1. Introduction
Along with the growing expansion of housing construction in rural areas in Poland, the volume of water used by residents increases, thus also the volume of generated sewage increases. Unfortunately, in many cases, professional water supply and sewage systems are not provided along with the expansion of housing. In Poland, in areas where there are no collective sewage systems, residents use the so-called septic tanks (tanks for liquid waste) in which sewage is collected for disposal to a collective treatment plant. The numbers show how big this problem is in Poland.
In rural areas, according to current Central Statistical Office (CSO) data [
1], in 2018 there were 21,775.50 thousand tanks for liquid waste, commonly known as cesspool. It still is the most common way of wastewater disposal in Poland, in rural areas where there are no collective or individual sewage systems. The correct operation of such tanks consists in regular emptying of wastewater, which should then be transported by a professional slurry tanker to a collective wastewater treatment plant, whose technological line is suitable to treat this type of wastewater. Unfortunately, according to national literature reports, many such tanks are improperly used by residents, because the wastewater coming from them goes illegally to the environment, i.e., land or flowing water [
2,
3,
4]. Such practices lead to pollution of land, surface and underground waters [
5]. The main reason for this is the intention of users to reduce the costs of exporting and utilizing wastewater from cesspits in a collective treatment plant. To prevent this type of practice, it is necessary to control the leak-tightness of this type of tanks and the regularity of their emptying by a professional company with the appropriate slurry tanker.
Another problem posed by wastewater from cesspool is their utilization in a collective wastewater treatment plant, because the concentrations of pollutants contained in them are often several times, and sometimes several dozen times, higher than typical wastewater flowing into the sewage system [
6,
7,
8]. High concentrations of pollutants in wastewater from cesspool result from the saving habits of residents to save and thus use a small amount of water, which results in an increase in the concentration of pollutants [
9,
10]. In addition, the long-term storage of wastewater in tanks creates the conditions for anaerobic decomposition of organic pollutants and the occurrence of sewage rotting, which causes the release of an unpleasant odour of hydrogen sulfide [
11,
12,
13]. The volume of liquid waste tanks should be designed to be emptied with a 3–4 week time interval. Too long intervals between emptying these tanks result in the wastewater being rotten and similar in composition to sewage sludge with very high hydration [
14]. In practice, wastewater from cesspools is transported to the area of the collective wastewater treatment plant on an irregular basis on each day of the week [
8,
15], and this type of practice is conducive to disruption of biological wastewater treatment processes, as these processes are sensitive to major changes in both the quantity and quality of treated wastewater. Therefore, the amount of wastewater from cesspool should be dosed (batched) with great caution to the total amount of wastewater subjected to treatment [
16,
17,
18,
19,
20]. Adding (mixing) in a short time, e.g., directly from the slurry tanker sewage from cesspool to sewage flowing into the sewage network, will result in a sudden increase in the amount of treated sewage, as well as a sudden increase in the so-called “impact” of the pollutant load in the wastewater mixture subjected to the treatment process [
21].
As it has been shown in the publications concerning the problem of neutralization of sewage from septic tanks, there are alternative treatment systems to transport to collective treatment plants, which is an expensive process and may have a negative impact on biological processes in collective treatment plants. As Forbis-Stokes et al. [
21] stated, it is possible to treat sewage from septic tanks with a mobile installation at the place where they are generated. Another solution in this aspect is the treatment of sewage from septic tanks in wetland sewage treatment plants. As Jong and Tang [
22] stated, wetland wastewater treatment plants, while maintaining an appropriate operating regime, demonstrate high efficiency in septic tank waste material treatment. In addition, as Mancl and Rosencrans indicated, it is possible to drain sewage from septic tanks into properly prepared fields, which allows water to be retained in the soil, thus increasing water retention, which is crucial nowadays [
23].
However, in Poland, now and in the following years, the basis for the disposal of sewage from septic tanks is their selection and transport to a collective treatment plant. This applies to the majority of collective treatment plants in rural communes in Poland. Therefore, there is an urgent need to indicate the optimal proportion (volume) of sewage delivered by the slurry rolling stock, so that other sewage does not adversely affect the treatment of all sewage.
The overall aim of the analysis was to calculation an optimal organic and biogenic load into the wastewater treatment plant (WWTP) by blending strong cesspool wastewater with weak sanitary wastewater.
