3.1. Physicochemical parameters and diesel removal efficiency
The variations in ambient temperature, precipitation, soil organic matter content, and soil pH with time during the pot experiments are presented in Fig. S2. The ambient temperature and precipitation information were obtained from the Korea Meteorological Administration. The average ambient temperature ranged from 13.3 to 30.2 ℃, and the maximum and minimum temperatures were 35.4 ℃ and 8.5 ℃, respectively (Fig. S2a). Among the total of 52 rainfall events, there were 3 intensive rainfalls (> 90 mm) between days 50 and 69 (Fig. S2b). The average organic contents of the soil samples increased with increasing compost addition, and were 0.64, 2.50, 3.19, and 6.07% for the S-C0, S-C5, S-C10, and S-C20 samples, respectively. These values did not change significantly during the experimental period (Fig. S2c). Meanwhile, the average pH values of the soil samples decreased with increasing amount of added compost, being 9.01, 8.10, 7.80, and 7.76 for the S-C0, S-C5, S-C10, and S-C20 samples, respectively, and did not significantly vary during the experiment (Fig. S2d).
The changes in the residual diesel concentrations of the various samples with time are presented in Fig. 1. Thus, the initial TPH concentration was 9,432 mg-TPH·kg-dry-soil− 1, and did not change significantly until day 12 in all samples. Thereafter, the residual diesel concentration decreased significantly, and the diesel removal rate was proportional to the amount of compost added. On day 103, the diesel removal efficiencies of the S-C0, S-C5, S-C10, and S-C20 samples were 54.6, 77.5, 80.7, and 85.7%, respectively. Notably, the residual diesel concentration of the S-C20 sample on day 76 was below the 2,000 mg-TPH·kg-soil− 1 pollution risk criterion for oil-contaminated soil in Korea.
3.2 Soil enzyme activity
Dehydrogenase is known to be involved in the initial decomposition of soil organics, catalyzing the removal of hydrogen from organic molecules; hence the dehydrogenase activity (DHA) is used as an index for evaluating the degradation activity of soil organics (Casida Jr, 1977; Bolton et al., 1985). The results in Fig. 2a indicate that the DHA of the S-C0 did not significantly change during the initial 33 days, but increased slightly to 19.5 µg-TPF·g-dry-soil− 1·h− 1 on day 103. In the soils amended with compost, the initial DHA increased with increasing amount of added compost, being 203.8, 333.6, and 462.2 µg-TPF·g-dry-soil− 1·h− 1 in the S-C5, S-C10, and S-C20, respectively (Fig. 2a). During the experimental period, the DHAs of the amended soils decreased gradually as the residual diesel concentration decreased (Fig. 1 and Fig. 2a).
Urease promotes the mineralization of organic nitrogen to hydrogen-bound nitrogen, thereby providing the soil microorganisms with ammonia as an available nitrogen source (Lloyd and Sheaffe 1973). Although the urease activity (UA) cannot explain all of the biological mechanisms, it can be used as a good indicator of TPH metabolism in the soil under various soil conditions (Ceccanti et al., 1977; Li et al., 2007; Guo et al., 2012). As with the DHA, the UA of the S-C0 sample did not change significantly during the early stages of the experiment, but increased slightly during the mid-late period (Fig. 2b). In the amended soils, the initial UA increased with increasing amount of added compost, and further increased with time until the 33rd day, decreasing gradually thereafter (Fig. 2b).
3.3 CH4-oxidation and N2O-reduction potentials
The results in Fig. S3 and Table 2 indicate that there was no significant difference in the initial CH4-oxidation potentials of the various soil samples, which ranged from 1.40 to 1.95 µmol·g-dry-soil–1·h–1. During the experiment, however, the CH4-oxidation potential of the S-C0 increased significantly until around day 33, and remained relatively constant thereafter. Except on day 12, the CH4-oxidation potentials of the soils amended with compost (7.00–9.83 µmol·g-dry-soil–1·h–1) were higher than that of the non-compost amended soil (5.39–5.80 µmol·g-dry soil–1·h–1), and continued to increase significantly with time up until at least day 51. Further, the CH4-oxidation potential of the S-C20 sample (8.49–9.83 µmol·g-dry-soil–1·h–1) was slightly higher than those of the S-C5 and S-C10 samples (7.00–8.21 µmol·g-dry-soil–1·h–1).
