The original experiment took place during the 2019–20 cropping season at the Chitala, Chitedze, and Ngabu Agricultural Research Stations, located in the Lilongwe, Salima, and Chikwawa districts, respectively (51). Table 1 provides details about the locations of these experimental sites. All three sites experience a unimodal rainfall pattern, with rain typically occurring from November to April. In Malawi, maize is usually planted in November at the start of the rainy season and harvested in May or June, with planting beginning earlier in the southern and low-lying regions. The sites generally undergo a cool, dry period from May to mid-August, with temperatures ranging between 16°C and 27°C, followed by a warmer phase from mid-August to October, when temperatures range from 20°C to 40°C. From November to April, the climate is warm and wet, with temperatures fluctuating between 25°C and 36°C. A detailed study area map that illustrates the research’s geographical location, sampling sites, and spatial context is presented in Figure 1.

Before setting up the original experiment, soil samples were collected across the entire experimental area for each soil type at each site (51). The soil sampling procedure was carefully designed to capture the variability across each experimental area by selecting five evenly spaced, random points per site. Soil was collected uniformly from the top 0–20 cm layer using clean tools to prevent contamination. Samples from these points were combined into a 500-gram composite sample representing the average soil conditions for each soil type and site. These composites were then analyzed in the laboratory for key soil properties—such as pH, organic carbon, total nitrogen, DTPA Zn, cation exchange capacity, and exchangeable bases (calcium, magnesium and potassium) to establish baseline fertility and nutrient status before planting. The analysis results presented in Table 2 revealed considerable variation in soil characteristics between the different soil types. Notably, Vertisols consistently showed higher average values for pH, organic carbon, total nitrogen, and exchangeable bases compared to Lixisols across all sites. This trend suggested that Vertisols generally possess greater inherent fertility, making them more favorable for crop growth under the conditions studied. Understanding these baseline differences was critical for interpreting experiment outcomes.

As previously stated, the original study was conducted at Chitedze, Chitala and Ngabu Agricultural Research Stations. Since larger Zn application rates were considered in the original study, this experiment was conducted to examine residual effects in a second cropping season at the same locations. The experiment involved three Zn fertilizer rates;1, 30, and 90 kg Zn ha^-1 applied as ZnSO₄.7H₂O in the previous season, arranged in a randomized complete block design (RCBD) with 10 replications per site. Subsequent trials were, however, not successful at Ngabu Agricultural Research Station due to drought and failure of trial establishment.

The residual benefit of soil-applied Zn to subsequent crops has previously been noted (39, 43, 53, 54). Measurement of residual Zn in the soil prior to another crop being planted can determine the extent of its availability for the next crop. Samples were analyzed as described by (52). Soil samples from the depth of 0–20 cm were collected at the final harvest in 2020 from all the plots at Chitedze Research Station. The samples were collected at ten points along the summit of one of the peripheral ridges, which were selected at random from each net plot, using a Dutch soil auger with a flight length of 15 cm and a diameter of 3.5 cm, and the 10 samples from each plot were bulked. The samples were air-dried, sieved (<2 mm) and homogenized before determination of extractable Zn as a measure of plant-available Zn using the diethylenetriaminepentaacetic acid (DTPA) method (55). The extraction procedure was undertaken on duplicate subsamples from each plot, using 5 g of soil extracted with 10 mL of 0.005 M DTPA, 0.1 M triethanolamine and 0.01 M CaCl₂ at pH = 7.3 shaken for 2 h on an end-over-end shaker. Thereafter, the samples were centrifuged at 3000 rpm for 15 minutes and the supernatant filtered through <0.22 µm syringe filters prior to analysis using Inductively Coupled Plasma-Mass Spectrometry (ICP-MS).

