Spatial Variation and Sources of Heavy Metals in Soils Near Waste Disposal Points on University Farms in Rivers State Nigeria

material were 95–105%, and replicate relative standard deviations were below 5%. Detection limits were set at three times the standard deviation of the blank. Data analysis The triplicate readings of each composite were averaged, and the 48 composite values were used in all tests. Concentrations were summarised as mean ± standard deviation (SD). They were compared with the Dutch target and intervention values (Ministry of Housing, Spatial Planning and the Environment, 2000), adopted in Nigeria’s Environmental Guidelines and Standards for the Petroleum Industry (EGASPIN) (Department of Petroleum Resources, 2018), and the number of samples exceeding each target value was counted. No guideline values exist for K and Mg. The effects of farm (four levels), sampling station ST1, ST and month (April, May, June) on each element were tested together in a three-factor main-effects analysis of variance , with the two sampling rounds in each month as replicates (n = 48; 41 residual degrees of freedom). Testing the three factors in one model removes the variation due to the other two from the error term. Effect sizes are given as partial eta squared (η2p). Significant farm and month effects were followed by Tukey’s honestly significant difference test using the residual mean square of the model. For each element, the number of matched pairs (same farm, month and round) in which ST1 exceeded ST2 was also counted. Normality of residuals was checked with the Shapiro–Wilk test, and the ANOVA was repeated on log-transformed concentrations as a sensitivity check. Relationships among the eight elements were examined with Pearson correlation (n = 48) and with PCA of the correlation matrix of standardised concentrations. The suitability of the data for PCA was checked with the Kaiser–Meyer–Olkin measure and Bartlett’s test of sphericity. Loadings with an absolute value of 0.50 or more were treated as strong. Analyses were performed in Python (NumPy and SciPy) and Microsoft Excel 365, with significance set at p < 0.05. Results Concentrations and comparison with guideline values Table 1 shows that the mean concentrations of Zn (2.07–3.76 mg/kg), Pb (1.03–1.36 mg/kg), Cr (0.57–0.72 mg/kg), Co (0.50–0.82 mg/kg) and As (0.36–0.46 mg/kg) at the four farms were one to two orders of magnitude below their target values, and no sample exceeded any of these targets. Cd was the exception. Its overall mean of 0.72 mg/kg was close to the 0.8 mg/kg target value, and the Etche mean (0.82 ± 0.58 mg/kg) was marginally above it. Eighteen of the 48 samples (37.5%) exceeded the Cd target; these came from all four farms, and 15 of the 18 were from ST1. The highest Cd value, 2.09 mg/kg, was recorded at Etche ST1 in April. No sample approached any intervention value. K (106.50–116.89 mg/kg) and Mg (29.21–35.51 mg/kg) occurred at far higher concentrations than the heavy metals, as expected for essential plant nutrients. Table 1: Mean Concentrations of Heavy Metals and Macro Elements in University Farm Soils Compared With Soil Guideline Values (mg/kg) Element Emohua Etche Ndele Choba Overall mean Target value Intervention value Samples > target (of 48) Zn 2.07 ± 0.54 2.80 ± 1.03 3.76 ± 2.00 2.75 ± 1.38 2.85 140 720 0 Pb 1.15 ± 0.38 1.03 ± 0.37 1.36 ± 0.51 1.36 ± 0.56 1.23 85 530 0 Cd 0.71 ± 0.27 0.82 ± 0.58 0.72 ± 0.32 0.62 ± 0.23 0.72 0.8 12 18 Cr 0.60 ± 0.24 0.57 ± 0.26 0.61 ± 0.16 0.72 ± 0.32 0.63 100 380 0 Co 0.50 ± 0.19 0.72 ± 0.21 0.82 ± 0.43 0.80 ± 0.40 0.71 9 240 0 As 0.46 ± 0.13 0.43 ± 0.23 0.44 ± 0.21 0.36 ± 0.10 0.42 29 55 0 K 106.50 ± 18.53 113.53 ± 9.27 116.89 ± 18.90 114.35 ± 14.32 – – – – Mg 34.65 ± 8.99 29.21 ± 8.21 32.02 ± 7.23 35.51 ± 6.38 – – – – Note. Farm values are mean ± SD of 12 composite samples (2 stations × 2 rounds × 3 months), mg/kg dry weight. Target and intervention values are the Dutch values (Ministry of Housing, Spatial Planning and the Environment, 2000) adopted in EGASPIN (Department of Petroleum Resources, 2018); none exist for K and Mg. After station and month were accounted for, only three elements differed significantly between farms Table 2: Zn F3, 41 = 8.45, p < 0.001, η2p = 0.38), Co F 3, 41 = 4.04, p = 0.013, η2p = 0.23 and Mg F 3, 41 = 3.59, p = 0.022, η2p = 0.21). Tukey’s test (Table 5) showed that Zn was higher at Ndele than at Emohua (difference 1.69 mg/kg, p < 0.001), Choba (1.01 mg/kg, p = 0.023) and Etche (0.96 mg/kg, p = 0.033). Co was lower at Emohua than at Ndele (0.32 mg/kg, p = 0.019) and Choba (0.31 mg/kg, p = 0.027), and Mg was higher at Choba than at Etche (6.30 mg/kg, p = 0.025). Pb (p = 0.090), K (p = 0.143), As (p = 0.259), Cr (p = 0.295) and Cd (p = 0.479) did not differ between farms. Differences between farms were therefore limited and smaller than the differences between stations described next. Table 2: Three-Factor Analysis of Variance of Element Concentrations by Farm, Sampling Station and Month Farm (df = 3, 41) Station (df = 1, 41) Month (df = 2, 41) Element F p η2p F P η2p F p η2p Zn 8.45 < 0.001 0.38 41.23 < 0.001 0.50 17.10 < 0.001 0.45 Pb 2.32 0.090 0.15 18.47 < 0.001 0.31 4.02 0.025 0.16 Cd 0.84 0.479 0.06 17.47 < 0.001 0.30 5.73 0.006 0.22 Cr 1.28 0.295 0.09 12.72 < 0.001 0.24 5.87 0.006 0.22 Co 4.04 0.013 0.23 12.04 0.001 0.23 9.83 < 0.001 0.32 As 1.39 0.259 0.09 39.28 < 0.001 0.49 3.94 0.027 0.16 Farm (df = 3, 41) Station (df = 1, 41) Month (df = 2, 41) Element F p η2p F P η2p F p η2p K 1.91 0.143 0.12 32.76 < 0.001 0.44 6.95 0.003 0.25 Mg 3.59 0.021 0.21 47.28 < 0.001 0.54 4.97 0.012 0.20 Note. Model: element = farm + station + month (additive main effects; n = 48; residual df = 41); each effect is adjusted for the other two. η2p = partial eta squared. Values in bold are significant at p < 0.05. Spatial variation between stations in the study area Every element was significantly more concentrated at ST1, near the waste disposal points, than at ST2 after farm and month were allowed for (Tables 2 and 3; Figure 2). The station effect was the largest of the three factors for seven of the eight elements (all except Co). Mean Zn was 3.61 mg/kg at ST1 against 2.08 mg/kg at ST2, Cd 0.90 against 0.54 mg/kg, Pb 1.46 against 0.99 mg/kg and as 0.53 against 0.31 mg/kg; overall, the six heavy metals were 40–74% higher at ST1. ST1 exceeded ST2 in 20 to 24 of the 24 matched pairs, depending on the element, and in all 24 for K, Zn and As. The station effects stayed significant when the analysis was repeated on log-transformed data. Table 3: Concentrations of Heavy Metals and Macro Elements at Stations Near (ST1) and Away From (ST2) the Waste Disposal Points Element ST1 mean ± SD (mg/kg) ST2 mean ± SD (mg/kg) ST1 higher by (%) F(1,41) P Pairs with ST1 > ST2 (of 24) K 122.03 ± 14.75 103.60 ± 10.51 18 32.77 < 0.001 24 Zn 3.61 ± 1.57 2.08 ± 0.73 74 41.23 < 0.001 24 Mg 38.00 ± 4.35 27.69 ± 7.32 37 47.28 < 0.001 22 Pb 1.46 ± 0.40 0.99 ± 0.42 47 18.47 < 0.001 22 As 0.53 ± 0.17 0.31 ± 0.08 71 39.28 < 0.001 24 Cd 0.90 ± 0.39 0.54 ± 0.24 67 17.48 < 0.001 20 Cr 0.73 ± 0.24 0.52 ± 0.21 40 12.72 < 0.001 20 Co 0.84 ± 0.35 0.58 ± 0.28 45 12.04 0.001 20 Note. n = 24 samples per station. F and p are for the station effect in the three-factor ANOVA (element = farm + station + month), adjusted for farm and month; η2p = 0.23–0.54. Pairs are matched by farm, month and sampling round. Figure 2. Mean (± SD) concentrations of (a) heavy metals and (b) macro elements at stations within 10 m of the waste disposal points (ST1) and 150–200 