Optimization of Yield and Fiber Yield of Cotton Cultivars Under Water Regimes in the Tropical Dry Season

Abstract

This study pioneers the integration of the water sensitivity coefficient (Ky) with cotton yield performance under varying water regimes in the Brazilian Cerrado. The objective was to evaluate the productive performance and fiber yield of cotton cultivars under different water regimes during the tropical dry season. The experiment followed a randomized block design in a 5 × 4 factorial scheme with four replications, totaling 80 plots. Treatments consisted of five irrigation levels based on crop evapotranspiration (25%, 50%, 75%, 100% and 125% of ETc) and four cultivars (TMG44B2RF, FM944GL, IMA5801B2RF and IMA709B2RF). Increasing water supply enhanced cotton lint yield, reaching 3209.4 kg ha⁻¹ at the highest regime. Water regimes between 25 and 125% of the ETc significantly improved yield components, leading to an increase of up to 221% in lint yield. Fiber quality remained stable across irrigation levels and was mainly genotype-dependent. Among the cultivars, FM944GL showed high productivity and fiber yield, while IMA5801B2RF demonstrated greater water resilience (Ky = 0.73), making it suitable for water-limited environments. The findings reflect the specific conditions of the evaluated growing season. Thus, long-term studies under diverse environmental conditions are recommended to confirm these trends and enhance understanding of cotton responses to water regimes in the Cerrado.

1. Introduction

Cotton (Gossypium hirsutum L.) has strategic importance in tropical agriculture, especially in the Brazilian Cerrado, which is characterized by seasonal rainfall. However, one of the greatest physiological limitations for the crop is its dependence on water during dry periods, when water stress becomes more intense, decisively affecting plant growth, development, productivity and fiber quality [1].

Over the years, Brazil, particularly in the Central-West, North, and Northeast regions, has faced extreme weather events such as drought, which has proven to be a major challenge for agricultural production, and cotton is no exception. This difficulty is exacerbated when severe droughts or dry spells coincide with the phenological phases of the crop. Cotton has a water demand that varies between 500 and 1500 mm of precipitation, which needs to be well distributed throughout the production cycle. However, under arid or dry conditions, or during dry spells, especially in critical phases such as flowering and boll formation, the lack of rain and low soil moisture can lead to floral abscission, resulting in a decrease in the number of bolls per plant [2].

Given this scenario, the intensification of agricultural activity in water-restricted environments, such as the Cerrado, reinforces the urgent need to adopt irrigation regimes during the dry season that promote the efficient use of water without compromising cotton productivity [3,4,5].

Although water regimes play a central role in the productive and qualitative performance of cotton, optimizing yield and fiber production also depends on integrated agronomic management. As highlighted by Vitale et al. (2024) [6], cotton cultivation requires high mineral fertilization, particularly with essential macronutrients (N, P, K e S), in addition to significant water supplies to meet its high physiological demand. However, conventional irrigation methods, such as furrow irrigation, often exhibit low application efficiency, resulting in water and nutrient losses that can compromise both productivity and fiber quality. Therefore, understanding cultivar performance under different water availabilities is essential to mitigate these effects and improve the efficiency of water use in tropical systems.

Adjusting irrigation depths on crop evapotranspiration (ETc) according to cultivar requirements is an efficient alternative to mitigate water deficit [7,8]. Nevertheless, the agronomic performance of cotton under distinct water regimes is not uniform, as it depends on the interaction between the cultivar’s genetic traits and environmental conditions [9].

Studies by Althof et al. (2021) [4], Daniel et al. (2021) [3], Guedes et al. (2023) [7], and O’Shaughnessy et al. (2023) [8] indicate that while increases in seed and lint yield are generally proportional to water availability, fiber yield often responds less sensitively to irrigation management, being primarily determined by cultivar genetics [10,11,12].

In this context, a widely used parameter for quantifying crop response to water deficit is the water sensitivity coefficient (Ky) [13]. This coefficient expresses the relationship between the relative reduction in productivity and the relative reduction in crop evapotranspiration (ETc), allowing the degree of sensitivity of each species and cultivar to water stress to be estimated. High Ky values indicate high sensitivity to water shortage, while low values suggest greater tolerance. In the case of cotton, studies have used Ky as a tool to define irrigation management strategies, evaluate water use efficiency, and identify phenological stages that are most critical to water deficiency [14,15,16].

