__________________________________________________________________________________________ Water and Environmental Engineering Department of Chemical Engineering Upgrading alternatives for a wastewater treatment pond in Johor Bahru, Malaysia Master’s Thesis by Alexander Szabo and Oscar Engle March 2010
124
Embed
Upgrading alternatives for a wastewater treatment pond … · Water and Environmental Engineering Department of Chemical Engineering Upgrading alternatives for a wastewater treatment
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Water and Environmental Engineering Department of Chemical Engineering
Upgrading alternatives for a wastewater treatment pond in Johor Bahru, Malaysia
Master’s Thesis by
Alexander Szabo and Oscar Engle
March 2010
I
__________________________________________________________________ Vattenförsörjnings- och Avloppsteknik Water and Environmental Engineering Institutionen för Kemiteknik Department of Chemical Engineering Lunds Universitet Lund University, Sweden
Upgrading alternatives for a wastewater treatment pond in Johor Bahru, Malaysia
Master Thesis number: 2010-X
Alexander Szabo and Oscar Engle Water and Environmental Engineering Department of Chemical Engineering
April 2010
Supervisor: Associate Professor Dr. Karin Jönsson Co-Supervisor: Associate Professor Dr. Azmi Bin Aris
Examiner: Professor Jes la Cour Jansen
Picture on Front Page:
Inlet distribution pipe at Treatment Pond during de-sludging operation, UTM, Johor Bahru, Malaysia.
3.1 Retention time .................................................................................................................................... 9
Short Circuiting ................................................................................................................................... 10
Building several ponds in series .......................................................................................................... 10
Results and conclusions ...................................................................................................................... 23
5.3 Previous Study on the studied treatment pond at UTM .................................................................. 24
6. Study Area .............................................................................................................................................. 25
BOD5 and COD in effluent from WSP .................................................................................................. 35
TSS in effluent ..................................................................................................................................... 36
Average effluent values ...................................................................................................................... 38
Algae in effluent .................................................................................................................................. 38
Reduction in pond system .................................................................................................................. 39
Water flow .......................................................................................................................................... 48
Total flow ............................................................................................................................................ 48
11.2 Results and discussion .................................................................................................................... 53
BOD5 and COD ......................................................................................................................................... 53
Total Suspended Solids ........................................................................................................................... 56
15. Future work ........................................................................................................................................... 83
The retention time of the complete-mix system will approach the retention time of a plug-flow system
by increasing the numbers of ponds (Shilton et al., 2005). It can be observed that the reduction of
necessary retention time is greatest with the first additional ponds. Maturation ponds will also benefit
from multiple pond systems in series. Short-circuiting sewage water may result in large amounts of E.
coli bacteria in the effluent. If another following maturation pond is used, the risk that untreated short-
circuiting water from the first pond, will be part of the water that immediately gets to the outlet in the
second pond, is minimized.
In Christchurch wastewater treatment plant (New Zeeland), two series of 3 ponds each were used after
a biological trickling-filter process for treatment of domestic wastewater. In 2004, the two series were
joined to one train of 7 ponds in total (some rearrangements were done with additional baffles). The
reduction of E. coli bacteria was reduced from an average of 22.400 CFU/100 ml before down to 340
CFU/100 ml after reconstruction. The BOD5 average decreased from 25 to 14 mg/l with the new
arrangement (Masterton District Council, 1999).
Baffles
Baffles are “walls” within a pond that force the water to go through certain patterns. The baffles are
preferably built as a part of the original design. It is however possible to add “retro-fitting” baffles using
plastic-fabric sheeting, anchored in the pond bottom and connected to a floating device at the top.
12
13
4. Activated sludge process technology
Introduction to activated sludge process technology The use of activated sludge systems is a relatively new and modern treatment technology and was not
used to treat municipal wastewater in large scale until the 1950’s in the industrialized part of the world
(Alleman & Prakasam, 1983). A modern treatment facility uses a combination of mechanical, biological
and chemical treatment in order to purify wastewater. The activated sludge basin is the biological
treatment stage in which treatment is achieved by active microorganisms in wastewater. Because the
microorganisms in the activated-sludge basin use oxygen to degrade biodegradable material, the basin
must be supplied with aerators. The aeration consumes a lot of energy which is costly but at the same
time the treatment efficiency is high.
