Farmyard manure improves the productivity and profitability of cotton (Gossypium hirsutum L.) under semi-arid conditions through integrated nutrient management Authors: Cevher İlhan Cevheri Published: 2026 ABSTRACT Cotton cultivation in semi-arid areas is often limited by insufficient soil organic matter content and increasing fertilizer costs. The integrated use of composted farmyard manure (FYM) with mineral fertilizers may enhance crop yields and agricultural profitability in irrigated, calcareous soils. This study evaluated the potential of integrating composted FYM with mineral fertilizers to optimize cotton productivity, reduce chemical inputs, and promote circular agriculture. Nine fertilizer treatments were applied to the ‘Candia’ cotton variety, and data related to morphological, physiological, yield, and fiber quality traits were collected. Treatments consisted of control (no fertilizer application), 350 kg ha-1 urea (46% N), 100, 200, and 300 kg ha-1 diammonium phosphate (DAP) + 350 kg ha-1 urea, 20, 40, and 60 t ha-1 FYM + 350 kg ha-1 urea, and 100 t ha-1 FYM without urea. Fertilizer treatments significantly influenced the number of bolls, boll weight, boll seed cotton weight, seed cotton yield, ginning outturn, and fiber yield. However, fiber quality traits remained unaffected by fertilizer treatments. The highest seed cotton yield (5207.3 kg ha⁻¹) was recorded with 60 t ha⁻¹ FYM combined with urea (32.9% improvement compared to the control 3918.5 kg ha⁻¹). This was associated with increased values of boll weight (6.70 g vs. 5.05 g), ginning outturn (44.03% vs. 40.06%), and fiber output (2292.9 kg ha⁻¹ versus 1569.8 kg ha⁻¹). Conversely, the highest values for plant height (82.40 cm) and NDVI (0.50) were observed in the 300 kg ha-1 + urea treatment. The economic analysis revealed the highest net income ($693.29 ha⁻¹) with FYM at 60 t ha⁻¹ + urea and a benefit-cost ratio of 1.57. It is concluded that integrating FYM at 60 t ha⁻¹ with 350 kg ha-¹ urea increased cotton yield and economic returns without affecting fiber quality. This study was conducted over a single growing season at a single location; therefore, multi-year testing that reports on soil and plant nutrient dynamics is needed to support broader recommendations.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12870-026-09387-9. Introduction Global cotton production faces significant sustainability challenges, including environmental degradation, excessive agrochemical inputs, and economic volatility in fertilizer markets [1, 2]. Although cotton represents only 2.4% of global agricultural output [3] it accounts for ~ 27% of global pesticide use, making it one of the most chemically intensive crops [3, 4]. The production of 1 kg of textile fiber requires > 100 L of water [5]. Annual global textile production generates approximately 5 billion cubic meters of wastewater [6]. These environmental concerns are coupled with significant economic pressures on cotton farmers (particularly in developing nations), as input costs constitute a significant share of total production costs [1, 2]. The Southeastern Anatolia Project (GAP) region of Türkiye is an important cotton-producing zone [7]. Şanlıurfa province in the region hosts approximately 200,000 head of cattle and 2 million head of small ruminants, collectively generating an estimated 6–7 million tons of livestock manure annually [8, 9]. This massive livestock waste stream represents an underutilized nutrient resource [10] with nitrogen content potentially exceeding 200,000 tons annually [11]. However, existing agricultural practices have largely failed to exploit this potential, instead investing in expensive synthetic fertilizers while livestock manure remains inadequately valorized [12]. Recent studies have shown that farmyard manure (FYM) significantly improves soil organic carbon, nutrient availability, and enzymatic activities, thereby enhancing soil health and crop productivity. Long-term FYM applications also increase microbial biomass and nutrient cycling efficiency [12]. Soil organic matter is crucial to agricultural sustainability, regulating key physical, chemical, and biological processes in soil that support nutrient cycling, water retention, soil structure, and crop productivity [13, 14]. Intensive agriculture in the Harran Plain has reduced soil organic matter below 1%, significantly decreasing aggregate stability, water infiltration rates, nutrient availability, and crop productivity [15, 16]. The decline in soil quality has required increased chemical fertilizer use, but nitrogen use efficiency has fallen from 67% to less than 47% over the last 50 years, indicating diminishing returns from synthetic nitrogen fertilizer [17]. At the same time, excessive use of nitrogen fertilizer contributes to greenhouse gas emissions and other reactive nitrogen losses. Reactive nitrogen cycling can increase ammonia (NH₃) volatilization and nitric oxide (NO) and nitrous oxide (N₂O) emissions, thereby altering atmospheric composition and affecting climate regulation [18]. Nitrogen-use inefficiency also results in economic losses due to reduced product value and increased costs of environmental remediation [19]. Integrated nutrient management (INM) systems combining organic and inorganic nutrient sources offer compelling alternatives to production models based solely on synthetic fertilizers [20, 21]. Recent studies have indicated that FYM increases soil biological activity and nutrient-cycling efficiency and maintains crop productivity equal to or better than synthetic-fertilizer-only methods [22, 23]. The use of livestock waste for valorization in agriculture aligns with the principles of the circular economy. This approach