Components of High-Yielding Cotton Grown in Rain-Fed Conditions in the Brazilian Cerrado

Abstract

Brazil leads globally in achieving high lint yields for rain-fed cotton in large-scale fields, with about 92% of its cotton area unirrigated. This study hypothesized that cotton could achieve high yields when favorable climate conditions and management practices favor high fruit load. The objective was to analyze the impact of these factors on cotton yields by examining two commercial fields in Brazil in the same climatic zone (Aw, Koppen)—one in Sapezal (SPZ) and the other in Riachão das Neves (RN). The SPZ field (cv. TMG 47B2RF) spanned 20 hectares, while the RN field (cv. FM 974GLT) covered 90 hectares. The soils of both fields were classified as oxisols, with SPZ possessing a clayey texture and RN a sandy loam texture. The findings indicate that the high lint cotton yields—3111 kg·ha⁻¹ in SPZ and 3239 kg·ha⁻¹ in RN—were achieved through a combination of ideal weather conditions, high-quality soil, and effective management practices, which favored boll retention, and an optimal plant architecture with short stature (<1.1 m), 19–22 nodes, and ~165 bolls m⁻². Boll weights averaged 1.85–1.91 g of lint, and fruit retention rates were 61.6% in SPZ and 66.2% in RN. The study reveals a significant yield gap compared to Brazil’s average lint cotton yield (~1900 kg·ha⁻¹) and other high-yield commercial fields (~3500–3900 kg·ha⁻¹ of lint). The results underscore that bridging this gap—ranging from 1200 to 2000 kg·ha⁻¹—could enhance the sustainability of cotton farming in Brazil by maximizing existing cultivated areas. Ultimately, the insights from this study highlight the role of combining climate suitability, management practices, and soil quality improvement to achieve higher cotton productivity and reduce environmental pressures from agricultural expansion.

1. Introduction

Cotton is the most widely used natural fiber globally, although it represents less than 30% of total global fiber consumption [1]. During the 2023/2024 season, cotton was cultivated across 31.7 million hectares worldwide. Brazil ranks as the third-largest cotton producer with 3.67 million tons produced, the largest exporter with 2.6 million tons exported, fifth in cultivated area (1.9 million hectares), and eighth in cotton consumption (0.75 million tons) [1]. Notably, Brazil leads the world in yields from rainfed fields, at 1890 kg of fiber per hectare during the 2022/2023 season [2].

Brazilian cotton production is concentrated in the Cerrado region, primarily in states of Mato Grosso (70%) and Bahia (20%). In Mato Grosso, cotton is often grown as a second crop following soybeans without irrigation, while in Bahia, cotton is either the main rainfed crop or grown following soybeans in irrigated systems. The Cerrado, initially known for its low soil fertility, has two distinct seasons: a rainy season from September to April and a dry season from May to August. Its predominant soil type is an oxisol, and the climate is tropical savanna (Aw), according to the Köppen classification.

Forty-five years ago, Brazil cultivated about 4 million hectares of cotton, producing only 0.5 million tons of lint with a yield of 140 kg·ha⁻¹. Over the last 20 years, Brazilian cotton production has shifted from the Southeast (São Paulo and Paraná) to the Midwest (Mato Grosso) and Northeast (Bahia) regions, with yields rising from 1000 kg·ha⁻¹ in the early 2000s to 1900 kg·ha⁻¹ in the 2022/2023 season (Figure 1).

Several factors have driven this transformation, including advancements in breeding and biotechnology, mechanization, improved soil fertility, integrated pest and disease management, and favorable climatic conditions in key regions. The absence of extreme temperatures and dry weather during the harvest period (June to August) further supports high yields, since temperatures below 30 °C and the high amount of rain early in the season avoids damage during pollen pollination and thus increases boll retention [3].

Brazil’s cotton success can be attributed largely to these yield increases, which have improved the crop system’s profitability. As a perennial shrub with an indeterminate growth habit, cotton varieties grown in Brazil are photoperiod-insensitive. Given adequate temperature and water, these plants continue to grow and flower, which is an advantage in tropical regions where extended growth periods allow plants to reach optimal node numbers (fruiting sites) and maximize yield potential.