In this context, optimizing the agronomic performance and fiber yield of cotton cultivars subjected to ascending irrigation regimes in the dry season of the Brazilian Cerrado can contribute to identifying optimal irrigation strategies for each cultivar. This approach aligns with the Sustainable Development Goals (SDGs), particularly regarding the efficient use of water resources (SDG 6) and the promotion of resilient and sustainable agriculture (SDGs 2, 12 and 13).

Thus, the objective of this study was to evaluate the productive performance and fiber yield of four cotton cultivars under five water regimes during the tropical dry season, providing a basis for irrigation management strategies that maximize water efficiency and productivity under water-restricted conditions in the Brazilian Cerrado.

This study was conducted during a single growing season, which represents a limitation for obtaining more robust agronomic data. Therefore, the results presented here should be considered preliminary and require confirmation through additional experimental years.

2. Materials and Methods

2.1. Experimental Site and Conditions

The experiment was conducted during the dry season, between June and November 2021, in the experimental irrigation area of the Federal University of Rondonópolis (UFR), which is located in the municipality of Rondonópolis, state of Mato Grosso, Brazil (16°46′43″ S, 54°58′88″ W; 290 m altitude) (Figure 1). The region has a tropical climate of the Aw type, according to the Köppen–Geiger classification, with a well-defined dry season in winter. Meteorological data were obtained by an automatic agrometeorological station installed 260 m from the experimental area (Figure 1).

The average temperature during the period was 30.9 °C, with a maximum of 43.2 °C and a minimum of 18.5 °C. The average relative humidity was 65.8%, and the total rainfall recorded was 205.3 mm (Figure 2).

The soil of the experimental area was classified as a dystrophic Red Latosol (Oxisol) with a clayey texture.

Table 1. Chemical and granulometric analysis of the dystrophic Red Latosol (Oxisol) of the experimental area before the installation of the experiment with five water regimes (25, 50, 75, 100 and 125% of ETc) with four cotton cultivars (TMG 44 B2RF; FM 944 GL; IMA 5801 B2RF; IMA 709 B2RF).

pH	P	K	Ca	Mg	Al	CEC	V	m	OM	Sand	Silt	Clay
CaCl2	mg kg−1		cmolc kg−1				%		g kg−1	g kg−1-
4.4	9.6	37.0	0.8	0.4	0.5	3.6	26.4	27.9	19.5	390	133	477

pH: hydrogen potential; P: phosphorus; K: potassium; Ca: calcium; Mg: magnesium; Al: aluminum; CEC: cation exchange capacity; V%: base saturation; m%: saturation in aluminum. OM: organic matter. P: Mehlich-1.

presents the results of the chemical and granulometric analyses of the soil before the installation of the experiment.

2.2. Experimental Design and Treatments

The experimental design was randomized blocks, with four replications, in a 5 × 4 factorial scheme, totaling 80 experimental plots. The treatments consisted of five levels of water regimes based on crop evapotranspiration (25%, 50%, 75%, 100% and 125% of ETc) and four cotton cultivars (TMG 44 B2RF; FM 944 GL; IMA 5801 B2RF; IMA 709 B2RF). Each plot was 5 m × 3 m (15 m²), and the total area of the experiment was 1200 m².