The possible reduction of nitrogen and phosphorous is dependent on which activated sludge system is
used. Figure 4.1 shows a typical arrangement of a primary treatment followed by biological treatment.
Figure 4.1: Typical arrangement of an activated sludge system (printed with permission from Kemira, 2003)
Primary treatment Before the wastewater enters the activated-sludge basin course contaminants are removed by a screen.
The gaps of the screen normally range between 3 and 20 mm in width (Kemira, 2003). The screen is
usually followed by a grit chamber in which heavier contaminants such as sand and gravel sink to the
bottom. Sometimes the grit chamber is combined with aeration to reduce courser particles and to
achieve grease removal. After the grit chamber the remaining settleable material can be collected in a
sedimentation tank.
The mechanical treatment is referred to as a primary treatment and the screen, the grit chamber and
the sedimentation tank together remove around one third of the oxygen demanding particles (BOD)
from wastewater (Kemira, 2003). According to Federal Wastewater Treatment Requirements (United
States) at least 30% of BOD must be removed in the primary treatment in order to meet the
requirements (EPA, 2009).
14
Biological treatment After mechanical treatment, wastewater is normally treated biologically and mechanical treatment is
therefore known as secondary treatment. With the help of micro-organisms, organic contaminants are
broken down to biological sludge. As wastewater contains a wide range of different contaminants there
are also a number of different microorganisms specialized in breaking down specific substances in
wastewater. The most common type of microorganism in wastewater treatment water is bacteria. The
bacteria are naturally occurring in nature and in a biological treatment basin these bacteria are brought
together in large numbers to break down unwanted substances (see chapter 2.3 - Degradation
processes). Apart from bacteria a biological treatment basin should also include other types of
microorganisms which help to improve treatment efficiency. There are for example worm-like organisms
which help to improve treatment efficiency by permeating the sludge and thus facilitate a better flow of
wastewater through the sludge.
Nitrification and denitrification
All wastewater contains nitrogen which is an important nutrient for biological growth. A small part of
the nitrogen is removed during the primary treatment. Most of the nitrogen in wastewater is in the form
of ammonium ( ) which is most easily removed in a biological twostage process called nitrification
and denitrification. This process ensures an effective removal of nitrogen in which nitrogen in form of
ammonium is finally removed from wastewater as nitrogen gas ( ). The first step in this process is
called nitrification and in two separate chemical reactions, autotrophic bacteria use oxygen in order to
convert ammonium into nitrate ( ).
The two chemical reactions can be seen below.
(Eq. 4.1)
(Eq. 4.2)
If reaction (1) and (2) are combined the total chemical equation can be written as:
(Eq. 4.3)
From the first chemical reaction (Eq. 4.1) it can be seen that acid in form of hydrogen ions ( ) is
produced. The hydrogen ions will react with the carbonate in water, if it is available. So if the amount of
carbonate is low, the pH in wastewater could drop and thus inhibit the nitrifying bacteria. If the pH
drops below 5.5 the nitrification process stops completely. But at the same time, the bacteria oxidizing
nitrite into nitrate in formula (Eq. 4.2) prefer a lower pH and so the pH cannot be kept too high.
The nitrification process is also temperature dependent. At lower temperatures, the sludge age has to
be higher in order to achieve nitrification (Figure 4.2).
15
Figure 4.2: Design curve for nitrification (Jansen, n.d)
In Eq. 4.4 below the equation for sludge age is given. The sludge age is the time the average sludge
particle spend in the activated sludge basin before it is removed.
(Eq. 4.4)
where:
SA = sludge age, days
V = volume of activated sludge basin, m3
SSa = average sludge content of aeration basin, kg SS/m3
Qex =excess sludge flow rate, m3
SSex =sludge content of excess sludge, kg SS/m3
Qef = effluent water flow rate
SSef = suspended content of effluent water, kg SS/m3
In the next step microorganisms called facultative anaerobes will reduce the nitrate into nitrogen gas.