utilizes waste streams from livestock production to create valuable agricultural inputs, thereby reducing reliance on external fertilizers and improving system-level sustainability [24]. However, limited quantitative studies are available on the optimal use of livestock manure and synthetic fertilizers in semi-arid cotton production. This gap is evident in the studies that simultaneously assess physiological responses, soil enzyme dynamics, and economic outcomes. This study evaluated the potential for integrating FYM with synthetic fertilizer to enhance cotton productivity and economic viability in the Harran Plain. The objectives of the study were to: (1) characterize the effects of varying FYM application rates combined with standardized nitrogen and phosphorus fertilization on cotton growth, physiological parameters, and yield components; (2) evaluate fiber quality traits in response to integrated nutrient management practices; and (3) conduct economic analysis to determine cost-benefit ratios of alternative fertilizer management strategies. These results would provide practical information to support farmers and policymakers in developing more sustainable nutrient management strategies for cotton production in water-limited regions. Materials and methods Experimental site and soil The experiment was conducted at the experimental research farm of Harran University, Faculty of Agriculture, Şanlıurfa, Turkey (37°23′N, 38°47′E, 270 m elevation) during 2021. The experimental site is characterized by a continental climate with hot, dry summers and, mild wet winters. Long-term average annual precipitation is 240 mm, whereas potential evapotranspiration exceeds 1.500 mm, resulting in severe water deficit that necessitates supplemental irrigation for crop production. Soil samples collected from the experimental site (0–20 cm depth) were analyzed for physicochemical properties using standard procedures. Soil reaction was assessed using a saturated soil sample, indicating that the soil was moderately alkaline (pH 8.33) [25]. Electrical conductivity was measured in the saturated paste extract, indicating the absence of salinity (EC: 0.81 mS cm-1). Total limestone content was high (29.71%), potentially limiting micronutrient availability. Soil organic matter content was critically low (0.82%) than the agronomic sufficiency threshold for semi-arid production systems. These physicochemical characteristics necessitate strategic supplementation of organic amendments to restore soil productivity and biological function. Plant materials and experimental design Candia cotton variety served as the plant material. This cultivar is characterized by early maturity, relatively sparse leaf hairiness, a low lint percentage (~ 60–70%), a conical plant architecture, and high productivity potential. The variety has demonstrated superior adaptation to the harsh environmental conditions of the Harran Plain, particularly in emergence vigor and early-season growth under water stress. The experiment was conducted using a randomized complete block design (RCBD) with three replications. Nine fertilizer treatment combinations i.e., (1) control (no fertilizer), (2) 350 kg ha-1 urea (%46), (3) 100 kg ha-1 DAP (%18) + 350 kg ha-1 urea (%46), (4) 200 kg ha-1 DAP (%18) + 350 kg ha-1 urea (%46), (5) 300 kg ha-1 DAP (%18) + 350 kg ha-1 urea (%46), (6) 20 tons ha-1 FYM + 350 kg ha-1 urea (%46), (7) 40 tons ha-1 FYM + 350 kg ha-1 urea (%46), (8) 60 tons ha-1 FYM + 350 kg ha-1 urea (%46), and (9) 100 tons ha-1 FYM without supplemental urea were included in the study [25]. Each experimental plot consisted of seven 12 m-long rows. The inter- and intra-row spacing was maintained at 70 cm and 10–12 cm, respectively. This configuration corresponded to a theoretical plant density of 119,000–143,000 plants ha⁻¹. The final planting density was maintained at ~ 120,000 plants ha⁻¹ after thinning. Blocks were separated by 3 m, and 3 m buffer zones were maintained between adjacent plots to minimize edge effects. In total, the experiment comprised 27 plots arranged as nine fertilizer treatments with three replications. Farmyard manure characterization and application The FYM was obtained from a cattle farm adjacent to the university’s experimental farm. Fresh manure was composted for 45 days prior to field application. The manure was regularly turned to promote decomposition and temperature stabilization. The composted manure was analyzed using standard procedures. Total nitrogen was determined by the Kjeldahl method [26]. Phosphorus was determined by spectrophotometric analysis following acid digestion [27]. Potassium was analyzed by flame photometry [28], and organic carbon by the Walkley–Black method [29]. Composted manure had nitrogen (1.2%), phosphorus (0.8%), potassium (0.9%), and organic carbon (18.5%). The manure was applied to the soil surface and incorporated by shallow cultivation 1 day before sowing. Agronomic and practical importance of the applied FYM rates The FYM application rates were chosen to represent an acceptable range of organic amendment under the semi-arid, calcareous soil conditions of the Harran Plain. The incorporation of 20–60 t ha− 1 FYM with urea was to explore integrated nutrient management strategies that would provide readily available nitrogen, increase soil organic matter, and improve nutrient cycling through manure addition. The FYM 100 t ha-1 treatment without urea was included to examine whether a high organic-input strategy could partially replace mineral nitrogen using locally available livestock manure. Thus, the selected rates aimed to represent both agronomic response and practical manure valorization potential in cotton production systems. Field preparation and