Influence of DOC under high NO 3 -N and reductive bottom sediment

Points of similarity can be seen between cases 3 and 4 in the onset of the ORP decline, which began with a 1 mg/L decrease in DO (Fig. 3.6). However, because case 4 showed a faster reduction in DO than case 3, Step 1 (which indicates when the drop in ORP begins) showed

- 23 -

Fig. 3.6 Continuous measurements of DO and ORP for 20 days in case 3 and case 4.

A five-step process of decline in ORP is represented by points a, b, c, and d with each step having the following characteristics: Step 1: a linear decline from a to b; Step 2: a point of inflection at approximately 0 mV; Step 3: a steep decline from b to c; Step 4: a gradual decline from c to d; and Step 5:

the state of equilibrium. The red broken line (case 3) and the black broken line (case 4) indicate results of linear regression formula for the drop from positive value to negative value in ORP. S is the slope of the linear line. R² is the coefficient of determination.

- 24 -

Fig. 3.7 Periodic measured results of NO3-N, NH4-N, PO4-P, sulfide, SO42-, TFe, and E254 during the experimental period in case 3 and case 4. Error bar stand-ards are one standard error from the mean. The red solid line (case 3) and the black solid line (case 4) indicate results of linear regression formula for temporal changes (increase or decrease) in concentrations of NO3-N, NH4 -N and PO4-P. S is the slope of the linear line. R² is the coefficient of deter-mination.

0 0.2 0.4 0.6

0 1 2 3

0 0.2 0.4 0.6

0 200 400

0 5 10 15

0 4 8

0 10 20 30 40 50 60 70

0 0.4 0.8 1.2 NO3-N (mg/L)PO4-P (mg/L)NH4-N (mg/L)Sulfide (μg/L)SO42- (mg/L)TFe (mg/L)E254

Lapsed times from the experiment start (d) Case 3: = 0.015 mg/L/d; = 0.70

Case 4: = 0.016 mg/L/d; = 0.61

Case 3: = 0.0016 mg/L/d;

Case 4: = 0.0028 mg/L/d;

Case 3 ( N_H -D_L -R) Case 4 ( N_H - D_H -R)

Case 4: = -0.0352 mg/L/d; = 0.94 Case 3: = -0.0212 mg/L/d; = 0.96

= 0.51 = 0.58

S R²

S R²

S R²

S R²

S R²

S R²

- 25 -

results approximately 5 days earlier for case 4 than for case 3. Moreover, because case 4 showed a rapid reduction in ORP alongside the decline in DO, Steps 2–5 were completed much sooner than in case 3. The results indicated clearly that time needed to attain equilibrium in Step 5 greatly influenced the increase of sulfide concentrations by the accompanying sulfate reduction.

Therefore, the difference in the ORP rate of decline resulting from the initial DOC conditions was the main variable used for quantitatively assessing water quality dynamics based on the redox reactions under anaerobic conditions.

The point of similarity between cases 3 and 4 was when ORP reached equilibrium (at about –400 mV) in Step 5, which was the same point in time when NO3-N decreased to zero, as well as the point that the linear decrease in NO3-N was accompanied by the decline in DO.

On the other hand, the time needed for NO3-N to drop to zero was approximately 9 days for case 3 and 14 days for case 4 (Fig. 3.7), which results in the high rate of decrease over time in case 4 compared to case 3. This indicates that a high concentration of DOC accelerated the occurrence of denitrification. Possible reasons for this are that the rapid development of anoxia may have accelerated the progression of denitrification, and that a high concentration of DOC may have induced denitrifying bacteria to perform denitrification. It may also be inferred that the early completion of denitrification in case 4 hastened the shift to reduction half-reactions with lower electric potential, which shortened the time required to reach Step 5, as described above.

Considering the time-dependent changes in NH4-N and PO4-P for cases 1 to 4 (Figs. 3.5 and 3.7), the impact of the initial NO3-N conditions on these increasing rates was studied. In both cases 3 and 4 with high NO3-N conditions, NH4-N increased immediately following the start of the DO decline, and PO4-P mainly increased after DO had decreased to zero (Figs. 3.6 and 3.7). The timings of these increases were similar for cases 1 and 2 with low NO3-N condi-tions, where the rates of NH4-N and PO4-P increase had similar logarithmic functions. On the other hand, for high NO3-N conditions, changes with time in NH4-N and PO4-P were represented by a linear regression formula. It was found that for this type of difference between proximate functions, the initial conditions of NO3-N affect the rates of increase in NH4-N and PO4-P. Fur-thermore, regardless of the DOC conditions, the increase rates in NH4-N and PO4-P under high

- 26 -

NO3-N conditions were low when compared to the increase rates during the early period of the PO4-P and NH4-N increases (about 15 days from starting the experiment) under low NO3-N conditions. This difference in the rates of increase is similar in the case of changes in total iron ion concentrations with time, which is related to phosphate ion elution, and also in the changes with time in E254, which is related to water quality that increases due to the organic matter decomposition by microorganisms.

