Mycotoxins represent a significant risk to global food safety and security. Each year, one billion metric tons of food products are contaminated, ranking mycotoxins as the third-highest hazardous category reported in the Rapid Alert System for Food and Feed of the European Union [1, 2]. The United Nations Food and Agriculture Organization states that 25% of crops produced worldwide are contaminated and wasted annually due to mycotoxins [3]. These mycotoxins include trichothecenes, fumonisins, ochratoxins, and aflatoxins [4], which are widely recognized for their harmful effects on humans and animals [5, 6], posing a serious threat to public health, agricultural productivity, and food safety worldwide.

Corn is recognized as one of the most susceptible crops to mycotoxins and contamination from mycotoxigenic fungi [7, 8]. Additionally, Aspergillus spp. prefer polysaccharides as its main substrate [9]. The high carbohydrate content in corn provides a secure environment for the fungus to thrive. Furthermore, its open husk structure and mechanical injuries may provide direct access to fungal spores. Among all mycotoxins, aflatoxins pose the greatest health risks due to their hepatotoxic and carcinogenic properties [10, 11]. Primarily, aflatoxins are produced by Aspergillus flavus and A. parasiticus found on cash crops, maize, and peanuts, particularly in warm and humid climates [12]. The principal aflatoxin types include B1, B2, G1, and G2, commonly designated as aflatoxin B1 (AFB1), AFB2, AFG1, and AFG2. Among these, AFB1 is regarded as the most toxic and highly carcinogenic [13]. In addition to corn, aflatoxin contamination is also frequently reported in cereals [14]. Aflatoxin production in corn begins when the corn kernels in the field are infected, particularly by Aspergillus spp. [15, 16]. Aspergillus spp. thrives in warm and humid environments, and Sri Lanka is an ideal location with warm, tropical, humid weather, which is favorable for growing and spreading aflatoxin contamination rapidly [17]. Moreover, Molnár et al. [18] demonstrated that elevated temperatures and drought conditions promote Aspergillus infection in maize kernels, particularly when exposure occurs during the silking stage.

The United States Food and Drug Administration (US FDA) has established a maximum limit of 20 parts per billion (ppb) for total aflatoxins in food [1]. Therefore, the accurate detection and quantification of AFB1 are essential for ensuring food safety and mitigating its impact on communities, including its monitoring in maize and associated soil samples. A range of analytical techniques has been employed to detect aflatoxin contamination, each with distinct advantages and limitations, including thin-layer chromatography (TLC), high-performance liquid chromatography (HPLC), and enzyme-linked immunosorbent assay (ELISA) [15]. Among these, ELISA is particularly well-suited due to its rapid and routine application in aflatoxin detection [19].

Corn is one of the major staple foods worldwide, especially in Sri Lanka, where it is regarded as the second most produced crop and is utilized for both human consumption and animal feed. Since 2007, the annual corn production in Sri Lanka has rapidly increased due to private sector involvement in cultivation [16]. According to the US Department of Agriculture, Sri Lanka comprises five major corn-growing districts: Ampara, Anuradhapura, Badulla, Monaragala, and Kurunegala (Figure 1). In 2022, these districts accounted for approximately 80% of the total corn production in Sri Lanka, with Anuradhapura contributing 33%, followed by Ampara (17%), Monaragala (16%), Badulla (11%), and Kurunegala (4%) [8]. Additionally, these districts belong to different agro-ecological regions due to variations in rainfall, temperature, and altitude [20]. A single study focused on corn in Sri Lanka, conducted in one district, found that corn kernel samples exhibited lower AFB1 levels compared to the soils [14]. However, as the study was confined to a single district, it provides valuable localized insights while highlighting the need for broader investigations to better understand AFB1 contamination in maize and associated soils across Sri Lanka. Most studies indicate that AFB1 contamination in maize is a multifaceted issue influenced by both environmental and agronomic factors. Building on identified gaps in the literature, the present study undertook a comprehensive evaluation of AFB1 contamination in maize kernels and their associated soils across Sri Lanka, with a focus on the country’s principal maize-producing districts. Accordingly, the study was designed to test the overarching hypothesis that AFB1 contamination in maize and its growing soils in Sri Lanka is not random but is significantly shaped by the interaction of environmental conditions (e.g., temperature and soil pH) and the role of soil as a reservoir for aflatoxigenic fungi. To address this, a multi-district survey was conducted across major maize-producing regions, incorporating systematic sampling of both soil and kernel samples, alongside robust nonparametric statistical analyses to evaluate the relationships between these factors and AFB1 contamination.