For example, Constable and Bange [4] estimate a theoretical yield potential of 5000 kg·ha⁻¹ of lint using modeling, requiring around 250 bolls per square meter, assuming a planting density of 10 plants per square meter and a boll weight of 2 g of lint. In 2022, a 700-hectare field in Australia achieved 4767 kg·ha⁻¹ of lint, and yields of over 3000 kg·ha⁻¹ are common in irrigated Australian fields [5]. However, high yields of 3000 to 5000 kg·ha⁻¹ are typically achieved only in small plots under experimental conditions, and comprehensive data on high-yield indicators at the farm level are limited in the literature.

Understanding production parameters is crucial for enhancing cotton yields and minimizing land demand, especially as global population growth drives the need for increased agricultural output. By analyzing factors such as soil fertility, water availability, plant density, and pest management, producers can identify strategies to optimize their practices and maximize yield potential. For instance, leveraging Brazil’s success in cotton production through improved breeding techniques, mechanization, and integrated pest control has led to significant yield increases. Recognizing the relationship between these parameters allows for more efficient use of resources, reducing the need for additional land while still meeting market demands. Ultimately, knowledge of production parameters not only supports sustainable agricultural practices but also contributes to food security and environmental preservation by minimizing the ecological footprint of cotton cultivation.

This research presents a case study on the production parameters of two commercial cotton fields in Brazil, where rainfed conditions resulted in yields exceeding 3000 kg per hectare of lint. By examining these fields, the study aims to identify and analyze the key factors contributing to such high productivity levels. Understanding the specific agricultural practices, environmental conditions, and management strategies employed in these successful operations can provide valuable insights for other cotton producers seeking to enhance their yields under similar rainfed systems. This case study not only highlights the potential for high performance in Brazilian cotton agriculture but also emphasizes the importance of optimizing production parameters to achieve sustainable growth in the sector. Despite that, this study describes only two specific locations, which may restrict the generalizability of the findings to other regions.

2. Materials and Methods

2.1. Characterization of the Study Area

This is a case study of two commercial cotton fields (Gossypium hirsutum L.) cultivated in Brazil: cultivar TMG 47B2RF was tested in Sapezal (SPZ, 12°59′22″ S; 58°45′52″ W 560 m asl) in Mato Grosso State (MT) during the 2021 season, and cultivar FM 974GLT was tested in Riachão das Neves (RN, 11°31′49″ S; 45°43′57″ W 795 m asl) in Bahia State (BA) in the 2021/2022 season. Sowing was carried out on 01/23/2021 in SPZ (rows spaced at 0.9 m) and on 12/13/2021 in RN (rows spaced at 0.76 m), using 10 seeds per meter, and emergence occurred 5 days later.

The field in SPZ was 20 hectares and was cultivated with soybean (1st crop) and cotton (2nd crop) over the last 10 seasons and represents a growing area of 266,000 ha (14% of Brazilian cotton area). In RN, the field size was 90 hectares, including soybean (2020/2021) and a mixture of cover crops (pearl millet, turnip, and Urochloa ruziziensis) after soybean harvest preceded cotton cultivation. Riachão das Neves had a cotton cultivation area of 33,500 hectares, which represents 9% of Bahia State cotton area and 2% of Brazilian cotton area. The soil is characterized as oxisol in both areas (Soil Survey Staff) with a clayey texture in SPZ and sandy loam texture in RN. The regional climate, according to Köppen and Geiger, is Aw (tropical savanna climate with dry winter). Meteorological parameters such as rainfall, maximum and minimum temperatures, and solar radiation (Figure 2) were recorded in a weather station 1 km from the experiment. Degree day ((DD)  =  ((T + t)/2) − 15) was used to calculate the effective daily temperature, and vapor pressure deficit (VPD) was calculated using average air temperature and air humidity (Figure 3).

2.2. Cotton Management

Soil fertilizers were applied following rates and sources described in

Table 1. Soil fertilization program of cotton field in Sapezal, MT, and Riachão das Neves, BA, Brazil. 2021/2022 season.

Nutrient	Rate (kg·ha−1)	Source	DAE	Application
Sapezal, MT
K	120	potassium chloride	0	Broadcast
S	30	elemental	0
N, S	50 + 15	urea + ammonium sulfate	20	Topdressing
N, P	11.7 + 62	monoammonium phosphate	40
N, S	33 + 10	urea + ammonium sulfate	40
B	3.5	ulexite	40
N	62.4	urea + ammonium sulfate	60
N	23	ammonium nitrate	80
	Riachão das Neves, BA
P	72	single superphosphate	−50	Incorporated in row
N, P, K	17 + 87 + 42	urea + triple superphosphate + potassium chloride	0	Incorporated in row
N	110	urea	25	Incorporated between rows
K + B	180 + 1.5	potassium chloride + ulexite	30	Topdressing
N	66	urea (NBPT)	70

DAE: days after emergence; N: nitrogen; K: potassium; P: phosphorus; S: sulfur; B: boron.