2.2.1. Cotton Cultivars

The cultivars were selected according to the cultivars most commonly used by cotton growers in the region where the Brazilian Cerrado Biome predominates. Therefore, four cultivars were selected: (i) FM 944 GL (BASF-FiberMax^®), which stands out for its fiber and seed quality, excellent pointer formation and aggressive root development. Allows the application of the herbicides glyphosate and glyphosate (GL biotechnology, Glytol Liberty Link). It is resistant to blue disease (virus) and does not have the Bt2 biotechnology that controls caterpillars (BASF-FiberMax^®, 2021); (ii) TMG 44 B2RF (Tropical Breeding & Genetics), which allows the application of glyphosate (RF biotechnology, Roundup Ready Flex) and has technologies that allow caterpillar control, namely, Bt2 and BollgardTM II BIOTECHNOLOGY (TMG, 2021); and (iii) IMA 5801 B2RF (Mato Grosso Cotton Institute), which is a cultivar that has a modern plant architecture, good fiber quality and is resistant to ramularia, one of the diseases that most harms cotton. It allows the application of glyphosate herbicide (RF biotechnology, Roundup Ready Flex) and has technology for caterpillar control (Bt2 biotechnology, BollgardTM II) (IMAmt, 2021), and (iv) IMA 709 B2RF, which enables the application of glyphosate herbicide (RF biotechnology, Roundup Ready Flex) and has Bt2 and BollgardTM II biotechnology technology for caterpillar control.

The agronomic characteristics of the selected cultivars (FM 944 GL, TMG 44 B2RF, IMA 5801 B2RF and IMA 709 B2RF) are presented in

Table 2. Morphophysiological characteristics of the cotton cultivars.

Cotton Cultivars	Cycle	Bulk Weight (g)	Fiber Yield (%)	Fiber Length (mm)
FM 944 GL	Medium	4.5–5.5	40–42	30
TMG 44 B2RF	Medium–Early	4.87	43.1	30.3
IMA 5801 B2RF	Medium–Early	5.0–5.6	38–40	29–30
IMA 709 B2RF	Super early	4.0–4.6	41–42	29–30

.

2.2.2. Irrigation

The pressurized irrigation system used in the experimental area had a Branco motor pump (model: B4T-5.5H, maximum power of 5.5 hp.). The irrigation system used was of the drip type, with the use of polyethylene drip hoses of the NaanDanJain—Jain Irrigation Company brand, Top Drip PC AS model of 16 mm. The emitters used were inline and spaced 0.3 m apart, with an average flow rate of 1.6 L h⁻¹ and a maximum pressure of 1.4 bar.

Irrigation management in each water regime was carried out in an irrigation shift fixed on alternate days (every other day) on the basis of the maximum crop evapotranspiration (ETc), which was obtained by the product of the reference evapotranspiration (ETo) and the cultivation coefficient (Kc) based on the FAO [17]. The treatments with values less than 100% refer to deficient irrigation water regimes, and the treatment with 125% ETc corresponds to the level of excess irrigation.

To calculate ETo, the Penman—Monteith method, recommended by the FAO [16], was used (Equation (1)):

$$\mathrm{ETo} = {\displaystyle \frac{0.408 \times \mathsf{\Delta } \times \left(\mathrm{Rn}-\mathrm{G}\right) + \mathsf{\gamma } \times \frac{900}{\mathrm{T} + 273} \times {\mathrm{U}}_2 \times ({\mathrm{e}}_{\mathrm{s}}{- \mathrm{e}}_{\mathrm{a}})}{\mathsf{\Delta } + \mathsf{\gamma } \times (1 + 0.34 \times {\mathrm{U}}_2)}}$$

(1)

where

ETo: Reference evapotranspiration (mm d⁻¹);

Δ: Slope of the vapor pressure curve (kPa °C⁻¹);

Rn: Net radiation (MJ m⁻² d⁻¹);

G: Soil heat flux density (MJ m⁻² d⁻¹);

γ: Psychrometric constant (kPa °C⁻¹);

T: Mean daily air temperature (°C);

U₂: Wind speed at 2 m (m s⁻¹);

e_s: Saturation vapor pressure (kPa);

e_a: Actual vapor pressure (kPa).

The meteorological data necessary for the use of this model were obtained from the automatic agrometeorological station, which belongs to the Graduate Program in Agricultural Engineering, at the Federal University of Rondonópolis.

2.3. Soil Preparation and Sowing

Owing to previous cotton crops and the respective soil corrections carried out in the experimental area of this research, for this experiment, it was not necessary to lime the soil, as it is possible to perform chemical analysis of the soil (Table 1); however, soil preparation was carried out by means of simple harrowing with the passage of a light harrow to a depth of 0.20 m.