The detailed aim of the study was to determine the optimal percentage (amount) of sewage from cesspool in the mixture of wastes subjected to treatment processes, at which the unit load (Ul) of organic and biogenic pollutants will be at the level of the unit load assumed in the project for the wastewater treatment plant. The unit load of the analysed indicators assumed in the project is: 60 g·I−1·d−1 for Biochemical Oxygen Demand (BOD5), 120 g·I−1·d−1 for Chemical Oxygen Demand (COD) and 13 g·I−1·d−1 for Total Nitrogen (TN). The study results have an important practical aspect, as many small, collective wastewater treatment plants in rural areas in Poland have problems with setting the daily limit of sewage delivered by the slurry tanker, which will not adversely affect the functioning of the facility. The novelty of this work is the answer to an important question asked by the operators of wastewater treatment plants in Poland: “How much sewage from cesspool that is transported by slurry tanker can be introduced into the technological system of the wastewater treatment plant, so as not to disturb the purification processes?” The test results presented in this publication indicate the need to build or modernize technological systems for the collection and uniform dosing of collective wastewater from cesspools, so that this wastewater does not adversely affect biological treatment processes in wastewater treatment plants with bioreactors with activated sludge.
3. Results and Discussion
During the tests, the average daily sewage inflow from the sewage network amounted to Q
1 = 238.5 m
3·d
−1 to the said sewage treatment plant, while the average daily sewage delivered by the slurry tanker was Q
2 = 15.1 m
3·d
−1, which constituted 5.7% of their share in the total amount of wastewater subjected to the treatment process. However, the amount and frequency of wastewater from cesspool and transported by slurry tanker to the treatment plant was irregular on each day of the week. Wastewater from cesspools was delivered only on weekdays, i.e., from Monday to Friday. In the examined period, on weekdays, the amount of sewage delivered ranged from 5 to 28 m
3·d
−1. This represented from 2.1% to 12.4% (median 13.5 m
3·d
−1) of their share in the total amount of treated wastewater. This type of irregular delivery of wastewater from cesspools to the treatment plant is undesirable, as the irregularity of the amount of sewage flowing in and the load of pollutants contained in it causes disruption of wastewater treatment processes [
7,
16,
17,
18].
In the introductory part of the analysis on the quality of treated wastewater, the study presents pollutant indicators in wastewater flowing in from the sewage network and in wastewater from cesspool supplied by slurry tanker. In inflowing wastewater, the BOD
5 median value was 236.0 mg·dm
−3, the COD median value was 390.0 mg·dm-3, while the median of TN concentration was 57.5 mg·dm
−3. In inflowing sewage, the coefficient of variation for BOD
5 was Cv = 18%, for COD it was Cv = 14%, and for Total Nitrogen it was Cv = 13%, which indicates in all cases a small differentiation of this indicator according to the scale proposed by Mucha [
24]. Based on the results of analyses regarding the size of pollutant indicators in inflowing wastewater, it was found that their values corresponded to typical domestic sewage described in the literature [
25,
26,
27,
28]. The median value of the analysed indicators in sewage from cesspool was as follows: for BOD
5—3825.0 mg·dm
−3, for COD—7750.0 mg·dm
−3, and for TN—585.0 mg·dm
−3. As demonstrated, the values of indicators in wastewater from non-drainage tanks are much higher than the inflowing wastewater, which is confirmed by literature reports on the quality of wastewater from cesspool [
6,
8]. The variability of the values of the analysed indicators in wastewater was delivered at the level of mean variability according to the Mucha scale [
24]. The coefficient of variation Cv oscillated between 20% and 25%. Characteristic values of the analysed indicators in inflowing and delivered wastewater are presented in
Table 1.
To determine the optimal, i.e., design unit load of the analysed indicators, i.e., BOD5, COD and TN in the wastewater mixture subjected to the treatment process, an analysis was carried out in the following stages:
- −
Determination of the concentration of indicators in the mixture of inflowing sewage;
- −
Determination of the unit load of indicators in inflowing sewage;
- −
Determination of the unit load of indicators in the mixture of inflowing and delivered wastewater;
- −
Determining the optimal share of the amount of wastewater delivered in the total wastewater mixture so as to obtain the concentrations assumed in the project.
Based on the values of the analysed indicators, i.e., BOD
5, COD and TN in inflowing sewage and delivered sewage from cesspools as well as taking into account the quantitative balance of both types of sewage, the value of these indicators was calculated in the mixture of sewage subjected to the treatment process. In order to calculate the values of the analysed indicators in the wastewater mixture, the weighted average formula was used (1):
where:
Wa—value of the indicator in the wastewater mixture (g·m−3);
W1—value of the indicator in inflowing wastewater (g·m−3);
W2—value of the indicator in delivered wastewater (g·m−3);
Q1—amount of inflowing sewage (m3·d−1);
Q2—amount of sewage delivered (m3·d−1).
