Bioeconomic optimization of zinc and iron agronomic biofortification in cowpea (Vigna unguiculata [L.] Walp.) in Mozambique under tropical field conditions Authors: Tancredo José Carlos, Leonel Domingos Moiana, Omar Jorge Sabbag, Boniface Hussein J. Massawe, Akwilina Wendelin Mwanri, Kallunde Pilly Sibuga, Muhammad Amjad Ali, Muhammad Amjad Ali, Muhammad Amjad Ali Published: 2026 ABSTRACT Agronomic biofortification with zinc (Zn) and iron (Fe) has emerged as a promising approach to enhance crop productivity while improving food and nutritional security. However, fertilizer recommendations based exclusively on maximum biological yield may not maximize farmers’ economic returns. This study integrated agronomic and economic analyses to identify the biologically and economically optimal Zn and Fe fertilization strategies for cowpea (Vigna unguiculata [L.] Walp.) under tropical field conditions in Mozambique. A randomized complete block design with a 3 × 4 × 4 factorial arrangement was used to evaluate three fertilization methods (soil, foliar, and combined soil + foliar) and four application rates of elemental Zn and Fe. Grain yield response was adequately described by a second-order polynomial regression model (R2 = 0.767), with maximum technical efficiency predicted at 18.92 kg ha ⁻ ¹ Zn and 19.17 kg ha ⁻ ¹ Fe. In contrast, maximum economic efficiency was achieved at lower application rates (16.94 kg ha ⁻ ¹ Zn and 10.09 kg ha ⁻ ¹ Fe), reducing Fe input by 47.4% while maintaining more than 97% of the maximum predicted grain yield and increasing net profit. Across the evaluated treatments, the combined soil + foliar application produced greater grain yield and agronomic efficiency than either application method alone. Economic robustness was further evaluated through break-even, sensitivity, and stochastic risk analyses, which indicated positive economic performance across the evaluated treatnebts under the agronomic, environmental, and economic conditions considered in this study. Within the agronomic, environmental, and economic conditions evaluated in this single-site, single-season study, the economic optimum was reached before the biological optimum, demonstrating that integrating agronomic and economic optimization can substantially reduce micronutrient inputs while maintaining high productivity and profitability. These findings provide an evidence-based framework for developing economically efficient Zn and Fe fertilization strategies for cowpea under tropical smallholder farming systems. Introduction Cowpea (Vigna unguiculata (L.) Walp.) is one of the most important grain legumes cultivated in tropical and subtropical regions, particularly in Sub-Saharan Africa (SSA), where it plays a key role in food security, household income, and sustainable agricultural production [1,2]. The crop is widely grown by smallholder farmers because of its remarkable adaptation to drought-prone environments, short growth cycle, and ability to fix atmospheric nitrogen, thereby improving soil fertility while reducing dependence on synthetic nitrogen fertilizers [1,3]. In addition to its agronomic importance, cowpea grains constitute an important source of dietary protein, vitamins, and essential minerals for millions of people whose diets rely heavily on plant-based foods [4,5]. Despite its importance, cowpea productivity in many tropical production systems remains considerably below its genetic potential. Among the major constraints, deficiencies of essential micronutrients, particularly zinc (Zn) and iron (Fe), are common in highly weathered tropical soils characterized by low organic matter content, poor nutrient retention, and limited micronutrient availability [6]. These nutritional limitations reduce plant growth, impair physiological processes associated with biomass accumulation and reproductive development, and decrease nutrient-use efficiency, ultimately reducing grain yield and farm profitability [6,7]. Similar micronutrient constraints have been widely reported for tropical agricultural systems across Sub-Saharan Africa, emphasizing the need for locally adapted nutrient management strategies. Micronutrient deficiencies in agricultural soils are closely linked to human nutrition. Zinc and iron deficiencies are among the leading causes of hidden hunger worldwide, affecting more than two billion people and contributing to impaired immune function, anemia, reduced cognitive development, and increased health risks, particularly among women and children in low-income countries [8,9]. Previous research further emphasized that adequate dietary Fe and Zn are essential for cognitive development, neurological function, healthy growth, and overall human well-being, reinforcing the importance of increasing the micronutrient density of staple foods through sustainable agricultural strategies [10]. Therefore, improving both the productivity and nutritional quality of staple crops has become a major component of global food and nutrition security initiatives. Agronomic biofortification, based on the application of mineral fertilizers containing essential micronutrients, has emerged as a practical and rapidly deployable strategy for simultaneously improving crop performance and increasing the concentration of micronutrients in edible plant tissues [7,11,12]. Unlike conventional breeding, agronomic biofortification can be implemented immediately within existing production systems and therefore provides rapid benefits to farmers while complementing long-term genetic improvement programs. Younas et al. (2025) showed that integrated nitrogen and iron fertilization can improve grain Fe concentration and bioavailability while maintaining agronomic performance in bread wheat. Numerous studies have reported positive effects of Zn and Fe fertilization on crop growth, grain yield, nutrient-use efficiency, and grain micronutrient concentration [9,14,15].Likewise, Alkarawi [16] reported that the combined application of Fe- and Zn-based fertilizers significantly improved vegetative growth, nutrient uptake, biomass production, and grain yield in soybean, demonstrating the complementary physiological roles of these micronutrients in legume productivity. Nevertheless, crop responses remain highly dependent on fertilizer source, application method, nutrient interactions, soil characteristics, and environmental conditions, highlighting the importance of developing site-specific fertilizer recommendations. Recent studies have further demonstrated that balanced micronutrient nutrition enhances several physiological processes associated with crop productivity, including photosynthesis, antioxidant defense, enzyme activity, and nutrient metabolism. For example, Muhae-Ud-Din et al. [17] reported that micronutrient-based nanofertilizers improved photosynthetic performance, antioxidant enzyme activity, nitrogen metabolism, seed quality, and grain yield, providing further evidence that integrated micronutrient management promotes multiple physiological processes involved in crop productivity. Although the magnitude of these responses varies among crops and production environments, these findings collectively reinforce the importance of balanced Zn and Fe nutrition for sustainable crop production. Although the agronomic benefits of Zn and Fe fertilization are well documented, fertilizer recommendations are still predominantly based on maximizing biological yield. For smallholder farmers, however, fertilizer management decisions depend not only on productivity but also on production costs, expected economic returns, and investment risks. This consideration was particularly relevant in the present study because Fe-EDTA was substantially more expensive than the Zn fertilizer source and therefore represented a major component of the total micronutrient fertilization cost. Younas et al. [13] emphasized that fertilizer management should simultaneously consider agronomic performance and grain nutritional quality to maximize the practical benefits of biofortification programs. However, fertilizer rates that maximize biological productivity do not necessarily maximize economic return when input costs and crop prices are explicitly considered [18]. The distinction between Maximum Technical Efficiency (MTE) and Maximum Economic Efficiency (MEE) has therefore become increasingly important for developing economically viable fertilizer recommendations. While MTE identifies the nutrient combination that maximizes crop yield, MEE identifies the combination that maximizes economic return by simultaneously considering crop response and production costs [18]. Despite its practical importance, relatively few studies have integrated agronomic and economic optimization to establish micronutrient fertilizer recommendations for grain legumes cultivated under tropical field conditions. Response Surface Methodology (RSM) provides a robust analytical framework for evaluating nonlinear crop responses, nutrient interactions, and optimization criteria simultaneously. By integrating biological and economic response functions, RSM allows the identification of fertilizer combinations that maximize both productivity and profitability, providing more realistic recommendations for smallholder farming systems than yield-based approaches alone [19]. Furthermore, response surface models enable the quantification of nutrient interactions and the estimation of biologically and economically optimal fertilizer rates, thereby supporting evidence-based nutrient management strategies under contrasting production conditions. Therefore, this study aimed to integrate agronomic and economic analyses to identify the biologically and economically optimal combinations of Zn and Fe fertilization for cowpea production under tropical field conditions in Mozambique using Response Surface Methodology (RSM). Specifically, the study sought to quantify the divergence between Maximum Technical Efficiency (MTE) and Maximum Economic Efficiency (MEE) and to identify fertilization strategies that maximize farm profitability while maintaining high grain productivity and promoting more sustainable micronutrient management for smallholder farming systems. Materials and methods Study area description The field experiment was conducted during the 2025 rainy season at the experimental station of the Instituto de Investigação Agrária de Moçambique (IIAM), located in Namacurra District, Zambezia Province, Mozambique (17°19′29″ S, 37°01′00″ E; 35 m above sea level) (Fig 1). Location of the experimental site at the Instituto de Investigação Agrária de Moçambique (IIAM), Namacurra, Zambézia Province, Mozambique. According to the Köppen–Geiger climate classification, the study area has a tropical savanna climate (Aw), characterized by a warm rainy season and a distinct dry season. Meteorological conditions were monitored throughout the crop cycle using data from the Quelimane meteorological station, the nearest station to the experimental site. During the growing season, the mean maximum and minimum air temperatures were 30.18 and 22.11 °C, respectively. Relative humidity ranged from 62% to 94%, while cumulative rainfall reached 1,471.8 mm. Monthly variations in air temperature, rainfall, and relative humidity during the experimental period are presented in Figs 2 and 3. To assess whether the 2025 growing season was climatically representative of the region, monthly mean temperature and rainfall during the crop cycle (April–July) were compared with the long-term climate normals of the Quelimane meteorological station. Monthly variation of maximum and minimum air temperature (°C) during the cowpea growing season. Monthly variation of relative humidity (%) and total precipitation (mm) during the cowpea growing season. Soil Analysis Prior to crop establishment, composite soil samples were collected from the 0–30 cm soil layer to characterize the initial physicochemical properties of the experimental area. Sampling was performed using a Dutch auger following a systematic zigzag pattern to obtain representative samples across the field. Soil analyses were conducted according to standard laboratory procedures. Organic carbon was determined by the Walkley–Black method, total nitrogen by the Kjeldahl method, available phosphorus by the Bray I extraction method, exchangeable potassium by ammonium acetate (NH₄OAc) extraction, available sulfur by CaCl₂ extraction, and available Zn and Fe by DTPA extraction. The experimental soil had a sandy-loam texture with a pH of 6.9. Chemical analysis indicated low concentrations of organic carbon (<17 g kg ⁻ ¹), available phosphorus (<2.0 mg kg ⁻ ¹), and DTPA-extractable Zn (0.1 mg kg ⁻ ¹), indicating Zn deficiency. In contrast, exchangeable potassium (41 mg kg ⁻ ¹) and DTPA-extractable Fe (31.0 mg kg ⁻ ¹) were considered adequate, whereas total nitrogen ranged from 1.0 to 2.0 g kg ⁻ ¹. According to the National Soil Map of Mozambique, the study region is predominantly characterized by Arenosols, Ferralsols, Lixisols, and Acrisols. However, no site-specific pedological survey or WRB/USDA soil classification was available for the experimental field. Therefore, the soil was characterized based on its measured texture and physicochemical properties. The complete physicochemical characterization of the soil, together with the analytical methods used, is presented in Table 1. 