, according to the farmer’s practices.

Crop management followed procedures adopted in commercial cotton farming for monitoring and control of pests, weeds, and disease; plant growth monitoring and plant growth regulator management; and harvest aids, including defoliant and boll openers. Mepiquat chloride was used as a growth regulator, and rates were calculated following Echer and Rosolem’s method [6] using the Cotton Apps^® platform. Harvesting was carried out at 195 DAE in Sapezal and 180 DAE in Riachão das Neves, about 14 days after physiological maturity and defoliant application (Diuron + thidiazuron: 30 and 60 g·ha⁻¹).

2.3. Data Collection, Evaluation and Analysis

At harvest, soil samples were collected for chemical (0–60 cm), physical (0–40 cm in SPZ and 0–60 cm in RN), and soil enzyme activity (0–20 cm). The chemical analysis was carried out using the methodologies described by Raij et al. [7]. The particle size was analyzed [8]. Soil pH was measured in a 0.01 mol/L⁻¹ CaCl₂ suspension, and soil organic carbon (OC) was determined using potassium dichromate (K₂Cr₂O₇). Available phosphorus (P) was determined colorimetrically using a spectrophotometer and exchangeable basic cations calcium (Ca), magnesium (Mg), and potassium (K) were extracted using an ion-exchange resin by atomic absorption spectrometry (PerkinElmer analyst 200, Shelton, CT, USA). Exchangeable aluminum (Al³⁺) was extracted using 1 mol/L⁻¹ KCl in a 1:10 soil-to-solution ratio. The potential acidity (H + Al) was measured immediately after measuring soil pH by adding 5 mL of SMP buffer solution (pH = 7.0) to the suspensions. Sulfur (S) determination was performed at a 0.01 mol/L ⁻¹ Ca(H₂PO₄)₂ solution, and available micronutrients iron (Fe), copper (Cu), manganese (Mn), zinc (Zn), and boron (B) were extracted with DTPA and determined by inductively coupled plasma optical emission spectroscopy (ICP-OES).

For soil physical properties, undisturbed soil samples were collected from the midpoint of 0–20 and 20–40 cm depth intervals in SPZ and 0–20, 20–40, and 40–60 cm depth intervals in RN. Small trenches were dug, and undisturbed samples were collected using stainless steel cylinders. The soil samples were wrapped with aluminum foil, carefully transported to the laboratory, and maintained at 5 °C to reduce biological activity [9]. Subsequently, the samples were placed in trays filled with distilled water and left until they reached saturation. The samples were then weighed and placed on a tension table [10] at matric potential (ψ) of −10 kPa. Upon reaching hydraulic equilibrium, the samples were weighed again and dried in an oven at 105 °C for 48 h. Soil bulk density (ρ_b) was then calculated as the ratio between oven-dried soil mass and sample volume [11]. Total porosity (TP) of soil was obtained as the saturated soil water volume [12]. Field capacity (FC) was equated with the soil water content retained at ψ = −10 kPa, and the air-filled porosity (ε_a) was computed by the difference between TP and FC [9]. Other soil physical indicators were obtained as per Reynolds et al. [13], including (1) the storage capacity of soil water (dimensionless), that is, the ratio between the soil water content at ψ = −10 kPa (FC) and TP, and (2) the storage capacity of soil air (dimensionless), that is, the ratio between air-filled porosity at ψ = −10 kPa and TP.

After equilibration at ψ = −10 kPa, air permeability (Kair) was measured using a constant head permeameter [14]. A mass flow controller was used to set the airflow values that flowed into each sample, in addition to a differential pressure manometer to identify the pressure differences, modified according to the structural soil conditions. Kair (µm²) was calculated using Equation (1).

$$Kair=\left[{\displaystyle \frac{Q\eta }{A_s}}\right]x\left[{\displaystyle \frac{z}{P}}\right]$$

(1)

where Q is the mass flow (m³/s); η is the air viscosity at 20 °C (1.84 × 10⁻⁵ N s/m²); As is the area perpendicular to the air movement (m²); z is the soil column height (m); and P is the differential air pressure (Pa).

