Chapter 4. Results and discussion
Section keywords: adverse or benefitial effects, magnitude and importance, main activities or conditions worsening the situation, effects on health, on biodiversity, selection of most favorable scenario, possible pollution control strategy.
The results obtained by this model show that in general the local effects on the water quality of the river caused by the wastewater discharge in terms of organic matter contribution are very small. The BOD and DO concentration within the river reaches do not vary widely, basically due to the large size of the river and the abundant flow compared to the wastewater flow, even during the dry seasons. The most significant change that was observed in the results was the increase in the BOD concentration in reach 2 during an extreme scenario of very low flow. Here, the wastewater discharge increased the CBOD concentration in 26%, raising it from 4 to 5 mgO2/L, and the NBOD in 19%, raising it from 3.9 to 4.6 mgO2/L. However, since the water quality in this season is very good, this increase barely changes it from class 1 (4 mgO2/L or lower) to class 2 (between 4 and 6 mgO2/L) and for a very short time.
Something that is interesting in the curves obtained from the BOD-DO model is the evident variation in the river behavior between seasons. The main difference is the initial organic matter content in the river and therefore the water quality class. Regarding the BOD, during the high water season the river has a water quality class 5 (15 mgO2/L or higher), while during the low water season it has a class 1. In terms of dissolved oxygen, even though the initial concentration in the river does not vary from one season to the other, remaining mostly in class 2 (between 6 and 7 mgO2/L) or slightly below, the resulting oxygen sag curves present large differences.
During the high water season the dissolved oxygen is expected to sink to much more critical values that go down to 0 mgO2/L, while during the low water seasons it reaches values that are still above 4 mgO2/L, considered a tolerable water quality. But not only the critical values for DO are different between seasons, also the duration of this critical stage and the total time required for a self-purification process in the river vary, being these much longer during the high water season. These differences in assimilation capacity between seasons are not only due to the BOD content in the river, but also to hydraulic characteristics, such as water depth and flow velocity, which affect the value of the rate constant coefficients. During the high water season, the deoxygenation and reaeration rate constants presented values below 1 day-1, while during the low water season values above 1.2 and 1.6 were observed.
Regarding water quality standards and regulations in Colombia, it can be said that the minimum DO concentration for fauna and flora preservation established in the decree 1584 (Ministerio de Salud, 1984) as 4 mgO2/L is maintained within the river reaches, but is expected to be violated downstream of the municipality, during the high water season. Unfortunately, for Colombia no BOD limiting values were found, to which the results of the present study could be compared to.
Transversal mixing model
For the first two seasonal cases, reach 1 could not be modeled due to the very low concentrations resulting from the wastewater discharge. Only in the extreme scenario some effects started to show, however with still very low concentrations (0.957 mg/L). Reach 2 instead could be modeled for the three cases, all of them showing similar dispersion patterns, but with increasing concentrations towards the lowest water level case (extreme scenario). According to the graphs (see Table 22) it can be said that the river geometry, combined with the location of the pipe outlet and the wastewater flow and concentration, result in a dispersion that always takes place in the first half across the river, usually 5 to 8 meters from the left riverbank, with the higher concentrations in the first 70 or 80 meters downstream of the pollution source.
The concentrations obtained for the case of Puerto Berrío are not very high, but since the model assumes that the background concentration in the river is cero, the real concentrations after the discharges are expected to be somewhat higher than the ones obtained by the model. In the case of the extreme scenario, the BOD concentration in reach 2 at 80 m from the source (13.3 mg/L) is close to reaching the critical level set by the model as 15 mgO2/L. On the other hand, the mixing of the wastewater near the left riverbank is not desired since this is the part of the river where most river related activities, such as water and sand extraction and recreation are practiced by the inhabitants of the municipality.
One aspect, in which the model does not fit the reality completely, is the fact that the pipe outlet in the model is always assumed to be in touch with the water, which means that the mixing zone will happen right after the wastewater is discharged. In the case of Puerto Berrío this would only be true for the high water season, since during the low water season, due to the reduction in the water level and river width (about 3 to 8 meters), the pipes outlet are not covered by water anymore. The wastewater being discharged runs on the ground without dilution for several meters before it reaches the river and the mixing process finally takes place. Based on this, an extra adverse effect should be added to the results of the model, in that the people are exposed to direct contact with the wastewater and there is a favorable environment for vectors proliferation, which leads to health problems besides the aesthetic aspects of landscape modification and bad smells.
Figure 39 Difference between reality and model assumptions
Water Quality Index
The results obtained by the calculation of the WQI make the seasonal water quality variations evident, complementing what had already been found by means of the BOD-DO model. Due to higher concentrations of most of the water quality parameters (or lower values in the case of pH and oxygen concentration) during the high water season, the river water quality is described as bad, while during the low water season it is described as medium.