AFB1 concentrations across all selected districts, encompassing both maize kernel and soil samples, exhibited a range of minimum and maximum values that reflect the overall contamination levels in Sri Lanka (Table 2). The mean AFB1 concentrations in maize samples collected from six districts (n = 30) are presented in Figure 2. Numerous locations showed average AFB1 levels exceeding the safe limit established by the US FDA for corn consumption, which is 20 ppb. Additionally, 13 locations showed corn concentrations below the detection limit, with three for soil (Figure 3).

The lowest mean AFB1 concentration in maize was recorded at Kandaudapanguwa in the Monaragala District (1.64 ppb), while the highest was observed at Kekirawa in the Anuradhapura District (39.811 ppb). For soil samples, the lowest mean concentration occurred at Kanulwewa in the Monaragala District (1.161 ppb), whereas the highest was recorded at Giribawa in the Kurunegala District (181.81 ppb). In comparison, Jayaratne et al. [14] previously reported substantially higher AFB1 levels, with maize kernels from the Anuradhapura District ranging from 60 ppb to 70 ppb and corresponding soils between 350–400 ppb. In contrast, the present study recorded lower levels at Kekirawa, with 7.27 ppb in maize and 80.95 ppb in soil.

The Shapiro-Wilk normality test showed that the data were not normally distributed for both the corn kernel and the corn growing soil. The test indicated non-normality for the corn kernel samples [df = 29, w = 0.69345, p = 9.315 × 10^–7 (p < 0.05)] and for the soil samples [df = 29, w = 0.78348, p = 3.326 × 10^–5 (p < 0.05)].

Statistical analysis using Spearman’s correlation demonstrated a significant positive association between recorded average air temperature and Log₁₀(X + 1)-transformed mean AFB1 concentrations in corn [p-value = 0.00014 (p < 0.05); correlation coefficient (r) = 0.5578; degrees of freedom (df) = 28]. In contrast, no significant association was observed for soil samples [p-value = 0.3968 (p > 0.05); correlation coefficient (r) = 0.1605; degrees of freedom (df) = 28].

Mean AFB1 concentrations in maize-growing soils also showed a positive correlation with AFB1 levels in maize kernels [p-value = 0.0261 (p < 0.05), r = 0.4056, degrees of freedom (df) = 28] (Figure 4).

Environmental parameters, particularly temperature and soil pH, play a significant role in AFB1 production by influencing the metabolic activity of Aspergillus spp. Optimal AFB1 production has been reported at a pH of 5.5–6.5 and temperatures between 20°C and 26°C, while contamination decreases between 26°C and 30°C and is completely inhibited at 30°C [4]. In this study, temperatures recorded across maize-growing regions of Sri Lanka fell around this range. The mean air temperatures observed in each district—Ampara (32.01°C), Anuradhapura (31.51°C), Badulla (21.79°C), Kurunegala (29.70°C), Monaragala (32.91°C), and Polonnaruwa (31.85°C) underscore the relevance of these environmental conditions to the risk of aflatoxin contamination in maize. The tropical climate of Sri Lanka is particularly favorable to Aspergillus spp., making it important to examine the relationship between environmental parameters and AFB1 levels [7]. Importantly, maize is particularly vulnerable during silk emergence, when prevailing environmental conditions critically determine the extent of AFB1 infection [5]. However, contamination by Aspergillus spp. may be influenced by rainfall, agronomic practices, and post-harvest handling factors, which were not captured in the present study. These variables will be considered in future studies.