The soil pore organization index, k₁ (μm² m³ m⁻³), was calculated according to Groenevelt et al. [15]:

$$k_1={\displaystyle \frac{k_{air}}{{\varepsilon }_a}}$$

(2)

The degree of compactness (DC) in both fields was estimated using Equation (3).

$$DC={\displaystyle \frac{\mathsf{\rho }\mathrm{b}}{\mathsf{\rho }\mathrm{r}\mathrm{e}\mathrm{f}}} \ast 100$$

(3)

The soil bulk density reference (ρ_ref) was calculated using the pedotransfer function [16].

Soil enzyme activity was determined by sampling soil at 0–10 cm and 10–20 cm soil depths. Acid phosphatase was determined by the method described by Tabatabai and Bremner [17]. We added 4 mL of MUB pH 6.5, and 1 mL of PNF was diluted in MUB pH 6.5 in 1 g of soil. All samples were placed in the bath for 1 h at 37 °C; after this, 1 mL of CaCl₂ 0.5 mol/L⁻¹ and 4 mL of NaOH 0.5 M were added. The samples were then shaken and filtered. The samples were recorded at 400 nm absorbance using a spectrophotometer. Arylsulfatase enzyme activity was measured [17] after weighing 1 g of soil and adding 4 mL of acetate–acetic acid buffer 0.5 mol/L⁻¹ (pH 5.8) and 1 mL of solution p-nitrophenyl sulfate (potassium sulfate 0.025 M) PNS. The samples were shaken and incubated for one hour under a constant temperature of 37 °C. After incubation, 1 mL of CaCl₂ 0.5 M and 4 mL of THAM (pH = 12) were added to the samples. The samples were filtered and recorded at 400 nm absorbance. β-glucosidase enzyme activity was measured [18]. A total of 1 gram of soil was homogenized in 4 mL of MUB buffer (pH = 6.0) and 1 mL of PNG (p-Nitrophenol 1000 ppm). After incubating for one hour, 1 mL of 0.5 mol/L⁻¹ CaCl₂ and 4 mL of 0.1 M THAM (pH = 12) were added to the samples, which were then shaken and filtered for readings at 400 nm absorbance.

Plant mapping was performed in five plants per replication by counting and weighing each boll and indicating node and fruit position (P1, P2, and P3+) right before mechanical harvest. Boll retention in each fruit position (P1, P2, and P3+) and the total boll setting (considering all positions) were calculated by dividing the number of bolls retained by the total fruiting sites issued by the plant. Accumulated yield was calculated to determine the contribution of each node to yield using the data from plant mapping and consisted of the boll weight multiplied by the number of bolls in each node multiplied by plant density. The production of vegetative branches (monopodial) was summed to the following fruiting branch and so on until the last node at the top of the plant.

At harvest, 190 DAE in SPZ and 180 DAE in RN, yield was determined by a mechanical machine (John Deere CP 690, Moline, IL, USA) in all fields. Yield components (boll weight and boll number) were determined in 2 m of a row (1.8 m⁻² in SPZ and 1.52 m⁻² in RN) with 20 replications by counting, handpicking, and dividing the total boll weight by the boll number to obtain mean boll weight (Figure 4). A sample of 150 g of seed cotton was used for fiber quality determination (High Volume Instruments, Uster Technologies, Uster, Switzerland). The gin turnout used to calculate lint yield was from the commercial field.

2.4. Climate Conditions

Rainfall from sowing to harvest was about 880 mm in SPZ and 903 mm in RN, and 90% and 85% of the total rainfall, respectively, occurred until peak flowering (~90 DAE) (Figure 2;

Table 2. Summary of the climate data considering cotton phenological stages.