10.1371/journal.pone.0357473.t001Table 1Initial physicochemical characteristics of the experimental soil (0–30 cm depth).Soil propertyUnitValueAnalytical methodTexture class—Sandy loamHydrometer methodpH (H₂O)—6.9PotentiometricOrganic carbong kg ⁻ ¹<17Walkley–BlackTotal nitrogeng kg ⁻ ¹1–2KjeldahlAvailable phosphorusmg kg ⁻ ¹<2.0Bray IAvailable sulfurmg kg ⁻ ¹5.0CaCl₂ extractionExchangeable potassiummg kg ⁻ ¹41NH₄OAc extractionDTPA-extractable Znmg kg ⁻ ¹0.1DTPA extractionDTPA-extractable Femg kg ⁻ ¹31.0DTPA extraction Initial physicochemical characteristics of the experimental soil (0–30 cm depth). Experimental design and treatments The experiment was established using a randomized complete block design (RCBD) in a 3 × 4 × 4 factorial arrangement, with four replications, resulting in a total of 192 experimental plots. Treatments were randomly allocated within each block using a table of random numbers to minimize the influence of spatial variability across the experimental area [20]. Three experimental factors were evaluated: (i) fertilization mode, consisting of soil application, foliar application, and combined soil plus foliar application; (ii) zinc fertilization, applied as zinc sulfate (ZnSO₄·H₂O, 35% Zn) at rates of 0, 7.5, 15.0, and 22.5 kg Zn ha ⁻ ¹; and (iii) iron fertilization, applied as Fe–EDTA (13% Fe) at rates of 0, 5.0, 15.0, and 25.0 kg Fe ha ⁻ ¹. The selected Zn and Fe application rates were established based on the initial soil fertility assessment and previous agronomic recommendations for micronutrient management under tropical conditions. The range of application rates was designed to correct the identified Zn deficiency while capturing crop responses across agronomically relevant fertilizer levels, thereby enabling the estimation of both Maximum Technical Efficiency (MTE) and Maximum Economic Efficiency (MEE) using Response Surface Methodology. Crop establishment and management The experimental area was prepared by conventional tillage, consisting of plowing followed by harrowing to provide a uniform seedbed for crop establishment. Meteorological conditions were monitored throughout the experimental period to characterize the environmental conditions under which the study was conducted. Because the experiment was carried out under rainfed tropical conditions, soil moisture was not continuously monitored. However, rainfall distribution was recorded during the growing season, and supplemental irrigation was applied during the pod development stage whenever necessary to prevent water stress and ensure adequate grain filling. Each experimental plot consisted of 11 rows, each 3.4 m long, spaced 0.30 m apart, with 0.20 m between plants within the row, resulting in a total plot area of 5.95 m2. To minimize border effects, grain yield and all agronomic measurements were obtained from the net plot area (4.02 m2), which excluded the two outer rows and 0.30 m from both ends of each row. Cowpea (Vigna unguiculata [L.] Walp.) cultivar IT-16 was manually sown on 12 April 2025. Two seeds were placed per planting hole at a depth of approximately 2 cm. Basal fertilization was applied at sowing according to the results of the pre-planting soil analysis and consisted of 155 kg ha ⁻ ¹ monoammonium phosphate (MAP), 40 kg ha ⁻ ¹ muriate of potash (MOP), and 5 kg ha ⁻ ¹ urea, ensuring that macronutrient supply was not limiting during crop development. Micronutrient treatments consisted of zinc supplied as zinc sulfate monohydrate (ZnSO₄·H₂O, 35% Zn) and iron supplied as Fe–EDTA (13% Fe). Soil applications were incorporated into the planting furrows at sowing, whereas foliar applications were divided into two equal applications at 40 and 45 days after planting (DAP), corresponding to the beginning (R1) and full flowering stages, respectively. This application schedule was adopted to synchronize micronutrient supply with periods of high physiological demand [15]. Micronutrient solutions were prepared using deionized water to avoid ionic interference and applied with a calibrated manual backpack sprayer. All foliar applications were performed during the late afternoon to reduce evaporative losses and minimize the risk of leaf injury under high solar radiation [21]. Weed control and thinning were performed manually at 30 DAP to maintain a uniform stand of two plants per planting hole throughout the experiment. Plants were harvested manually at physiological maturity following the standardized sampling procedures recommended by the HarvestPlus program for agronomic biofortification studies [22]. Grain yield was determined from the net plot area and subsequently converted to kilograms per hectare. To ensure comparability among treatments, grain yield was standardized to 14% moisture content before statistical and economic analyses. Economic analysis The economic performance of Zn and Fe agronomic biofortification was evaluated under field conditions representative of tropical cowpea production in Mozambique. To facilitate international comparison and interpretation, all production costs originally recorded in Mozambican Metical (MZN) were converted to United States dollars (USD) using the prevailing exchange rate during the 2025 cropping season. The baseline production cost (Cbase) included all standard agronomic practices required for cowpea cultivation, including land preparation, seed purchase, basal fertilization, crop establishment, field management, harvesting, and post-harvest operations, excluding micronutrient fertilization. The baseline production cost was estimated at USD 234.72 ha ⁻ ¹, based on production records and operational cost estimates provided by the Instituto de Investigação Agrária de Moçambique (IIAM), where the field experiment was conducted. The use of IIAM records ensured that the economic evaluation reflected production conditions and management practices representative of experimental cowpea production under tropical conditions in Mozambique. This cost excluded all expenses associated with Zn and Fe fertilization, which were calculated separately according to the fertilizer source, application rate, and fertilization method for each treatment. The additional cost of micronutrient fertilization was calculated separately for zinc and iron according to fertilizer source, application rate, and market price. Zinc fertilizer cost (CZn) and iron fertilizer cost (CFe) were estimated as: where DoseZn and DoseFe represent the application rates of zinc sulfate and Fe–EDTA (kg ha ⁻ ¹), respectively. Fertilizer prices were obtained from commercial agricultural suppliers in Mozambique during the 2025 cropping season, ensuring that the economic analysis reflected the prevailing market conditions during the experimental period. Application costs (Capp) included labor and operational expenses associated with each fertilization strategy. These costs varied according to the application method and were estimated at USD 43.82 ha ⁻ ¹ for soil application, USD 62.59 ha ⁻ ¹ for foliar application, and USD 106.41 ha ⁻ ¹ for the combined soil plus foliar application. The total production cost (TC) for each treatment was calculated as: Gross revenue (GR) was calculated by multiplying the corrected grain yield by the average farm-gate price of cowpea grain: where grain yield was expressed in kg ha ⁻ ¹ at 14% moisture content, and the grain selling price was fixed at USD 0.70 kg ⁻ ¹, corresponding to the average farm-gate price observed in local markets during the 2025 harvest period. Net profit (NP) and the benefit–cost ratio (B:C) were then calculated using the following expressions: where NP represents the economic return after deducting total production costs from gross revenue, whereas B:C expresses the economic return generated per unit of production cost. Treatments with B:C > 1 were considered economically feasible, indicating that gross revenue exceeded total production costs. Together, Net Profit and the Benefit: Cost ratio provide complementary indicators of economic performance by quantifying both the absolute financial return and the economic efficiency of fertilizer investment. To evaluate the robustness of the proposed fertilization strategies under variable production and market conditions, complementary break-even, sensitivity, and stochastic risk analyses were performed using the observed experimental dataset. Break-even analysis was used to determine the minimum grain yield required to offset production costs, sensitivity analysis quantified the relative influence of the principal economic variables on net profit, and stochastic risk analysis estimated the probability of obtaining positive economic returns under the evaluated production scenarios. Statistical analysis and response surface optimization Grain yield data were first subjected to analysis of variance (ANOVA) according to a randomized complete block design arranged in a 3 × 4 × 4 factorial scheme, with three fertilization modes, four Zn application rates, and four Fe application rates. The statistical model included the effects of block, fertilization mode, Zn rate, Fe rate, and all possible interactions among the treatment factors. When significant treatment effects were detected by the F-test (P < 0.05), means for fertilization modes were compared using Tukey’s honestly significant difference (HSD) test at the 5% probability level. where Yijkl is the observed response; μ is the overall mean; Bi is the block effect; Mj is the fertilization mode effect; Zk is the zinc dose effect; Fl is the iron dose effect, and εijkl is the experimental error. Prior to statistical analysis, model assumptions were evaluated by examining residual normality and homogeneity of variances. Normality was assessed using the Shapiro–Wilk test, whereas homogeneity of variances was evaluated using Levene’s test. Residual plots were also visually inspected to verify model adequacy before proceeding with subsequent analyses. When significant treatment effects were detected by the F-test (p < 0.05), comparisons among fertilization modes were performed using Tukey’s honestly significant difference (HSD) test at the 5% probability level. To quantify the combined effects of Zn and Fe fertilization on grain yield, response surface methodology (RSM) was applied by fitting a second-order polynomial regression model including linear, quadratic, and interaction terms: where Y represents grain yield (kg ha ⁻ ¹), and Zn and Fe correspond to the applied fertilizer rates (kg ha ⁻ ¹). Model adequacy was evaluated using the coefficient of determination (R2), adjusted coefficient of determination (adjusted R2), statistical significance of regression coefficients, and residual diagnostics. The maximum technical efficiency (MTE) was determined from the stationary point of the fitted response surface, corresponding to the nutrient combination that maximized predicted grain yield within the experimental domain. To complement the agronomic evaluation, economic performance was assessed using two derived indicators: net profit (NP) and benefit–cost ratio (B:C), both calculated from grain yield and treatment-specific production costs. These variables were analyzed separately