crop management practices Land preparation consisted of autumn moldboard plowing to a depth of 20–25 cm and spring secondary tillage with a cultivator to a depth of 10–12 cm. Pendimethalin herbicide (300 g ha-1) was applied 5 days before sowing to control broadleaf and grassy weed species. The soil was disk-harrowed following herbicide application to incorporate it and ensure uniformity. The final field preparation consisted of a light harrowing to obtain seedbed conditions suitable for seeding. Cotton seeds were sown with a pneumatic seed drill on 5 May 2021. The initial planting density was 140,000 seeds ha-1. Thinning was conducted at the 2–4 true-leaf stage to achieve target plant populations of 120,000 seeds ha-1. The crop was irrigated through a drip irrigation system. Irrigation scheduling initiated when soil water tension reached 45–50 centibars (kPa), monitored by tensiometers installed at 30 cm soil depth. It is consistent with recommended thresholds for cotton in semi-arid environments. A total of 13 irrigations were applied throughout the growing season. The crop was irrigated 3 times during the establishment phase (emergence to squaring), 5 times during peak flowering and boll setting, and 5 times during boll development and maturation. A total of 1,200 mm of water was applied throughout the growing season. The dominant weed species in the experimental area were Johnson grass [Sorghum halepense (L.) Pers.] and common cocklebur (Xanthium strumarium L.). Slight variations were noted among experimental plots, which are owed to soil seed bank distribution. Manual and mechanical hoeing were performed 4 times during the growing season to control the weeds. The first hoeing was conducted 4 weeks after emergence (4–6 true-leaf stage), followed by subsequent operations at 5-week intervals. Weed control practices were uniform across all fertilizer treatment plots. Insect pests were controlled based on economic threshold levels of the observed species. The main pests recorded during the study were onion thrips (Thrips tabaci), cotton aphid (Aphis gossypii Glover), two-spotted spider mite (Tetranychus urticae), cotton bollworm (Helicoverpa armigera), tobacco whitefly (Bemisia tabaci), and plant bugs (Creontiades pallidus). A total of 4 insecticide sprays were carried out to manage the recorded pests. Imidacloprid (200 g/L SL; 0.25 L ha-1) was applied at early squaring for aphid and thrips. Abamectin (18 g/L EC; 1.0 L ha-1) was sprayed at peak boll development for spider mite. Emamectin benzoate (19 g/L EC; 0.75 L ha-1) was applied twice at initial flowering and peak boll-setting for bollworm and whitefly. All applications were uniform across treatments. Morphological and growth parameters Morphological and growth traits were recorded on standard phenological stages (60, 75, 90, 105, and 120 days after planting) from ten randomly selected plants per plot. Plant height was measured from the soil surface to the terminal growing point with a ruler. The total number of nodes on the main stem was counted from the cotyledon to the terminal apex. Measurements were conducted in the early morning to standardize plant turgidity. Leaf dry matter accumulation was determined on ten randomly selected plants harvested at each measurement interval. All leaves on the plants were separated, and the fresh weight was recorded immediately after harvest. Samples were subsequently dried at 70 °C for 72 h to constant weight to determine dry matter content. Dried samples were weighed using an analytical balance with 0.01 g precision. Spectral and physiological traits Chlorophyll content was quantified using a Minolta SPAD-502 chlorophyll meter (Minolta Camera Co., Ltd., Osaka, Japan) at different intervals described above. Measurements were conducted on the uppermost fully expanded fifth leaf from the terminal apex on ten plants per plot. Five readings were taken from each leaf and averaged across all plants within a replication to get treatment means. All spectral readings were conducted during mid-morning hours (09:00–11:00) to minimize diurnal variation in chlorophyll fluorescence. The Normalized Difference Vegetation Index (NDVI) was measured using a Trimble GreenSeeker sensor (Trimble Navigation Limited, Sunnyvale, CA), maintained at a height of 76–91 cm above the plant canopy. Readings were recorded at the five intervals on ten randomly selected plants per plot within each replication. High-resolution multispectral imagery was acquired using a Parrot Bluegrass drone equipped with a Micasense camera at four-week intervals throughout the growing season. Imagery was georeferenced and processed to generate plot-specific NDVI and normalized pigment chlorophyll index (NPCI) datasets. Yield and yield components Seed cotton yield was quantified through manual harvest at maturity. Seed cotton yield was determined from manually picking the two central rows of each plot after excluding border rows and the first 1 m at both ends of the sampled rows. The harvested seed cotton was weighed, and the yield was expressed as kg ha⁻¹. The total number of bolls was counted on ten randomly selected plants within each plot at maturity. Boll weight (grams per boll) was determined by weighing individual bolls from representative plants and calculating mean values. Seed cotton weight per boll was determined by weighing harvested seed cotton from measured bolls. Ginning outturn was calculated as the mass of lint fiber divided by total seed cotton mass, expressed as a percentage. Fiber quality traits Lint samples were evaluated for physical and chemical quality traits in accordance with American Society for Testing and Materials (ASTM) standards [30]. High Volume Instrument (HVI) testing was