Consequently, when high NO3-N concentrations existed at the start time of the DO de-cline, the reaction rates of anaerobic respiration, such as ferric reduction, did not rise drastically.

However, throughout the approximately two-month anaerobic period, anaerobic reactions gen-erally advanced at a fixed rate. For this reason, NH4-N and PO4-P with high NO3-N conditions experienced dissolution over a longer period than with low NO3-N conditions. The causes of this are considered to be that the denitrifying bacteria was activated as a result of high NO3-N conditions (Sørensen et al., 1979), promoting the decomposition of anaerobic organic matter, and as a result, the amount of ammonium ion released from the organic matter increased and the NH4-N elution amounts increased. Furthermore, the increase in low-molecular organic mat-ter by denitrification is tied to the increase in the respiratory substrate for iron-reducing bacmat-teria (Ellis-Evans and Lemon, 1989; Lovley, 1991; DiChristina, 1992), while it can be assumed that PO4-P elution amounts increased along with ferric reductions.

Next, the impact of the initial conditions for DOC on time-dependent changes in NH4 -N and PO4-P was examined by comparing cases 3 and 4. In both cases, there was a linear in-crease until the experiment ended. Furthermore, the inin-crease in NH4-N release rates, which was estimated as the gradient of linear regression equations, had the same value in both cases. As a result, properties accompanying the increase in NH4-N elution from the substratum are influ-enced by the NO3-N concentrations at the time of the DO decline, but are not influenced by DOC concentrations. On the other hand, cases 3 and 4 both experienced linear increases in PO4 -P concentrations until the experiment ended with the rate of increase being quite a bit higher for case 4, which had high DOC concentrations. As noted above, the temporal rate of change in the NO3-N decline with the decrease in DO was also more significant for case 4 than for case 3. Therefore, in the case of high DOC conditions, such as those in case 4, the activation of

- 27 -

denitrifying bacteria and iron-reducing bacteria could be promoted by increasing the available respiratory substance (Müller et al., 2016). Thus, the PO4-P elution, which was attributed to the oxidation-reduction reaction by anaerobic bacteria, increased at a high rate without any tem-poral attenuation. In other words, the high DOC conditions were a factor that led to higher PO4 -P increases compared to those under low DOC conditions. The point in time at which sulfide in cases 3 and 4 began to increase coincided with the ORP transitioning into Step 5 and equilib-rium, as well as the point when NO3-N decreased to zero. Similar results were obtained in cases 1 and 2. Moreover, despite sulfide having increased linearly in both cases, they started to de-crease approximately 30 days after the beginning of the inde-crease for case 3 and after approxi-mately 20 days for case 4. Although time-dependent changes in sulfide did not demonstrate a clear convex peak value in cases 3 and 4, case 2 exhibited the same point in time where the maximum values shifted from increasing to decreasing in sulfide.

The temporal rate of change in linearly increasing concentrations was approximately the same for cases 3 and 4, with few DOC condition-related differences. In case 4, ferrous ion had increased monotonically due to ferric reduction until it started to decrease slowly at the point in time when sulfide concentrations shifted from increasing to decreasing. This decrease in ferric irons reflects their reaction with sulfide to form iron sulfides (Omoregie et al., 2013).

Furthermore, the point in time when sulfide peak generally is the same point when sulfate ions were reduced to zero as a result of sulfate reduction. It can be concluded that the decreases in dissolvable sulfides could be caused by the termination of sulfate reduction because their amount has reached zero, and by the formation of non-soluble sulfides.

In case 4, the initial concentration of SO42 was about 3 mg/L less than with the other cases. This caused sulfate ion to drop to less than 1 mg/L within a 30-day period. In contrast, the initial concentration of SO42 for case 3 was high (8 mg/L), which causes sulfate to drop to near zero after a 50-day period. Accordingly, the maximum concentration of sulfide in case 4 was less than in case 3, and sulfide in case 4 reached its maximum value faster than in case 3, because the initial level of sulfate concentration was low in case 4, and this accelerated the sulfate reduction response time and was not related to the experimental conditions such as initial concentrations of DOC and NO3-N.

- 28 -

In the case of low NO3-N, a comparison between cases 1 and 2 showed that the rate of increase in sulfide by means of high DOC was significant. In contrast, a comparison between cases 3 and 4 did not confirm differences in the temporal rate of change in sulfide due to DOC conditions. For case 4, these factors also involved low concentrations of initial SO42, and there-fore, the effect of high DOC on the rate of sulfide increase is assumed to be minor.