The field trials were conducted at Chitedze and Chitala Agricultural Research Stations on two contrasting soil types: Lixisols and Vertisols. The original experiment was laid out in a randomized complete block design (RCBD) with each Zn fertilizer treatment replicated 10 times for each soil type at the experimental sites (51, 52). The recommended planting pattern was followed as described in the Guide to Agricultural Production and Natural Resource Management of the Ministry of Agriculture (32). The detailed procedures of Good Agronomic Practices (GAP) followed in the current study generally included the following key steps: site selection and preparation, seed selection and sowing, crop management, pest and disease control, and harvesting and post-harvest handling of the grain. The residual benefit of Zn to the maize crop was assessed by growing the crop in the subsequent cropping season (2020–2021) on the same plots and ridges without ploughing or any added Zn. The gross plot size was 5 ridges, each 5 m long, with the net plot being the 3 middle ridges, each 3 m long. The ridges were spaced at 75 cm apart. Maize seeds were sown at 25 cm spacing along the ridge, giving an expected plant density of 53,333 plants ha^-1. Good agronomic practices were followed except for avoiding creating new ridges. The SC 403 maize variety, locally known as Kanyani, was used. General information about the maize variety and climatic conditions of the sites is provided in Botoman et al. (2020). Kanyani is a F₁ hybrid variety widely grown in Malawi, can mature in ~90 days and adapts to a wide range of environmental conditions. Critical nutrients including nitrogen (N), phosphorus (P), potassium (K) and sulphur (S) were adequately applied as straights to avoid extra Zn coming in following the guidelines outlined in the Guide to Agricultural Production and Natural Resource Management of the Ministry of Agriculture (32).

Maize was sown in December 2020 when effective rain started and harvested in April 2021 at Chitala, and in May 2021 at Chitedze. At harvest, grain and stover samples were collected. Grain yield (kg) and dry weight of stover (kg) were recorded from the net plots and used to calculate Zn uptake and harvest index of the crop. The Zn harvest index is a ratio between Zn accumulated in the grain to the sum of the Zn accumulated in the grain and stover (56), expressed as a percentage. Daily rainfall (mm) was also recorded using rain gauges stationed in each of the research stations where the experiment was conducted (Supplementary Figure S1). Generally, rainfall was adequate at both Research Stations and additional irrigation was not used given that rain-fed agriculture is the common practice in Malawi.

Grain and stover samples were prepared and Zn concentrations determined as described by (52, 57). A portion of finely ground plant material (c. 0.2 g of grain or stover) was digested with 6 mL of 70% HNO₃ (trace analytical grade) using a microwave system comprising a Multiwave Prom 41HVT56 Rotor and pressure-activated-venting vessels made of modified polytetrafluoroethylene (PTFE-TFM, 56 mL ‘SMART VENT’, Anton Paar GmbH, Graz, Austria). A Certified Reference Material, (CRM; Wheat flour SRM 1567b, National Institute of Standards and Technology, Gaithersburg, MD, US; 11.61mg kg^-1) and 12 operational blank digestions were used to determine the accuracy of the analyses and the limit of detection (LOD) for quality control. The Zn elemental recovery for SRM 1567b was 105%. After the extraction procedure was complete, grain and stover samples were analyzed for Zn by inductively coupled plasma mass spectrometry (ICP-MS).

Data analyses were conducted using the linear and non-linear mixed effects (nlme) package for the R platform (58). The analysis of data was done after validating the assumptions of normal distribution of residuals and homogeneity of variances by checking the model plots. After estimation of the model parameters, histograms were plotted of the random effects estimates at each level, the marginal residuals were plotted against the fitted values (Supplementary Figures S2–S7) and summary statistics (Supplementary Tables S1–S6) were computed. The outputs for maize grain yields, grain and stover Zn concentrations and uptake met these assumptions. For harvest index, these assumptions were not valid, and data were transformed using a natural log. A linear mixed model (LMM) was used with a random effects structure to reflect how the fertilizer rate was randomized among plots within sets of blocks all within one sub-site of a single soil type. A fixed effects model was used comprising main effects of fertilizer rate, soil type and their interaction. Further, the main effect of fertilizer rate was partitioned into linear and non-linear components with an appropriate choice of orthogonal polynomials, and the soil type by fertilizer rate interaction was similarly partitioned into contrasts between the linear and non-linear responses to Zn applicate rate on the different soils. The output of the analysis tested the hypothesis concerning the differences between soil types and Zn fertilizer rates with respect to the response variable.