After the harrowing operation and chemical analysis of the soil, fertilization was carried out according to the recommendation of applying 135 kg ha⁻¹ P₂O₅ in the form of simple superphosphate and 140 kg ha⁻¹ K₂O as potassium chloride [18]. At sowing, 25 kg ha⁻¹ N was applied in the form of urea, with subsequent complementation of 130 kg ha⁻¹ N in topdressing, which was divided into five equidistant applications every 10 days after plant emergence. All nitrogen applications were performed via fertigation via a portable dosing system (Dosmatic^® MiniDos 1% BSD) coupled to the irrigation control head.

Sowing was carried out manually in two rows, with a spacing of 0.7 m between rows and 0.30 m between plants, totaling approximately 12 plants per linear meter.

During the experiment, the following phenological stages of cotton were observed: emergence at 4 days after sowing, appearance of flower buds at 38 days after emergence, flowering at 60 days, boll formation at 89 days, and boll opening at 105 days after emergence.

Phytosanitary management for weed control was carried out with a systemic herbicide (glyphosate), with two applications during the vegetative period and four during the reproductive period. Pest control, in turn, was conducted through the application of systemic and contact insecticides (acetamiprid and pyriproxyfen), acaricides (diafenthiuron), and fungicides (pyraclostrobin), totaling eleven applications in the vegetative phase and fifteen in the reproductive phase.

2.4. Variables Analyzed

At harvest, plant height was measured with a graduated ruler from the soil surface to the apex of the plants, and leaf area was determined using a leaf area meter (LI-3100C, LI-COR, Lincoln, NE, USA), based on the average of three plants per plot. The bolls were manually counted, collected with pruning shears, and dried in a forced-air oven at 60 °C for 48 h. After drying, the average boll weight was determined using a semi-analytical balance.

Seed cotton yield was obtained from material harvested within a 1 m² area per plot, after separating the dehiscent capsules from the seed cotton. Lint yield was determined after separating the fiber from the seed and weighing it on a semi-analytical balance. Both seed and lint yields were expressed in kg ha⁻¹. Fiber yield was calculated as the ratio of lint weight to seed cotton yield (Equation (2)).

$$\mathrm{FY}=\left({\displaystyle \frac{\mathrm{WF}}{\mathrm{SY}}}\right) \times 100$$

(2)

where

FY: Fiber yield (%);

WF: Weight from fiber (kg ha⁻¹);

SY: Seed cotton yield (kg ha⁻¹).

The evaluation of the effects of water deficit regimes was based on the effects of water stress on cotton yield via the linear model of water productivity [15,19,20], according to Equation (3):

$$\left(1-{\displaystyle \frac{{\mathrm{Y}}_{\mathrm{r}}}{{\mathrm{Y}}_{\mathrm{m}}}}\right) = {\mathrm{K}}_{\mathrm{y}}\left(1-{\displaystyle \frac{{\mathrm{ET}}_{\mathrm{r}}}{{\mathrm{ET}}_{\mathrm{c}}}}\right)$$

(3)

In which,

Y_r: Real yield from culture (kg ha⁻¹);

Y_m: Maximum or potential crop yield (kg ha⁻¹);

K_y: Crop sensitivity factor to water deficit;

ET_r: Actual crop evapotranspiration (mm);

ET_c: Maximum crop evapotranspiration (mm).

The term 1 − Y_r/Y_m corresponds to yield reduction, and 1 − ET_r/ET_c corresponds to the relative evapotranspiration deficit.

2.5. Statistical Analysis

Statistical analysis was performed with the aid of the SISVAR software, version 5.8, to evaluate the interactive effects of two factors, soil water regime levels versus cultivar, on yield parameters. The data were tested for normality via the Shapiro—Wilk and Kolmogorov—Smirnov tests and for homoscedasticity via the Cochran, Hartley, and Bartlett tests. Once the basic statistical assumptions were met, analysis of variance (ANOVA) was performed for qualitative effects (cultivars), and the means were compared via Tukey’s test at a confidence level of 95%. The quantitative effects (water regimes on the basis of crop evapotranspiration) were analyzed via regression [21]. The symbols for the different levels of significance: ^ns = not significant, p > 0.05; * = p < 0.05; ** = p < 0.01; *** = p < 0.001. All the data are presented as the means ± standard errors, and groups with different letters indicate differences between the combinations studied and those cultivated. Then, Principal Component Analysis (PCA) was used, employing the R software [22] version 4.1.2, with the help of the FactorMinieR [23] and factoextra [24] packages, with the aim of summarizing the multivariate variability of the variables analyzed.