Table 1
The physicochemical properties of the barren soil, the compost, and the various combinations.
| Barren soil | Compost | S-C0 | S-C5 | S-C10 | S-C20 |
Total nitrogen (%) | BDL* | 2.6 ± 0.3 | BDL* | 0.13 | 0.26 | 0.52 |
NH4+-N (mg·kg-soil− 1) | 1.8 ± 0.9 | 465.3 ± 0.2 | 1.8 | 24.98 | 48.15 | 94.50 |
NO3−-N (mg·kg-soil− 1) | 0.7 ± 0.0 | 34.5 ± 3.8 | 0.7 | 2.39 | 4.08 | 7.46 |
Total phosphorus (mg·kg-soil− 1) | 70.8 ± 3.6 | 11443.8 ± 603.5 | 70.8 | 639.45 | 1208.10 | 2345.40 |
pH | 8.54 ± 0.14 | 9.18 ± 0.00 | 8.52 ± 0.04 | 8.13 ± 0.01 | 8.63 ± 0.07 | 8.87 ± 0 .06 |
Water content (%) | 2.89 ± 0.71 | 55.33 ± 0.52 | 3.81 ± 0.53 | 7.11 ± 0.81 | 7.66 ± 0.87 | 14.86 ± 0.64 |
Organic matter (%) | 0.24 ± 0.11 | 36.73 ± 0.25 | 0.58 ± 0.08 | 2.80 ± 0.33 | 4.04 ± 0.51 | 8.24 ± 0.45 |
Water holding capacity (%, v/w) | - | - | 24.17 ± 2.50 | 31.67 ± 2.36 | 38.89 ± 1.57 | 46.67 ± 1.36 |
Soil texture | Sand | Sandy loam | – | – | – | – |
Sand (%) | 96.68 | 75.8 | – | – | – | – |
Silt (%) | 1.32 | 6.44 | – | – | – | – |
Clay (%) | 2 | 17.76 | – | – | – | – |
*Below detection limit |
Table 2
Comparison of soil CH4 oxidation potentials
Time (d) | Soil CH4 oxidation potential (µmol·g-dry-soil− 1·h− 1) |
S-C0 | S-C5 | S-C10 | S-C20 |
0 | 1.40 ± 0.14 I | 1.80 ± 0.04 I | 1.72 ± 0.21 I | 1.95 ± 0.29 I |
12 | 3.31 ± 0.17 H | 1.94 ± 0.50 I | 1.60 ± 0.02 I | 2.04 ± 0.01 I |
33 | 5.09 ± 0.25 F, G | 4.50 ± 0.61 G | 5.02 ± 0.27 F, G | 7.12 ± 0.69 D, E |
51 | 5.80 ± 0.62 F | 7.85 ± 0.20 C, D, E | 7.00 ± 1.37 E | 8.88 ± 0.31 B |
76 | 5.50 ± 0.35 F, G | 8.00 ± 0.11 I | 8.21 ± 0.08 B, C | 9.83 ± 0.15 A |
103 | 5.39 ± 0.18 F, G | 7.90 ± 0.06 C, D | 7.78 ± 0.22 D, C, E | 8.49 ± 0.17 B, C |
The results in Fig. S4 and Table 3 indicate that the initial N2O-oxidation potential of the non-compost amended soil (S-C0) was insignificant (< 56.82 nmol·g-dry-soil–1·h–1), while those of the S-C5, S-C10, and S-C20 were 868.03, 1,399.57 and 1,757.76 nmol·g-dry-soil–1·h–1, respectively. Moreover, while the N2O-oxidation potentials gradually decreased with time during bioremediation, a relatively high activity was maintained when the amount of compost added was large. In the S-C20 sample, the N2O-oxidation potential decreased from 838.14 nmol·g-dry-soil–1·h–1 on day 12, to 224.08 nmol·g-dry-soil–1·h–1 on day 103. In the S-C5 sample, it decreased from 328.57 nmol·g-dry-soil–1·h–1 on day 12, to 140.37 nmol·g-dry-soil–1·h–1 on day 103.
Table 3
Comparison of the soil N2O reduction potentials
Time (d) | Soil N2O reduction potential (nmol·g-dry-soil− 1·h− 1) |
S-C0 | S-C5 | S-C10 | S-C20 |
0 | < 56.8 ± 4.8 | 868.0 ± 44.4C | 1,399.6 ± 18.5B | 1,757.8 ± 7.3A |
12 | < 56.8 ± 4.8 | 329.6 ± 54.7F, G, H | 372.1 ± 19.7F | 838.1 ± 36.9C |
33 | < 56.8 ± 4.8 | 316.2 ± 19.5I, G, H | 281.7 ± 38.7I, J, H | 603.6 ± 4.1E |
51 | < 56.8 ± 4.8 | 265.4 ± 17.4I, J, K | 363.4 ± 20.1F, G | 661.8 ± 50.3D |
76 | < 56.8 ± 4.8 | 250.3 ± 4.8J, K | 256.0 ± 14.9J, K | 275.3 ± 31.0I, J, H |
103 | < 56.8 ± 4.8 | 140.4 ± 3.4J, K | 214.9 ± 9.1K | 224.1 ± 5.7J, K |
3.4 Functional gene dynamics
The functional gene dynamics during bioremediation of various diesel-contaminated soil samples are indicated in Fig. 3. Thus, the 16S rRNA gene copy number of the S-C0 sample increased from 103 to 105·g-dry-soil–1, while those of the S-C5, S-C10, and S-C20 were maintained at around 106·g-dry soil-1 during bioremediation (Fig. 3a).
