syst. 8:1508913. doi: 10.3389/fsufs.2024.1508913

Crossref Full Text | Google Scholar

Tsakmakis, I. D., Kokkos, N. P., Gikas, G. D., Pisinaras, V., Hatzigiannakis, E., Arampatzis, G., et al. (2019). Evaluation of aquacrop model simulations of cotton growth under deficit irrigation with an emphasis on root growth and water extraction patterns. Agric. Water Manag. 213, 419–432. doi: 10.1016/j.agwat.2018.10.029

Crossref Full Text | Google Scholar

Vanuytrecht, E., Raes, D., Steduto, P., Hsiao, T. C., Fereres, E., Heng, L. K., et al. (2014). Aquacrop: FAO’S crop water productivity and yield response model. Environ. Model. Softw. 62, 351–360. doi: 10.1016/j.envsoft.2014.08.005

Crossref Full Text | Google Scholar

Wu, F., Guo, S., Huang, W., Han, Y., Wang, Z., Feng, L., et al. (2023). Adaptation of cotton production to climate change by sowing date optimization and precision resource management. Ind. Crop. Prod. 203:117167. doi: 10.1016/j.indcrop.2023.117167

Crossref Full Text | Google Scholar

Wu, F., Guo, S., Huang, W., Han, Y., Wang, Z., Feng, L., et al. (2024). Soil water movement may regulate soil water consumption and improve cotton yields under different cotton cropping systems. Ind. Crop. Prod. 211:118278. doi: 10.1016/j.indcrop.2024.118278

Crossref Full Text | Google Scholar

Wu, F., Qiu, Y., Huang, W., Guo, S., Han, Y., Wang, G., et al. (2022). Water and heat resource utilization of cotton under different cropping patterns and their effects on crop biomass and yield formation. Agric. For. Meteorol. 323:109091. doi: 10.1016/j.agrformet.2022.109091

Crossref Full Text | Google Scholar

Zhang, J., Li, K., Gao, Y., Feng, D., Zheng, C., Cao, C., et al. (2022). Evaluation of saline water irrigation on cotton growth and yield using the AquaCrop crop simulation model. Agric. Water Manag. 261:107355. doi: 10.1016/j.agwat.2021.107355

Crossref Full Text | Google Scholar