3. Results

For the cotton cultivation cycle in the tropical dry season under the conditions of the Brazilian Cerrado, 41 irrigations were carried out, in which the water regimes applied on alternate days ranged from 0.29 to 18.71 mm, reflecting the climatic fluctuations and rainfall variabilities characteristic of the study region, as well as the occurrence of accumulated precipitation of approximately 205 mm in the period from September to November 2021. The total water regime applied in each treatment was adjusted on the basis of the fractions of the maximum evapotranspiration of the crop, which resulted in total volumes of 437.08 mm for the treatment corresponding to 125% of the ETc, 348.79 mm for 100% of the ETc, 263.6 mm for 75% of the ETc, 174.10 mm for 50% of the ETc and 86.21 mm for 25% of the ETc.

The water regime significantly influenced the variables plant height (PH), seed cotton yield (SY), and cotton lint yield (LY). The variables leaf area (LA), cotton lint yield (LY), and fiber yield (FY) were significantly affected by the cultivar. No significant interaction was detected between the applied water regimes and the evaluated cultivars (

Table 3. Summary of analysis of variance with F values for height (PH), leaf area (LA), seed yield (SY), cotton lint yield (LY) and cotton fiber yield (FY) as a function of water regime (25, 50, 75, 100 and 125% of ETc) and cotton cultivar (TMG44B2RF, FM944GL, IMA5801B2RF and IMA709B2RF).

SV	DF	Statistic F
		PH	LA	SY	LY	FY
Block	3	15.24 *	4.05 *	13.15 ***	9.68 ***	1.73 ns
Irrigation (I)	4	47.02 *	2.38 ns	20.45 ***	19.70 ***	1.81 ns
Cultivar (C)	3	1.84 ns	23.45 *	1.12 ns	4.94 **	15.20 ***
I × C	12	0.45 ns	0.69 ns	0.49 ns	0.72 ns	0.49 ns
Error	57
CV (%)		11.36	45.38	34.56	37.87	16.80
Overall average		50.41 (cm)	204.19 (cm2)	3744.86 (kg ha−1)	2104.13 (kg ha−1)	18.34 (%)

SV: Sources of variation; DF: Degrees of freedom; CV: coefficient of variation; Significance levels: *** = p < 0.001, ** = p < 0.01, * = p < 0.05, ns = not significantly (p > 0.05).

).

A linear increase in height was observed depending on the water regime adopted, with an average of 63.08 cm per plant in the water regime corresponding to 125% of the maximum crop evapotranspiration (ETc) (Figure 3a).

The cultivar TMA 44 B2RF presented the largest average leaf area of the plants, corresponding to 353.32 cm², which differed statistically from the cultivars FM 944 GL, IMA 5801 B2RF and IMA 709 B2RF, which presented the lowest values (169.98, 156.55 and 136.92 cm², respectively) (Figure 3b).

The yield of seed cotton was positively influenced by the water regime, adjusting to the increasing linear model, reaching an estimated yield of 5572.4 kg ha⁻¹ in the water regime corresponding to 125% of the ETc (Figure 4).

There was a linear increase in cotton lint yield as a function of the water regime, with an estimated value of 3209.45 kg ha⁻¹ in the highest water regime applied (125% of the ETc) (Figure 5a). Among the cultivars, FM 944 GL and IMA 709 B2RF presented the highest lint yields (2458.32 and 2241.66 kg ha⁻¹, respectively), whereas the IMA 5801 B2RF cultivar presented the lowest yield, with 1537.86 kg ha⁻¹ (Figure 5b).

Significant variations related to cotton fiber yield were observed only in relation to the cultivars studied. The highest fiber yields ranged from 18.6 to 20.1% for the cultivars TMG 44 B2RF, FM 944 GL and IMA 709 B2RF, which differed significantly from the cultivar IMA 5801 B2RF, which had the lowest yield, approximately 14.4% (Figure 6).