The residual Zn availability, at the end of the growing season in the year of application, typically increased with an increase in applied Zn fertilizer rate (Figure 2). There were no significant differences in the concentration of DTPA-extractable Zn between the application rates of 1 and 30 kg ha^-1. However, the differences were significant when the rate increased to 90 kg ha^-1. There was an apparent drastic increase in soil zinc concentration from 0.76 mg kg^-1 to over 300 mg kg^-1 following the application of 90 kg ha^-1. It is important to note that the values represent DTPA-extractable zinc, which reflects the readily available zinc fraction rather than total soil zinc content. The high zinc application rate likely created localized zones of elevated zinc availability, which the DTPA extraction sensitively detected. Additionally, it is possible that zinc fertilizer residues on the soil surface contributed to the inflated extractable zinc values immediately post-application. No observable toxic effects of Zn on the maize crops were noticed.

The maize grain yields obtained over all experimental sites are presented in Figure 3a, error bars show the standard error. Some of the soil-applied Zn appeared to remain in an available form to the succeeding maize crop resulting in a positive grain yield response. Soil type is not replicated within sites, and so we can make inferences only about an additive soil effect over all the sites. A LMM framework was used to fit the data as proposed by (52). The main effect of Zn fertilizer rate was partitioned into linear and non-linear components. A positive response of maize grain yield to residual Zn for each Zn fertilizer rate was observed at each site except at Chitala. The observed decrease in yield at Chitala with 90 kg Zn application compared to 30 kg is likely due to antagonistic nutrient interactions and nutrient imbalance rather than simply excessive Zn. The interaction effects vary strongly with soil type, nutrient status, and local environmental factors. For example, calcareous soils with high pH commonly limit Zn solubility, and excessive Zn application may not translate linearly into higher plant availability or yield. Because of these factors, the large yield difference between 30 kg and 90 kg Zn application suggests that the optimal Zn rate lies somewhere between these values, where the beneficial residual Zn effect enhances yield without provoking antagonistic deficiencies of P, N, or K. Overall, the mean grain yield increased by 1490 kg ha^-1 in response to residual Zn from the 30 kg ha^-1 Zn fertilizer rate relative to the 1 kg Zn ha^-1 rate (24.6% higher). However, no further increases in yield was observed at Chitedze when the Zn application rate was increased to 90 kg ha^-1.

The analysis of variance (ANOVA) for the maize grain yield is shown in Table 3. There is strong evidence for an effect of residual Zn in soil for both linear and non-linear components. The linear component (p < 0.05) represents the positive effect of residual Zn on grain yield, while the non-linear component (p < 0.05) shows the diminishing marginal returns of 90 kg ha⁻¹ rate, relative to the response at 30 kg ha⁻¹. However, there was no significant differences among the soil types (p > 0.05). Furthermore, an interaction of linear response of Zn fertilizer rate with soil type (p > 0.05) and the non-linear response with soil type (p >0.05) was not significant. This, therefore, suggests that maize grain yield response to Zn fertilizer rate did not depend on soil type.