Regarding the changes in the water quality within the river reaches due to the effects of the wastewater discharge, it can be said that the most noticeable change is also found during the high water season, where the WQI presents a reduction of 6.4%, going from 47 to 44. This reduction is caused by an increase in the oxygen deficit and the fecal coliform bacteria. During the low water season, the WQI does not vary within the reaches, which means that the effects of the wastewater on the river water quality parameters are not high enough to modify the index. As for the extreme scenario, even though the background concentration of the river is the same as in the previous case, the wastewater discharge does slightly affect the water quality, with an index reduction of 1.6%, going from 62 to 61. This reduction is due to a lower dilution flow and a faster degradation process that lead to an increase in the BOD and a sooner decrease in the dissolved oxygen.
Since the characteristics of the wastewater were assumed to be same all year long (see Table 1), it is clear that the magnitude of the effects in the different seasons is directly related to the river water quality, i.e., the better the river water quality, the higher the assimilation capacity. Calculating the water quality index demonstrated the importance of other parameters in the analysis of water quality and proved that a bad quality means a lower assimilation capacity.
High water season Vs. Low water season
According to the results obtained by the different analysis tools, the high water season has the advantage of diluting the wastewater right at the outlet of the pipe and to such an extent, that almost no change in the BOD concentration in the river takes place. The disadvantage of this season is the bad quality of the river, with values lying over the critical limits. This should be a reason to avoid discharges into the river. Moreover, in the high water season the river has not only a bad quality, but also a slow self-purification process due to a much lower reaeration rate and a higher content of organic matter, which leads to high amounts of oxygen consumption for its degradation.
The low water season presents a higher assimilation capacity in terms of water quality since the concentration of pollutants is much lower than in the high water season, that is why in spite of the larger contribution to BOD and fecal coliform of the wastewater compared to the high water season, it remains within the limits of good water quality classes. The low water season has the disadvantage of level reduction entailing consequences with respect to the location of the pipes, which leads to bad smells and health problems due to an eventual direct contact of the people with the wastewater before it reaches the river. Figure 40 summarizes the results of the analysis tools applied in the present study.
Figure 40 Summary of results
Possible improvements of the situation
Wastewater treatment plant Lagunas
As it has been mentioned in Section 2.1.3, in April 2008 a new wastewater treatment plant had just been built in Puerto Berrío. By that time, the plant was not being operated yet. This new plant is expected to treat around 80% of the municipal wastewater being currently discharged into the Magdalena River. The plant consists of one anaerobic and two facultative ponds with a total capacity of the plant of 110 L/s. It will treat initially 80% of the municipal wastewater, i.e., around 96 L/s. The expected efficiencies in BOD and suspended solids removal are 60% for the anaerobic pond and 80% for the facultative ponds, with a 90% total efficiency for the whole system (Aguas del Puerto, 2007). This reduction in suspended solids and organic matter and the oxygenation of the wastewater through the oxidation ponds before its discharge, will mitigate significantly the effects caused on the receiving water body. By removing at least 80% of the BOD, and 50% suspended solids from the wastewater, the municipality will be also meeting the requirements of the decree 1594 (Ministerio de Salud, 1984).
Pipe extension: relocation of pipe outlet
According to the results obtained by the transversal mixing model, one of the problems related to the wastewater discharges in Puerto Berrío is the location of the pipes outlets, which together with the river geometry cause a plume distribution near the riverbank, where people have the most contact with the river. Moreover, the situation is worsened during the low water season when the pipes outlets remain uncover. Based on the above, and considering that the wastewater treatment plant may not start functioning in the near future, and also for the rest of the discharges in the municipality that are not being treated yet, this situation needs to be taken care of. A simple way of improving the mixing of the wastewater in the river arises from the results and observations of the present work. It consists of the extension of the pipes, so that the outlet is not right at the riverbank, but some meters inside the river. To see how this change would affect the distribution plume in the river, the transversal mixing model was run again, this time changing the position of the pipe outlet. From the three water level cases analyzed previously, the extreme scenario in reach 2 was chosen, since it showed the most significant effects. The pipe outlet position was defined by trial and error, running the model until the lowest concentration possible was found and the plume distribution was favorable.. The most favorable plume distribution in the model was obtained by extending the pipe until the outlet is located in the middle of the riverbed. In this case, the concentrations in the river after the discharged are reduced to half, compared to the current situation, and the zone near the riverbanks, both left and right, are kept free of pollutants. For this to be achieved, a pipe somewhat larger than 20 m should be added to the current outlet.
Figure 41 Pipe outlet in the middle of the river
Nevertheless, such a long extension might not be necessary and for practical matters it might even be unfavorable, affecting the navigation of small boats and canoes, especially under low water level conditions. Considering this, another possible solution with a shorter pipe was tried out, resulting in a location of the pipe outlet 5 m across the river. By doing this, the concentration is also reduce to half, as in the previous case, with the only difference that the zone near the left riverbank is not completely free of pollutant. However, this pipe extension is enough to reduce the concentrations in this zone below 4 mg/L, which is the limiting value for the class 1 water quality.
Figure 42 Pipe outlet 5 m across the river
In both cases, an extra pipe length needs to be taken into account due to the distance between the real pipe outlet and the one assumed by the model (see Figure 39).