Principal component analysis (PCA) identified that the first two components explained 86.8% of the total variability of the data, with 66.6% attributed to PC1 and 20.2% to PC2 (Figure 7). PC1 had greater negative contributions to plant height (PH), seed yield (SY), leaf yield (LY), and fiber yield (FY). PC2 was mainly influenced by leaf area (LA) and fiber yield (FY) (Figure 7).

Treatments with lower water regimes (25 and 50% of ETc) showed positive PC1 values, associated with lower development and yield, while higher regimes (100 and 125% of ETc) showed negative PC1 values, indicating higher productive performance (Figure 7). Among the cultivars, TMG 44 B2RF and IMA 709 B2RF stood out under full irrigation, while FM 944 GL and IMA 5801 B2RF showed lower performance under water deficit.

With respect to the effects of the water regimes on the yield of the cotton cultivars, the values of the water deficit sensitivity coefficient (Ky), represented by the slope coefficient of the adjusted equations (Figure 8), allowed us to classify the cultivars regarding their tolerance to water stress as follows: low sensitivity for the cultivar IMA 5801 B2RF (Ky = 0.73), medium/high sensitivity for IMA 709 B2RF (Ky = 1.05) and high sensitivity for the cultivars FM 944 GL (Ky = 1.20) and TMG 44 B2RF (Ky = 1.28).

4. Discussion

Among the various biotic and abiotic stresses that affect cultivated plants, water stress is considered one of the most limiting factors for the growth, development and production of agricultural crops, and it is not different for cotton [25]. In the case of cotton, different studies have shown that water stress negatively affects production components, resulting in significant reductions in fiber and seed productivity [3,7,8,26,27,28].

The results of this study highlight the close relationship between water regime and the vegetative and productive performance of cotton plants. A positive linear response in plant height was observed as irrigation increased, indicating the crop’s response to water availability. This behavior reinforces that maintaining adequate irrigation levels, adjusted to the crop’s evapotranspirative demand, favors vegetative expansion and contributes to better conditions for radiation interception and assimilation accumulation.

Furthermore, these gains translated into increases in productivity, with increases of 190.6% in the production of seed cotton and 221.26% in the production of lint, showing that irrigation strategies adjusted to ETc are decisive for optimizing the water efficiency and agronomic performance of cotton. However, the fiber yield showed lower sensitivity to different water regimes, corroborating the hypothesis that this parameter is influenced primarily by genetic factors intrinsic to the cultivars.

These results provide valuable technical support to overcome obstacles in cotton production with a view to more sustainable production systems that are resilient to water scarcity in the Brazilian Cerrado, which contributes to more efficient agriculture adapted to tropical climatic conditions.

The reduced yield of cotton under water deficit compared with the highest water regimes (125% of ETc or the water regime of excess irrigation) may have occurred because of the high sensitivity of the crop between the stages of flower bud formation and full flowering, a finding that corroborates the data obtained by Lin et al. (2024) [29] investigating the effects of water deficit (80–90% of the water regime) in the flowering and boll setting phase of cotton in a dry area. The authors reported strong correlations (R² > 0.95) between the maximum leaf area index and yield, as well as high levels of malondialdehyde, which are indicators of water stress, and yield in terms of cotton yield was significantly reduced under water deficit. In addition, they reported that maintaining 90% of the water regime can be a tool to save water without affecting productivity, whereas larger deficits negatively affect the critical phase for the crop.

On the other hand, Ferrit et al. (2022) [30] reported that reduced irrigation during periods of greater water demand reduces the number of green bolls in the lower middle nodes, the region where most of the production is concentrated. Reproductive development was significantly affected by water limitation during this period. After full flowering, the water supply had less influence. This information confirms that irrigation management should prioritize the initial phases of reproduction to ensure a greater number and weight of bolls, with a direct effect on productivity.

Although there was a significant reduction in cotton yield under the lowest water regime (25% of ETc), the average yields of seed and lint cotton for this treatment reached 1674.5 kg ha⁻¹ and 855.2 kg ha⁻¹, respectively. These values are close to the Brazilian national average recorded at the 2023/24 harvest, which was 2681.3 kg ha⁻¹ for seed cotton and 1879.4 kg ha⁻¹ for cotton lint. The results show the productive potential of the cotton cultivars studied when managed under adequate conditions of water availability, highlighting the importance of supplementary irrigation to maximize agronomic performance, especially in regions with water deficit [31].