Aspergillus spp. exhibit optimal growth at temperatures ranging from 25°C to 30°C, with growth reduced to half the maximum at 20°C and five times lower at 15°C [11]. This supports the idea that mean corn AFB1 levels have a significant positive correlation [for corn, p-value = 0.00014 (p < 0.05)] with the recorded average air temperature (Figure 5), indicating AFB1 contamination may be significantly driven by average air temperature. Moreover, the statistical analysis showed a positive correlation between mean corn kernel AFB1 levels and mean soil AFB1 levels [p-value = 0.0261 (p < 0.05)]. Previous studies suggest that soil may act as a primary inoculum source for aflatoxin-producing Aspergillus spp. in corn kernels, with spores being transferred into the kernels via wind and rain [14]. While the positive correlation observed between mean soil AFB1 levels and maize kernel AFB1 levels supports a potential association, it does not confirm soil as the primary source of contamination, as the origin of the inoculum was not directly assessed. Additionally, this study found no significant correlation between AFB1 concentrations in maize kernels and the corresponding soil pH. However, earlier research demonstrated that Aspergillus spp. producing AFB1 is more prevalent in high pH soils than in low pH soils, such as those used for growing peanuts [21–27], providing clear evidence of the impact of Aspergillus spp. in the soil. Although ELISA assays indicate these AFB1 contaminations, these findings are reported with caution due to potential cross-reactivity and matrix effect associated with the assay; these values may reflect slight over- or underestimation, therefore these findings should be considered as preliminary evidence and future studies should use more specific analytical techniques.

The recovery results ranged from 73.9% to 82.3% for maize samples and 82.8% to 99.1% for soil samples, which largely fall within the acceptable range specified by the assay kit. However, the use of only duplicate measurements represents a methodological limitation, and the reported recovery values should therefore be regarded as preliminary. Furthermore, key analytical performance parameters—including the limit of detection (LOD), linear range, intra- and inter-assay precision, and validation using certified reference materials—were not independently established in this study and instead relied on the manufacturer’s specifications. As a result, the analytical output should be interpreted as semi-quantitative. Future studies should therefore incorporate full method validation with increased replication and independent determination of performance characteristics.

Despite its broad scope, this study was subject to several limitations. Seasonal rainfall patterns driven by the monsoon system in Sri Lanka posed challenges in synchronizing sampling across the study period. In addition, marked variations in ambient temperature among the six maize-growing districts complicated the formulation of unified recommendations for AFB1 management, given the strong influence of temperature on fungal proliferation. Heterogeneity in land use and land preparation practices further limited the implementation of a fully standardized sampling strategy.

Post-harvest handling practices also varied considerably across study areas, and the generally limited infrastructure and minimal adoption of value-adding processes may contribute to increased AFB1 risk, although these factors could not be quantitatively assessed in the present study. Moreover, reliable field humidity data were not available for the sampling sites during the study period. Potential influences of UV-related AFB1 degradation, as well as detailed soil physicochemical properties and microbial activity, were also not assessed.

From an analytical standpoint, future studies should incorporate confirmatory techniques such as LC-MS/MS to ensure specificity and to rule out potential interference from degradation products or co-occurring mycotoxins. Although no significant association was observed between soil pH and AFB1 levels in the present study, previous reports have indicated a possible relationship; thus, this interaction should be re-examined using improved environmental datasets and refined analytical approaches.

To address current limitations, multi-seasonal sampling is recommended to account for monsoon-driven rainfall variability, alongside agroecological analyses that better capture the combined influence of rainfall patterns and temperature regimes. Future research should also consider key agronomic and environmental variables, including land preparation practices, fertilizer application, crop management, humidity, and soil organic matter, as these are critical for developing district-specific AFB1 mitigation strategies.

Continuous monitoring of field humidity, air temperature, and soil temperature would further strengthen the understanding of environmental drivers of contamination. In addition, the potential effects of UV-related AFB1 degradation at sampling sites and detailed soil properties, particularly organic matter content, should be included to better interpret soil microbial activity and toxin degradation dynamics.

From a methodological perspective, the use of larger, more balanced datasets across districts is needed to better disentangle geographic and agronomic effects. Incorporating agroecological stratification and applying multivariate or mixed-effects models would further improve the ability to separate interacting environmental influences on AFB1 levels. Moreover, future studies should explicitly consider district-level interactions with key variables such as temperature and soil pH.

Finally, strengthening post-harvest systems through farmer education on best handling practices and promoting value addition of surplus produce could significantly reduce aflatoxin risks in the food chain. Collectively, these improvements are essential for developing sustainable, regionally tailored strategies for AFB1 mitigation.