m away (ST2), across four university farms in Rivers State (n = 24 per station). Spatial variation across months in the study area Tukey’s test identified which months differed (Table 4). Three patterns were seen. For K, Mg, Pb, As, Cd and Cr, the lowest mean occurred in May, and the significant differences lay mainly between April and May: April exceeded May by 13.19 mg/kg for K, 5.75 mg/kg for Mg, 0.33 mg/kg for Pb, 0.11 mg/kg for As, 0.35 mg/kg for Cd and 0.24 mg/kg for Cr (p = 0.005–0.046). K also rose again from May to June (12.22 mg/kg, p = 0.010), and Cr was higher in April than in June (0.19 mg/kg, p = 0.036). Zn fell steadily, from 3.76 mg/kg in April to 2.70 mg/kg in May and 2.07 mg/kg in June, a decline of 45%, with April significantly higher than both May (p = 0.002) and June (p < 0.001). Co followed the opposite course and was highest in June (0.93 mg/kg), significantly above both April (by 0.26 mg/kg, p = 0.017) and May (by 0.39 mg/kg, p < 0.001). The largest single increases in Co-occurred at ST1 of Choba and Ndele in June. When the analysis was repeated on log-transformed data, the month effects for Mg (p = 0.069) and As (p = 0.073) were no longer significant, so these two should be read with caution. All other month effects were unchanged. Table 4: Spatial Variations in Heavy Metal Concentrations (Mean ± SD, mg/kg) in University Farm Soils across Stations and Months Month Location Station K Zn Mg Pb As Cd Cr Co Month 1 Emohua ST1 143.43 ± 0.08 2.45 ± 0.12 37.48 ± 0.44 1.41 ± 0.05 0.49 ± 0.03 1.06 ± 0.05 0.88 ± 0.06 0.72 ± 0.04 Emohua ST2 93.38 ± 0.17 1.64 ± 0.12 45.45 ± 0.08 0.55 ± 0.03 0.25 ± 0.02 0.78 ± 0.02 0.48 ± 0.04 0.42 ± 0.03 Etche ST1 122.87 ± 1.53 4.41 ± 0.09 37.14 ± 0.05 1.16 ± 0.04 0.78 ± 0.03 1.79 ± 0.04 0.52 ± 0.03 0.82 ± 0.06 Etche ST2 100.29 ± 0.08 2.33 ± 0.03 15.22 ± 0.09 0.78 ± 0.02 0.26 ± 0.04 0.49 ± 0.04 0.80 ± 0.11 0.52 ± 0.07 Ndele ST1 141.27 ± 0.03 7.22 ± 0.20 39.64 ± 0.27 0.80 ± 0.03 0.69 ± 0.07 0.61 ± 0.03 0.37 ± 0.01 0.42 ± 0.03 Ndele ST2 120.04 ± 0.02 3.93 ± 0.19 27.33 ± 0.21 1.83 ± 0.16 0.39 ± 0.02 0.27 ± 0.01 0.38 ± 0.03 0.46 ± 0.05 Choba ST1 106.19 ± 0.70 5.17 ± 0.16 41.25 ± 0.29 1.82 ± 0.09 0.35 ± 0.02 0.86 ± 0.04 1.17 ± 0.05 0.37 ± 0.04 Choba ST2 105.07 ± 0.06 1.34 ± 0.07 34.19 ± 0.07 1.39 ± 0.03 0.20 ± 0.01 0.38 ± 0.04 0.64 ± 0.02 0.78 ± 0.04 Month 2 Emohua ST1 106.59 ± 2.34 2.44 ± 0.05 27.26 ± 0.15 1.44 ± 0.14 0.57 ± 0.03 0.61 ± 0.03 0.46 ± 0.04 0.48 ± 0.04 Emohua ST2 92.10 ± 0.11 1.79 ± 0.20 23.81 ± 0.79 0.74 ± 0.08 0.34 ± 0.05 0.43 ± 0.01 0.35 ± 0.05 0.33 ± 0.04 Etche ST1 111.32 ± 0.42 3.40 ± 0.34 32.62 ± 1.52 0.93 ± 0.04 0.27 ± 0.03 0.50 ± 0.02 0.60 ± 0.06 0.75 ± 0.05 Etche ST2 106.48 ± 2.33 2.01 ± 0.04 27.54 ± 0.49 0.42 ± 0.04 0.25 ± 0.02 0.31 ± 0.03 0.37 ± 0.04 0.44 ± 0.05 Ndele ST1 106.27 ± 2.74 4.07 ± 0.14 37.47 ± 0.87 1.98 ± 0.24 0.56 ± 0.04 0.97 ± 0.05 0.71 ± 0.04 0.92 ± 0.12 Ndele ST2 84.02 ± 0.32 2.16 ± 0.15 24.94 ± 0.67 0.88 ± 0.02 0.24 ± 0.03 0.34 ± 0.03 0.67 ± 0.06 0.26 ± 0.05 Month Location Station K Zn Mg Pb As Cd Cr Co Choba ST1 121.45 ± 0.05 3.24 ± 0.06 37.50 ± 0.11 0.92 ± 0.14 0.52 ± 0.03 0.42 ± 0.03 0.58 ± 0.03 0.67 ± 0.05 Choba ST2 104.42 ± 0.91 2.37 ± 0.07 25.62 ± 0.09 0.64 ± 0.10 0.30 ± 0.09 0.56 ± 0.06 0.43 ± 0.03 0.37 ± 0.06 Month 3 Emohua ST1 102.69 ± 0.94 2.54 ± 0.12 43.54 ± 0.32 1.38 ± 0.07 0.56 ± 0.07 0.68 ± 0.02 0.80 ± 0.03 0.50 ± 0.01 Emohua ST2 97.49 ± 0.60 1.21 ± 0.06 27.95 ± 1.91 1.08 ± 0.02 0.39 ± 0.04 0.47 ± 0.07 0.38 ± 0.02 0.30 ± 0.03 Etche ST1 125.35 ± 2.16 2.42 ± 0.10 37.49 ± 0.40 1.50 ± 0.24 0.51 ± 0.02 1.01 ± 0.13 0.52 ± 0.12 0.91 ± 0.04 Etche ST2 113.00 ± 2.51 1.78 ± 0.05 23.60 ± 0.73 1.11 ± 0.04 0.34 ± 0.05 0.49 ± 0.02 0.25 ± 0.04 0.59 ± 0.09 Ndele ST1 128.30 ± 5.70 2.70 ± 0.25 37.13 ± 2.65 1.30 ± 0.03 0.34 ± 0.03 0.80 ± 0.03 0.82 ± 0.08 1.42 ± 0.09 Ndele ST2 118.64 ± 0.50 1.86 ± 0.07 22.60 ± 1.45 1.00 ± 0.02 0.20 ± 0.03 1.12 ± 0.09 0.50 ± 0.09 1.18 ± 0.15 Choba ST1 141.83 ± 1.14 2.06 ± 0.12 42.36 ± 2.18 2.05 ± 0.06 0.37 ± 0.04 0.85 ± 0.07 0.81 ± 0.04 1.51 ± 0.15 Choba ST2 104.09 ± 2.52 1.80 ± 0.05 29.79 ± 2.85 1.03 ± 0.02 0.29 ± 0.05 0.42 ± 0.05 0.42 ± 0.05 0.86 ± 0.11 Values are mean ± SD of triplicate determinations of the first sampling round (Sample 1) of each month. Results for both sampling rounds are given in Appendix I, and the statistical tests in this section use all 48 samples. Table 4 shows that heavy metal concentrations varied across the four university farms. Ndele recorded the highest mean concentration of Zn (3.76 ± 2.00 mg/kg) and Emohua the lowest (2.07 ± 0.54 mg/kg). Etche had the highest mean Cd (0.82 ± 0.58 mg/kg) and Choba the lowest (0.62 ± 0.23 mg/kg), while Pb was highest at Ndele and Choba (1.36 mg/kg). The macro elements K and Mg occurred at much higher concentrations than the heavy metals, with K ranging from 106.50 ± 18.53 mg/kg at Emohua to 116.89 ± 18.90 mg/kg at Ndele. Compared with the Dutch target values adopted in EGASPIN (Table 1), the mean concentrations of Zn, Pb, As, Cr and Co were one to two orders of magnitude below their targets, and no individual sample exceeded them. The overall mean Cd concentration (0.72 ± 0.37 mg/kg) was close to its target value of 0.8 mg/kg but not significantly above it, although the farm mean at Etche (0.82 mg/kg) was marginally above it. However, 18 of the 48 samples (37.5%) exceeded 0.8 mg/kg; 15 of these were from ST1 stations, and they occurred at all four farms. The highest value was 2.09 mg/kg, at Etche ST1 in Month 1. Table 5: Tukey’s Honestly Significant Difference Test of Differences in Element Concentrations between Farms and Between Months (mg/kg) Comparison Zn Pb Cd Cr Co As K Mg (a) Farm Ndele − Emohua 1.69 (< 0.001) – – – 0.32 (0.019) – – −2.63 (0.606) Ndele − Etche 0.96 (0.033) – – – 0.10 (0.763) – – 2.81 (0.552) Ndele − Choba 1.01 (0.023) – – – 0.01 (0.999) – – −3.49 (0.366) Choba − Emohua 0.68 (0.203) – – – 0.31 (0.027) – – 0.86 (0.977) Choba − Etche −0.05 (0.999) – – – 0.09 (0.838) – – 6.30 (0.025) Etche − Emohua 0.73 (0.154) – – – 0.22 (0.169) – – −5.44 (0.065) (b) Month April − May 1.06 (0.002) 0.33 (0.040) 0.35 (0.005) 0.24 (0.006) 0.13 (0.314) 0.11 (0.046) 13.19 (0.005) 5.75 (0.009) April − June 1.69 (< 0.001) 0.02 (0.986) 0.13 (0.414) 0.19 (0.036) −0.26 (0.017) 0.10 (0.055) 0.98 (0.967) 2.24 (0.450) June − May −0.63 (0.091) 0.31 (0.057) 0.22 (0.108) 0.05 (0.777) 0.39 (< 0.001) 0.00 (0.996) 12.22 (0.010) 3.51 (0.149) Note. Values are mean differences (first level minus second level, mg/kg) with Tukey HSD p values in parentheses, based on the residual mean square of the three-factor ANOVA (df = 41). Farm means are based on 12 samples and month means on 16 samples. Values in bold are significant at p < 0.05. –, Tukey’s test not applied because the farm effect was not significant (Table 2). Figure 3. Monthly mean (± SD) concentrations of the eight elements in university farm soils from April to June 2026 (n = 16 samples per month). The y-axis scale differs between panels. Inter-element relationships and principal component analysis Table 5 shows the most pairs of elements were positively correlated, and no pair was significantly negatively correlated. The strongest relationships were As–Zn (r = 0.67), Co–K (r = 0.61), As–Cd (r = 0.58) and Cr–Pb r = 0.51, all p < 0.001. The full correlation matrix is given in Table 5. Table 6: Pearson Correlation Matrix of Heavy Metals and Macro Elements in University Farm Soils (n = 48) K Zn Mg Pb As Cd Cr Co K 1 Zn 0.44** 1 Mg 0.41** 0.38** 1 Pb 0.42** 0.28 0.41** 1 As 0.49** 0.67** 0.44** 0.36* 1 Cd 0.45** 0.32* 0.42** 0.34* 