The grain Zn concentrations and uptake for each fertilizer rate at all experimental sites are presented in Figures 3b, c, along with the standard errors calculated for each treatment level. The uptake of residual Zn from soil was clearly observed in the subsequent maize crop. As observed for grain yield, positive responses of grain Zn concentration and uptake to residual Zn fertilizer were apparent. The overall mean grain Zn concentrations at 1, 30 and 90 kg ha^-1 Zn fertilizer rate were 22.6, 26.1 and 27.4 mg kg^-1 respectively, with their standard errors, for the three Zn fertilizer rates as estimated in the LMM. Similarly, maize grain Zn uptake at 1, 30 and 90 kg ha^-1 Zn fertilizer rate were 149, 195 and 205 g ha^-1 respectively. The estimated additional grain Zn concentration and uptake arising from residual soil Zn following the 30 kg ha^-1 were 3.4 mg kg^-1 (12.5% higher than for 1 kg ha^-1) and 44 g ha^-1 (29.6% higher than for 1 kg ha^-1), respectively; no further increases were observed when the Zn fertilizer rate was increased to 90 kg ha^-1.

The ANOVA for maize grain Zn concentration and Zn grain uptake are presented in Table 1. There was a significant response of maize grain Zn concentration and uptake to Zn fertilizer rate for the linear (p < 0.05) and non-linear (p < 0.05) components of the response. Over all sites, there was no evidence for differences in grain Zn concentration (p > 0.05) and grain Zn uptake (p > 0.05) between soil types. The linear response was noticed when the Zn fertilizer rate increased from 1 to 30 kg ha^-1 whereas increasing Zn application from 30 to 90 kg ha^-1 resulted in a non-linear response. Thus, increasing the Zn fertilizer rate from 1 to 30 kg ha^-1 results in a proportional increase in maize grain Zn concentration and uptake from residual Zn in the subsequent growing season, while an increase from 30 to 90 kg ha^-1 results in a proportionally smaller increase in grain Zn concentration and uptake. Furthermore, the interaction of soil type and linear response for grain Zn concentration (p < 0.05) and grain Zn uptake (p < 0.05) was significant. For both response variables, no significant differences (p > 0.05) in grain Zn concentration and (p > 0.05) in grain Zn uptake were observed with the interaction of non-linear response and soil types. This suggests that maize grain Zn concentration and uptake depended on soil type when the rate was increased from 1 to 30 kg ha^-1 while from 30 to 90 kg ha^-1, soil type did not have any effect over all sites.

The results on the effect of residual available Zn for each Zn fertilizer rate on stover Zn concentration, uptake, and harvest index at all experimental sites are shown in Figures 3d–f. When the main effect of Zn fertilizer rate was partitioned into linear and non-linear components, a positive response of stover Zn concentration and uptake to Zn fertilizer rate was observed at all sites. The stover Zn concentrations at applications of 1, 30 and 90 kg ha^-1 were 27.8, 56.5 and 92.3 mg kg^-1, respectively while the stover Zn uptake at these rates were 254, 528 and 900 g ha^-1, respectively. Thus, over all sites, increasing the Zn fertilizer rate from 1 to 90 kg ha^-1 resulted in a linear correlation between the fertilizer rate and stover Zn concentration and stover Zn uptake.

The ANOVA output for stover Zn concentration and uptake are presented in Table 1. There was a significant response of stover Zn concentration and uptake to Zn fertilizer rate for the linear (p < 0.05) response. However, the non-linear response was not statistically significant for both stover Zn concentration (p > 0.05) and stover Zn uptake (p > 0.05). Similarly, over all sites, there was no evidence for differences in stover Zn concentration (p > 0.05) and stover Zn uptake (p > 0.05) between soil types. There was no significant interaction of the linear response (p > 0.05) and non-linear component of the response (p > 0.05) with soil type in stover Zn concentration over all sites. Similarly, there was no evidence for an effect of the interaction of linear (p > 0.05) and non-linear responses (p > 0.05) with soil type in stover Zn uptake. This suggests that stover Zn concentration and uptake did not depend on soil type when the rate was increased from 1 to 90 kg ha^-1.