		Sapezal (SPZ), MT
DAE	Phenological Stage	Rainfall	Solar Radiation	Tmax	Tmin	DD	VPD
		mm	MJ m−2 day−1	°C		-	kPa
0–33	E-PHS	415	14.61	30.52	20.97	10.75	0.46
34–55	EF	159	14.49	28.70	20.47	9.59	0.33
56–90	PF	228	16.05	29.59	20.23	9.91	0.41
91–130	BF	76	14.30	29.07	17.40	8.24	0.55
131–190	BM	2	11.99	30.17	14.77	7.47	1.00
DAE		Riachão das Neves (RN), BA
0–41	E-PHS	514	18.45	28.41	19.02	8.53	0.27
42–61	EF	78	19.94	30.17	18.79	9.91	0.44
62–94	PF	175	20.44	31.24	19.57	10.36	0.62
95–130	BF	135	21.62	31.37	19.04	10.06	0.66
131–180	BM	0	19.61	30.53	15.32	7.84	0.85

E: emergence; PHS: pinhead square; EF: early flowering; PF: peak flowering; BF: boll filling; BM: boll maturation. DD: degree day; VPD: vapor pressure deficit.

). Solar radiation was, on average, 13.8 (peaking at 16) and 20 (peaking at 21.6) MJ m⁻²/day⁻¹ in SPZ and RN, respectively (Supplementary Materials). Maximum temperatures were above 30 °C early (vegetative stage) and late in the season (boll maturation stage) in SPZ, and, in general, temperatures were in the adequate range for photosynthesis. In RN, Tmax was above 30 °C during early, middle, and late flowering, but it was always followed by a week of temperatures below 30 °C. Minimum temperatures were close to 20 °C until the peak flowering (~90 DAE) in SPZ and boll-filling stage (~125 DAE) in RN. Degree day accumulation was between 8 and 10 until the boll-filling stage and about 7 during boll maturation in both fields (Figure 3; Table 2). In general, VPD was lower in SPZ, but at the end of the cycle, it started to increase (from July) in both SPZ and RN (Figure 3).

A descriptive analysis was used to calculate the mean and the standard deviation considering 20 replications taken randomly inside an area of 20 hectares in Sapezal and 90 hectares in Riachão das Neves.

3. Results

3.1. Soil Fertility

In general, both fields had high fertility, with values shown in

Table 3. Soil fertility parameters from 0 to 60 cm layers in Sapezal (clayey) and Riachão das Neves (sandy loam). OM = organic matter.

Sapezal, MT
Depth	pH	OM	P	S	Al3+	H + Al	K	Ca	Mg	SB	CEC
cm	CaCl2	g dm−3	mg dm−3		mmolc dm−3
0–20	5.59	35	40	17	0.0	41	2	51	14	66	107
20–40	4.99	31	15	66	0.4	52	1.9	20	7	29	81
40–60	4.73	26	7	84	1.1	55	1.6	9	4	15	71
	m	BS	B	Cu	Fe	Mn	Zn	Sand	Silt	Clay
	%		mg dm−3					g kg−1
0–20	0	61	0.9	1.2	27	1.5	5	88	194	718
20–40	2	36	0.8	0.2	21	0.6	3	77	111	812
40–60	9	22	0.8	0.2	13	0.4	2	74	76	851
Riachão das Neves, BA
Depth	pH	O.M.	P	S	Al3+	H + Al	K	Ca	Mg	SB	CEC
cm	CaCl2	g dm−3	mg dm−3		mmolc dm−3
0–20	5.5	16	29	7	0.1	16	2	19	8	30	46
20–40	5.6	14	11	6	0.1	15	1.8	15	6	22	37
40–60	5.1	15	7	17	0.8	20	1.6	8	4	14	34
	m	BS	B	Cu	Fe	Mn	Zn	Sand	Silt	Clay
	%		mg dm−3					g kg−1
0–20	0.6	64	0.6	0.6	16	1.2	1.5	666	49	263
20–40	0.4	60	0.4	0.2	10	0.4	0.5	757	31	212
40–60	9.6	41	0.3	0.1	6	0.1	0.1	737	33	231

for pH range (5–5.5); low content of toxic aluminum (Al³⁺) in both fields; and high content of phosphorus (P), calcium (Ca), and boron (B) in the deeper layers. The highest values in cation exchange capacity (CEC) in Sapezal are explained by soil type (clayey) and soil organic matter content.

3.2. Soil Physics Parameters

The magnitude of soil physical parameters in SPZ and RN are due to the distinct granulometric composition between the soils: SPZ soils are around 80% clay and RN soils are 25% clay (

Table 4. Soil physical parameters from distinct layers in Sapezal (clayey) and Riachão das Neves (sandy loam).