Meanwhile, the relative copy numbers of the alkB gene in the S-C0 increased with bioremediation time to match that of the compost amended soils (gene copy number = 105 g-dry-soil–1) at day 51, and remained constant thereafter (Fig. 3b). The relative alkB gene copy numbers of the S-C5, S-C10, and S-C20 samples also varied during the initial period (0–33 days), but did not vary significantly after day 51. However, while the relative CYP153 gene copy number in the S-C0 increased from 102 to 104 g-dry-soil–1 during bioremediation, those of the S-C5, S-C10, and S-C20 samples decreased from 106 to 104·g-dry-soil–1 (Fig. 3c). Ultimately, on day 103, the CYP153 gene copy numbers were similar in all soil samples regardless of compost addition.
The pmoA/16S rRNA ratio of the S-C0 was always higher than that of the S-C5, S-C10, and S-C20 during bioremediation (Fig. 3d). The mcrA/16S rRNA ratio of the S-C0 sample increased from 101 to 103, whereas that of the S-C5, S-C10, and S-C20 samples increased from 102 to 106 during the initial 12 days, and then gradually decreased to 103 (Fig. 3e). The nosZ I/16S rRNA and cnorB/16S rRNA ratios in the S-C0 increased until day 51, and remained constant thereafter, while those of the compost amended soils increased slightly, with some exceptions (Figs. 3f and g).
3.5 Bacterial community dynamics
The dynamics of the bacterial communities during bioremediation of the diesel-contaminated soil are characterized by the Miseq analysis in Table 4 and Fig. 4. All samples showed good coverages of 0.99 or higher, thereby indicating that the results explain the actual bacterial communities of diesel-contaminated soil effectively (Table 4). The richness and diversity indices of all samples were increased during the bioremediation process, with those of the compost amended soils being slightly higher than those of the non-compost amended soil. However, the indices of the compost amended soil samples were largely identical, regardless of the amount of compost added.
Table 4
The richness and diversity of the bacterial community
Soil sample | Time (d) | OTU | Chao1a | Shannonb | Inverse Simpsonc | Good's Coveraged |
S-C0 | 0 | 297 ± 1 | 1249.4 ± 0.8 | 7.67 ± 0.03 | 0.989 ± 0.000 | 0.997 ± 0.000 |
12 | 201 ± 10 | 954.9 ± 1.3 | 6.29 ± 0.02 | 0.965 ± 0.002 | 0.996 ± 0.001 |
33 | 251 ± 1 | 1180.7 ± 1.6 | 6.69 ± 0.00 | 0.974 ± 0.000 | 0.997 ± 0.000 |
51 | 322 ± 40 | 1484.3 ± 148.9 | 6.81 ± 0.05 | 0.955 ± 0.003 | 0.995 ± 0.002 |
76 | 554 ± 32 | 2534.4 ± 114.0 | 8.03 ± 0.02 | 0.987 ± 0.001 | 0.996 ± 0.001 |
103 | 485 ± 45 | 2097.0 ± 146.1 | 8.03 ± 0.04 | 0.988 ± 0.000 | 0.994 ± 0.002 |
S-C5 | 0 | 176 ± 11 | 828.1 ± 7.5 | 6.35 ± 0.01 | 0.964 ± 0.001 | 0.998 ± 0.001 |
12 | 149 ± 18 | 691.5 ± 85.4 | 4.89 ± 0.13 | 0.913 ± 0.009 | 0.998 ± 0.001 |
33 | 205 ± 9 | 927.1 ± 47.1 | 6.56 ± 0.01 | 0.974 ± 0.001 | 0.998 ± 0.000 |
51 | 265 ± 6 | 1216.4 ± 3.7 | 6.39 ± 0.01 | 0.963 ± 0.000 | 0.997 ± 0.000 |
76 | 553 ± 34 | 2361.9 ± 102.6 | 8.06 ± 0.03 | 0.987 ± 0.000 | 0.995 ± 0.001 |
103 | 638 ± 53 | 2626.5 ± 203.0 | 8.42 ± 0.01 | 0.990 ± 0.000 | 0.995 ± 0.002 |
S-C10 | 0 | 172 ± 9 | 786.0 ± 45.8 | 6.48 ± 0.05 | 0.969 ± 0.002 | 0.998 ± 0.000 |
12 | 156 ± 7 | 689.1 ± 43.8 | 5.20 ± 0.05 | 0.930 ± 0.004 | 0.999 ± 0.000 |
33 | 219 ± 17 | 975.1 ± 28.7 | 6.41 ± 0.04 | 0.962 ± 0.002 | 0.998 ± 0.001 |
51 | 341 ± 6 | 1440.1 ± 24.6 | 6.89 ± 0.01 | 0.972 ± 0.000 | 0.999 ± 0.000 |