The mean Zn harvest indices (Zn_HI) for each fertilizer rate at all sites are presented in Figure 4, accompanied by their standard errors estimated for each treatment level. The effects of soil type, Zn fertilizer rate and their interaction on Zn_HI were analyzed using the LMM. Prior to analysis, Zn_HI was tested for normality of the residuals and the outputs showed a skewed distribution and, therefore the response variable was transformed into natural logarithm values. After the transformation, the assumption of a normal distribution and homogeneity of variances of the residuals were valid. Generally, mean Zn_HI decreased for all the soil types in response to the increase of Zn fertilizer rate (Figure 4). Note that no statistical inference about soil type at each site could be made since soil type was not replicated within each experimental site. There was no observed effect in Zn_HI between soil types when the rate was increased from 1 to 30 kg ha^-1 while at 90 kg ha^-1 the decrease was statistically different. The observed variations in Zn_HI response to Zn fertilizer rate over all sites might be due to differences in soil physical and chemical behavior.

Table 1 shows the ANOVA output for the natural log of Zn harvest index. There was strong evidence for an effect on Zn_HI of Zn fertilizer rate for the linear (p < 0.05) component of the response. However, there was no evidence for an effect on Zn_HI of Zn fertilizer rate for the non-linear (p > 0.05) component of the response. Normally, when the rate was increased from 1 to 90 kg ha^-1, there was a negative response for both linear and non-linear components. The observed reduction in Zn_HI with Zn fertilizer rate shows that Zn partitioning efficiency to the grain was negatively affected by the physiological response of the crop to Zn availability in the soil through reduction of Zn uptake by the crop roots. Over all sites, there were no differences in Zn_HI reduction as the Zn fertilizer rate was increased from 1 to 30 kg ha^-1 while noticeable differences were observed as the rate was increased from 30 to 90 kg ha^-1.

Increased maize grain yield resulting from residual available Zn in the soil has been documented in previous studies (59). For instance, research conducted in Brazil (39) demonstrated that maize grain yield increases were consistently observed during the second year of maize cropping when Zn fertilizer was applied at rates of 2, 4, and 8 kg Zn ha⁻¹. These results highlight the persistence of Zn in the soil and its continued availability to crops beyond the initial application year. Further supporting evidence comes from another study (43), which reported pronounced residual effects in terms of both plant Zn uptake and increased plant extractable Zn, with benefits lasting up to six years after Zn application at rates of 5.6, 11.2, 16.8, and 22.4 kg ha⁻¹. These findings underscore the long-term value of Zn fertilization in enhancing soil fertility and crop productivity.

Similarly, the findings of the present study confirm the positive impact of residual Zn fertilizer on maize grain yields. When Zn fertilizer was applied at a rate of 30 kg ha⁻¹, maize grain yields increased by 24.6% compared to the national recommendation rate of 1 kg ha⁻¹. This improvement translates to an additional 1,500 kg ha⁻¹ of maize grain produced because of increasing the Zn fertilizer rate from 1 to 30 kg ha⁻¹. Such results emphasize the importance of adequate Zn management in maize production systems, particularly in regions where Zn deficiency is a limiting factor for crop yields. The sustained benefits of Zn fertilization not only boost immediate yields but also contribute to improved soil health and long-term agricultural productivity.

The use of residual Zn fertilizer offers potential benefits for enhancing the food security of farmers by improving crop yields over time. However, research findings indicate that the maize grain yields obtained at a low Zn application rate of 1 kg ha^-1 were noticeably lower than yields achieved with a higher Zn application rate of 30 kg ha^-1 during the subsequent cropping season. This difference suggests that the amount of plant-available residual Zn remaining in the soil after applying 1 kg ha^-1 in the previous season is very limited and insufficient to sustain optimum crop growth.