Based on the results of the calculated weighted average (1), it was found that the median BOD5 value in the wastewater mixture was 438.5 mg·dm−3, the median COD was 773.1 mg·dm−3 and the median TN was 89.3 mg·dm−3. With regard to organic indicators in the wastewater mixture, the range of values for BOD5 ranged from 247.0 to 815.6 mg·dm−3, while for COD it was from 415.6 to 1495.8 mg·dm−3. In both cases, the variability of the values of these indicators expressed by the coefficient of variation Cv was 28%, which indicates their average differentiation. The range of TN concentration in the wastewater mixture oscillated from 57.4 to 126.0 mg·dm−3 and was characterized by a small variation at the level of Cv = 18%.
The next stage of the analysis was to determine in the wastewater flowing from the sewage network a unit load of organic pollutants expressed as BOD
5 and COD and a unit load of biogenic pollutants expressed as TN. In order to calculate the unit load (per one inhabitant) of individual pollution indicators, Formula (2) was used:
where:
ULi—unit load of the indicator in inflowing wastewater (g·I−1·d−1);
Wx—concentration of indicator in inflowing wastewater (g·m−3);
Q1—amount of inflowing wastewater (m3·d−1);
In—total number of inhabitants connected to the sewage network (In = 1200).
In wastewater flowing from the sewage network, the median BOD
5 unit load was 46.1 g·I
−1·d
−1 and it was a 23.2% lower BOD
5 unit load than assumed in the project. The median COD load was 75.7 g·I
−1·d
−1 and was lower by 36.9% for the load assumed in the project. In the case of unit load of TN in inflowing wastewater, it was found that the median of this parameter was 11.4 g·I
−1·d
−1 and it was smaller than assumed by 18.6%. As can be seen in all 3 cases, the unit load of the analysed indicators was lower as a result of design assumptions. Characteristic unit loads of the analysed indicators in the inflow wastewater are presented in
Table 2. Based on the calculated coefficients of variation Cv for the unit loads of the analysed indicators in the inflow wastewater, it was found that they were at the level of average differentiation during the test period.
The reason for the unit load lower than assumed in the design of the organic and biogenic indicators tested is the inflow of accidental (rainfall) and infiltration waters to the sewage system. According to the interview with the operator of the sewage system in question, rainwater is discharged into the sewage network from illegally connected roof gutters from residential buildings. As Kaczor et al. [
29] and Nowobilska-Majewska and Bugajski [
30] indicated, the inflow of rainwater to the sewage network significantly reduces the concentration of organic and biogenic pollutants in the wastewater subject to treatment. Rainwater entering the sewage system intended for the disposal of only domestic sewage affects periodic disruptions in the operation of sewage treatment plants and causes higher costs of wastewater treatment [
29,
31]. In addition, the analysed sewage network is partly located under the groundwater occurrence level, which causes groundwater to flow into the sewer collectors through leaks in their connections. The foundation of sewer collectors below the groundwater level causes that these waters infiltrate the collectors through all kinds of leaks and causes an increase in the amount of wastewater flowing into the treatment plant [
32]. As Madryas et al. described in their research [
33], the intensity of the infiltration water inflow to sewage channels is directly proportional to the height of the groundwater table above the pipe.
Because the treatment process is subjected to a mixture of inflowing and delivered wastewater from cesspool, the size of unit loads for BOD
5, COD and TN in the wastewater mixture is analysed using Formula (3). To calculate the unit load in the wastewater mixture, the sum of Q
1 inflows and Q
2 supplied wastewater was adopted, and its weighted average was used as the value of the given indicator.
where:
ULmix.—unit indicator load in the wastewater mixture (g·I−1·d−1);
Wa—weighted average indicator in the wastewater mixture (g·m−3);
Q1—amount of inflowing wastewater (m3·d−1);
Q2—amount of wastewater delivered (m3·d−1);
In—total number of inhabitants connected to the sewage network (In = 1200);
Ic—number of inhabitants served by the slurry tanker-average per day (Ic = 10).
