Environmental Engineering Reference
In-Depth Information
be considered as point discharges, because of the clearly defined locations. However,
considering the character of the pollution loads they could be defined as diffuse pollution
sources because of the high uncertainty of both qualitative and quantitative parameters
involved. The most complex case is the evaluation of pollution loads from agricultural
areas, which are complex in both directions - regarding pollution loads estimation and
regarding their spatial variation and identification. In the vast majority of such cases the
estimation of pollution loads requires the application of models based on the whole
catchment area analysis.
Pollution loads, contributed during the study described in section 3, were calculated as
the product of the runoff generated and the average pollutant concentration for each
specific drainage area. The runoff has been estimated by the simple procedure, based on
the total volume of rainfall during this wet season. The corresponding runoff coefficients
have been assigned more or less arbitrarily, based on engineering discretion and
information from maps. Unit pollution loads with respect to all parameters tested, have
been determined for each specific drainage area and corresponding sampling point.
Results are presented in Magombeyi et al. (in press). These loads could be regarded as
annual total and unit pollution loads from the corresponding areas for this specific wet
season. However, these results should be seen as a rough estimation, given the limitations
of the method applied and the time limitation of the project.
In order to illustrate the impact of the accuracy of determination of the individual
quantitaxtive and qualitative components of the pollution load on the final result, an
example has been solved (Fig. 4.5). The basic data, referring to an imaginary unit
drainage area is shown in the text box (Fig.4.5a). Figure 4.5b shows the magnitude of
variation of the pollution load, as a result of the deviation from the “true value”
(expressed in %), in the case when only one parameter is varying and the other is kept
constant at its “true value”. The pollution load variation is expressed also in terms of the
equivalent population, which would generate the same annual load in the form of raw
sewage. The same percentage errors for the quantitative and qualitative parameters were
selected. The errors with respect to the quantitative parameter (net rainfall) would be
generated by error in the determination of the runoff coefficient. With respect to the
qualitative parameters, COD was selected to represent a mean concentration (300 mg/l),
and the impact of possible errors of determination, causing deviation from the “true”
value, was computed. Figure 4.5b shows that the same error percentage, e.g. + 50%,
would account for a pollution load equivalent to the one released by 2465 persons, and
would be due to the use of a wrong runoff coefficient of 0.9 instead of 0.6. The same
error would be implied by a wrongly determined COD mean annual value of 450 mg/l
instead of 300 mg/l.
Usually, errors with the estimation of the runoff coefficients are not of such
magnitude, while the difference in the estimation of the EMC value and mean
concentrations regarding a single storm event (Table 4.3) show that the expected mean
concentration could vary significantly from its true value and a considerable amount of
continuous data is necessary in order to determine correct mean concentrations (EMCs).
In most cases, errors in the estimation of both quantitative and qualitative parameters
occur simultaneously and this type of effect is shown on Figure 4.5c, where both
parameters have errors of the same magnitude and direction. It should be noted that errors
in the negative zone (underestimation of actual quantity and quality parameters) lead to a