	Sapezal-MT		Riachão Das Neves-BA
Parameter/Depth (cm)	0–20	20–40	0–20	20–40	40–60
Soil Bulk density (Mg m−3)	1.10 ± 0.06	1.04 ± 0.11	1.71 ± 0.05	1.60 ± 0.12	1.57 ± 0.03
Total porosity (m3 m−3)	0.59 ± 0.02	0.61 ± 0.04	0.36 ± 0.02	0.40 ± 0.05	0.41 ± 0.01
Soil water content (ψ = 10 kPa) (m3 m−3)	0.45 ± 0.04	0.41 ± 0.03	0.23 ± 0.02	0.21 ± 0.02	0.21 ± 0.01
Air-filled porosity (m3 m−3)	0.14 ± 0.04	0.19 ± 0.06	0.13 ± 0.02	0.18 ± 0.06	0.20 ± 0.02
Storage capacity of soil water	0.77 ± 0.09	0.68 ± 0.08	0.65 ± 0.04	0.54 ± 0.10	0.52 ± 0.03
Storage capacity of soil air	0.24 ± 0.06	0.30 ± 0.07	0.38 ± 0.04	0.45 ± 0.10	0.49 ± 0.03
Air permeability (µm2)	2.79 ± 3.17	10.23 ± 11.75	27.40 ± 29.91	68.11 ± 36.58	40.23 ± 21.85
Pore continuity index (µm2/m3 m−3)	16.67 ± 14.67	45.85 ± 43.05	214.72 ± 230.17	366.59 ± 116.68	204.26 ± 108.22
Degree of compactness (%)	80.28 ± 4.77	79.21 ± 8.4	95.98 ± 2.59	87.69 ± 6.69	86.80 ± 1.44

). In SPZ, the Bd and DC average values indicate good physical conditions for plant growth, which are corroborated by the air-filled porosity of >0.10 m³ m⁻³, an air permeability of >1 μm², and a storage capacity of soil water of close to 70%. Although the storage capacity of soil water and soil air values are slightly outside the limits proposed as ideal for plant growth [9], the other soil physical parameters measured suggest that the physical functionality of the soil allows plants to access the resources necessary to support high cotton yields. For RN, the DC values indicate the occurrence of soil compaction in the 0–20 cm layer, but the air-filled porosity values suggest that, at these levels, they are not restrictive to cotton growth. Although the 20–40 cm and 40–60 cm layers have reduced storage capacity of soil water values, the Kair and pore index continuity values suggest good conditions for root proliferation and access to water and nutrients, especially in the 20–40 cm layer (Table 4).

3.3. Soil Enzyme Activity

Acid phosphatase (AcidP) activity was similar in SPZ and RN for both soil layers (231 mg kg⁻¹ h⁻¹ on average); however, arylsulfatase (Aryl) and ß-glucosidase (ß-Glu) were higher in SPZ (clayey soil with higher soil organic matter (SOM)), showing values of 45.5 in SPZ compared to 15 mg·kg⁻¹ h⁻¹ in RN for arylsulfatase and 86 in SPZ compared to 28 mg·kg⁻¹ h⁻¹ in RN for ß-glucosidase, respectively (

Table 5. Soil enzymatic activity from high-yielding cotton fields in Sapezal (SPZ) and Riachão das Neves (RN), Brazil.

Soil Layer (cm)	Acid Phosphatase		Arylsulfatase		ß-Glucosidase
	mg kg−1 h−1
	SPZ	RN	SPZ	RN	SPZ	RN
0–10	234 ± 53	215 ± 31	50 ± 11	16 ± 3	93 ± 23	30 ± 8
10–20	252 ± 56	225 ± 24	41 ± 5	14 ± 3	79 ± 25	26 ± 7

).

3.4. Yield and Fiber Quality Parameters

Plant density was 8.9 and 9.9 m⁻² in SPZ and RN, respectively. Short plants (<1.1 m height) with 19 to 22 nodes (13–14 sympodial branches) (

Table 6. Yield components, yield, and fiber quality of cotton in Sapezal, MT (2021 season), and Riachão das Neves, BA (2021/2022 season).