0.58** 1 Cr 0.16 0.25 0.39** 0.50** 0.13 0.38** 1 Co 0.61** −0.03 0.30* 0.47** 0.07 0.49** 0.31* 1 Note. Values are Pearson correlation coefficients (r) calculated across the 48 composite samples. * p < 0.05; ** p < 0.01. The data were suitable for PCA (KMO = 0.64; Bartlett’s test, χ2 (28) = 165.1, p < 0.001). The first three components explained 74.7% of the total variance Table 6; Figure 4. PC1 (eigenvalue 3.65; 45.6% of variance) loaded positively and strongly on all eight elements 0.56–0.75, most of all on K, Cd, As, Mg and Pb, and therefore represents their joint enrichment, in line with the positive correlations between most pairs of elements. It separated the two stations clearly: ST1 samples had a mean PC1 score of 1.49 and ST2 samples a mean of −1.49. The highest PC1 scores were at Etche ST1 in April (3.48), Choba ST1 in June (2.74) and Emohua ST1 in April (2.53), and the lowest at Emohua ST2 (−2.68) and Ndele ST2 (−2.62) in May. PC2 (eigenvalue 1.36; 17.0%) contrasted Co (0.67), Cr (0.33) and Pb (0.28) with Zn (−0.63) and As (−0.57). The correlation matrix (Table 5) supports this split: Zn and As were strongly correlated with each other (r = 0.67, p < 0.01) but not with Co (r = −0.03 and 0.07), whereas Co was correlated with K (r = 0.61), Cd (r = 0.49), Pb (r = 0.47) and Cr (r = 0.31), and Pb with Cr (r = 0.51). PC2 picked out two distinct sample groups. Ndele ST1 in April had the most negative PC2 score (−3.47), reflecting its high Zn and As, while Choba ST1 in June had the highest score (2.48), reflecting its high Co. PC3 (eigenvalue 0.97, just below the Kaiser criterion of 1; 12.1%) was dominated by Cr (0.66), set against K (−0.50) and Co (−0.37), and was highest at Choba ST1 in April. This agrees with the weak, non-significant correlations of Cr with K (r = 0.16), As (r = 0.13) and Zn (r = 0.25) in Table 5, which show that Cr varied partly independently of the other elements. Table 6: Principal Component Loadings of Element Concentrations in University Farm Soils (n = 48) Element PC1 PC2 PC3 K 0.751 0.038 −0.496 Cd 0.747 0.033 −0.129 As 0.715 −0.570 −0.106 Mg 0.701 −0.044 0.209 Pb 0.696 0.284 0.250 Zn 0.615 −0.633 0.118 Co 0.591 0.665 −0.368 Cr 0.555 0.332 0.663 Eigenvalue 3.65 1.36 0.97 Variance explained (%) 45.6 17.0 12.1 Cumulative (%) 45.6 62.6 74.7 Note. Loadings are correlations between each element and the component, from the correlation matrix of standardised concentrations. Bold values: |loading| ≥ 0.50. KMO = 0.64; Bartlett’s test χ2(28) = 165.1, p < 0.001. Figure 4. PCA biplot of element concentrations in university farm soils (n = 48). Points are sample scores, coloured by station (red, ST1 near the waste disposal point; blue, ST2) and shaped by farm; arrows are element loadings (scaled ×3.4). Discussion Metal levels in context By Nigerian and international standards, metal levels on the four farms were low. They were far below those of mining areas in south-eastern Nigeria, where Obiora et al. (2016) found Zn above 200 mg/kg in about 31% of arable soils around the Enyigba Pb–Zn mines and Chukwu and Oji (2018) reported mean Zn of 210 mg/kg and Cd of 8.57 mg/kg around an abandoned mine at Mkpuma Ekwoku. The parameters were also below farmland in Girei, Adamawa State (Zn 15.0–30.1 mg/kg; Cd 3.38–7.58 mg/kg) (Hope et al., 2025), and were closest to farmland around the Ashangwa rice mill in Nasarawa State (Zn 0.10–0.81 mg/kg; Cd 0.014–0.565 mg/kg) (Augustine et al., 2025). Farmland near industry in Anxin County, China, held far more Zn (171.7 mg/kg) but a similar level of Cd (0.73 mg/kg) (Zhao et al., 2022). The most direct comparison is with Iyama et al. (2022), sampled farms of the same three institutions. The Cd means of the present study (0.62–0.82 mg/kg) fall within same range of 0.46–1.42 mg/kg, which