performed on a Uster HVI 1000 A in accordance with NEPS (Neps Extraction and Packing System) protocols. Fiber quality traits, including micronaire (fineness and maturity index, micrometers), fiber length (upper half mean length, mm), fiber strength (tenacity, g/tex), fiber elongation (%), uniformity index, and color parameters (degree of yellowing, reflectance %) were recorded. Measurements were conducted in triplicate on representative fiber samples from each treatment replication. Short fiber index (SFI) and fiber length uniformity were assessed as secondary quality parameters indicative of preparation quality and processing suitability. Economic analysis An economic analysis was conducted to evaluate the cost-benefit relationships of fertilizer treatments. Treatment costs were calculated based on 2021 market prices, i.e., diammonium phosphate ($ 0.45 kg-1), urea ($ 0.28 kg-1), and FYM ($ 3.12 ton-1). Total input costs were compared with yield improvements relative to the control treatment to determine benefit-cost ratios and return on investment for each treatment. Fertilizer prices, labor costs, and variability in market conditions may influence the relative profitability of the treatments. Therefore, the benefit–cost ratios reported in this study should be evaluated within the context of the local economic conditions prevailing during the study period. Statistical analysis All data were analyzed with R statistical software [31]. The experiment used a randomized complete block design (RCBD) for traits assessed at a single time point (yield components, yield traits, fiber quality, and color characteristics). Treatments were taken as fixed effects, and replication as a random blocking factor. Traits measured at different intervals during crop development were analyzed using a repeated-measures analysis of variance (ANOVA). Fertilizer treatment, sampling time, and their interaction were treated as fixed effects, with replication as a blocking factor and plot as the repeated experimental unit. The ANOVA was performed for each trait separately [32]. Treatment means were separated using Tukey’s honestly significant difference (HSD) test, where ANOVA indicated a significant treatment effect. The coefficient of determination (R²) was computed to quantify the fraction of variance elucidated by the model. Data recorded at different time intervals were plotted against time. Pearson correlation coefficients were computed using treatment means to investigate correlations among attributes and were presented as a correlation heatmap. Furthermore, principal component analysis (PCA) was conducted on standardized trait means to capture multivariate variance and identify the characteristics most strongly associated with treatment differences. Results Most of the yield-related traits were significantly altered by fertilizer treatments, whereas fiber quality and color traits remained unaffected (Table 1). Significant treatment effects were recorded for number of bolls (F = 3.19, p = 0.023; R² = 0.62), boll weight (F = 23.54, p < 0.001; R² = 0.92), boll seed cotton weight (F = 12.37, p < 0.001; R² = 0.86), seed cotton yield (F = 100.05, p < 0.001; R² = 0.98), ginning outturn (F = 13.57, p < 0.001; R² = 0.87), and fiber yield (F = 73.28, p < 0.001; R² = 0.97). In contrast, micronaire, fiber strength, fiber length, fiber uniformity, elongation, reflectance, and yellowness were not significantly affected by fertilizer treatments (p > 0.05). Fertilizer management mainly improved yield formation without causing statistically noticeable changes in fiber quality (Table 1). Table 1Analysis of variance (ANOVA) for yield and yield components, and fiber quality traits of cotton under different fertilizer treatmentsTraitF valuep value R 2 Number of Bolls3.190.023*0.62Boll Weight (g)23.540.000*0.92Boll Seed cotton (g)12.370.000*0.86Seed Cotton Yield (kg ha− 1)100.050.000*0.98Ginning outturn (%)13.570.000*0.87Fiber Yield (kg ha− 1)73.280.000*0.97Micronair0.850.578NS0.38Fiber Strength (g tex− 1)2.330.071NS0.59Fiber Length1.320.303NS0.51Fiber Uniformity (%),2.420.063NS0.58Elongation (%)0.580.776NS0.33Reflectance1.370.279NS0.43Yellowness2.330.071NS0.56R² indicates the proportion of the total variation in each trait that the model explains. Traits with p < 0.05 (*) are considered significantly affected by fertilizer treatments. NS non-significant Table 1Analysis of variance (ANOVA) for yield and yield components, and fiber quality traits of cotton under different fertilizer treatmentsTraitF valuep value R 2 Number of Bolls3.190.023*0.62Boll Weight (g)23.540.000*0.92Boll Seed cotton (g)12.370.000*0.86Seed Cotton Yield (kg ha− 1)100.050.000*0.98Ginning outturn (%)13.570.000*0.87Fiber Yield (kg ha− 1)73.280.000*0.97Micronair0.850.578NS0.38Fiber Strength (g tex− 1)2.330.071NS0.59Fiber Length1.320.303NS0.51Fiber Uniformity (%),2.420.063NS0.58Elongation (%)0.580.776NS0.33Reflectance1.370.279NS0.43Yellowness2.330.071NS0.56R² indicates the proportion of the total variation in each trait that the model explains. Traits with p < 0.05 (*) are considered significantly affected by fertilizer treatments. NS non-significant Analysis of variance (ANOVA) for yield and yield components, and fiber quality traits of cotton under different fertilizer treatments R² indicates the proportion of the total variation in each trait that the model explains. Traits with p < 0.05 (*) are considered significantly affected by fertilizer treatments. NS non-significant Growth and physiological traits showed treatment-dependent variation across the season (Fig. 1). Treatment effects were significant at all sampling dates for plant height, SPAD chlorophyll content, leaf fresh weight, and leaf dry weight (Tables S1 and