Published on International Journal of Agriculture & Agribusiness
Publication Date: June 21, 2019
Mango, L. & Kugedera, A. T.
Deparrtment of Agriculture Management, Zimbabwe Open University, Bindura
Deparrtment of Agriculture Management, Zimbabwe Open University, Masvingo
Disposal of mine tailings on an ecosystem can cause alterations in plant productivity and soil physico-chemical and biological properties. Vertical distribution of soil bulk density, SOC and soil pH to a depth of ninety (90) cm on nickel mine tailings dumps re-vegetated with in terraces with 14, 17, 19 and 21-year old Acacia saligna (Labill) trees at Trojan Nickel mine, Zimbabwe. Systematic sampling was used to collect vegetation and soil data. Soil samples were collected at 0 – 15, 45 – 60 and 75 – 90 cm depths. Soil organic carbon (SOC) was quantified using the loss on ignition method, bulk density was determined using the core method and pH was measured using 0.01M CaCl2 method. Analysis of variance (ANOVA) was used to determine the effect of depth and age on SOC, bulk density, pH and aboveground biomass C. The Tukey’s honestly significant difference (HSD) tests were used to separate means. Bulk density and pH did not differ significantly with soil depth and age of A. saligna (Labill) trees whilst SOC differed significantly (p < 0.05) with soil depth. Results from this study suggest that the age of A. saligna trees has an impact on SOC, bulk density and pH. Re-vegetation of mine tailings should be considered in ecological restoration and as a climate change mitigation strategy.
Keyword: Plant productivity, soil physico-chemical, biological properties, Acacia saligna (Labill) trees, soil organic carbon & re-vegetation.
Soils are the largest pool of terrestrial organic carbon in the biosphere, storing more C than the sum that is contained in both plants and the atmosphere combined (Schlesinger, 1999). Carbon sequestration rates vary by tree species, soil type, regional climate and topography and management practice (Lal, 2007; Srivastava and Ram, 2009). Shrestha and Rattan (2006), established that the rate of soil organic carbon (SOC) sequestration ranges from 0.1 to 3.1 Mg ha-1 year -1 in grassland and 0.7 to 4 Mg ha-1 year-1 in forested reclaimed mine soil ecosystem. Mining activities has been noted to cause changes in soil organic carbon, bulk density and pH leading to decreased soil fertility due to increased soil pH as a result of metals being deposed on the soil. Soil organic carbon will be greatly altered causing a reduction in microbial population in the soil reducing decomposition of plant and animal remains.
2.1 Study area
The study was conducted at Trojan Nickel Mine tailings dumps (31o17ˈE; 17o19ˈS, Altitude), west of Bindura town. The area is generally hilly with some valleys from where a small stream flows in an easterly direction. The terrace with 21-year old A. saligna trees is at the base of the dumps and is parallel to the flowing stream. Tailings dump 7 has a south facing slope whilst dump 3 has a northerly facing slope. Geologically, the area is a typical greenstone configuration consisting of mainly basaltic rocks with komatiitic basalt, tholeitic basalt, gabbro, banded iron formations and volcanic tuffs (Nyakudya et al., 2011). The ore mainly consist of Pyrrhotite (Fe, Ni) S. The intrusive rocks consist of dolerite, gabbroic rocks and quartz feldspar porphyry (Trojan mine report; 2007). Associated with pyrrhotite are two nickel-bearing minerals namely Pentlandite and Polydymite (Ni3S4). The ore also consist of Pyrite (FeS2) and Chalcopyrite (Cu, Fe) S2 (Nyakudya et al., 2011). Soils are generally red clays (fersiallitic) in the hills having moderate amounts of active clay (Nyamapfene, 1991), and vertisols occupy the lower catenary positions.
2.2 Experimental design
Systematic sampling was used for the collection of soil and surface litter data after every 20 m along the terrace at the middle of the catena with the first point being randomly selected. Three points were sampled from each terrace with three samples per pit from the three respective depths, 0–15 cm; 45–60 cm and 75–90 cm. Soil samples for bulk density were collected using 100 cm3 core rings at each depth. A total of nine samples for each terrace were collected giving a sum of 36 for all the four terraces, from which bulk density, pH and soil organic carbon were determined.