using the same factorial ANOVA model adopted for grain yield to evaluate the effects of fertilization mode, Zn rate, Fe rate, and their interactions. Subsequently, a second-order polynomial regression model was fitted using net profit as the response variable to determine the economically optimal combination of Zn and Fe application rates. where NP is expressed in USD ha ⁻ ¹. Model adequacy was evaluated using the coefficient of determination (R2), adjusted coefficient of determination (adjusted R2), statistical significance of the regression coefficients, and residual diagnostics. The stationary point of the fitted response surface was used to estimate the maximum economic efficiency (MEE), defined as the combination of Zn and Fe application rates that maximized net economic return. Agronomic efficiency (AE) of Zn and Fe fertilization was calculated as the increase in grain yield per unit of nutrient applied according to the following equation: where Yf is the grain yield obtained under fertilized conditions, Y0 is the grain yield of the corresponding unfertilized control treatment, and Nutrient applied represents the amount of Zn or Fe applied (kg ha ⁻ ¹). Agronomic efficiency was expressed as kilograms of grain produced per kilogram of nutrient applied (kg grain kg ⁻ ¹ nutrient). All statistical analyses, analysis of variance, regression modelling, response surface optimization, and graphical visualization were performed using R software (version 4.5.2; R Core Team, 2025). Statistical significance was declared at P < 0.05 throughout the study. Results Climatic conditions during the experimental season The climatic conditions observed during the 2025 cowpea growing season were generally representative of the regional climate, although rainfall distribution differed from the long-term average in some months (Table 2). Rainfall during May was substantially higher than the historical average, whereas June and July were slightly drier than normal. Mean air temperature remained close to the long-term climatic normals throughout the experimental period, indicating that the experiment was conducted under conditions broadly representative of the local production environment. 10.1371/journal.pone.0357473.t002Table 2Comparison between the climatic conditions observed during the 2025 cowpea growing season and the long-term climate normals at Quelimane Meteorological Station.MonthMean temperature normal (°C)Mean temperature 2025 (°C)Rainfall normal (mm)Rainfall 2025 (mm)InterpretationApril25.927.25124.6124.7TypicalMay23.924.7061.5169.4Wetter than averageJune22.023.3540.132.7Slightly drier than averageJuly21.521.7052.725.3Drier than averageSource: Experimental weather data recorded at the Quelimane Meteorological Station during the 2025 growing season and long-term climate normals obtained from the Japan Meteorological Agency (JMA). Comparison between the climatic conditions observed during the 2025 cowpea growing season and the long-term climate normals at Quelimane Meteorological Station. Source: Experimental weather data recorded at the Quelimane Meteorological Station during the 2025 growing season and long-term climate normals obtained from the Japan Meteorological Agency (JMA). Analysis of variance for grain yield Analysis of variance showed that cowpea grain yield was significantly affected by block, fertilization mode, Zn application rate, Fe application rate, and all interaction terms among the experimental factors (P < 0.001; Table 3). The significant interactions indicate that the effects of Zn and Fe on grain yield depended on the fertilization mode, demonstrating that the responses to micronutrient application were not independent. The significant block effect confirms that the randomized complete block design effectively accounted for spatial variability across the experimental area. Overall, grain yield was strongly influenced by both the main effects of fertilization mode and micronutrient application rates and by their interactions. 10.1371/journal.pone.0357473.t003Table 3Analysis of variance for cowpea grain yield as affected by fertilization mode, zinc (Zn) rate, iron (Fe) rate, and their interactions.Source of variationdfSSMSFPBlock316,632.495,544.16365.14<0.001Fertilization mode (M)2576,142.53288,071.2618,972.71<0.001Zinc rate (Zn)32,456,502.23818,834.0853,929.37<0.001Iron rate (Fe)3499,419.15166,473.0510,964.11<0.001M × Zn695,889.8215,981.641,052.57<0.001M × Fe621,576.923,596.15236.85<0.001Zn × Fe9152,749.4116,972.161,117.81<0.001M × Zn × Fe1815,286.84849.2755.93<0.001Residual error1412,140.8715.18——Total1913,836,340.24———Notes: df, degrees of freedom; SS, sum of squares; MS, mean square; F, F statistic; M, fertilization mode; Zn, zinc application rate; Fe, iron application rate. The residual mean square (MS error) was used as the denominator for all F-tests. Analysis of variance for cowpea grain yield as affected by fertilization mode, zinc (Zn) rate, iron (Fe) rate, and their interactions. Notes: df, degrees of freedom; SS, sum of squares; MS, mean square; F, F statistic; M, fertilization mode; Zn, zinc application rate; Fe, iron application rate. The residual mean square (MS error) was used as the denominator for all F-tests. Response surface model for grain yield The combined effects of Zn and Fe application rates on cowpea grain yield were adequately described by a second-order polynomial regression model (Table 4). The fitted equation was: 10.1371/journal.pone.0357473.t004Table 4Estimated coefficients of the second-order polynomial regression model describing the response of cowpea grain yield to zinc (Zn) and iron (Fe) fertilization.ParameterEstimatep-valueIntercept1473.68<0.001Zn29.15<0.001Zn2−0.90<0.001Fe13.20<0.001Fe2−0.47<0.001Zn x Fe0.256<0.001Notes: Zn, zinc application rate (kg ha ⁻ ¹); Fe, iron application rate (kg ha ⁻ ¹). The fitted model included linear, quadratic, and interaction terms. Positive coefficients indicate an increase in grain yield with increasing nutrient application, whereas negative quadratic coefficients indicate diminishing returns at higher application rates. Estimated coefficients of the second-order polynomial regression model describing the response of cowpea grain yield to zinc (Zn) and iron (Fe) fertilization. Notes: Zn, zinc application rate (kg ha ⁻ ¹); Fe, iron application rate (kg ha ⁻ ¹). The fitted model included linear, quadratic, and interaction terms. Positive coefficients indicate an increase in grain yield with increasing nutrient application, whereas negative quadratic coefficients indicate diminishing returns at higher application rates. All linear, quadratic, and interaction terms were statistically significant (P < 0.001; Table 4), demonstrating that both micronutrients contributed significantly to grain yield variation within the experimental range. The model explained 76.7% of the total variation in grain yield (R2 = 0.767), with an adjusted coefficient of determination (adjusted R2) of 0.761, indicating good predictive performance under the experimental conditions. The positive linear coefficients for Zn and Fe indicate that grain yield increased with increasing micronutrient supply at lower application rates, whereas the negative quadratic coefficients demonstrate diminishing marginal responses as fertilizer rates increased. In addition, the positive Zn × Fe interaction coefficient indicates that the simultaneous application of both micronutrients produced a synergistic effect on grain productivity. The fitted response surface was subsequently used to estimate the stationary point corresponding to the maximum technical efficiency (MTE), as presented in the following section. Maximum technical efficiency (MTE) Optimization of the fitted response surface identified an interior stationary point corresponding to maximum technical efficiency. The predicted maximum grain yield was 1,875.97 kg ha ⁻ ¹ at application rates of 18.92 kg Zn ha ⁻ ¹ and 19.17 kg Fe ha ⁻ ¹. Because these optimum rates were within the experimental range, the estimated maximum was obtained by interpolation, supporting the biological interpretation of the fitted model. Predicted grain yield increased progressively with increasing Zn and Fe application rates, reaching a maximum near the stationary point before declining at higher nutrient rates. This response pattern and the well-defined optimum region are illustrated by the response surface and contour plot (Figs 4 and 5). Response surface for the combined effects of zinc and iron application rates on predicted cowpea grain yield. The marker indicates the maximum technical efficiency, estimated at 18.92 kg Zn ha ⁻ ¹ and 19.17 kg Fe ha ⁻ ¹, corresponding to a predicted grain yield of 1,875.97 kg ha ⁻ ¹. Contour plot of predicted cowpea grain yield across zinc and iron application rates. The red triangle indicates the estimated point of maximum technical efficiency (18.92 kg Zn ha ⁻ ¹ and 19.17 kg Fe ha ⁻ ¹), corresponding to a predicted grain yield of 1,875.97 kg ha ⁻ ¹. Effect of fertilization mode Response surface analysis revealed clear differences in predicted grain yield among fertilization strategies. The combined soil + foliar application consistently produced the highest predicted yields, outperforming both soil-only and foliar-only fertilization (Fig 6). These results indicate greater nutrient-use efficiency when both application pathways were combined. Response surface of cowpea grain yield under different fertilization modes (soil, foliar, and soil + foliar). Agronomic efficiency of zinc and iron Agronomic efficiency differed among fertilization modes for both Zn and Fe (Fig 7). In both cases, the combined soil + foliar strategy produced the highest agronomic efficiency, followed by foliar-only and soil-only application. Mean AE-Zn values were 13.3, 15.1, and 22.4 kg grain per kg Zn applied for soil, foliar, and soil + foliar application, respectively. Corresponding AE-Fe values were 6.24, 7.34, and 11.5 kg grain per kg Fe applied. These findings indicate that integrating soil and foliar fertilization enhanced nutrient use efficiency and its conversion into grain yield under field conditions. Agronomic efficiency of zinc (Zn) and iron (Fe) under different fertilization modes. Analysis of variance of economic indicators Analysis of variance showed that fertilization mode, Zn rate, Fe rate, and all interaction terms significantly affected both the benefit–cost ratio (B:C) and net profit (NP) (Table 5). Block effects were also significant, indicating some spatial variation in economic performance across the experimental area. However, the substantially larger F-values associated with fertilization mode and micronutrient application rates indicate that treatment effects greatly exceeded block-to-block variability. Overall, the economic performance of cowpea biofortification was strongly influenced by micronutrient management, reflecting the combined effects of grain yield response and treatment-specific production costs. 