S2). Similarly, NDVI and the number of nodes also differed significantly among treatments at most sampling dates. Plant height increased progressively with crop development. The strongest late-season responses were generally observed for DAP300 + U and FYM-based treatments. The NDVI increased until 105 days after sowing and declined thereafter, reflecting canopy maturation. Fresh and dry leaf biomass also increased significantly over time, confirming that nutrient application enhanced vegetative development compared with the unfertilized control. Fig. 1Seasonal changes in cotton growth and physiological traits under different fertilizer treatments. Values represent treatment means, and error bars indicate ± standard error. Asterisks above sampling dates indicate significant differences among fertilizer treatments at the corresponding sampling time based on repeated-measures ANOVA/post hoc comparison: *p < 0.05, **p < 0.01, and ***p < 0.001. Treatment abbreviations: U = urea, DAP = diammonium phosphate, and FYM = farmyard manure Seasonal changes in cotton growth and physiological traits under different fertilizer treatments. Values represent treatment means, and error bars indicate ± standard error. Asterisks above sampling dates indicate significant differences among fertilizer treatments at the corresponding sampling time based on repeated-measures ANOVA/post hoc comparison: *p < 0.05, **p < 0.01, and ***p < 0.001. Treatment abbreviations: U = urea, DAP = diammonium phosphate, and FYM = farmyard manure Yield traits exhibited significant responses to fertilizer treatments (Fig. 2). The highest seed cotton yield (5207.3 kg ha⁻¹) was recorded under FYM60 + U, representing a 32.9% increase over the control. The lowest yield was recorded for the control treatment. Fiber yield followed a similar pattern. The FYM60 + U produced the highest (2292.9 kg ha⁻¹) fiber yield, followed by FYM40 + U and FYM100 treatments, indicating that higher FYM inputs consistently improved lint production. Fig. 2Effects of fertilizer treatments on cotton yield components and yield-related traits, including number of bolls, boll weight, boll seed cotton weight, seed cotton yield, ginning outturn, and fiber yield. Different lowercase letters above the boxes indicate statistically significant differences among treatments based on Tukey’s HSD test at p < 0.05. The p-value displayed in each panel represents the overall treatment effect from one-way ANOVA Effects of fertilizer treatments on cotton yield components and yield-related traits, including number of bolls, boll weight, boll seed cotton weight, seed cotton yield, ginning outturn, and fiber yield. Different lowercase letters above the boxes indicate statistically significant differences among treatments based on Tukey’s HSD test at p < 0.05. The p-value displayed in each panel represents the overall treatment effect from one-way ANOVA Boll-related traits also exhibited the yield advantage of FYM60 + U treatment. The highest boll weight (6.70 g) was recorded for FYM60 + U treatment. Similarly, the highest boll seed cotton weight (4.82 g) was also noted for this treatment. Likewise, the FYM60 + U treatment resulted in the highest ginning outturn, although it was statistically comparable to several FYM-based treatments. The number of bolls showed a weaker but significant response. Overall, FYM60 + U resulted in heavier bolls, higher boll seed cotton weight, improved ginning outturn, and higher fiber yield (Fig. 2). Fiber quality traits were statistically similar across fertilizer treatments (Fig. 3). Although small differences were observed in micronaire, fiber strength, fiber length, fiber uniformity, and elongation, none of these traits showed a significant treatment response (Table 1). Therefore, the variation among fertilizer treatments should not be interpreted as a true fertilizer effect. These results indicate that the yield gains obtained with FYM-based fertilization (particularly FYM60 + U) did not compromise fiber quality. Fig. 3Effects of fertilizer treatments on cotton fiber quality traits, including micronaire, fiber strength, fiber length, fiber uniformity, and elongation. The p-value displayed in each panel represents the overall treatment effect from one-way ANOVA. Since fertilizer treatments did not significantly affect any fiber quality trait (p > 0.05), observed numerical differences among treatments should be interpreted as non-significant Effects of fertilizer treatments on cotton fiber quality traits, including micronaire, fiber strength, fiber length, fiber uniformity, and elongation. The p-value displayed in each panel represents the overall treatment effect from one-way ANOVA. Since fertilizer treatments did not significantly affect any fiber quality trait (p > 0.05), observed numerical differences among treatments should be interpreted as non-significant Fiber color traits also remained unaffected by fertilizer treatments (Fig. 4). Reflectance varied within a narrow range and showed no significant treatment effect (p = 0.279). Similarly, yellowness was not significantly affected (p = 0.071), despite minor numerical differences among treatments. Thus, fertilizer-induced increases in productivity were achieved without detectable changes in fiber color attributes. Fig. 4Effects of fertilizer treatments on cotton fiber color attributes, including reflectance and yellowness. The p-value displayed in each panel represents the overall treatment effect from one-way ANOVA. Since fertilizer treatments did not significantly affect reflectance or yellowness (p > 0.05), the observed numerical differences among treatments should be interpreted as non-significant Effects of fertilizer treatments on cotton fiber color