Component	Sapezal, MT	Riachão Das Neves, BA
Plants m−2	8.9 ± 1.12	9.91 ± 0.12
Plant height (cm)	107 ± 9.6	97.4 ± 2.18
Node number	21.7 ± 0.9	19.5 ± 0.27
Simpodial node no.	14.4 ± 0.9	13.13 ± 0.7
Node of the first sympodia	8.30 ± 0.3	6.00 ± 0.5
Bolls per plant−1	18.50 ± 2.44	17.35 ± 0.50
Bolls per m−2	165.3 ± 11.9	167.8 ± 3.6
Boll weight (g of lint)	1.84 ± 0.07	1.91 ± 0.02
Seed cotton yield (kg·ha−1)	7211 ± 476	7356 ± 174
Fiber yield	3111 ± 213	3239 ± 85
Gin turnout (%)	43.31 ± 0.83	44.00 ± 0.17
Micronaire (µg in−1)	4.14 ± 0.18	4.31 ± 0.34
Length (mm)	30.24 ± 0.61	30.87 ± 0.52
Strength (gF/Tex)	29.1 ± 0.99	30.59 ± 0.73
Short fiber index (%)	8.90 ± 0.62	7.61 ± 0.41
Elongation (%)	6.84 ± 0.18	7.51 ± 0.21
Uniformity (%)	81.7 ± 1.11	83.8 ± 0.76
Maturity ratio	85.4 ± 0.51	85.0 ± 0.11

) were reported in both fields. The first sympodia node of SPZ plants were at a higher position than RN plants due to genetic differences. Boll number per plant was 17 to 18, and bolls were very similar in both fields (Table 6). Boll weight was, on average, 4.3 g of seed cotton (1.85 g of lint in SPZ and 1.91 in RN), which resulted in a seed cotton yield of 7211 (3111 kg·ha⁻¹ of lint) in SPZ and 7356 kg·ha⁻¹ (3239 kg·ha⁻¹) in RN (Table 6).

Despite this high yield for rainfed environments, fiber quality parameters were classified as premium, especially for parameters such as micronaire, fiber length, and strength (Table 6).

Plant mapping analysis showed lower nodes (VB to 10th) accumulating on average 250 kg·ha⁻¹ of lint per node, while the 11th, 12th, and 13th nodes accumulated about 300 kg·ha⁻¹ of lint. From the 14th node, the accumulation of yield decreases up to the 21st node in Sapezal, MT (Figure 5). In Riachão das Neves, vegetative branches yielded about 480 kg·ha⁻¹, and reproductive branches from the bottom of the plant (5th to 9th node) accumulated 300 kg·ha⁻¹ per node, decreasing thereafter; after the 18th node, yield accumulation stopped (Figure 5).

Additionally, plant mapping also indicated the heaviest bolls in the middle third of the plant in Sapezal and in the bottom of the plant in Riachão das Neves (

Table 7. Boll weight (g of lint) per fruiting position and node in Sapezal, MT, and Riachão das Neves, BA, Brazil.

	Sapezal, MT			Riachão Das Neves, BA
Node	P1	P2	P3+	P1	P2	P3+
VB	1.51			1.46
5				1.99	1.87
6				2.25	1.68
7	1.63	1.70	1.64	2.48	1.72
8	1.85	1.90	1.52	2.40	1.86
9	1.98	1.86	1.25	2.05	2.19
10	2.10	1.92	1.84	1.75	1.79
11	2.18	1.71	1.30	2.20	1.79
12	2.06	1.83	1.14	1.80	1.47
13	2.28	1.85	1.73	2.01	1.35
14	2.22	1.49		1.64	0.94
15	2.20	1.70		1.84
16	2.01	1.39		1.68
17	2.17	1.73		1.91
18	1.98	1.30
19	2.09	1.73
20	1.76	1.73
21	0.95

).

Total fruit retention was 61.6% in SPZ and 66.2% in RN. Bolls from the first fruit position were retained at a rate of 82.9% (SPZ) and 74.8% (RN), and their contributions to yields were 64% (SPZ) and 71% (RN), while fruits from the second position were retained, on average, at a rate of 60% and their contributions to yields were around 28% (28.8% in SPZ and 27.2% in RN) (Figure 6). Third position bolls had a low retention rate (~33%), contributing to 6.8% (SPZ) and 1.6% (RN) of the total yield (Figure 6).

4. Discussion

The high cotton yields observed in the commercial fields of Sapezal (SPZ) and Riachão das Neves (RN) can be attributed to a synergistic combination of favorable environmental conditions, superior soil quality, and effective agricultural management practices. This study underscores the significance of these factors in optimizing yield potential for rainfed cotton systems in Brazil.