10.1371/journal.pone.0357473.t005Table 5Analysis of variance for the economic indicators, benefit–cost ratio (B:C) and net profit (NP), as affected by fertilization mode, zinc (Zn) rate, iron (Fe) rate, and their interactions.Source of variationdfBenefit–cost ratio (B:C)Net profit (NP)SSMSFPSSMSFPBlock30.0570.019545.20<0.00181502717365.14<0.001Mode23.0831.54144536.93<0.00129048145241952.20<0.001Zn34.5271.50943608.07<0.001102016934005645707.15<0.001Fe326.4118.804254392.42<0.001215260717539644.39<0.001Mode × Zn60.2170.0361045.39<0.0014698678311052.57<0.001Mode × Fe60.6690.1123223.47<0.001105731762236.85<0.001Zn × Fe90.3960.0441272.01<0.0017484783161117.81<0.001Mode × Zn × Fe180.0570.00391.27<0.001749141655.93<0.001Residual1410.0050.000——10497——Notes: df, degrees of freedom; SS, sum of squares; MS, mean square; F, F statistic; B:C, benefit–cost ratio; NP, net profit; Zn, zinc application rate; Fe, iron application rate; M, fertilization mode. The residual mean square (MS error) was used as the denominator for all F-tests. Analysis of variance for the economic indicators, benefit–cost ratio (B:C) and net profit (NP), as affected by fertilization mode, zinc (Zn) rate, iron (Fe) rate, and their interactions. Notes: df, degrees of freedom; SS, sum of squares; MS, mean square; F, F statistic; B:C, benefit–cost ratio; NP, net profit; Zn, zinc application rate; Fe, iron application rate; M, fertilization mode. The residual mean square (MS error) was used as the denominator for all F-tests. Economic response surface for net profit Net profit response to Zn and Fe fertilization was adequately described by a second-order polynomial regression model, which explained 86.99% of the total variation in net profit (R2 = 0.8699). The fitted regression equation was: All model coefficients were statistically significant. The positive linear coefficients for Zn and Fe indicate that net profit initially increased with increasing micronutrient rates, whereas the negative quadratic terms show diminishing economic returns beyond the optimum. The positive interaction term further indicates that combined Zn and Fe application enhanced profitability up to a certain threshold. The economic response surface is shown in Fig 8. Response surface showing the effect of Zn and Fe fertilization on net profit (USD ha ⁻ ¹). Maximum economic efficiency (MEE) Optimization of the economic response surface identified the maximum economic efficiency (MEE) at 16.94 kg ha ⁻ ¹ Zn and 10.09 kg ha ⁻ ¹ Fe. At this combination, the predicted net profit reached USD 909.57 ha ⁻ ¹. This result indicates that the economic optimum occurred at lower fertilizer rates than the agronomic optimum, particularly for Fe. The lower economic optimum for Fe was mainly associated with the relatively high cost of Fe fertilizer, which reduced profitability at higher application rates despite continued increases in grain yield. Thus, maximum economic return was achieved before the point of maximum biological productivity was reached. The location of the economic optimum is illustrated in Fig 9. These findings were further reinforced by sensitivity and stochastic risk analyses, which confirmed the stability and robustness of the economic optimum under variable production and market conditions. Contour plot illustrating the economic optimum (maximum economic efficiency, MEE) for Zn and Fe fertilization. Agronomic versus economic optimization Comparison between agronomic and economic optimization revealed distinct micronutrient requirements for maximizing grain yield and maximizing economic return (Table 6). Maximum technical efficiency (MTE) was achieved at 18.92 kg Zn ha ⁻ ¹ and 19.17 kg Fe ha ⁻ ¹, corresponding to a predicted grain yield of 1875.97 kg ha ⁻ ¹. In contrast, maximum economic efficiency (MEE) occurred at 16.94 kg Zn ha ⁻ ¹ and 10.09 kg Fe ha ⁻ ¹, producing a predicted grain yield of 1838.21 kg ha ⁻ ¹ and a predicted net profit of USD 909.57 ha ⁻ ¹. 10.1371/journal.pone.0357473.t006Table 6Comparison between the maximum technical efficiency (MTE) and maximum economic efficiency (MEE) for Zn and Fe fertilization in cowpea.Optimization criterionZn rate (kg ha ⁻ ¹)Fe rate (kg ha ⁻ ¹)Predicted grain yield (kg ha ⁻ ¹)Net profit (USD ha ⁻ ¹)Fertilizer cost (USD ha ⁻ ¹)Maximum technical efficiency (MTE)18.9219.171875.97883.14124.46Maximum economic efficiency (MEE)16.9410.091838.21909.5771.58Change relative to MTE (%)−10.48−47.39−2.01+2.99−42.50Notes: MTE, maximum technical efficiency; MEE, maximum economic efficiency. Relative changes were calculated using MTE as the reference. Fertilizer cost includes only Zn and Fe fertilizer costs and excludes baseline production and application costs. Comparison between the maximum technical efficiency (MTE) and maximum economic efficiency (MEE) for Zn and Fe fertilization in cowpea. Notes: MTE, maximum technical efficiency; MEE, maximum economic efficiency. Relative changes were calculated using MTE as the reference. Fertilizer cost includes only Zn and Fe fertilizer costs and excludes baseline production and application costs. Relative to the agronomic optimum, the economic optimum reduced Zn and Fe application rates by 10.48% and 47.39%, respectively. This reduction resulted in only a 2.01% decrease in predicted grain yield while reducing fertilizer costs by 42.50%. Consequently, net profit increased by 2.99%, from USD 883.14 ha ⁻ ¹ at MTE to USD 909.57 ha ⁻ ¹ at MEE. These results demonstrate that maximizing economic return required substantially lower micronutrient inputs than maximizing biological productivity, particularly for Fe, highlighting the value of bioeconomic optimization for fertilizer recommendation under the conditions evaluated Economic risk and sensitivity analysis To further evaluate the robustness of the proposed fertilization strategies, complementary break-even, risk, and sensitivity analyses were performed using the observed experimental dataset. Break-even analysis indicated that all treatments produced grain yields substantially above the minimum threshold required to offset production costs, confirming the economic feasibility of all fertilization strategies evaluated (Fig 10). Although the combined soil plus foliar application (SF) involved higher operational costs, it consistently generated the highest economic returns, whereas the soil-only (S) application required the lowest break-even yield to achieve profitability. Break-even analysis showing the minimum grain yield required to offset production costs under different fertilization strategies. The empirical distribution of net profit showed a mean value of USD 812.87 ha ⁻ ¹ with a standard deviation of USD 86.03 ha ⁻ ¹, indicating relatively low variability in economic returns. All evaluated treatments generated positive net profits within the observed experimental dataset, and none of the evaluated treatment combinations resulted in a negative net profit under the agronomic and economic conditions considered in this study. This pattern is illustrated by the cumulative distribution function (Fig 11), in which the distribution of net profits remained above the break-even threshold. Cumulative distribution function of net profit (USD ha ⁻ ¹), showing the probability distribution of economic returns across fertilization treatments. Sensitivity analysis was conducted using the predicted maximum economic efficiency (baseline net profit = USD 909.57 ha ⁻ ¹) as the reference scenario. Grain price, grain yield, zinc fertilizer cost, and iron fertilizer cost were independently varied by ±20% relative to their baseline values while maintaining the remaining variables constant. Thus, each variable was evaluated at 80% (low) and 120% (high) of its baseline value to quantify its influence on net profit. The tornado diagram (Fig 12) showed that grain price and grain yield were the variables exerting the greatest influence on net profit, whereas the costs of zinc and iron fertilizers had comparatively smaller effects. These results indicate that the economic performance of the proposed fertilization strategies is primarily driven by market price and productivity fluctuations, while remaining relatively insensitive to plausible variations in micronutrient fertilizer costs. Tornado diagram showing the relative influence of ±20% changes in grain price, grain yield, zinc fertilizer cost, and iron fertilizer cost on net profit. Discussion Soil micronutrient limitation as a primary driver of cowpea productivity The results clearly demonstrate that cowpea productivity in the studied environment was primarily constrained by severe zinc deficiency, as evidenced by the extremely low DTPA-extractable Zn concentration (0.1 mg kg ⁻ ¹). This value is well below the critical threshold required for optimal legume production, confirming Zn limitation as the dominant yield-restricting factor [23]. Accordingly, these findings are consistent with previous studies highlighting widespread Zn deficiency in highly weathered tropical soils of Sub-Saharan Africa [9,24]. Physiologically, Zn plays a central role in enzymatic activation, protein synthesis, and hormonal regulation, particularly auxin metabolism, which directly influences root growth and reproductive development [11,25]. Thus, the strong yield response observed most likely reflects the restoration of key metabolic functions impaired under Zn deficiency. Although soil Fe levels were relatively high, the significant Zn × Fe interaction indicates that plant performance depends on nutrient balance rather than the absolute availability of individual micronutrients. Collectively, these findings reinforce the concept that crop nutritional responses arise from integrated interactions among soil nutrient availability, nutrient acquisition, and plant physiological processes rather than from isolated nutrient effects. Accordingly, the agronomic responses observed in the present study are best interpreted as the outcome of coordinated soil–plant interactions regulating Zn and Fe availability, uptake, and utilization under the experimental conditions [26–28]. Synergistic interaction between zinc and iron The positive Zn × Fe interaction identified by the response surface model suggests a synergistic effect between these micronutrients under the experimental conditions. Although antagonistic interactions between Zn and Fe have been reported under excessive nutrient supply, previous studies have shown that their combined application under deficient conditions may enhance nutrient acquisition, improve physiological performance, and increase crop productivity [16,17]. The physiological mechanisms underlying this response were not directly evaluated in the present study. Nevertheless, previous research suggests that Zn may contribute to improved root development and membrane integrity, potentially enhancing Fe acquisition, whereas Fe is involved in chlorophyll synthesis and electron transport, thereby supporting photosynthetic activity [29–31]. Furthermore, interactions between Zn and Fe may influence nutrient uptake and homeostasis at the soil–plant interface [26]. Accordingly, the positive interaction observed in this study is interpreted as a plausible consequence of complementary physiological processes associated with balanced micronutrient nutrition rather than as direct evidence of specific physiological mechanisms. Likewise, synergistic responses have been reported in cereals and grain legumes, where integrated Zn and Fe management enhanced nutrient uptake, physiological performance, grain micronutrient accumulation, and crop productivity [11,27,28]. Moreover, previous studies have shown that root-induced rhizosphere modifications, including localized pH changes, alterations in rhizosphere chemistry, and the release of root exudates, may influence Zn and Fe availability and uptake [32,33]. Although these processes were not investigated in the present study, they may also have contributed to the positive interaction observed under the experimental conditions. Fertilization strategy and nutrient use efficiency The superior performance of the combined soil + foliar application highlights the importance of synchronizing nutrient supply with crop physiological demand. Soil application provides a continuous nutrient supply during the early stages of crop establishment, promoting root development and initial plant growth, whereas foliar application rapidly supplies nutrients during periods of high demand, particularly at the reproductive stage [33]. Consequently, the combination of these application methods improves the temporal synchronization between nutrient availability and crop requirements, thereby increasing nutrient recovery and agronomic efficiency. Comparable responses have been reported in studies demonstrating that integrated fertilization strategies improve nutrient-use efficiency by better synchronizing nutrient availability with crop demand throughout the growing season [34,35]. Moreover, the combined strategy may improve nutrient-use efficiency by complementing soil nutrient availability with direct foliar uptake during critical growth stages, thereby potentially reducing nutrient losses associated with soil fixation and leaching. Although these processes and nutrient uptake pathways were not directly evaluated in the present study, the present interpretation is consistent with previous reports showing that integrated micronutrient management enhances nutrient uptake, physiological performance, biomass accumulation, and grain productivity compared with individual fertilization strategies [16,36]. Bioeconomic optimization and diminishing returns A major contribution of the present study is the clear distinction between the agronomic and economic optima for Zn and Fe fertilization in cowpea. Whereas maximum technical efficiency (MTE) was achieved at relatively higher micronutrient application rates, maximum economic efficiency (MEE) was attained at substantially lower fertilizer inputs, particularly for Fe. This divergence demonstrates that the fertilizer combination maximizing biological productivity does not necessarily maximize economic return, especially when fertilizer costs are explicitly incorporated into the optimization process. This pattern is consistent with the principle of diminishing marginal returns, whereby successive increases in fertilizer application produce progressively smaller increments in grain yield that eventually become insufficient to compensate for the additional production costs. Similarly, fertilizer optimization studies have shown that recommendations based exclusively on maximum biological yield frequently exceed economically optimal fertilizer rates, particularly under conditions of increasing fertilizer prices and fluctuating grain market values [18,37]. The relatively high cost of Fe fertilizer was the principal factor responsible for the divergence between the agronomic and economic optima observed in the present study. Although grain yield continued to increase with Fe application until reaching the biological optimum, the fitted response surface indicated progressively smaller yield gains as Fe rates increased. As a result, the economically optimal solution was achieved at a substantially lower Fe application rate, reducing Fe input by approximately 47% while maintaining more than 97% of the maximum predicted grain yield. Overall, these findings demonstrate that moderate reductions in fertilizer inputs can substantially improve economic performance while causing only negligible reductions in biological productivity. From a practical perspective, the results emphasize the importance of integrating agronomic and economic criteria when developing fertilizer recommendations for smallholder production systems. Recommendations based solely on maximum yield may encourage fertilizer application rates that are agronomically effective but economically inefficient. Conversely, bioeconomic optimization provides a more robust decision-support framework by simultaneously considering crop response, fertilizer costs, and market returns, thereby maximizing farm profitability while maintaining high levels of productivity [38,39]. Nevertheless, the economic optimum identified in this study should be regarded as inherently context-specific. Variations in fertilizer prices, grain market values, labor costs, and local production conditions may alter the economically optimal combination of Zn and Fe application rates. Therefore, although the methodological framework presented here is broadly applicable, optimum fertilizer recommendations should be locally calibrated to reflect the economic conditions of different production environments [40,41]. Robustness of the economic optimum under uncertainty Under the agronomic, environmental, and economic conditions evaluated, all fertilization treatments generated positive net returns within the observed dataset, indicating consistently favorable economic performance across the scenarios analyzed. Collectively, these results indicate that agronomic biofortification with Zn and Fe represents a relatively low-risk investment under the conditions evaluated in this study. Furthermore, sensitivity analysis identified grain price and grain yield as the principal determinants of economic performance, whereas variations in Zn and Fe fertilizer costs had comparatively smaller effects on net profitability. These findings suggest that market fluctuations and crop productivity exert a substantially greater influence on economic returns than moderate changes in micronutrient fertilizer prices. Likewise, recent studies have demonstrated that uncertainty in crop yield and market prices constitutes one of the major sources of variability in fertilizer recommendation strategies and farm profitability, emphasizing the importance of incorporating economic risk into fertilizer management decisions [42]. From an applied perspective, the combined soil + foliar application of approximately 16–17 kg Zn ha ⁻ ¹ and 10 kg Fe ha ⁻ ¹ provided the most favorable balance between grain productivity and economic return under the conditions of the present study. Nevertheless, because the economically optimal fertilizer combination depends on prevailing fertilizer prices, grain market values, labor costs, and local production conditions, these recommendations should be regarded as site-specific rather than universally applicable. Accordingly, fertilizer recommendation strategies should be periodically refined to reflect local agronomic and economic conditions while also considering farmers’ access to agricultural inputs, extension services, and institutional support to facilitate adoption. Such an integrated approach is essential for improving farm profitability, reducing production risks, and promoting the long-term sustainability of nutrient management in smallholder farming systems [43]. The significant block effects observed for both grain yield and the economic indicators indicate that the randomized complete block design successfully accounted for spatial variability across the experimental area. Although some environmental heterogeneity was present among blocks, treatment effects remained substantially greater than block-to-block variation, as demonstrated by the markedly higher F-values associated with fertilization mode and micronutrient application rates Implications for sustainable intensification and policy Beyond the agronomic and economic benefits demonstrated at the farm level, the present findings have broader implications for sustainable agricultural intensification and food and nutrition security in Sub-Saharan Africa. Agronomic biofortification with Zn and Fe represents a practical strategy to simultaneously increase crop productivity and improve the nutritional quality of staple foods, thereby contributing to efforts aimed at reducing micronutrient deficiencies while strengthening the resilience of smallholder farming systems [8,44]. Importantly, the present study demonstrates that economically optimized micronutrient fertilization can substantially reduce fertilizer inputs without significantly compromising grain yield. Under the conditions evaluated, the economically optimal fertilization strategy required considerably lower Fe application than the agronomic optimum while maintaining more than 97% of the maximum predicted yield. These findings underscore the value of integrating agronomic and economic criteria to improve nutrient-use efficiency, increase farm profitability, and promote more sustainable fertilizer management [39,45]. From a policy perspective, the results indicate that fertilizer recommendation programs should move beyond yield maximization and incorporate economic optimization to better reflect farmers’ production realities. Decision-support approaches integrating crop response, fertilizer costs, and market prices can provide more efficient and economically viable recommendations, particularly for resource-constrained smallholder farmers [46,47]. Equally important, the successful implementation of agronomic biofortification at larger scales will depend not only on scientifically sound fertilizer recommendations but also on farmers’ access to micronutrient fertilizers, effective agricultural extension services, reliable input supply systems, and supportive public policies. Therefore, strengthening these institutional and technological components will be essential to facilitate the adoption of integrated nutrient management practices and maximize their contribution to sustainable agricultural intensification, improved livelihoods, and food and nutrition security particularly in Sub-Saharan Africa [8,43]. Conclusion This study demonstrates that integrating agronomic and economic optimization provides a more efficient framework for defining Zn and Fe fertilization strategies than approaches based solely on maximum biological yield. Under the tropical field conditions of Mozambique evaluated in this study, the economically optimal fertilization strategy required lower micronutrient inputs than the agronomic optimum, particularly for Fe, while maintaining more than 97% of the maximum predicted grain yield and improving economic return. The combined soil + foliar fertilization strategy consistently produced the greatest grain yield and agronomic efficiency, highlighting the importance of synchronizing micronutrient supply with crop demand. These findings indicate that bioeconomic optimization can substantially improve fertilizer recommendations by reducing unnecessary micronutrient inputs while maintaining high productivity and profitability under the conditions evaluated. Because this study was conducted at a single experimental site during one growing season, the proposed fertilizer recommendations should be interpreted as specific to the environmental, agronomic, and economic conditions investigated. Additional multi-location and multi-season studies across contrasting soil types, climatic conditions, and production environments are needed to validate the robustness of the proposed optimum fertilization strategy. Future research should integrate agronomic biofortification with genotype selection, grain micronutrient accumulation, and multi-environment validation to further improve the efficiency, robustness, and scalability of Zn and Fe fertilization strategies for cowpea production under tropical conditions. Supporting information Experimental dataset used for the statistical analyses, including block, fertilization mode, zinc (Zn) and iron (Fe) application rates, and grain yield. (XLSX) 29 Jul 2026 Dear Dr. Carlos, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Sep 27 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. As the corresponding author, your ORCID iD is verified in the submission system and will appear in the published article. PLOS supports the use of ORCID, and we encourage all coauthors to register for an ORCID iD and use it as well. Please encourage your coauthors to verify their ORCID iD within the submission system before final acceptance, as unverified ORCID iDs will not appear in the published article. Only the individual author can complete the verification step; PLOS staff cannot verify ORCID iDs on behalf of authors. We look forward to receiving your revised manuscript. Kind regards, Muhammad Amjad Ali, PhD Academic Editor PLOS One Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section. 3. Thank you for stating the following financial disclosure: [Regional Scholarship and Innovation Fund (RSIF), (B8501N10019), Mr Tancredo Jose Carlos; -Partnership for Skills in Applied Sciences, Engineering, and Technology (PASET), I ampleased to inform you that, following a highly competitive selection process, you have been selected to receive a Regional Scholarship and Innovation Fund (RSIF),]. Please state what role the funders took in the study. If the funders had no role, please state: "The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript." If this statement is not correct you must amend it as needed. Please include this amended Role of Funder statement in your cover letter; we will change the online submission form on your behalf. 4. In the online submission form, you indicated that [All relevant data underlying the results presented in this study are available from the corresponding author upon reasonable request. All summarized results generated from the dataset are included within the manuscript.]. All PLOS journals now require all data underlying the findings described in their manuscript to be freely available to other researchers, either 1. In a public repository, 2. Within the manuscript itself, or 3. Uploaded as supplementary information. This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If your data cannot be made publicly available for ethical or legal reasons (e.g., public availability would compromise patient privacy), please explain your reasons on resubmission and your exemption request will be escalated for approval. 