Chapter 4 Estimation of Water Quality Dynamics in Organically Polluted Reservoir by Field Observations and Improved Ecosystem Model
4.4 b). The factors behind such a difference in sediment surface would include the following:
4.2.3 Characteristics of biochemical dynamics of water quality near the bottom
The seasonal changes in the vertical profiles of water quality parameters included a decrease in DO and NO3-N and an increase in NH4-N, PO4-P, and sulfide levels above the bottom bed. These concentration variations resulted from the biochemical reactions under anaerobic conditions and the numerical modeling of these processes would require a quantitative evaluation of the temporal changes (Fig. 4.5 and Fig. 4.6), considering the interrelationship among the water quality param-eters.
I evaluated the increase in the properties of NH4-N, PO4-P, and sulfide above the sedi-ment during the anoxic period by conducting beaker-scale experisedi-ments. The findings showed that 1) NO3-N decreased due to denitrification and NH4-N increased by organic matter decom-position immediately following the start of the DO decline, 2) PO4-P increased due to iron reduction after DO had decreased to zero, 3) sulfide began to increase exponentially due to sulfate reduction as soon as NO3-N decreased to zero. These results imply that the stepwise anaerobic respiration of the bacteria occurs by the reductions in high ORP, in the order of deni-trification, iron reduction, and sulfate reduction. The initial concentration of NO3-N and the initial redox state of the bottom sediment impacted the increase in PO4-P, NH4-N, and sulfide under anaerobic conditions. The temporal changes in high NO3-N conditions were represented by a linear regression equation, while the increase in low NO3-N conditions had similar loga-rithmic functions. Besides, the rates of increase in these inorganic ions under high NO3-N con-ditions were low compared to the rates during the sudden growth period under low NO3-N conditions. There were no detectable differences in the dynamic properties of NO3-N and NH4 -N between oxidative and reductive states of the bottom sediment or in the rates of decrease of NO3-N and increase of NH4-N, as well as in the timings of the start of the NO3-N decrease and
- 49 -
Fig. 4.5 Field observations of DO, ORP, NO3-N, NH4-N, PO4-P, sulfide, SO42-, E254, and TFe at the bottom in 2015. S is the slope of linear line. R2 is the coefficient of determination. Dates in rec-tangle frames denote the data period used for calculating regres-sion lines.
0 2 4 6 8
-400 -200 0 200 400
0 0.2 0.4 0.6
0 1 2 3
0 0.1 0.2 0.3 0.4
0 200 400 600 800
0 2 4 6 8
0.1 0.2 0.3 0.4
4 5 6 7 8 9 10 11 12
0 2 4 6
Months in 2015
NH4-N (mg/L)PO4-P (mg/L)NO3-N (mg/L)Sulfide (μg/L)DO (mg/L) E254
= 0.015 mg/L/d = 0.88
SO42- (mg/L)TFe (mg/L)
E254
DO
ORP (mV)
ORP
= 6.1 μg/L/d
SO4 2-S R2
= 0.0020 mg/L/d = 0.87
S R2
= 0.82 S R2
= 0.025 mg/L/d = 0.70
S R2
7/8 ~11/25
8/5 ~11/25
8/5 ~11/25
7/8 ~11/25
- 50 -
Fig. 4.6 Field observations of DO, ORP, NO3-N, NH4-N, PO4-P, sul-fide, SO42-, E254, and TFe at the bottom in 2016. S is the slope of linear line. R2 is the coefficient of determination.
Dates in rectangle frames denote the data period used for calculating regression lines.
