24 Sampling Assistance
on-line devices is increasing, the great majority of wastewater quality measurements
is carried out in the laboratory, after sampling. Thus, before considering analytical
methods for wastewater quality monitoring, based on either standard or alterna-
tive procedures, the sampling step must be considered because of its importance
as a source of potential errors. With the aim of getting a representative volume of
effluent, sampling has to face a lot of specific constraints related to wastewater char-
acteristics. Thus, wastewater sampling is difficult, considering the heterogeneity and
variability of effluents, and moreover the evolution of samples during transportation
from sampling site to laboratory, related to sample aging.
1.2.1.1 Heterogeneity
As for water, there are several types of wastewater. All types are characterized by
their composition heterogeneity. A wastewater is composed of water, carrying a lot
of suspended solids and dissolved substances which were not present originally (the
pollutants). Wastewater types depend on the nature and concentration of solids and
pollutants.
The most frequent type is urban wastewater, mixing municipal wastewater and
industrial ones. The composition of municipal wastewater is rather well known
and does not vary a lot from one human being to another or one town to another.
Typical compositions of urban wastewater have been published (Muttamara, 1996;
Metcalf and Eddy, 2003; Degr´emont, 2005). The concentration of total suspended
solids (TSS) varies from 200 to 600 mg/l, the volatile suspended solids from 200 to
600 mg/l, the biological oxygen demand (BOD) from 100 to 500 mg/l, the chemical
oxygen demand (COD) from 200 to 1200 mg/l, the total organic carbon (TOC) from
50 to 300 mg/l, the total nitrogen from 50 to 100 mg/l, and the total phosphorous
from 10 to 20 mg/l. These values can be decreased in the case of combined sewer
(effect of dilution of rainfall) or increased, depending on the proportion and nature
of industrial wastewater collected in the urban area.
Thus, the heterogeneity is related to the diversity of soluble pollutants’ nature, and
increased when considering emergent pollutants, but also to the nonsoluble fractions
distribution: colloids, supra-colloids and settleable suspensions. Table 1.2.1 presents
the size distribution of particulates and the coarse chemical composition of the
soluble fraction.
The composition of industrial wastewaters is obviously related to the industrial
activity (Eckenfleder, 2001; Metcalf and Eddy, 2003; Degr´emont, 2005), but above
all, to the existence of environmental equipments (e.g. wastewater treatment plant)
and investments (e.g. recycling process). Contrary to wastewater of domestic origin,
which increases with number of inhabitants, industrial loads are more and more
controlled and reduced under regulatory pressure. However, some problems remain
for industrial discharges in urban sewers, when the industrial fraction of wastewater
is dominant, leading to toxic effect and increasing the heterogeneity.
Wastewater Monitoring Constraints 25
Table 1.2.1 Dispersion characteristics of the main fractions of wastewater. (Adapted from
Sophonsiri and Morgenroth, 2004)
Standard Results
Fraction Min. Max. Mean deviation RSD (%) calculated from
Settleable (%) (>100 µm) 7 45 26.3 13.2 50 8 studies
Supra-colloidal (%) (1–100 µm) 12 50 27.4 12.1 44 9 studies
Colloidal (%) (0.1–1 µm) 7 48 15.6 12.6 81 9 studies
Soluble (%) (<0.1 µm) 9 64 37.2 17.4 47 10 studies
Composition
COD (mg/l) 203 967 496 292 59 7 studies
Protein (% COD) 8 31 19.3 9.1 47 8 studies
Carbohydrate (% COD) 6 18 11.3 4.6 40 9 studies
Lipid (% COD) 7 82 33.2 28.1 87 6 studies
Unidentified (% COD) 8 78 51.4 26.0 51 7 studies
1.2.1.2 Variability
Wastewater variability is due to its composition, changing along the sewer system
under the influence of several factors (see Chapter 2.1) and with the mixing of efflu-
ents of different origin (municipal and industrial). For an industrial sewer network,
the wastewater composition varies from downstream units or workshops to treat-
ment plant, with a decrease in variability under homogenisation effects of mixing
and storage tanks. Another variability factor is time, the wastewater production be-
ing generally less during the night for domestic activities, or during weekends and
holidays for some industries.
For all fractions and chemical compound groups of Table 1.2.1, the variability,
expressed as the residual standard deviation (RSD), is around 50 %, except for the
colloid fraction and for lipids. It should be noted that, for the soluble fraction, half
of the chemical compounds are not actually identified.
The variability can also be estimated from nonparametric measurement like UV
absorption spectra, giving qualitative information on the global composition of
wastewater (linked to UV absorbing substances). This approach will be explained
in Chapter 4.2 on industrial wastewater and discharges.
The heterogeneity and variability of wastewater quality must be taken into account
when a monitoring programme is planned.
1.2.1.3 Sampling Ageing
As in sewers, wastewater composition can vary very quickly when sampled. This
phenomenon, known as sample ageing, occurs under the influence of at least
26 Sampling Assistance
three factors:
rFirstly, as a heterogeneous medium, agitated in a sewer, suspended solids settle
rapidly in the sampling flask modifying the distribution of the fraction size by
flocculation, adsorption, etc.
rThe second factor is of a chemical nature, with reactions of reduction, complex-
ation, modification of acidic–basic equilibria, etc., occurring when the depletion
of dissolved oxygen leads to anaerobic conditions and to variation of redox po-
tential and pH. For example, the adsorption of surfactants on suspended solids, is
responsible, in raw or physico-chemically treated wastewater, for colloidal frac-
tion aggregation and, thus, for the increase of suspended solids (Baur`es et al.,
2004).
rThe third factor is probably the most important with the biodegradation effect
by microorganisms present in wastewater (coming from domestic waste). The
consequence is principally a degradation of organic matter, under aerobic or
anaerobic conditions, as it is the case in sewers. This will be explained in Chap-
ter 2.1.