5. Please upload a new copy of Figure 1 as the detail is not clear. Please follow the link for more information: https://journals.plos.org/plosone/s/figures 6. We note that Figure 1 in your submission contains map images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright. We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission: 1. You may seek permission from the original copyright holder of Figure 1 to publish the content specifically under the CC BY 4.0 license. We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text: “I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.” Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission. In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].” 2. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only. The following resources for replacing copyrighted map figures may be helpful: USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/ The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/ Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/ Landsat: http://landsat.visibleearth.nasa.gov/ USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/# Natural Earth (public domain): http://www.naturalearthdata.com/ 7. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 2 in your text; if accepted, production will need this reference to link the reader to the Table. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? Reviewer #1: Yes Reviewer #2: Yes ********** 2. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #2: Yes ********** 3. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy Reviewer #1: Yes Reviewer #2: Yes ********** 4. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ********** Reviewer #1: Dear All, The manuscript evaluates the agronomic and economic optimization of zinc (Zn) and iron (Fe) fertilization for cowpea biofortification under tropical field conditions. Using response surface methodology, the authors investigated the effects of different Zn and Fe application rates and methods on grain yield, grain Zn and Fe concentrations, and economic return. The study demonstrates that the combined management of Zn and Fe improves crop productivity, nutritional quality, and profitability, allowing the identification of fertilization strategies that optimize both agronomic and economic performance. The topic is timely and relevant because agronomic biofortification has become an important strategy to alleviate micronutrient deficiencies while maintaining crop productivity. Overall, the manuscript presents valuable field data and an appropriate experimental approach. Nevertheless, several physiological interpretations require further refinement, and some conclusions should be moderated because they extend beyond the direct evidence generated by the study. * Strong points * 1) The manuscript addresses a highly relevant issue concerning agronomic biofortification and food security. 2) The study combines agronomic, nutritional, and economic evaluations, increasing its practical significance. 3) The field experiment provides information obtained under realistic production conditions. 4) Response Surface Methodology is appropriate for optimizing fertilization strategies. 5) The manuscript is logically organized and generally well written. 6) The practical recommendations are potentially useful for tropical cowpea production systems. * Weaker aspects * 1) Several physiological explanations regarding Zn–Fe interactions are presented as demonstrated mechanisms rather than plausible interpretations. 2) The distinction between nutrient accumulation and nutrient use efficiency deserves further clarification. 3) Soil processes influencing Zn and Fe availability are only briefly discussed. 4) The limitations associated with evaluating a single growing season should be more explicitly acknowledged. 5) The Discussion would benefit from a broader integration of plant nutrition concepts with the bioeconomic analysis. So, I have annotated below an attempt to clarify certain ideas, but the authors should examine my suggested wording changes carefully, to be sure that I have not misinterpreted what they wanted to say. 1. The manuscript concludes that the combined application of Zn and Fe promotes synergistic effects that improve grain yield and biofortification. Although the response surface analysis clearly demonstrates positive interactions between the two nutrients, the physiological mechanisms responsible for these responses were not directly investigated. Therefore, the Discussion should distinguish more carefully between statistically detected synergistic responses and the underlying physiological mechanisms, which remain hypothetical under the present experimental conditions. 2. Throughout the Discussion, increased grain Zn and Fe concentrations are frequently interpreted as evidence of improved nutrient uptake efficiency. However, grain nutrient accumulation reflects the combined effects of nutrient acquisition, internal redistribution, remobilization during grain filling, and biomass production. Consequently, the Discussion should avoid equating increased grain concentration exclusively with greater uptake efficiency. 3. The bioeconomic optimization is one of the principal strengths of the manuscript. Nevertheless, the economic recommendations should acknowledge that optimum fertilizer rates may vary according to fertilizer prices, grain market values, soil fertility, and regional production systems. A brief discussion of these limitations would improve the applicability of the recommendations. 4. The interaction between Zn and Fe is discussed mainly from the perspective of plant physiology. However, soil chemical processes—including adsorption, precipitation, pH effects, and competition for sorption sites—also regulate the availability of these micronutrients. Incorporating this broader soil–plant perspective would strengthen the interpretation of the observed responses. 5. The experiment was conducted during a single cropping season at one experimental location. While the results are promising, the Conclusions should recognize that additional validation across contrasting soil types, climatic conditions, and production environments is necessary before recommending generalized fertilization strategies. - Specific comments - P1 L20–26: The abstract concludes that the optimized Zn–Fe fertilization strategy improves both productivity and profitability. I suggest indicating that these conclusions are valid under the environmental and economic conditions evaluated in this study, avoiding broader generalization. P1 L30–36: The term "bioeconomic optimization" is central to the manuscript. Consider briefly defining this concept in the Abstract so that readers unfamiliar with the approach immediately understand that both biological and economic responses were simultaneously optimized. P3 L115–128 (Soil Analysis): The authors provide a useful baseline characterization of the experimental soil, including texture class (sandy-loam), pH, organic carbon, available P, exchangeable K, DTPA-extractable Zn and Fe, and the analytical methods used (Table 1). These data are appropriate for interpreting micronutrient biofortification responses. However, the manuscript does not identify the soil according to an internationally recognized soil classification system (e.g., WRB or USDA Soil Taxonomy). Because soil class strongly influences Zn and Fe availability through differences in mineralogy, buffering capacity, drainage, and sorption processes, I recommend that the authors report the formal soil classification together with the corresponding reference. P4 L92–99: Please report whether the experimental season was climatically representative of the long-term average for the region. This information would help readers evaluate the broader applicability of the proposed fertilization strategy. P4 L100–108: Include the soil classification (WRB or USDA Soil Taxonomy), since soil class strongly influences Zn and Fe availability and facilitates comparison with similar field studies. P4 L109–117: Please clarify whether DTPA-extractable Zn and Fe were determined before fertilizer application. These values are important for interpreting crop responsiveness to micronutrient fertilization. P5 L118–127: Indicate whether the micronutrient fertilizers were surface-applied or incorporated into the soil, as fertilizer placement may influence Zn and Fe availability. P5 L130–138: Please specify whether soil moisture was monitored during the growing season, since water availability directly affects micronutrient diffusion and plant uptake. P6 L145–153: The description of the experimental design is clear. Nevertheless, briefly indicate whether possible border effects were minimized between treatment plots. P6 L160–169: Clarify whether grain yield was corrected to a standard moisture content before statistical analysis. P7 L176–186: The response surface methodology is appropriate. However, please indicate whether residual analyses confirmed the assumptions of normality and homogeneity of variance. P8 L200–208: The manuscript reports significant Zn × Fe interactions. Please provide the corresponding interaction P-values within the text or refer readers directly to the ANOVA table. P9 L218–226: Consider reporting confidence intervals for the predicted optimum nutrient rates, which would provide an indication of the robustness of the optimization model. P10 L236–245: The response surfaces are informative. Adding contour values or predicted yield labels would facilitate interpretation of the biological optimum. P11 L252–260: The grain Zn and Fe concentrations increased substantially. Please discuss whether dilution effects associated with higher grain yield were considered during data interpretation. P12 L270–279: The manuscript attributes increased grain micronutrient concentration to improved nutrient uptake. This interpretation should be moderated because internal remobilization during grain filling was not evaluated. P13 L285–294: Please clarify whether nutrient accumulation was calculated on a concentration basis or total nutrient content per unit area, since these parameters provide different physiological information. P14 L300–308: The economic analysis is one of the strengths of the manuscript. Nevertheless, indicate the fertilizer prices and grain market prices used, together with their reference period. P15 L309-317: It would be useful to mention whether sensitivity analyses were conducted for different fertilizer price scenarios. P20 L313-316: The Discussion appropriately emphasizes that plant performance depends on nutrient balance rather than the absolute availability of Zn or Fe. This section could be strengthened by highlighting that plant nutritional responses arise from integrated interactions among nutrient acquisition, soil processes, and plant physiology rather than from isolated nutrient effects. NOTICE: Suggested insertion: "This observation reinforces the concept that plant nutritional performance depends on integrated interactions among nutrient acquisition, soil processes, and physiological regulation, rather than on the isolated availability of individual nutrients (Thiengo et al., 2026 - https://doi.org/10.1590/1678-992X-2026-0067 )." P20 L313-321: The authors discuss the increase in grain Zn concentration following micronutrient fertilization. This discussion would benefit from acknowledging that the effectiveness of agronomic Zn biofortification depends not only on fertilizer supply but also on the timing of application and genotype-specific responses. NOTICE: Suggested insertion: "Previous studies have demonstrated that successful agronomic Zn biofortification depends on the interaction between application timing and genotype, reinforcing that crop responses to foliar Zn fertilization are strongly influenced by physiological differences among cultivars (Morais et al, 2025 - https://doi.org/10.1590/1678-992X-2024-0191." P21 L318–325: The discussion of Zn × Fe synergism is appropriate. However, the physiological mechanisms should be presented as plausible explanations because