0 2 4 6 8
-400 -200 0 200 400
0 0.2 0.4 0.6
0 1 2 3
0 0.1 0.2 0.3 0.4
0 200 400 600 800
2 4 6 8
0.1 0.2 0.3 0.4
4 5 6 7 8 9 10 11 12
0 2 4 6
Months in 2016
NH4-N (mg/L)PO4-P (mg/L)NO3-N (mg/L)Sulfide (μg/L)DO (mg/L) E254
SO42- (mg/L)TFe (mg/L) ORP (mV)
E254 SO4
2-DO ORP
= 0.019 mg/L/d = 0.88
S R2
= 0.0019 mg/L/d = 0.86
S R2
= 7.2 μg/L/d = 0.96
S R2
= 0.96 S R2
= 0.045 mg/L/d 6/1 ~10/7
7/6 ~10/7
7/6 ~10/7
7/6 ~10/7
- 51 -
the NH4-N increase. However, the condition of the oxidative initial bottom sediment delayed the timing points of PO4-P and sulfide increases and this condition elevated the rates decrease and the NH4-N increase. However, the condition of the oxidative initial bottom sediment delayed the initiation of PO4-P and sulfide increases and this condition elevated the rates of temporal changes in both concentrations. Based on these findings, the changes in NO3-N, NH4-N, PO4 -P, and sulfide with time above the bottom in the actual stratified water area were statistically estimated and this study particularly focused on the influence of the difference in the redox state at the bottom surface between 2015 and 2016 spring on the property of these temporal decreases or increases.
In mid-April of 2016 and the beginning of May 2015, DO=0 was reached and continued until early December of both years. Common to both years, ORP began to decrease from approximately the time when DO=0 was reached and it attained a strong reductive state of about 200 mV in early June. In 2015, the bottom sediment changed from a brown color of oxidative state to a gray color of reductive state when the strongest reductive state was reached. NO3-N began to de-crease in early June 2015 and in mid-April 2016. Because DO=0 at the time of this dede-crease, NO3-N decreased with denitrification occurring under anaerobic conditions. Although NO3-N decreased one month later than the time of DO=0 in 2015, NO3-N in 2016 decreased at the same time as DO decreased to zero. The timing of the start of NO3-N reduction was different in two years. The start of the decrease in NO3-N was consistent with the point of time when the oxide layer on the sediment surface disappeared in 2015; thus, NO3-N began to decrease after a lapse of time from the point of DO=0 because the aerobic state of the sediment inhibited the activation of anaerobic denitrifying bacteria. Because the sediment was in a reductive state at the time of the DO decline in 2016, denitrifying bacteria were activated with the decrease in DO to the anoxic state and a reduction of nitrate ions promptly occurred. Such a difference between the two years regarding the timing of the NO3-N decrease had a significant impact on the water quality dynamics caused by the gradual reductive half-reaction after denitrification.
Upon examining the dynamics of TFe and PO4-P because of the change in NO3-N with time, both TFe and PO4-P began to increase rapidly when denitrification under anaerobic con-ditions had completed. Because an increase in Fe3+ under anaerobic conditions is unlikely from
- 52 -
a chemical standpoint, the increase in TFe would be mainly resulted from the elution of Fe2+. The concentration Fe2+ increased with the reduction of the ferric ion at the bottom sediment. In addition, the timing of the beginning of the PO4-P and TFe increases coincided with each other and there was a strong correlation between the increases in PO4-P and TFe (Tables 4.1 and 4.2).
The increase in PO4-P was regarded as the elution from the bottom sediment, which resulted from iron reduction. The linear increase of PO4-P and TFe in 2015 was confirmed in the long term when the anoxic state in mid-December was resolved. Both PO4-P and TFe in 2016 increased sharply from July to September and remained constant after October. Although the increase rate of PO4-P in 2015 was almost the same as the one in 2016, the increase rate of TFe in 2016 (0.045 mg/L/d) was larger than that in 2015 (0.025 mg/L/d), suggesting that iron reduction was more active in 2016. In addition, when no oxide layer was formed on the sediment surface, the activity of iron-reducing bacteria was not suppressed by maintaining its reductive state and the reaction rate of iron reduction increased with the completion of denitrification; thus, PO4-P reached its peak value at an earlier time.