attributes, including reflectance and yellowness. The p-value displayed in each panel represents the overall treatment effect from one-way ANOVA. Since fertilizer treatments did not significantly affect reflectance or yellowness (p > 0.05), the observed numerical differences among treatments should be interpreted as non-significant Correlation analysis confirmed that yield improvement was mainly associated with boll development and lint production capacity (Fig. 5). Seed cotton yield was strongly and positively correlated with fiber yield (r = 0.99, p < 0.001), ginning outturn (r = 0.88, p < 0.01), boll seed cotton weight (r = 0.82, p < 0.01), and boll weight (r = 0.78, p < 0.05). Fiber yield was also strongly correlated with ginning outturn (r = 0.93, p < 0.001), boll seed cotton weight (r = 0.83, p < 0.01), and boll weight (r = 0.80, p < 0.05). Among vegetative traits, plant height was positively correlated with the number of nodes and NDVI, while NDVI was strongly associated with node development. In contrast, most fiber quality traits showed weak or non-significant relationships with yield traits, further supporting that productivity gains were not accompanied by deterioration in fiber quality. Fig. 5Pearson correlation heatmap showing relationships among cotton growth, physiological, yield, fiber quality, and fiber color traits under different fertilizer treatments. Values within cells represent Pearson correlation coefficients (r), with positive correlations shown in warm colors and negative correlations in cool colors; stronger color intensity indicates a stronger association. Asterisks indicate statistical significance: *p < 0.05, **p < 0.01, and ***p < 0.001. Non-significant correlations are shown without asterisks Pearson correlation heatmap showing relationships among cotton growth, physiological, yield, fiber quality, and fiber color traits under different fertilizer treatments. Values within cells represent Pearson correlation coefficients (r), with positive correlations shown in warm colors and negative correlations in cool colors; stronger color intensity indicates a stronger association. Asterisks indicate statistical significance: *p < 0.05, **p < 0.01, and ***p < 0.001. Non-significant correlations are shown without asterisks The PCA further supported the separation of treatments according to productivity-related traits (Fig. 6). The first two principal components explained 61.8% of the total variation, with PC1 accounting for 36.2% and PC2 for 25.6%. The FYM60 + U treatment was positioned in the direction of seed cotton yield, fiber yield, ginning outturn, boll weight, and boll seed cotton weight, confirming its strong association with yield performance. FYM100 was more closely associated with fiber length and reflectance, whereas the control was separated from high-yielding FYM treatments and aligned more closely with fiber strength. These multivariate patterns indicate that treatment differentiation was driven mainly by yield and boll-related traits rather than fiber quality attributes. Fig. 6Principal component analysis (PCA) biplot showing the multivariate relationships among fertilizer treatments and recorded traits. Points represent fertilizer treatments based on standardized trait means, while arrows indicate the direction and relative contribution of individual traits to treatment separation. Longer arrows indicate stronger contributions to the principal components, and smaller angles between arrows indicate stronger positive associations among traits Principal component analysis (PCA) biplot showing the multivariate relationships among fertilizer treatments and recorded traits. Points represent fertilizer treatments based on standardized trait means, while arrows indicate the direction and relative contribution of individual traits to treatment separation. Longer arrows indicate stronger contributions to the principal components, and smaller angles between arrows indicate stronger positive associations among traits Economic performance significantly differed among fertilizer treatments (Table 2). The FYM60 + U generated the highest gross income (1903.80 $ ha⁻¹) followed by FYM40 + U and FYM100. Net income was also highest under FYM60 + U (691.66 $ ha⁻¹). The lowest net incomes were recorded under DAP200 + U and DAP300 + U. The highest BCR was obtained with FYM60 + U (1.57). The FYM60 + U achieved the highest yields, fiber yields, gross income, and net income, making it the most favorable treatment in agronomic and economic terms. Overall, FYM60 + U provided the best balance between productivity gain and economic return under the conditions of this study. Table 2Economic performance of cotton production under different fertilizer management strategiesTreatmentsGross Income$ (ha)Expenditures$ (ha)Net Income$ (ha)BCRControl1432.60 f925.24507.36 bc1.55 abUrea (350 kg ha− 1)1588.53 cd1033.57554.96 b1.54 abDAP 100 kg ha− 1 + Urea1518.92 de1076.43442.49 cd1.41 deDAP 200 kg ha− 1 + Urea1506.27 ef1119.29386.98 d1.35 efDAP 300 kg ha− 1 + Urea1555.69 cde1162.14393.55 d1.34 fFYM 20 t ha− 1 + Urea1600.72 c1093.10507.62 bc1.46 cdFYM 40 t ha− 1 + Urea1711.62 b1152.62559.00 b1.48 bcFYM 60 t ha− 1+ Urea1903.80 a1212.14691.66 a1.57 aFYM 100 t ha− 11774.62 b1222.86551.76 b1.45 cdBCR benefit: cost ratio, Means followed by different letters within a column are statistically different from each other (p < 0.05) Table 2Economic performance of cotton production under different fertilizer management strategiesTreatmentsGross Income$ (ha)Expenditures$ (ha)Net Income$ (ha)BCRControl1432.60 f925.24507.36 bc1.55 abUrea (350 kg ha− 1)1588.53 cd1033.57554.96 b1.54 abDAP 100 kg ha− 1 + Urea1518.92 de1076.43442.49 cd1.41 deDAP 200 kg ha− 1 + Urea1506.27 