This step is referred to as denitrification and it is a process in which bacteria is oxidizing organic matter
without dissolved oxygen. In an activated-sludge system the nitrogen gas is ventilated into the air. The
facultative anaerobes can also utilize oxygen when breaking down organic matter so it is important to
exclude oxygen during the denitrification process. Instead, the bacteria have to utilize oxygen which is
bound up as nitrate or nitrite (Kemira, 2003). The formula for the denitrification process can be seen
below.
(Eq. 4.5)
16
As can be seen in equation 4.5, facultative bacteria need a carbon source (organic matter) for the
denitrification process. Some of the carbon is taken from the wastewater but this is usually only
sufficient for denitrifying 50 % of the nitrogen (Kemira, 2003). If the wastewater does not contain
enough organic content for the denitrification process an external carbon source can be added, for
example in the form of methanol or glycol.
Chemical treatment Chemical treatment is a reliable method for removal of BOD and phosphorous. It can be used in many
parts of the treatment system where mechanical or biological treatment is not considered to be
sufficient. In chemical treatment, a coagulant is mixed with water and suspended flocs develop. The
suspended flocs are removed as sludge by sedimentation, flotation or filtration (Kemira, 2003).
The coagulants used in order to precipitate biological material, nutrients and minerals can be used
directly after the grit chamber in the primary treatment. This will give a reduction of phosphorous of
around 90% and a reduction of organic matter of around 75% (Kemira, 2003). The problem of using
chemical precipitation before the biological treatment stage is that more untreated sludge will be
produced.
Another stage where chemical coagulants can be used is during the biological activated-sludge process.
If a coagulant is added during the activated sludge process the chemical-biological sludge created there
will be removed in the subsequent sedimentation stage. By adding a coagulant in the activated-sludge
basin around 90% of the BOD, 90% of the phosphorous and 25% of the nitrogen will be removed
(Kemira, 2003). Because more sludge is created from the coagulant, the time the sludge spends in the
activated sludge basin (sludge age) will be lowered and the possibility for nitrification is reduced.
In order to maintain an effective nitrification and denitrification process in the activated-sludge basin,
coagulants can be added after the biological step. This will ensure an effective removal of phosphorous
without reducing the biological removal of nitrogen. Around 95% of the phosphorous will be removed
by this method (Kemira, 2003).
17
Activated sludge system – arrangements Practically, the activated sludge system can be arranged in a way which ensures efficient nitrification,
denitrification and organic breakdown of BOD. For example, sludge collected from the subsequent
sedimentation stage is recycled back to the activated-sludge basin in order to achieve a high
concentration of bacteria in the in sludge basin (Figure 4.3).
Figure 4.3. In this activated sludge system, sludge is recirculated back from the sub-sequent pre-settler in
order to achieve a more efficient breakdown of organic matter.
The method of recycling sludge could also serve as an extra carbon source for denitrification. If the first
basin in the activated sludge is kept anoxic, which means that no oxygen is available, the denitrification
bacteria can utilize carbon from incoming wastewater and recycled sludge. The last basins in the
activated sludge are kept aerated in order to achieve nitrification. The nitrate produced during the
nitrification stage is needed for the denitrification process and nitrate from the aerobic basin is
therefore transferred to the anoxic basin together with sludge from the aerobic zone. The amount of
nitrogen removed with this method is dependent on the degree of recirculation. This arrangement for
achieving both nitrification and denitrification with the help of recirculation of nitrate and sludge is
called pre-denitrification and is illustrated schematic by Figure 4.4.
18
Figure 4.4. In this activated-sludge arrangement both nitrification and denitrification are achieved with the
help of recirculation of nitrite and sludge to the anoxic zone. This method is called pre-denitrification.