One possible explanation for the poor residual effect at the lower Zn rate is the chemical interactions in the soil. Zinc tends to form stable complexes with soil organic matter, which can reduce its availability to plants (60). Additionally, Zn ions are prone to adsorption onto soil minerals such as iron (Fe) and manganese (Mn) oxides as well as aluminosilicate clays (61). These adsorption processes effectively immobilize Zn, preventing it from being easily taken up by plant roots (61). Such mechanisms largely explain why the application of 1 kg ha^-1 Zn fertilizer does not result in a meaningful residual yield benefit because much of the Zn applied becomes fixed in unavailable forms shortly after application. In contrast, the impact of Zn adsorption on residual availability appears to be less pronounced at higher Zn fertilizer rates such as 30 kg ha^-1. At this elevated level, the quantity of Zn applied likely exceeds the soil’s capacity to fix or adsorb it entirely, leaving more plant-available Zn in the soil for the following cropping season. This increased residual availability contributes to the improved maize grain yields observed at the higher fertilizer rate.

When the initial Zn fertilizer rate was set at 30 kg ha-1, the grain Zn concentration measured in the residual year showed a notable increase of 12.5% compared to the grain Zn concentration under the lower Zn fertilizer rate of 1 kg ha^-1. This sustained enhancement in Zn concentration is particularly significant because it demonstrates that the effects of higher Zn application persist beyond the initial cropping season. Interestingly, this difference is quite consistent with the 15% increase in grain Zn concentration observed between these two treatments during the very first year of Zn application, as documented in our previous study (51). This consistency suggests that initial Zn application rates can have lasting impacts on Zn bioavailability in the soil, thereby maintaining grain Zn enrichment over multiple seasons.

In parallel, grain Zn uptake by the plants was also substantially influenced by the initial Zn fertilizer rates. Specifically, grain Zn uptake was 29.6% greater following application of 30 kg ha^-1 Zn fertilizer compared to the minimal application rate of 1 kg ha^-1. This increase in Zn uptake exceeds the previously reported difference of 23% observed during the initial application period, implying that plants not only concentrate more Zn in their grains but also assimilate higher total Zn amounts when higher initial fertilization rates are employed. This enhancement in Zn uptake could be indicative of improved Zn availability and root uptake mechanisms persisting into subsequent crop cycles.

Overall, these findings reinforce the agronomic value of applying Zn fertilizer at adequate rates, highlighting that an initial Zn input of 30 kg ha-1 not only maximizes Zn enrichment in harvested grains during the first year but also sustains this nutritional benefit in subsequent seasons. This residual effect is crucial for strategies aiming to improve Zn biofortification in staple crops, especially in Zn-deficient soils prevalent in many regions. Furthermore, the pronounced increase in Zn uptake emphasizes the soil-plant Zn dynamics’ responsiveness to fertilization, which could inform more efficient and cost-effective Zn management practices in cereal production systems.

The data presented in this study clearly demonstrate that the application of residual Zn fertilizer not only enhances grain Zn levels but also significantly increases the Zn concentration and uptake in crop stover. Specifically, raising the Zn fertilizer rate from 1 to 90 kg ha^-1 resulted in a proportional increase in Zn content within the stover biomass. This finding holds considerable practical implications for livestock farmers, as the stover commonly used as fodder for ruminants can serve as an enriched source of dietary Zn. Improved Zn intake in animals is essential for maintaining optimal immune function, growth, and reproduction, ultimately enhancing overall livestock health and productivity.

Moreover, the elevated Zn levels in crop residues carry further importance for farmers practicing conservation agriculture (CA) systems. CA emphasizes maintaining and incorporating organic residues into the soil to improve soil health and fertility. The higher Zn concentration in the stover means that when these residues are returned to the soil, they contribute to replenishing soil Zn levels, which is critical for sustaining soil nutrient balance and long-term fertility. This cycle of residue incorporation can gradually enhance soil Zn availability without the immediate need for additional Zn fertilizer inputs each season.