ef1119.29386.98 d1.35 efDAP 300 kg ha− 1 + Urea1555.69 cde1162.14393.55 d1.34 fFYM 20 t ha− 1 + Urea1600.72 c1093.10507.62 bc1.46 cdFYM 40 t ha− 1 + Urea1711.62 b1152.62559.00 b1.48 bcFYM 60 t ha− 1+ Urea1903.80 a1212.14691.66 a1.57 aFYM 100 t ha− 11774.62 b1222.86551.76 b1.45 cdBCR benefit: cost ratio, Means followed by different letters within a column are statistically different from each other (p < 0.05) Economic performance of cotton production under different fertilizer management strategies BCR benefit: cost ratio, Means followed by different letters within a column are statistically different from each other (p < 0.05) Discussion The current study indicated that the integrated nutrient management improved cotton productivity under semi-arid, calcareous soil conditions by increasing yield without altering fiber quality. Fertilizer treatments had a significant effect on the number of bolls, boll weight, boll seed cotton weight, seed cotton yield, ginning outturn, and fiber yield. However, micronaire, fiber strength, fiber length, fiber uniformity, elongation, reflectance, and yellowness remained unchanged under applied fertilizer treatments. This suggests that nutrient management mainly influenced reproductive development and lint production capacity, whereas fiber quality traits remained stable under the tested fertilization regimes. Cotton nutrient-management studies have reported that yield components and lint yield are more sensitive to fertilization than fiber quality traits when nutrient supply is in acceptable ranges from an agronomic perspective [33, 34]. Seasonal growth responses indicated that fertilizer treatments improved vegetative and physiological performance, but the final yield ranking could not be explained solely by the canopy development. The DAP300 + U and several FYM-based treatments resulted in strong responses in late-season growth parameters (plant height, NDVI, SPAD chlorophyll content, and leaf biomass). However, FYM60 + U produced higher seed cotton and fiber yields than the highest DAP rate. This suggests that greater canopy greenness or vegetative growth may not necessarily increase lint yield without adequate reproductive sink strength, nutrient synchrony, and boll development. NDVI and SPAD are useful indicators of canopy vigor and nitrogen status, but their relationship with yield may depend on nutrient source, soil constraints, crop stage and assimilate partitioning between vegetative and reproductive organs [35–37]. The variation observed between canopy vigor and final yield indicates that nitrogen-driven vegetative growth may not translate into reproductive efficiency when soil limitations and nutrient synchrony constrain nutrient partitioning and sink strength [35–37]. The variation in canopy development between treatments might be partly explained by differences in phosphorus availability as phosphorus plays an important role in reproductive organ formation, assimilate partitioning and biomass allocation in cotton, although cultivar sensitivity to phosphorus supply might vary [38, 39]. However, this interpretation should be approached with caution because the current treatments were not designed to provide equivalent nitrogen and phosphorus inputs across fertilizer sources. For example, the total nutrient supply from DAP and urea was not directly offset by nutrients released through FYM mineralization and urea application. This distinction is particularly relevant in alkaline, calcareous soils where P availability is often limited by sorption and precipitation reactions [38, 40]. Under these conditions, organic amendments can enhance phosphorus mobility by releasing organic acids and chelating calcium, thus reducing phosphorus fixation and increasing plant-available phosphorus [38, 40]. However, the fiber quality traits were not significantly affected by the fertilizer treatments, despite the nutrient-related effects on productivity. This suggests that the yield improvements obtained through FYM and mineral fertilization were achieved without compromising standard fiber quality attributes. Similar responses have been reported in long-term studies of cotton nutrient management. Yield and lint production are usually more responsive to fertilization than fiber quality traits, although fiber quality may be affected under specific environmental conditions or at extreme nutrient rates [34, 41]. The greater efficiency of FYM60 + U could be attributed to the synergistic effect of organic and mineral nutrient sources. The immediate supply of nitrogen from urea was likely responsible for early vegetative growth and reproductive development. The FYM produced a slower release of nutrients and improved the soil environment. Under the alkaline and calcareous conditions at the experimental site, FYM may have improved soil structure, water retention, microbial activity, and nutrient cycling, thereby increasing root function and nutrient uptake during boll formation. Organic amendments might also affect phosphorus availability in calcareous soils by the release of organic acids and by the enhancement of chelation reactions that reduce calcium-induced phosphorus fixation. These mechanisms are particularly important in soils with low organic matter content, where the capacity to buffer nutrients, aggregate stability, and biological activity are limited [33, 42]. This aligns with a recent study on arid/semi-arid cotton cultivation, which reported significant improvements in yield and canopy indices with FYM applications [43]. These improvements are likely linked to increased soil organic carbon, aggregation, and water retention, as well as stimulated microbial activity, which leads to improved nutrient cycling and root nutrient