76 | 634 ± 48 | 2581.2 ± 87.8 | 8.24 ± 0.02 | 0.989 ± 0.000 | 0.996 ± 0.001 |
103 | 702 ± 9 | 2826.0 ± 31.0 | 8.57 ± 0.01 | 0.992 ± 0.000 | 0.996 ± 0.000 |
S-C20 | 0 | 174 ± 18 | 702.7 ± 57.4 | 6.30 ± 0.13 | 0.956 ± 0.007 | 0.998 ± 0.001 |
12 | 163 ± 12 | 627.8 ± 27.3 | 5.50 ± 0.03 | 0.952 ± 0.001 | 0.999 ± 0.000 |
33 | 217 ± 6 | 916.0 ± 48.8 | 6.24 ± 0.03 | 0.958 ± 0.001 | 0.998 ± 0.000 |
51 | 325 ± 17 | 1234.1 ± 78.1 | 6.83 ± 0.03 | 0.974 ± 0.001 | 0.997 ± 0.000 |
76 | 717 ± 18 | 2495.9 ± 31.5 | 8.38 ± 0.00 | 0.992 ± 0.000 | 0.995 ± 0.001 |
103 | 667 ± 14 | 2506.4 ± 38.7 | 8.58 ± 0.02 | 0.993 ± 0.000 | 0.994 ± 0.000 |
a The Chao1 index is used to evaluate the bacterial population richness. |
b The Shannon index is used to evaluate the diversity within the bacterial population. It accounts for both species abundance and evenness. |
c The Simpson diversity index is calculated as D = 1–[Σn(n–1)/N(N–1)], where n is the number of individuals of each species and N is the total number of individuals of all species. The Simpson diversity index is the probability that two randomly selected individuals in a given habitat will belong to the same species. |
d Good coverage is calculated as C = 1– (s/n), where s is the number of unique operational taxonomic units (OTUs) and n is the number of individuals of each species. The index gives a relative measure of how well the sample represents a large environment. |
The genus level analysis in Fig. 4 indicates that the structure of the bacterial community in the compost amended soil was significantly different from that of the control sample. In the S-C0 sample, Sphingomonas remained predominant (10.4–6.7%) throughout the bioremediation, whereas the other major genera changed over time. Thus, Stenotrophobacter (5.2%) and Sphingohabdus (3.0%) were dominant during the initial period, but were superseded by Alkanindiges (11.9%), Ralstonia (5.1%), and Pseudomonas (3.9%) during the intermediate period, and by Rugosibacter (6.8%), Chthoniobacter (4.3%), and Parvibaculum (3.8%) during the late period. By contrast, the dominant genera in the various compost amended soil samples were initially Membranicola (15.1–19.6%) and Truepera (6.5–7.6%), which were superceded by Immundisolibacter (7.3–16.9%), Dietzia (5.4–10.9%), and Paracoccus (2.3–4.8%) during the intermediate period, and by Sphingomonas (3.5–5.1%), Acidibacter (3.0–4.50%), Immundisolibacter (2.2–5.0%), Marinobacter (2.0–5.6%) and Terrimonas (2.0–4–0%) during the late period.
The PCA results in Fig. S5 indicate that the initial bacterial community was clearly divided into two groups according to compost addition, whereas the similarity between the two groups increased as the bioremediation progressed. These results suggest that bacterial community succession in the two groups progressed in a similar direction. Hence, to evaluate the effect of compost amendment upon the structure of the bacterial community, those genera having a close relationship with compost amendment were selected via network analysis (Fig. 5). The results indicate that compost amendment resulted in an increase in the relative abundances of Atopostipes, Halomona, Massilia, Membranicola, Paracoccus, Pseudogracilibacillus, Pusilimonas, Sphingorhabdus, and Truepera, and a decrease in the relative abundances of Stenotrophobacter, Sphingomonas and Massilia.