2.3 Determination of soil Properties
2.3.1 Soil bulk density
Soil bulk density is the ratio of the oven-dried mass of soil to its volume either at time of sampling or at specified moisture content. The bulk density samples were oven-dried for 24 hours at 105oC and each sample weight was recorded and expressed in g cm-3 (Equation 1).
BD (g/cm3) = Mass of Oven dry soil (g) ̸ Volume of core (cm3) 
2.3.2 Soil Organic Carbon
The dry oxidation (loss on ignition) method was used for C analysis (Ball, 1964; David, 1988). Weighed soil samples were oven-dried for 24 hours at 105oC. The samples were placed in a desiccator for cooling, then reweighed and placed in a muffle furnace at a temperature of 550oC for six hours (Davies, 1974). The difference in weight of oven dried sample and weight of the same sample after ignition at 550oC expressed as a fraction of the weight of oven dried sample is the C content of that particular soil sample (Allen et al., 1986) (Equation 2).
Carbon content (%) = (Wd – Wi) ̸ Wd x 100 
where Wd is the weight of oven dried sample;
Wi is the weight of sample after ignition at 550oC.
SOC in tonnes per hectare was then calculated using the following formula:
SOC (Mg ha-1) = Depth (cm) x Bulk density (g cm-3) x carbon content (%) 
2.3.3 Soil pH
Soil pH was determined using the 0.01M calcium chloride method as described by Blackmore (1987). Soil samples were air dried for 7 days and a sub sample of 20 grams for each depth was taken for analysis. A total of 36 samples for all the four terraces were analyzed for soil pH.
2.3.3 Data analysis
Data on SOC, bulk density, soil pH and aboveground biomass C were analyzed after testing for normality (Shapiro-Wilk’s test) and homogeneity of variance (Levene’s test). One way analysis of variance (ANOVA) in SPSS v. 21 was used to assess the effects of age of A. saligna and soil depth on SOC, soil pH, bulk density and aboveground biomass C. Tukey’s honestly significant difference (HSD) tests were used to test significant effects at p ≤ 0.05. This was done by comparing mean differences for soil characteristics (bulk density, organic carbon and pH) with depth and age, and aboveground biomass including surface litter C were compared with the age of A. saligna trees.
4.1 Variation of soil bulk density, organic carbon and pH with soil depths at Trojan Nickel mine tailings dumps in Bindura, Zimbabwe
The mean soil bulk densities and pH on nickel mine tailings dumps did not differ (p > 0.05) with soil depths whereas soil organic carbon contents varied significantly (p < 0.05) with soil depths (Table 1). Soil organic carbon contents were similar in the 0 -15 cm and 45 – 60 cm depths but higher than the 75 – 90 cm depth. Table 1: Variation of bulk density (BD), soil organic carbon (SOC) and pH with soil depths at Trojan Nickel mine tailings dumps in Bindura, Zimbabwe (n = 12). Means without common superscripts within each column are significantly different (p < 0.05) (Tukey’s HSD). Soil bulk density, organic carbon and pH did not differ (p > 0.05) with the age of A. saligna trees (Table 2). Soil organic carbon content had relatively high variability within individual terraces as reflected by the large standard deviations. The overall coefficient of variation for SOC with terrace was found to be 0.44 in terrace with 21-year old A. saligna trees, 0.42 in terrace with 19-year old A. saligna trees, 0.45 in terrace with 17-year old A. saligna and 0.63 in terrace with 14-year old A. saligna trees. The terrace with 14-year old A. saligna trees (T14) was highly variable relative to other terraces.