Finally, sample ageing occurs even if the samples are refrigerated (in this case the
kinetic of sample evolution is slowed down) and can lead to 20 % variation for
some parameters (COD, TSS) in a few hours (Baur`es et al., 2004). This implies that
samples must be transported to the laboratory for analysis as soon as possible after
sampling.
1.2.2 MAIN PROCEDURES FOR WASTEWATER
QUALITY MONITORING
1.2.2.1 Sampling
Wastewater sampling is generally performed by one of two methods; grab (manual or
spot) sampling or automatic (sequential or composite) sampling. The first method is
simple, cheap and largely used, whilst the second is better for monitoring relevance,
considering the heterogeneity and variability of wastewater. The choice of a sampling
procedure is related to the sampling objective, regulatory requirements, measuring
treatment plan efficiency, sewer management, knowledge. Grab sampling is useful
for detecting fluctuation in composition, and discharge of pollutants, especially in
industrial effluent and storm-sewage investigations (Muttamara, 1996; Metcalf and
Eddy, 2003), and automatic sampling is preferred for all other purposes (regulatory,
time variation, mass balance, etc.). In any case, the measurement of flow rate during
sampling is strongly recommended for pollution loads calculation.
Main Procedures for Wastewater Quality Monitoring 27
Grab sampling
Grab sampling is like a snapshot, giving instantaneously a volume of wastewater
in one point. The reliability of measurement and analysis carried out from a grab
sample is thus limited to the composition of wastewater for a given control point at one
moment. Nevertheless, grab sampling is extensively used for water and wastewater
quality monitoring, and can be very useful for rapid information on a ‘slug’ discharge,
intermittently flows, short term variations checking or analysis or very unstable
constituents (phenols, cyanides, volatile organic compounds) (WEF, 1996).
It can be thus complementary to composite sampling. However, even if the grab
sampling procedure seems to be simple, several recommendations have to be made,
namely the following:
ruse of clean and adapted flasks, depending on the analysis to be made;
rchoose a sampling site with a homogeneous section preventing wastewater quality
variability (as for flow measurement);
rpay attention for sludge, biofilm or sediment on bottom or sides of sampling site;
rbe aware to not modify the sample composition just after sampling;
rdo not agitate before dissolved oxygen on site measurement or fill up the flask for
laboratory measurement;
ruse relevant conservation procedure(s) depending on analysis;
ralways note the sampling conditions of air temperature and time.
Thus grab sampling is not so easy to do, and cannot be carried out by untrained
people.
Automatic sampling
For wastewater quality monitoring, an automatic sample is generally preferred be-
cause of the time variability of effluents. Automatic sampling can principally be
performed using sequential or integrated mode, depending on time or volume.
rEven if it is the simplest form of automatic sampling, because no other devices are
needed other than the automatic sampler, the sequential mode can be carried out
several ways. The first one is the full sequential sampling mode with sampling at
regular time intervals of a given volume collected in one flask. After one sample,
the distributing system moves inside the sampler in order to fill the next flask,
i.e. several flasks are placed into the sampler (generally 24 or 12), correspond-
ing to hourly or bi-hourly samples. The composite sequential sampling mode is
28 Sampling Assistance
preferred, when a higher sampling frequency is needed, with the collection of
equal volume sub-samples at regular time intervals. A selected volume is sampled
with a given frequency (e.g. 200 ml every 15 min) and samples are collected in
a same flask of large volume (e.g. 20 l) for a single daily composite sample or
in several flasks for hourly or bi-hourly composite samples. In this last case, the
collection system of the automatic sampler is constituted of 12 or 24 flasks of 1
or 0.5 l, each corresponding to a period of time of 2 or 1 h, if the sampling period
is one full day. This technique is used if the daily variation of effluent charac-
teristics has to be known and is obviously more representative than several grab
samples.
rThe integrated sampling mode is selected when the knowledge of the daily load
has to be known. Instead of sequential samples of fixed volume, taken at regular
intervals over a period of 24 h, the volume of each sample is proportional to the
mean flow rate of a given time interval. Thus a flow meter, generally a device
measuring the height of the water table in a control section where the relation
height/flow is known, has to be installed and coupled with the automatic sampler.
Samples are collected in a single container in order to have a sample representative
of the average of the daily composition of wastewater and the pollution load is
calculated as the product of a given parameter by the mean value of flow rate
during 24 h. If the evolution of composition and load has to be known, samples
are collected, as for hourly or bi-hourly sequential sampling, in 24 or 12 flasks. In
this case, the daily load can thus be calculated as the sum of hourly or bi-hourly
loads. Sometimes, the volume of samples remains constant, but the time interval
is automatically adjusted, inversely proportional to the flow rate (e.g. 200 ml are
sampled every 10 m3). The use of two composite sampling during 24 h, at the inlet
and outlet of a wastewater treatment plant, is the most common way to determine
the average efficiency of the plant.
In practice
The urban wastewater treatment European Directive (Council Directive of 21 May
1991) indicates in Annex I-D that flow-proportional or time-based 24-h samples
shall be collected at the same well-defined point in the outlet and if necessary in the
inlet of the treatment plant in order to monitor compliance with the requirements for
discharged wastewater laid down in this Directive (see Chapter 1.1). Good interna-
tional laboratory practices aiming at minimizing the degradation of samples between
collection and analysis shall be applied. The minimum annual number of samples
shall be determined according to the size of the treatment plant and be collected at
regular intervals during the year:
r2000–9999 p. e.: 12 samples during the first year with four samples in subsequent
years, if it can be shown that the water during the first year complies with the