Another solution for achieving denitrification is called post-denitrification. The wastewater will first
enter an aerobic zone where nitrification and breakdown of BOD occurs. After the aerobic zone
wastewater is transferred to a subsequent anoxic basin. Because the wastewater now contains smaller
amounts of BOD when entering the anoxic zone, an external carbon source is added (usually in the form
of hydrolyzed primary sludge or different kinds of alcohol) (Kemira, 2003). The advantage of this method
is that no recirculation of nitrate is needed. The disadvantage is that an external carbon source in form
of methanol or ethanol is expensive and large amounts may be necessary to add.
Sludge treatment Sludge created from wastewater treatment could easily become a public-health hazard if not handled
properly. What to do with the sludge is today considered as one of the main problems in wastewater
treatment. Still, there are several methods to choose from when selecting arrangements which can
process sludge. A careful selection of the sludge-treatment method has to be made, as costs for
stabilization, dewatering and disposal of sludge may be higher than the operating cost for the activated-
sludge treatment (Hammer, 1986).
In Malaysia facilities for treating sludge are limited and today most of the sludge is treated in sludge
lagoons or dried on large sludge-drying beds. Only in semirural areas are these methods seen as long
term solutions. A more long-term solution for urban areas would be to install digesters and mechanical
dewatering facilities. Then the gas produced from the digestion chamber could be captured and utilized
as fuel.
According to Malaysian standards presented in Sewerage Services Department (1998) a sludge strategy
consists of three main stages:
19
Stage 1 – Preliminary treatment and digestion
Stage 2 – Conditioning and dewatering
Stage 3 – Utilization and disposal
In stage 1, the sludge can be thickened in a centrifuge in order to increase the dry-solids content. After
the sludge has been thickened, it is sent to an aerobic or anaerobic digestion chamber.
In stage 2, the stabilized sludge is dewatered in a filter press and the dry-solids content in the sludge is
raised to more than 25% (Sewerage Services Department (1998). Sludge lagoons or drying beds can also
be used in rural areas.
In stage 3, a sludge tank which can handle sludge for more than 30 days must be provided (Sewerage
Services Department, 1998). The final step is composting or disposal of the sludge. The treated sludge
can for example be utilized as fertilizers on fields.
20
21
5. Previous studies on WSP systems
5.1 Example of computer simulation of WSP geometry
Method
In a test which was executed at Alexandria University, Egypt by Abbas et al. (2006), different
length/width ratios for a WSP were tested in a computer program in order to determine the degradation
of BOD and the amount of dissolved oxygen. In reality, there are always a lot of parameters affecting the
treatment pond which the engineers cannot control, for example temperature variations, amount of
sunlight and wind speed. In the computer model these parameters were set to a constant value.
The influent wastewater had the following characteristics: BOD concentration 300 mg/l, DO
concentration 0.0 mg/l and 150 kg/(ha*day) in surface organic loading rate of BOD. The retention time
was set to 31 days, the inflow was set to 0.18 m3/s and the pond depth to 1.5 meter (Abbas et al., 2006).
The simulation was executed for the four different length/width ratios which can be seen in Figure 5.1
below.
Figure 5.1: Different shapes of the waste stabilization pond simulated in the computer program (printed with
permission from Abbas et al., 2006).
For each shape, simulations were also executed with 2 or 4 baffles. The baffles will change parameters
such as the retention time and the flow path in the waste stabilization pond. In Figure 5.2 an example of
the 4 ponds with 2 baffles can be seen.
22
Figure 5.2: A schematic representation of each shape with two baffles (printed with permission from Abbas et
al., 2006).
Results and conclusions
The effluent amount of BOD5 was calculated for the different length/width ratios L1/L2 =1, 2, 3 and 4. In
the first case, simulations were executed for different length/width ratios with no baffles. The same
procedure was then done for each shape with two and four baffles respectively. In Table 5.1 the values
of BOD5, DO and water velocity can be seen from the simulations.