root architecture, transporter activity, and gene expression were not evaluated. P21 L322–326: The statement that Zn improves Fe acquisition through enhanced root development should be interpreted cautiously, since root morphological traits were not measured. P21 L324–330: Consider discussing the contribution of root-induced rhizosphere modifications to Zn and Fe acquisition, while acknowledging that these processes were not directly investigated. P21 L330–337: The distinction between nutrient acquisition efficiency and nutrient utilization efficiency should be clarified because these represent different physiological processes. P21 L338–343: Increased grain micronutrient concentration does not necessarily indicate greater uptake efficiency. Redistribution during grain filling may also contribute to the observed responses. P22 L344–350: The discussion of diminishing marginal returns is well presented. Nevertheless, indicate that the economic optimum may change with fluctuations in fertilizer prices and grain value. P22 L346–350: The recommendation for reduced Fe application should be presented within the context of the specific soil fertility conditions evaluated in this experiment. P22 L351–360: The uncertainty analysis strengthens the manuscript. Please clarify whether climatic variability between years was incorporated into the economic simulations. P22 L361–370: Consider discussing whether genotype × environment interactions may influence the robustness of the proposed optimum fertilization strategy. P23 L373–380: The Conclusions appropriately summarize the findings. However, the statement that Zn deficiency strongly constrained productivity should be limited to the experimental conditions evaluated. P23 L378–384: Please indicate that the optimized fertilization rates should be validated under contrasting soil types before broad recommendations are proposed. P23 L384–390: The manuscript would benefit from mentioning future research integrating micronutrient fertilization with genotype selection for biofortification. Reviewer #2: Overview Manuscript: Bioeconomic optimization of zinc and iron agronomic biofortification in cowpea (Vigna unguiculata [L.] Walp.) under tropical field conditions Manuscript ID: PONE-D-26-24166 Recommendation: Major revision The study addresses an important topic and the comparison between yield-maximizing and profit-maximizing fertilizer rates could be valuable. The factorial field experiment also appears to have generated a substantial dataset. However, the manuscript currently contains several conceptual, experimental, statistical, and economic inconsistencies that prevent the main conclusions from being accepted. Nevertheless, the absence of grain micronutrient measurements, ambiguity concerning treatment doses, incorrect or insufficiently reported statistical analysis, pooling of distinct application modes, and use of an average application cost undermine the central conclusions. A substantial reanalysis and major revisions are required before the manuscript can be considered scientifically reliable. Comments 1.The title should include Mozambique and indicate that this was a single-season field study. If grain mineral concentrations were not measured, “biofortification” should be removed from the title. 2.The Abstract should specify whether rates refer to elemental nutrients or fertilizer products. 3.“Kg” should be written as “kg” throughout. 4.Use consistent formatting for ha⁻¹, mg kg⁻¹, USD ha⁻¹ and percentages. 5.Avoid unnecessary capitalization of terms such as “Maximum Technical Efficiency” within running text. 6.The statement that Fe application continued to increase yield at high rates should be checked against the fitted surface, which ultimately predicts declining yield. 7.The references should be checked for relevance and completeness. Some citations do not directly support the economic claims, several entries are incompletely formatted, and Sheahan and Barrett appears to be duplicated as References 25 and 48. 8.Table 1: Correct inconsistent units for N and P and distinguish measured values from interpretation thresholds. “Physical properties” and “chemical properties” should not be included as empty data rows. 9.Table 2: Degrees of freedom are inconsistent with a four-level factorial analysis. Report sums of squares, mean squares, F statistics and the correct error structure. 10.The missing table should be added or all later tables renumbered. 11.Figures: The figures has low resolution and several labels are difficult to read. 12.Figures 8 and 9: These economic surfaces are based on an average application cost and therefore do not represent a specific application strategy. They should be replaced with mode-specific economic surfaces and uncertainty regions. 13.Figure 12: Define the baseline net profit, the ±20% perturbation procedure, the low and high values used for every variable, and whether variables were changed independently. Numerical outcomes should be included in the figure or a supplementary table. 14.The Introduction and Discussion would benefit from incorporating recent literature that is directly relevant to integrated Zn and Fe fertilization, crop physiological responses, grain micronutrient bioavailability, and the nutritional importance of these elements. In particular, the authors may consider citing Alkarawi (2025) when discussing the effects of combined Fe- and Zn-based formulations on legume growth and yield; Younas et al. (2025) in relation to integrated fertilization and grain Fe bioavailability; Muhae-Ud-Din et al. (2026) when discussing metal-mediated changes in ROS metabolism, nitrogen metabolism, yield, and seed quality; and Iftikhar et al. (2024) to support the broader nutritional significance of Fe and Zn. Alkarawi HH. The effect of integrating nano-formulated iron and zinc chelates and Biohealth® bio-fertilizer on some vegetative growth and yield characteristics of soybean (Glycine max [L.] Merrill). Pak J Agric Sci. 2025;62(4):717–728. doi:10.21162/PAKJAS/25.964. Younas, A., Farooq, M., Waqar, S., Ghafoor, I., Makhdoom, M., Parveen, A., Kashif, M., & Uzair, M. (2025). Improving grain iron bioavailability in bread wheat through integrated nitrogen and iron fertilization. Integrative Plant Biotechnology, 3(4), 343–350. https://doi.org/10.58257/pbiotech.v3i4.1680 Muhae-Ud-Din G, Zhong F, Jabran M, Wang Y, Smagghe G, Sun X, et al. Metal nanoparticles enhance ROS scavenging, nitrogen metabolism, and stress-responsive pathways to improve soybean yield and seed quality. Plant Physiology and Biochemistry. 2026:111363. Iftikhar H, Israr B, Butt MS, Pasha I. The impact of iron- and zinc-fortified snacks on cognitive performance in pre-adolescents. Pak J Agric Sci. 2024;61(4):1075–1083. doi:10.21162/PAKJAS/24.360. 15.Conclusions and recommendations are too broad for a single site-year experiment. Please revise. 16.The manuscript would benefit from professional English-language and formatting revision after the scientific and analytical issues have been resolved. ********** what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy Reviewer #1: No Reviewer #2: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation. NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications. 14 Aug 2026 Response to the Academic Editor and Reviewers Manuscript ID: PONE-D-26-24166 Title: Bioeconomic optimization of zinc and iron agronomic biofortification in cowpea (Vigna unguiculata [L.] Walp.) under tropical field conditions in Mozambique Dear Academic Editor and Reviewers, We sincerely thank the Academic Editor and both reviewers for their careful evaluation of our manuscript and for their constructive comments and insightful suggestions. We greatly appreciate the time and effort invested in reviewing our work, as well as the recognition of its scientific relevance, technical soundness, experimental design, and practical importance. In response to the comments received, we have thoroughly revised the manuscript. The major revisions include clarification of the experimental design and statistical methodology, complete reanalysis of the factorial analysis of variance (ANOVA), revision of the bioeconomic optimization framework, improvement of the soil and climatic characterization, moderation of physiological interpretations, acknowledgment of the limitations associated with a single-site and single-season experiment, incorporation of recent and directly relevant literature, revision of the figures and tables, and comprehensive editing of the English language and manuscript formatting. We have carefully addressed every comment raised by the reviewers. Below, each reviewer comment is reproduced in bold, followed by our detailed response and a summary of the corresponding revisions made to the manuscript. All modifications are highlighted in the revised manuscript submitted with tracked changes. We hope that the revisions have satisfactorily addressed all concerns and have substantially improved the quality, clarity, and scientific rigor of the manuscript. Sincerely, Tancredo José Carlos On behalf of all authors Responses to Journal Requirements Journal Requirement 1 Comment: Please ensure that your manuscript meets PLOS ONE’s style requirements, including those for file naming. Response: We sincerely thank the journal for this reminder. The revised manuscript has been carefully reviewed and reformatted to comply with the PLOS ONE style requirements. Terminology, units, abbreviations, table and figure presentation, reference formatting, and file naming have been standardized throughout the submission. Changes made: The manuscript formatting was revised according to the PLOS ONE guidelines. Units, abbreviations, tables, figures, and references were standardized. The revised files will be uploaded using the required designations: Manuscript, Revised Manuscript with Track Changes, and Response to Reviewers. Journal Requirement 2 Comment: The grant information provided in the “Funding Information” and “Financial Disclosure” sections does not match. Please provide the correct grant numbers. Response: We sincerely thank the journal for identifying this inconsistency. The funding information has been reviewed and corrected to ensure that the grant number and funding organization are reported consistently in the manuscript, cover letter, and online submission system. Revised funding statement: This study was supported by the Regional Scholarship and Innovation Fund (RSIF), a program of the Partnership for Skills in Applied Sciences, Engineering and Technology (PASET), under grant number B8501N10019 awarded to Tancredo José Carlos. Changes made: The Funding Information and Financial Disclosure statements were harmonized to report the same funding organization and grant number. Journal Requirement 3 Comment: Please state what role the funders took in the study. If the funders had no role, please state: “The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.” Response: We confirm that the funders had no role in the scientific design, execution, analysis, interpretation, or publication decisions associated with this study. Revised Role of Funder statement: The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Changes made: This statement has been included in the revised manuscript and will also be provided in the revised online submission information to ensure consistency between the manuscript and the submission system. Journal Requirement 4 Comment: All data underlying the findings must be freely available through a public repository, within the manuscript, or as Supporting Information. Response: We sincerely thank the journal for this clarification. The data underlying the findings of this study will be made freely available as Supporting Information accompanying the revised manuscript. The dataset includes the experimental factors, block structure, fertilization modes, Zn and Fe application rates, grain yield, and the variables used in the agronomic and economic analyses. Revised Data Availability statement: All data underlying the findings reported in this study are provided in the Supporting Information files accompanying the manuscript. Changes made: The underlying experimental dataset will be uploaded as Supporting Information. The Data