supports the view that Cd is the metal of most concern on these farms. The present Pb means 1.03–1.36 mg/kg, however, were about three to seven times lower than result obtained from that study ranging between 4.35–6.80 mg/kg, and the Cr means about eight to fifteen times lower. The earlier study sampled farms inside Port Harcourt metropolis, where traffic and urban runoff add Pb and Cr, and used X-ray fluorescence, which measures total content, including the silicate-bound fraction that aqua regia does not dissolve. Waste disposal points as hotspots of metal enrichment The clearest finding of the study is that soil within 10 m of each farm’s waste disposal point was richer in every element measured than soil 150–200 m away, a pattern seen across all four farms. The station effect was larger than the farm effect for almost every element, so where soil sits in relation to the disposal point mattered more than which farm it belonged to. This agrees with studies of soils beside municipal dumpsites in Nigeria (Ekere et al., 2020; Ihedioha et al., 2017; Omada et al., 2024) and with the near–far contrast reported by Jolly et al. (2025) in Bangladesh, where cropland next to a disposal site carried several times more Pb, Cd and Zn. The contrast on the university farms was smaller (40–74% for the heavy metals), which fits the smaller size and largely organic make-up of farm waste heaps compared with municipal dumps. The enrichment of K and Mg together with the heavy metals is informative. These nutrients are released as crop residues and other organic waste decompose. Their rise at ST1 alongside Zn, Cd, Pb and As suggests that the disposal points act as a combined source, delivering both nutrients and metals to nearby soil, rather than one particular item in the heap being responsible. Cadmium near the disposal points deserves particular attention. Although the overall Cd mean stayed just below the target value, 15 of the 18 exceedances came from ST1, and the highest value 2.09 mg/kg. Etche ST1 was more than two and a half times the target. The Dutch target assumes a standard soil with 25% clay and 10% organic matter (Ministry of Housing, Spatial Planning and the Environment, 2000), so the corrected target for the sandy loam soils of the study area would be lower than 0.8 mg/kg, and the exceedances are, if anything, understated. Because ST1 lies beside cultivated beds, crops grown nearest the disposal points are the ones most likely to take up Cd. Short-term spatial variation Significant changes over only three months show that soil metal levels on these farms are not static during the rainy season. For most elements, levels dropped from April to May and partly recovered in June. The sampling window covers the onset of the rains, and one plausible explanation is that early heavy rainfall leached or washed metals from the topsoil, after which new waste and farm inputs began to restore them. Zn, which fell steadily across all three months, may be more mobile under these conditions, while the rise in Co in June was concentrated at the ST1 stations of Choba and Ndele and may reflect a fresh input at those disposal points. These explanations remain tentative, because rainfall, soil pH and organic matter were not measured, and the month effects for Mg and As were not robust to log transformation. The practical point is clear. The mean Zn concentration in April was 1.8 times that in June, and Cd was 1.7 times higher in April than in May. A single round of sampling therefore over- or under-estimate soil metal levels considerably, depending on when it is done. Monitoring programmes and health risk assessments on these farms should use repeated sampling over time rather than one-off measurements. Likely sources of the elements The positive