In the mixture of inflowing and delivered wastewater, which were subjected to the treatment process, the median unit load of BOD
5 was 90.5 g·I
−1·d
−1, and it was higher than the designed value (60 g·I
−1·d
−1) by 50.8%. The median COD in mixed wastewater was 155.2 g·I
−1·d
−1. Compared to the designed (120 g·I
−1·d
−1) unit load of COD, it was higher by 29.3%. The median TN in mixed wastewater was 18.5 g·I
−1·d
−1 and it was higher than the value assumed in the design (13 g·I
−1·d
−1) by 42.3%. As stated in all 3 cases, the unit load volumes of the analysed indicators in the wastewater mixture were significantly higher than the load assumed in the project. At the same time, it should be noted that this type of situation only took place on days when wastewater was delivered by means of slurry tanker. In relation to the unit load of the analysed indicators in the wastewater mixture, a greater unevenness was observed compared to the unit load of these indicators in the inflowing sewage. The unevenness of the unit load in the wastewater mixture expressed as the coefficient of unevenness Cv was for BOD
5—35%, for COD—35% and for TN—26%. Characteristics of unit load of analysed indicators in mixed wastewater are presented in
Table 3.
Based on the analysis of the unit load of individual indicators in the inflow sewage and in the mixture of inflowing and delivered sewage, it was found that the unit load of the indicators in the inflowing sewage is lower than assumed in the project, while the unit load in the sewage mixture is too high in relation to the specified load in the sewage treatment plant design.
Because it was found that in the wastewater mixture, the unit load of the examined indicators increases with the increasing amount (percentage share) of added wastewater delivered by means of Pearson’s linear correlation analysis, the following was determined:
- −
impact of the percentage (%) of wastewater delivered in the wastewater mixture on the unit load of the examined indicators;
- −
impact of the concentration of the analysed indicators in the delivered sewage on the unit load of these indicators in the sewage mixture.
Using the data covering the percentage (%) of the amount of wastewater delivered in the total wastewater mixture (independent variable) and the unit load data of the examined indicators in the wastewater mixture (dependent variable), the strength of the relationship of these two variables was determined in
Figure 2,
Figure 3 and
Figure 4.
Based on the analysis of the impact of the percentage of sewage delivered to the BOD
5 unit load in the sewage mixture, a correlation of r
xy = 0.73 was found, which in the scale proposed by Stanisz [
34] defines this level of correlation as very high. In the analysed case, the correlation is statistically significant at the level of α = 0.05. From the equation describing the regression line presented in
Figure 2, it may be stated that a change (%) in the sewage supplied in the total sewage mixture by 1% causes a change in the BOD
5 unit load by 11.1 g·I
−1·d
−1. The dependence of the influence of the percentage of sewage delivered to the COD unit load in the sewage mixture was determined at the level of r
xy = 0.78, which also indicates the relationship of these variables at a very high level. In the case of the COD load from the equation describing the regression line shown in
Figure 3, it is stated that a change in the proportion (%) of sewage delivered in the total sewage mixture by 1% causes a change in the COD unit load by 20.1 g·I
−1·d
−1. In the case of the analysed biogenic indicator, i.e., TN, the impact of the share (%) of sewage delivered on the unit load of Total Nitrogen in the wastewater mixture was at a high level, as indicated by the calculated correlation coefficient of r
xy = 0.59. From the equation describing the regression line presented in
Figure 4, it can be stated that a change in the share (%) of sewage delivered in the total sewage mixture by 1% causes a change in the unit load of TN by 1.2 g·I
−1·d
−1. In all analysed cases of studied relationships, the correlation is statistically significant at the level of α = 0.05.
In relation to the analysis of the correlation relationship between the concentration of organic and biogenic impurities (BOD
5, COD and TN) in the supplied wastewater and the size of the unit load of these parameters in the wastewater mixture, it was found that the correlation between the BOD
5 size of the delivered wastewater and the unit load of this parameter in the wastewater mixture is r
xy = 0.61. As follows from the equation describing the regression line in
Figure 5, a change in BOD
5 value in the supplied sewage by 100 g·m
−3 causes a change in the BOD
5 unit load in the sewage mixture by 2.1 g·I
−1·d
−1. The correlation of the COD value in the supplied sewage and the COD unit load in the sewage mixture was r
xy = 0.52. The equation describing the regression line in
Figure 6 indicates that with a change in COD value of 100 g·m
−3 in delivered sewage, there is a change in the COD unit load in the sewage mixture by 2.1 g·I
−1·d
−1. Whereas the correlation of the Total Nitrogen concentration in the supplied sewage and the unit load of Total Nitrogen in mixed sewage was r
xy = 0.50, and as the equation describing the regression line in
Figure 7 shows, along with the change in the Total Nitrogen concentration in the sewage delivered by 100 g·m
−3, the unit load changes TN in sewage mixed by 1.8 g·I
−1·d
−1. The level of correlation of the analysed variables in all cases in the scale proposed by Stanisz [
34] was high. In the analysed cases, the correlation is statistically significant at the level of α = 0.05.