The climatic conditions during the growing season were pivotal to achieving high yields. The consistent and adequate rainfall, coupled with the absence of severe temperature stresses—both heat and cold—created an ideal environment for cotton growth. Additionally, the favorable radiation incidence during the flowering stage positively influenced boll formation and retention. These findings align with previous research [4], which highlights the critical role of weather patterns in cotton yield optimization. Temperatures exceeding 30 °C reduce boll retention and thus final yield, which can also be impaired due to continuous rain during flowering affecting the pollination [3]. The water requirement of cotton depends on climate and length of the total growing period [3]. We report here an amount of ~900 mm of rain for a 180–190 days long season, highly concentrated early in the season (~50%) (Table 2).

4.1. Boll Characteristics and Yield Potential

The analysis revealed that both fields maintained a similar boll count per square meter (~165–168) and per plant (18.5 in SPZ and 17.3 in RN). Notably, the boll weights measured (1.84 g in SPZ and 1.91 g in RN) suggest that enhancing this yield component could be a viable strategy for increasing overall productivity. These data are aligned with those reported earlier [4], which suggested a boll number of 175 boll m⁻² weighing 2 g of lint each to achieve a 3500 kg ha⁻¹ yield. Given that achieving higher boll numbers is challenging late in the season due to resource limitations for late-fruiting positions, focusing on improving boll weight may provide a more feasible pathway for yield enhancement.

4.2. Yield Gap Insights

This study identifies a substantial yield gap for rainfed cotton crops in Brazil, ranging from 1200 to 2000 kg·ha⁻¹ compared to irrigation systems in other countries, such as Australia, where yield potentials of up to 3500 kg·ha⁻¹ were reported. The average yields reported in this study (~3100 kg·ha⁻¹) and in commercial fields (3500–3900 kg·ha⁻¹) highlight the urgent need for targeted strategies to bridge this yield gap. Future studies could focus on the adoption of best management practices that align with the unique climatic and soil conditions across different Brazilian regions.

4.3. Soil Quality and Nutrient Management

The evaluated fields were characterized by deep soils with excellent soil physical properties, which supported extensive root growth—observed to reach depths of up to 4 m. This deep-root system was crucial for water uptake during dry periods following the rainy season. The high soil fertility levels, particularly regarding essential nutrients like phosphorus (P), calcium (Ca), and boron (B), contributed to robust root development and overall plant health. The nitrogen application rates (180 kg·ha⁻¹ in SPZ and 190 kg·ha⁻¹ in RN), supplemented by residual contributions from previous soybean crops and cover crops, provided a significant nutrient base for the cotton crops. This integrated nutrient management approach is essential for sustaining soil fertility and promoting high yields.

4.4. Soil Enzymatic Activity

This study observed similar levels of enzymatic activity for AcidP across both locations, indicating moderate soil health [19,20]. However, lower activity levels for Aryl and ß-Glu in RN may be attributed to its sandy loam texture and lower soil organic matter content. This highlights the importance of enhancing soil organic matter through cover cropping and other practices to improve microbial activity and nutrient cycling [19]. Given the inherent limitations of sandy soils in organic matter content, the implementation of sustainable agricultural practices is crucial for maintaining soil health and productivity.

4.5. Limitations and Future Research

While this study provides valuable insights, it is important to acknowledge its limitations. Conducting the research in only two specific locations may restrict the applicability of the findings to other regions of Brazil with different climatic and soil conditions. Moreover, variability in agricultural management practices, such as crop rotation and fertilization strategies, could influence yield outcomes. Therefore, further research should explore a broader range of conditions, including diverse soil types and management practices, to enhance the understanding of the factors influencing cotton fiber yield across Brazil.

5. Conclusions

The successful high yields achieved in the cotton fields of Sapezal and Riachão das Neves are the result of a complex interplay of edaphic factors, favorable climatic conditions, and effective management practices. A rainfall volume of 900 mm, low instances of temperature stress (30/20 °C), and high radiation rates during the critical phenological stages of cotton (flowering and boll filling), combined with high soil quality in deeper layers and appropriate plant management (including mepiquat chloride, pest control, and disease management), were crucial factors for achieving high fiber yield. A total of 165 bolls m⁻² weighing 1.85 g of lint per boll resulted in 3100 kg·ha⁻¹ of fiber. Continued research and innovation in cotton production strategies are essential to unlocking the full potential of Brazilian cotton agriculture, bridging the identified yield gaps, and promoting sustainability within the industry.