Table 2: Variation of soil characteristics across the three depths (0 -15, 45 -60 and 75 – 90 cm) over time in revegetated Trojan Nickel mine tailings dumps in Bindura, Zimbabwe (n = 12)
T = terrace and numbers represent years after re-vegetation with A. saligna trees. Means without common superscripts within each column are significantly different (Tukey’s HSD: p < 0.05). A comparison of SOC content for an individual depth among the terraces differed significantly (p = 0,012) for the 45 – 60 cm depth whereas 0 -15 cm and 75 – 90 were not significantly different (Figure 1). The cumulative SOC contents were 147.03 Mg ha-1 for the terrace with 14-year old A. saligna trees, 148.37 Mg ha-1 for the terrace with 17-year old A. saligna trees, 165.39 Mg ha-1 for the terrace with 19-year old A. saligna trees and 196.46 Mg ha-1 for the terrace with 21-year old A. saligna trees. Figure 1: Variation of soil organic carbon with depths among the terraces Different letters show significant differences (p < 0.05) for SOC within each depth across the terraces. Error bars ± 1 SD of the mean. 5. Discussion 5.1 Variation of bulk density, organic carbon and pH with soil depths at Trojan Nickel mine tailings dumps Bulk density did not differ (p > 0.05) with soil depths (Table 1). The mean bulk densities found in this study were higher than those of forest soils (1.13 – 1.3 g cm-3), but similar to those of productive soils (1.1 – 1.5 g cm-3) (Bradshaw, 2005; Brady, 2002). The relatively high bulk densities in this study may be due to the effect of heavy metals in mine tailings which masks the presence of organic matter and compaction (Henriques and Fernandes, 1991). Bulk densities found in this study are congruent to the 1.42 – 1.62 g cm-3 found by Seybold et al., (2004) who worked in reclaimed coal mine dumps and cultivated croplands in Indiana. Similarly, Nyakudya et al., (2014) found bulk density ranging from 1.5 – 1.6 g cm-3 in a medium sandy loam soil of Magaranhewe village in Rushinga, Zimbabwe.
The SOC stocks in this study ranged between 66.79 Mg ha-1 and 58.94 Mg ha-1 are within the range of SOC stocks found in woodlands of southern Africa, (32 – 133 Mg ha-1) (Walker and Desanker, 2004; Zingore et al., 2005), and higher than 53.9 Mg ha-1 in planted forests in Zimbabwe found by Mujuru et al., (2014). The soil organic carbon stocks did not differ with tree age (Table 2) and is therefore contrary to the findings by Jordan (2005) who concluded that soil C stocks differ with tree age. In this study there was little variation in tree ages (2 years) which can result in similar quantities of SOC (Šourková et al., 2004). Furthermore the terrace with 14-year old A. saligna trees is from a different tailings dump which might be characterized by different soil edaphic conditions (Table 1).
Generally, SOC content decreased with increasing soil depth. Thus, SOC differed (p < 0.05) significantly with soil depths in each revegetated terrace (Appendix C). The decrease in SOC with depth is attributed to the fact that organic carbon is directly proportional to organic matter content, which often decreases with soil depth (Baldock and Skjemstad, 2000). The high variability in SOC with depth may be attributed to heterogeneity in vegetation density within individual terraces. Soil organic carbon for individual depths differed in the 45 – 60 cm whilst 0 – 15 cm and 75 – 90 cm did not differ (Figure 1). The differences may be caused by high variations in bulk densities in the 45 – 60 cm depth than in the other depths. Soil pH was not significantly different (p > 0.05) with soil depth and age (years) of A. saligna trees (Table 1). The pH on nickel mine tailings dumps generally range from slightly to moderately alkaline (Brady, 2002). This high pH values are attributed to the chemical treatment of slimes with sodium hydroxide (NaOH) at the concentrator plant. Results suggest that alkalinity persisted in the mine tailings’ dumps for over 40 years (from deposition of tailings since 1968 and 1971 for dumps 3 and 7 respectively). The pH values found in this study are consistent to Cao et al., (2002) who found alkaline soil pH values greater than 7.20 in the weathering of lead bullets, Florida, USA, and with Neuman et al., (2005) who also found a pH values of 8.0 in the 0 – 15 cm and a pH of 7.9 in the 45 – 60 cm depths in the Keating iron tailings site in Broadwater County, Montana, USA.
Soil bulk density and pH did not differ with depth and age of A. saligna trees, whilst SOC was significantly higher in the upper soil layers than the lower layers in all the terraces. Soil organic carbon showed no significant differences with the age of A. saligna trees. The differences in surface litter C and species diversity and composition under mine tailings dumps has shown that diversity improves with the increase in age of A. saligna trees. Even though these terraces had high pH and bulk densities, their tailings soil quality produced relatively diverse vegetation with a composite of species. However, a slightly alkaline pH, high concentrations of organic carbon, low – medium bulk densities may result in enhanced establishment of grasses, shrubs and trees.