Because the variability of the unit load BOD5, COD and Total Nitrogen in the wastewater mixture depends on the percentage (%) in them of wastewater from cesspits delivered to the sewage treatment plant by the slurry tanker and on the size of these parameters in the supplied sewage, a partial correlation analysis was performed. Partial correlation analysis will allow to determine simultaneously the strength (relationship) of the relationship of two dependent variables to one independent variable.
Based on the partial correlation analysis regarding the BOD
5 unit load in the wastewater mixture, it was found that the unit load of this parameter in the wastewater mixture depends on the percentage (%) of sewage delivered to them, as well as the value of this parameter in the supplied wastewater. However, the results of the partial correlation analysis indicate that the unit BOD
5 load in the wastewater mixture is more dependent on the percentage share of wastewater delivered than on the value of this indicator contained in it. The impact of the percentage of delivered wastewater on the BOD
5 unit load in the wastewater mixture was determined at the correlation level R
c = 0.80, while the impact of the BOD
5 value in the wastewater on the BOD
5 unit load in the total wastewater mixture was determined at the level of R
c = 0.72. On the scale provided by Stanisz [
34], in both cases the relationship is at a very high level. The significance of the calculated correlation coefficients was tested by the Student’s
t-test at the significance level of α = 0.05. In both cases, the significance of the studied relationships was found. In the case of partial correlation analysis regarding COD, it was found that the share of supplied sewage has a greater impact on the COD unit load in the wastewater mixture than the value of this parameter contained therein. The impact of the percentage share of sewage delivered to the COD unit load in the wastewater mixture was determined at the correlation level of R
c = 0.87, while the effect of the value of COD in the wastewater on the COD unit load in the total wastewater mixture was determined at the level of R
c = 0.75. In both cases, the correlation relationship is at a very high level, and the examined relationships are statistically significant at the level of α = 0.05. With reference to the unit load of Total Nitrogen in the wastewater mixture, it was found that its concentration in the wastewater mixture at a very high level has a percentage (%) of the supplied wastewater, where R
c = 0.72, while at a high level, the effect of the concentration of this parameter in the wastewater was noted as delivered, where R
c = 0.66.
In order to indicate the optimal amount of supplied sewage, which was added to the inflowing sewage, so that the unit load of the analysed indicators was at the assumed level in the project, the nomograms are presented in
Figure 8,
Figure 9 and
Figure 10. Nomograms for individual indicators were developed based on the results of partial correlations. From the developed nomograms in
Figure 8,
Figure 9 and
Figure 10, it is possible to forecast (predict) the unit load of a given indicator in the wastewater mixture depending on the percentage (%) of wastewater delivered in the wastewater mixture and on the value (g·m
−3) of this indicator in the supplied wastewater and the share percentage (%) of sewage delivered. The optimal load of the analysed indicators can be described by the formulas below:
- −
ULmixBOD5 (g·I−1·d−1) = −28.7907 + 9.8888·BOD5 in wastewater form cesspool + 0.0174% share of delivered wastewater
- −
ULmixCOD (g·I−1·d−1) = −86.1077 + 19.4221·COD in wastewater form cesspool + 0.0190% share of delivered wastewater
- −
ULmixTN (g·I−1·d−1) = −1.2271 + 1.3171·TN in wastewater form cesspool + 0.0205% share of delivered wastewater
That the unit load in the wastewater subjected to the treatment process oscillated within the limits of the designed load, assuming that in delivered wastewater the median BOD5 value is 3825.0 g·m−3, the COD value is 7750.0 g·m−3 and the Total Nitrogen concentration is 585.0 g·m−3, and the percentage of wastewater transported in the mixture should be between 4% and 6% (average 5%). Assuming the average daily amount of treated wastewater, which during the study period was 253.5 m3·d−1, the amount of wastewater transported by the slurry tanker should be from 10.2 m3·d−1 to 15.2 m3·d−1 (average 12.7 m3·d−1).