Consequently, these improvements in soil Zn status through residue incorporation may lead to a reduction in Zn fertilizer requirements over the medium to long term. Less frequent or lower-dose fertilizer applications translate into lower production costs for farmers, which can directly boost their net economic returns. Thus, residual Zn fertilization coupled with conservation agriculture practices offers a synergistic approach that benefits both crop and livestock production systems while also promoting sustainability and cost-efficiency in smallholder farming contexts. This integrated strategy holds promise for enhancing nutrient cycling, farm resilience, and economic viability simultaneously.

Our findings on the residual effects of soil-applied Zn are consistent with previous studies from sub-Saharan Africa and globally, which indicate that applying zinc at higher rates has lasting benefits for following crops by enhancing grain yield, Zn concentration in grains, and Zn uptake (39, 40, 62). Comparable research confirms that these residual effects endure for at least two cropping seasons and occur across different soil types. These findings underscore the value of Zn fertilization for both improving crop productivity and achieving Zn nutritional biofortification in the region. Improving maize grain Zn concentration through agronomic biofortification in Malawi has direct potential to enhance dietary Zn intake among rural populations, who rely heavily on maize as a staple food and commonly experience Zn deficiency. Our previous studies show that Zn-enriched fertilizer application can increase grain Zn concentrations by 15% or more (51), which can meaningfully contribute to alleviating dietary Zn deficiency, supporting better immune function and growth. These improvements provide a cost-effective, food-based strategy to reduce Zn deficiency-related health risks in vulnerable communities without requiring dietary changes.

The Zn harvest index (ZnHI), which reflects the efficiency with which Zn is partitioned and accumulated in the grain relative to the total zinc taken up by the crop, was estimated in this study to understand how different Zn fertilizer application rates affect grain Zn loading. Notably, increasing Zn fertilizer application rates resulted in decreases in ZnHI, indicating that as more Zn was applied, the proportion of Zn ultimately allocated to the grain declined. Specifically, the grain Zn loading efficiencies at application rates of 1, 30, and 90 kg ha^-1 were found to be 40%, 30%, and 22%, respectively. This gradual decline in Zn loading efficiency suggests diminishing returns in grain Zn accumulation as fertilization intensifies. A closer examination reveals that when the Zn fertilizer rate increased from 1 to 30 kg ha^-1, there was a 10% reduction in grain Zn loading efficiency. This trend aligns well with observations made during the first season of the study, indicating a consistent response across different growing periods (51). However, with a further increase in fertilizer rate from 30 to 90 kg ha^-1, the loading efficiency decreased more sharply to 22%, a larger decline compared to the 30% reported in the previous study for the same increase. This sharper decline might be explained by changes in soil chemistry and zinc dynamics at higher application rates.

One plausible explanation for the decrease in ZnHI at higher fertilizer rates is related to the residual available Zn in soil. At lower Zn application rates, residual Zn that remains available in the soil possibly interacts with soil geo-colloids, components such as clay and organic matter that have high cation exchange capacity thereby affecting Zn retention and availability. However, as Zn inputs increase, the residual Zn may become susceptible to losses through leaching, especially in soils prone to water movement, and fixation where Zn becomes tightly bound to soil particles or converted into forms unavailable to plants. These processes reduce the pool of Zn accessible for root uptake and subsequent translocation to grains.

This study’s findings resonate with similar research conducted in China on maize, where ZnHI decreased from 74% to 52% when zinc fertilizer rates increased from 2.3 to 34.1 kg ha^-1 under field conditions (63). Similar results across different agroecological zones underscore a common physiological and soil chemical limitation in Zn uptake and partitioning under high fertilization regimes. Furthermore, an additional physiological mechanism may contribute to this observation. The delivery of Zn to the root xylem, which translocates nutrients from roots to shoots, could be adversely affected by excessive available Zn in the soil. High Zn concentrations have been reported to suppress both the loading of Zn into the xylem and its unloading within the plant tissues, possibly as a regulatory mechanism to prevent Zn toxicity (64, 65). This suppression effectively lowers the transport of Zn to the aerial parts and grains, further reducing ZnHI.