Sulfide began to increase when NO3-N and DO decreased to zero. SO42 decreased with the increase in sulfide; sulfide in the linear increasing process had a negative correlation with SO42 (Tables 4.1 and 4.2). These results suggest that as denitrification is terminated, an in-crease in sulfide due to sulfate reduction occurs. Furthermore, unlike beaker scale laboratory experiments, reduction of iron and sulfate occurs simultaneously under the condition of DO=0 and NO3-N =0 in the water body. In both 2015 and 2016, sulfide levels peaked at a certain point and then changed steadily or gradually decreased. Therefore, sulfate reduction reached the equi-librium state and the subsequent decrease in sulfide was attributed to insoluble metal sulfide precipitation in the sediment by manganese and iron binding. The increase in sulfides up to their peak value was approximated by a regression line. The rate of sulfide increase in 2016 (7.2 µg/L/d) was slightly higher than that in 2015 (6.1 µg/L/d). Moreover, sulfide reached the peak value earlier in 2016 than in 2015. These results mean that the activation of sulfate reduc-tion bacteria was promoted and finished early with the complereduc-tion of denitrificareduc-tion if the re-ductive state of the sediment was maintained in the spring of 2016. As sulfate reduction reached a state of equilibrium, sulfate concentrations remained unchanged in 2015 and increased in
- 53 -
2016 despite the anaerobic conditions. Since SO42 can be increased by the oxidation reaction of sulfur even under anaerobic conditions and such an oxidation reaction may be involved in methane fermentation, an anaerobic reductive half-reaction might shift from sulfuric acid re-duction to methane fermentation as the next step. In 2016, the reductive state of the sediment surface and the anaerobic state in water above the bottom would last longer than in 2015 and the strong reductive state could be maintained so that methane fermentation could occur be-cause of the progression of anaerobic respiration.
In both 2015 and 2016, NH4-N started to increase when denitrification completed and showed high correlation with TFe and E254 (Tables 4.1 and 4.2). The increase in NH4-N under the anaerobic state is caused by the separation of ammonia from organic matter. Accordingly,
Table 4.1 Correlation matrix of NH4-N, PO4-P, TFe, and E254 at the depth of 8 m in 2015.
PO4-P NH4-N TFe E254 SO42- Sulfide
PO4-P 1.00
NH4-N 0.91 1.00
TFe 0.95 0.95 1.00
E254 0.70 0.83 0.81 1.00
SO42- -0.94 -0.77 -0.86 -0.88 1.00
Sulfide 0.91 0.84 0.87 0.67 -0.87 1.00
Table 4.2 Correlation matrix of NH4-N, PO4-P, TFe, and E254 at the depth of 8 m in 2016.
PO4-P NH4-N TFe E254 SO42- Sulfide
PO4-P 1.00
NH4-N 0.98 1.00
TFe 0.95 0.93 1.00
E254 0.91 0.92 0.82 1.00
SO42- -0.45 -0.41 -0.22 -0.6 1.00
Sulfide 0.89 0.78 0.80 0.95 -0.74 1.00
- 54 -
the increase in NH4-N is strongly influenced by iron reduction rather than denitrification. There-fore, the redox state of the bottom sediment surface in spring would influence the timing of the increase of NH4-N as well as that of PO4-P and sulfides. As a result, NH4-N increased linearly throughout the observation period in 2015, while NH4-N increased linearly in 2016 with a slightly larger gradient than in 2015, and remained constant from October onward. In addition, NH4-N showed a very strong correlation with E254, which can be an indicator of humic acid in the environment. Dissolved organic matter from anaerobic organic matter decomposition in the sediment might be eluted together with NH4-N. The anoxic water near the sediment could lead to the increase in oxidizing ions such as Fe2+ and sulfide, as well as in organic matter, which could easily and instantly consume oxygen. In the case that the reductive state of the sediment surface was maintained for the long term (including the winter season), a quick recovery from the anoxic state around the bottom bed would be difficult owing to the existence of oxidizing substances, even if the vertical profile of water temperature is uniform due to destratification and DO could be easily transferred downward.
4.3 Construction of the water quality prediction model based on the ecosystem model