Availability statement has been revised accordingly. Journal Requirement 5 Comment: Please upload a new copy of Figure 1 because the details are not clear. Response: We sincerely thank the journal for this comment. Figure 1 has been replaced with a higher-resolution version. Font sizes, labels, geographical boundaries, symbols, and annotations were enlarged and standardized to improve readability. Changes made: Figure 1 was regenerated and replaced with a clearer, higher-resolution version that complies with the PLOS ONE figure preparation requirements. Journal Requirement 6 Comment: Figure 1 may contain copyrighted map images. Please provide permission or replace the figure with a map that complies with the CC BY 4.0 license. Response: We sincerely thank the journal for drawing our attention to this issue. The original map was replaced with a newly generated figure based exclusively on open-access or public-domain spatial data. No Google Maps, Google Earth, proprietary satellite imagery, or other copyrighted map material was used in the revised figure. Changes made: Figure 1 was reconstructed using spatial data compatible with open-access publication under the CC BY 4.0 license. The figure caption was revised to identify the spatial data source. The previous map containing potentially copyrighted material was removed. Figure caption source statement: Administrative boundaries and base-map data were obtained from open-access spatial datasets and the figure was created by the authors Journal Requirement 7 Comment: Table 2 is not referred to in the manuscript text. Please ensure that Table 2 is cited in the text. Response: We sincerely thank the journal for identifying this omission. In the revised manuscript, the factorial analysis of variance table, originally presented as Table 2, has been renumbered as Table 3 following the inclusion of a new table describing the climatic conditions during the experimental season. The Results section has been revised to refer explicitly to Table 3 when presenting the factorial analysis of variance and the significance of the main effects and interaction terms. Changes made: A direct citation to Table 3 was added to the Results section, ensuring that the factorial analysis of variance table is properly linked to the corresponding statistical description in the text. Reviewer #1 We sincerely thank Reviewer #1 for the thorough evaluation of our manuscript and for the constructive comments and insightful suggestions. We greatly appreciate the reviewer's careful assessment and the positive feedback regarding the scientific quality of our study. The comments have substantially improved the clarity, rigor, and interpretation of the manuscript. We have carefully considered each recommendation and revised the manuscript accordingly. Our point-by-point responses are presented below. Comment 1 Reviewer comment Please report the WRB or USDA soil classification of the experimental site. Response We sincerely thank the reviewer for this valuable comment. Unfortunately, a site-specific WRB or USDA soil classification was not available for the experimental field. Therefore, we chose not to assign a formal soil classification without direct pedological characterization, as doing so could result in an inaccurate or unsupported classification. Instead, the soil was characterized based on its measured physical and chemical properties obtained before crop establishment, including texture, pH, organic carbon, available phosphorus, exchangeable potassium, sulfur, DTPA-extractable zinc and iron, and total nitrogen. To provide regional pedological context, we also incorporated information from the National Soil Map of Mozambique, which indicates that the study area is predominantly characterized by Arenosols, Ferralsols, Lixisols, and Acrisols. We believe that this approach provides a scientifically sound description of the experimental soil while avoiding unsupported taxonomic classification. Changes made in the manuscript The description of the experimental soil in the Materials and Methods section was revised to clarify that the soil was characterized based on its measured physical and chemical properties rather than assigning an unsupported WRB or USDA classification. In addition, regional soil information from the National Soil Map of Mozambique was included to provide pedological context. Comment 2 Reviewer comment Please report whether the experimental season was climatically representative of the long-term average for the region. Response We sincerely thank the reviewer for this valuable comment. To address this suggestion, we compared the climatic conditions recorded during the 2025 growing season with the long-term climatic normals obtained from the Quelimane Meteorological Station, the closest meteorological station to the experimental site. To facilitate this comparison, we included a new table in the revised manuscript summarizing the observed and long-term climatic conditions. Mean monthly temperatures during the experimental period were very similar to the long-term averages, with only minor differences, while rainfall exhibited some monthly variability but maintained the characteristic seasonal distribution of the region. Overall, this comparison indicates that the experiment was conducted under climatic conditions that were broadly representative of the regional production environment. Changes made in the manuscript A new table comparing the climatic conditions observed during the 2025 growing season with the long-term climatic normals for the Quelimane Meteorological Station was added to the manuscript. The Materials and Methods and Results sections were revised accordingly to demonstrate that the experimental season was broadly representative of the climatic conditions of the study region. Comment 3 Reviewer comment Please clarify the statistical model used in the analysis and specify whether block was treated as a fixed or random effect. Response We sincerely thank the reviewer for this important comment. We carefully revised the Statistical Analysis section to provide a clearer and more accurate description of the statistical model used in this study. The treatment effects were evaluated using a factorial analysis of variance (ANOVA) for a randomized complete block design (RCBD) arranged in a 3 × 4 × 4 factorial scheme, including the effects of block, fertilization mode, Zn rate, Fe rate, and their interactions. Because a mixed-effects model was not fitted, we removed the previous statement indicating that block was treated as a random effect. The statistical model is now described in a manner that accurately reflects the analytical approach adopted in the study. Changes made in the manuscript The Statistical Analysis section was revised to provide a clear description of the factorial ANOVA model used in the study. The previous statement indicating that block was treated as a random effect was removed, and the statistical model was rewritten to accurately describe the effects included in the analysis and the experimental design Comment 4 Reviewer comment P8 L200–208: The manuscript reports significant Zn × Fe interactions. Please provide the corresponding interaction P-values within the text or refer readers directly to the ANOVA table. Response We sincerely thank the reviewer for this helpful suggestion. To improve the transparency and readability of the statistical results, we revised the Results section to explicitly report the corresponding P-values for the interaction terms discussed in the text. In addition, direct references to Table 3 were included wherever the ANOVA results are described, allowing readers to easily relate the narrative interpretation to the complete statistical analysis. Changes made in the manuscript The Results section was revised to include the corresponding P-values for the significant interaction terms discussed in the text. Direct references to Table 3 were also added to improve consistency between the narrative description and the ANOVA results. Comment 5 Reviewer comment P9 L218–226: Consider reporting confidence intervals for the predicted optimum nutrient rates, which would provide an indication of the robustness of the optimization model. Response We sincerely thank the reviewer for this valuable suggestion. We agree that confidence intervals for the estimated optimum nutrient rates could provide additional information regarding the uncertainty associated with the optimization results. However, estimating confidence intervals for the stationary point of a second-order response surface model requires specialized statistical procedures, such as bootstrap resampling or delta-method approaches, which were beyond the scope of the present study. To address the reviewer's concern, we clarified that the estimated optimum occurred within the experimental domain and strengthened the evaluation of model robustness by emphasizing the goodness-of-fit statistics, coefficient of determination (R²), and residual diagnostics used to assess model adequacy. Changes made in the manuscript The Statistical Analysis and Results sections were revised to clarify that the estimated optimum was located within the experimental domain. Additional information regarding model adequacy, including the coefficient of determination (R²), residual diagnostics, and model adequacy assessment, was incorporated to provide greater confidence in the robustness of the response surface model. Comment 6 Reviewer comment P10 L236–245: The response surfaces are informative. Adding contour values or predicted yield labels would facilitate interpretation of the biological optimum. Response We sincerely thank the reviewer for this valuable suggestion. We agree that adding contour labels could further facilitate the interpretation of the response surface. However, the contour plot already clearly identifies the optimum region corresponding to the maximum predicted grain yield. Rather than adding numerical labels that could reduce figure readability, we revised the figure legend to explicitly state that the contour plot represents the predicted grain yield and highlights the region corresponding to the Maximum Technical Efficiency (MTE). We believe this revision improves figure interpretation while preserving its visual clarity. Changes made in the manuscript The legend of Fig 5 was revised to clarify that the contour plot represents predicted cowpea grain yield and to explicitly identify the estimated Maximum Technical Efficiency (MTE), including the corresponding Zn and Fe application rates and predicted grain yield. This revision improves interpretation of the biological optimum while preserving the visual clarity of the figure. Comment 7 Reviewer comment P11 L252–260: The grain Zn and Fe concentrations increased substantially. Please discuss whether dilution effects associated with higher grain yield were considered during data interpretation. Response We sincerely thank the reviewer for this valuable comment. We agree that dilutio Submitted filename: Response to the Editor and Reviewers.docx 18 Aug 2026 Bioeconomic optimization of zinc and iron agronomic biofortification in cowpea (Vigna unguiculata [L.] Walp.) in Mozambique under tropical field conditions PONE-D-26-24166R1 Dear Dr. Carlos, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Muhammad Amjad Ali, PhD Academic Editor PLOS One Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #2: All comments have been addressed ********** 2. Is the manuscript technically sound, and do the data support the conclusions??> Reviewer #2: Yes ********** 3. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #2: Yes ********** 4. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy Reviewer #2: Yes ********** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #2: Yes ********** Reviewer #2: This is now well-written and executed study. I have no additional comments. I recommend the paper for publication. ********** what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy Reviewer #2: Yes: Muhammad Jabran, PhD. ********** PONE-D-26-24166R1 PLOS One Dear Dr. Carlos, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS One and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Muhammad Amjad Ali Academic Editor PLOS One