The absence of a significant interaction effect between the water regime and cultivar suggests that the genotypes evaluated responded in parallel with the increase in the water regime, without differences in response pattern, and a similar behavior was also observed. This finding indicates that irrigation itself has a major effect on productivity, regardless of genetic material. Therefore, this lack of interaction can be explained by the possible adaptive capacity of cotton plants, which may have efficient mechanisms for osmotic adjustment, maintenance of leaf water potential, and relative stability of the photosynthetic rate under moderate water deficit conditions. However, the effects of the cultivars on the productive variables evaluated demonstrated that the genetic basis significantly influences the productive efficiency of the genotypes, even under different water conditions [12].

Despite the lower yield observed, the cultivar IMA 5801 B2RF was the most tolerant to water deficit, with a sensitivity coefficient (Ky) of 0.72, whereas the other cultivars were less tolerant to the reduction in water availability imposed by the water regimes evaluated. The tolerance or resilience of the IMA 5801 B2RF cultivar can be understood from the physiological, biochemical, adaptive and evolutionary responses that these plants develop in the face of various environmental stresses [32]. Recent studies such as those by O’Shaughnessy et al. (2023) [8]; Virk et al. (2023) [10]; Zafar et al. (2023) [33] have indicated that cultivars that maintain their photosynthetic activity and the integrity of their cell membranes under water deficit conditions have greater tolerance to stress. In addition, its ability to accumulate compatible osmolytes and activate antioxidant enzymes, including superoxide dismutase (SOD), peroxidase (POD/APX), and catalase (CAT), is often considered a key mechanism for mitigating water stress-induced oxidative damage to plants [34].

Crivelari Costa et al. (2019) [14], evaluating the Ky of three cotton cultivars grown in the Brazilian Cerrado, reported values between 0.5, and 0.9. Yang et al. (2015) [16] observed a Ky of 0.65 in cotton cultivation with irrigation deficits of up to 45% of evapotranspiration.

These physiological responses are not only indicative of tolerance but also essential for agricultural management, which aims to maximize performance under adverse conditions of water availability [32]. The integration of physiological knowledge and agronomic practices allows for more effective strategies, such as the choice of cultivars with greater physiological plasticity and adjustments in the management of the water regime, as observed for the cultivar IMA 5801 B2RF, which enhances the resilience of the production system [34].

Therefore, these results reinforce the importance of selecting cultivars that tolerate different water regimes with considerable plume production potential, with the objective of ensuring stability and productive efficiency in tropical dry season environments.

5. Conclusions

The application of increasing water regimes, ranging from 90.4 mm to 432.9 mm, had a positive effect on the main productive components of cotton in the Cerrado.

Lint productivity recorded gains of up to 221%, with values ranging from 855.2 to 2741.5 kg ha⁻¹.

The fiber yield remained stable across the different irrigation regimes and was strongly affected by genotype. The average fiber yield of the cultivar FM 944 GL was 39% greater than that of IMA 5801 B2RF.

Despite lower productivity, IMA 5801 B2RF presented greater water resilience (Ky = 0.73), standing out as a viable alternative in scenarios of scarcity or deficit water regimes.

The lack of significant interactions between cultivar factors and water regimes reinforces that the effect of irrigation is predominant, regardless of genetic material, allowing the optimal depth to be defined on the basis of technical performance parameters.

Under conditions of full or moderate water availability, the use of cultivars such as FM 944 GL and IMA 709 B2RF, which are associated with water regimes between 100% and 125% of ETc, is recommended. In environments with water restriction or even the application of deficient water regimes, the cultivar IMA 5801 B2RF is the most suitable for maintaining stable productivity, even under stress. It should be noted, however, that the results reflect the specific conditions of the evaluated growing season. Therefore, longer-term studies conducted under different environmental conditions are recommended to confirm the observed trends and to deepen the understanding of cotton cultivar responses to water regimes in the Cerrado.