correlations among most elements and the strong loading of all eight on PC1 point to a shared source, and the clear separation of ST1 from ST2 samples along PC1 identifies the waste disposal points as that source. Similar positive inter-element correlations were attributed to common sources by Chukwu and Oji (2018) and Nuralykyzy et al. (2021). PC2 adds a second layer. It contrasted a Zn–As group with a Co–Cr–Pb group, and each was driven by particular samples: Zn and As peaked at Ndele ST1 in April, and Co at Choba ST1 in June. The strong As–Zn correlation r = 0.67 is consistent with agrochemical inputs, since some fertilizers and pesticides contain both elements (Wan et al., 2024), whereas the Co–Cr– Pb group may reflect discarded containers, metal scraps or other non-organic items in the heaps. PCA alone cannot confirm these sources. Analysis of the waste heaps themselves, or source-specific methods such as positive matrix factorization or isotope tracing, would be needed. Conclusion Heavy metal levels in the soils of four university farms in Rivers State were low and, except for Cd, well below Nigerian target values. Waste disposal points are nevertheless clear local hotspots: all eight elements measured were significantly more concentrated within 10 m of the disposal points than 150–200 m away, and PCA confirmed that the elements are enriched together at these points. Cd exceeded its target value in 37.5% of samples, mostly near the disposal points. Concentrations also changed significantly from month to month within a single rainy season. Recommendations 1. Farm managements should segregate waste at source, compost the organic fraction and remove plastics, fertilizer bags and agrochemical containers from the farm. 2. Waste disposal points should be relocated away from cultivated beds, and crops should not be grown in the immediate vicinity of existing heaps. 3. Soil Cd should be monitored routinely, with more frequent sampling near disposal points, and phosphate fertilizers should be used with caution. 4. Monitoring and risk assessments should rely on repeated sampling across both wet and dry seasons rather than a single round. Further studies should sample along transects at increasing distances from the disposal points, include an uncontaminated reference site, record rainfall, measure soil physicochemical properties and metal uptake by crops, and characterize the waste heaps directly to confirm the sources suggested here. References Akhter, P., Khan, Z., Hussain, M., Ahmad, K., Awan, M., Ashfaq, A., Chaudhry, U., Ullah, M., Abideen, Z., Almaary, K., Alwahibi, M., & Elshikh, M. (2022). Assessment of heavy metal accumulation in soil and garlic influenced by waste-derived organic amendments. Biology, 11(6), 850. https://doi.org/10.3390/biology11060850 Alengebawy, A., Abdelkhalek, S. T., Qureshi, S. R., & Wang, M. (2021). Heavy metals and pesticides toxicity in agricultural soil and plants: Ecological risks and human health implications. Toxics, 9(3), 42. https://doi.org/10.3390/toxics9030042 Augustine, U., Musa, S., & Abubakar, M. (2025). Determination of heavy metals concentration in soils from farmlands within Ashangwa rice mill, Lafia LGA, Nasarawa State, Nigeria. Journal of the Chemical Society of Nigeria, 50(4), 33–48. https://doi.org/10.4314/jcsn.v50i4.3 Chukwu, A., & Oji, K. K. (2018). Assessment of Pb, Zn, As, Ni, Cu, Cr and Cd in agricultural soils around settlements of abandoned lead-zinc mine in Mkpuma Ekwoku, southeastern Nigeria. Journal of Applied Sciences and Environmental Management, 22(9), 1485–1488. https://doi.org/10.4314/jasem.v22i9.21 Department of Petroleum Resources. (2018). Environmental guidelines and standards for the petroleum industry in Nigeria (EGASPIN) (Rev. ed.). Ekere, N. R., Ugbor, M. C. J., Ihedioha, J. N., Ukwueze, N. N., & Abugu, H. O. (2020). Ecological and potential health risk assessment of heavy metals in soils and food crops grown in abandoned urban open waste dumpsite. Journal of Environmental Health Science and Engineering, 18(2), 711–721. https://doi.org/10.1007/s40201-020-00497- 6 Hope, E., Usaku, R., Dowell, B., & Hauwa, A. (2025). Heavy metals levels in leguminous crops and associated soils from six farming communities in Girei Local Government Area, Adamawa State, Nigeria. Journal of Applied Sciences and Environmental Management, 29(11), 101–118. https://doi.org/10.4314/jasem.v29i11.3 Ihedioha, J. N., Ukoha, P. O., & Ekere, N. R. (2017). Ecological and human health risk assessment of heavy metal contamination in soil of a municipal solid waste dump in Uyo, Nigeria. Environmental Geochemistry and Health, 39(3), 497–515. https://doi.org/10.1007/s10653-016-9830-4 Iyama, W. A., Okpara, K., & Techato, K. (2022). Assessment of heavy metals in agricultural soils and plant (Vernonia amygdalina Delile) in Port Harcourt metropolis, Nigeria. Agriculture, 12(1), 27. https://doi.org/10.3390/agriculture12010027 Jolly, F. A., Pandit, D., Adikari, D., Saifullah, M. K., Uddin, G. T., Abir, S. T., Kumar, R., Dutta, A., & Roy, R. (2025). Comparison of heavy metal contamination in cropland soils adjacent to waste disposal site and other cropland soils in Khulna, Bangladesh. Discover Sustainability, 6(1), 745. https://doi.org/10.1007/s43621-025-01688-7 Ministry of Housing, Spatial Planning and the Environment. (2000). Circular on target values and intervention values for soil remediation (Ref. DBO/1999226863). Natural Earth. (n.d.). Free vector and raster map data at 1:10m scale. https://www.naturalearthdata.com Nuralykyzy, B., Wang, P., Deng, X., An, S., & Huang, Y. (2021). Heavy metal contents and assessment of soil contamination in different land-use types in the Qaidam Basin. Sustainability, 13(21), 12020. https://doi.org/10.3390/su132112020 Obiora, S., Chukwu, A., & Davies, T. (2016). Heavy metals and health risk assessment of arable soils and food crops around Pb-Zn mining localities in Enyigba, southeastern Nigeria. Journal of African Earth Sciences, 116, 182–189. https://doi.org/10.1016/j.jafrearsci.2015.12.025 Omada, J. I., Ogoko, E. C., Kelle, H. I., & Gideon, Y. B. (2024). Heavy metal pollution indices in soil and plants within the vicinity of the Gosa dumpsite in Abuja, Nigeria. Environmental Geochemistry and Health, 46, 223. https://doi.org/10.1007/s10653-024- 01979-4 Rashid, A., Schutte, B., Ulery, A., Deyholos, M., Sanogo, S., Lehnhoff, E., & Beck, L. (2023). Heavy metal contamination in agricultural soil: Environmental pollutants affecting crop health. Agronomy, 13(6), 1521. https://doi.org/10.3390/agronomy13061521 Taghavi, M., Bakhshi, K., Zarei, A., Hoseinzadeh, E., & Gholizadeh, A. (2024). Soil pollution indices and health risk assessment of metals in the agricultural soil of pistachio orchards. Scientific Reports, 14(1), 8971. https://doi.org/10.1038/s41598-024-59450-4 Tong, L., Yang, A., Fan, M., Zheng, D., Li, J., Liu, H., Li, C., Tang, Y., & Cheng, L. (2025). Assessment of heavy metal sources and health risks in soil-crop systems of fragmented farmland. Frontiers in Public Health, 13, 1637595. https://doi.org/10.3389/fpubh.2025.1637595 Wan, Y., Liu, J., Zhong, Z., Wang, Q., & Li, H. (2024). Heavy metals in agricultural soils: Sources, influencing factors, and remediation strategies. Toxics, 12(1), 63. https://doi.org/10.3390/toxics12010063 Zhao, G., Yang, Y., Liu, Y., Cheng, J., & Wang, X. (2022). Source analysis and ecological risk assessment of heavy metals in farmland soils around heavy metal industry in Anxin County. Scientific Reports, 12(1), 10562. https://doi.org/10.1038/s41598-022-13977-6