uptake efficiency in semi-arid conditions [33, 42]. The yield-component changes observed in the present study are consistent with the known agronomic responses expected when organic amendments improve the mineral nutrition in soils low in organic matter [44]. Synchronized nutrient release from FYM and rapid nitrogen availability from urea may enhance nutrient-use synchrony, especially during reproductive stages, improving assimilate partitioning towards boll development [44]. The higher yield obtained under FYM60 + U was mainly due to better boll development and lint production. Treatment increased boll weight, boll seed cotton weight, ginning outturn, seed cotton yield, d and fiber yield, indicating that combined organic–mineral nutrient supply improved both reproductive sink development and lint recovery. This interpretation was further strengthened by the correlation analysis, where seed cotton yield was highly and positively correlated with fiber yield, ginning outturn, boll seed cotton weight, and boll weight. Similarly, PCA separated FYM60 + U along the direction of seed cotton yield, fiber yield, ginning outturn, and boll-related traits. This indicates that treatment differentiation was primarily driven by productivity-related variables rather than fiber quality traits. The results showed that the major agronomic value of FYM-based integrated fertilization was its support for boll filling and lint yield rather than for vegetative biomass. Non-significant effects of applied treatments on fiber quality and color traits are important for production perspective. Small differences were observed among treatments, but none of the fiber quality traits responded significantly to fertilization. This suggests that the large yield increase under FYM60 + U was not associated with deterioration in micronaire, fiber strength, fiber length, uniformity, elongation, reflectance, or yellowness. Genotype, temperature, water status, boll position, and maturation conditions strongly influence fiber quality and could be less responsive than yield to moderate changes in nutrient management. Thus, the present findings suggest that FYM-based integrated nutrient management can increase productivity with commercially acceptable fiber quality under the tested conditions. The economic results were in line with the agronomic results. FYM60 + U treatment had the highest gross income, net income, and BCR, indicating that the yield advantage was sufficient to offset its higher input costs. However, the relatively high BCR values observed in the control and urea-only treatments suggest that profitability should not be determined by BCR alone. Lower input treatments may achieve acceptable BCR values due to lower costs, but their net income and total production value remain lower than those of the best-performing FYM treatment. From the practical point of view, FYM60 + U was the best in yield improvement and economic return. However, the profitability of FYM-based fertilization depends on the availability of manure, transport distance, labor costs, application costs, and local fertilizer and cotton prices. These factors need to be taken into account when recommending large-scale adoption [44]. There are several limitations to consider in interpreting these results. The experiment was conducted over a single growing season at a single location, and treatment performance may vary across years, soils, irrigation schedules, or climatic patterns. Second, the results suggest improved nutrient synchrony and soil biological functioning under FYM-based fertilization, but soil enzyme activity, microbial biomass, mineral N release, plant nutrient uptake, and soil water dynamics were not directly quantified. Thirdly, the treatments were not on a nutrient-equivalent basis, so differences between FYM, DAP, and urea treatments reflect practical fertilizer regimes rather than strictly balanced N and P supply. Finally, the economic analysis was based on local 2021 prices, and profitability may change with manure handling costs, fertilizer prices, labor costs, and cotton market value. Future multi-year and multi-location studies quantifying soil organic matter dynamics, plant-available nutrients, nutrient-use efficiency, microbial and enzyme activity and long-term economic performance should precede widespread recommendations for semi-arid cotton systems. Conclusions This study showed that fertilizer management significantly influenced cotton productivity under semi-arid, calcareous soil conditions, while fiber quality and color traits remained statistically unaffected. The FYM at 60 t ha⁻¹ combined with urea produced the highest seed cotton yield and fiber yield through improvements in boll weight, boll seed cotton weight, ginning outturn, and lint production. This treatment also generated the highest gross income, net income, and benefit–cost ratio, indicating that integrating FYM with mineral nitrogen provided the most favorable balance between agronomic performance and economic return. The findings suggest that FYM-based integrated nutrient management can improve cotton productivity and profitability without compromising fiber quality, while also supporting the agricultural reuse of locally available livestock manure. However, these results should be interpreted as site-specific, as the experiment was conducted over a single growing season at a single location. Multi-year, multi-location studies that quantify soil organic matter dynamics, plant-available nutrients, nutrient use efficiency, and long-term economic performance are needed before broad recommendations can be made for semi-arid cotton production systems. Supplementary Information Supplementary Material 1. Supplementary Material 1.