Table 5.1: The amount of BOD and DO in effluent water (re-drawn from Abbbas et al.,2006)
Without baffles
With two baffles With four baffles
Range of BOD5 concentration found in effluent water (mg/l) 235-252 22-233 13-54
Range DO found in effluent water (mg/l) 0.26-0.51 6.24-9.96 9.57-10.03
Range of minimum water velocity (x 10-3 m/s) 0.002-0.015 0.018-0.91 0.047-0.05
Range of maximum water velocity (x 10-3 m/s) 2.69-3.01 3.19-169 105-765
The analysis of the results shows that the BOD5 reduction increases significantly by introducing two or
four baffles. The best removal efficiency of BOD5 was achieved with four baffles and a value of 4:1 in
length/width ratio. Within each case it was reported that increasing the length/width ratio and
introducing baffles slightly increases the water velocity and DO in the effluent water (Abbas et al., 2006).
23
The conclusion drawn from these data simulations is that a pond should have a length/width ratio of 4:1
and at least two baffles.
5.2 Facultative and maturation ponds in Sri Pulai, Johor Bahru, Malaysia.
Method
An analysis of the performance of a wastewater stabilization pond in Sri Pulai, Johor Bahru was done in
2002. The treatment system uses a primary facultative pond followed by a maturation pond (Ujang et
al., 2002). This is the same pond arrangement as used at UTM. The system in Sri Pulai serves a
population equivalent (PE) of 10.327 from a residential area of approximately 0.7 km2. The ponds
together cover an area of 17.725 m2, with pond volumes of 16.275 m3 for the facultative pond and
10.115 m3 for the maturation pond. The volume of the facultative pond should result in a retention time
of about 30 days. Around 40 % of the incoming water is assumed to origin from infiltrating ground water
(Ujang et al., 2002).
Table 5.2: Average wastewater characteristics at Sri Pulai WSP. The influent results is based on 24 samples,
the results from the effluent facultative pond and the effluent maturation pond is based on 14 respectively 7
samples. (re-drawn from Ujang et al., 2002)
COD (mg O2 / l) SS (g/l) NH4+ (mg/l) NO3
- (mg/l)
Influent
Effluent facultative pond
Effluent maturation pond
Removal in entire pond system
Removal in facultative pond
Removal in maturation pond
446
139
114
74%
69%
18%
146
48
41
72%
67%
15%
23.1
19.7
16.8
27%
15%
15%
1.5
1.6
1.2
20%
- 7%
25%
Results and conclusions
Compared with the effluent standard for Malaysia, the treatment process was not sufficient, since the
total COD concentration of the effluent was on average 114 mg O2/l (see Table 5.2). The Malaysian
effluent standard B is set to 100 mg COD/l (Ujang et al., 2002).
The authors´ conclusion of the unexpected low treatment efficiency (the pond should be able to meet
standard B) could partly be caused by hydraulic short-circuiting. The low treatment efficiency could also
be caused by the specific growth rates of biological degrading bacteria not being so temperature
dependent as expected. The authors recommended aeration and installation of baffles to improve the
treatment.
24
5.3 Previous Study on the studied treatment pond at UTM In year 1999, an under-graduate report was produced by Zahari and Zain (Faculty of Civil Engineering,
UTM) analyzing the same WSP at UTM. The report is written in Malay language, and our focus has been
on the wastewater composition data found in this report. The measurements have been conducted
without flow measurements, hence, the results only represent average values, that is the values were
not weighted against the influent water flow. When the waterflow into the treatment pond is high, it is
likely that more BOD, COD and TSS will end up in the pond, and this is not considered in the report. The
results from Zahari and Zain can be found in appendix A.
25
6. Study Area
6.1 Climate Johor Bahru lies in a tropical region very close to the equatorial line with temperatures ranging between
21 to 32°C all year around. The climate is humid with an average relative humidity around 90%
(Richmond et al., 2007). The precipitation comes in form of short but heavy rain showers and the
average rainfall in Johur Bahru is around 2400 mm each year (World Meteorological organization, n.d.).
The rainiest periods are from March to May when Johor Bahru is influenced by the southwest monsoon
and November to December when the northeast monsoon arrives.
As Johor Bahru lies in a region with a typical tropical climate with heavy thunderstorms, the storm water
is usually not connected to the wastewater treatment systems. This is also the case at UTM where the
storm water is led through channels and ditches directly to the local rivers.
6.2 The wastewater treatment facilities at UTM Within the Universiti Teknologi Malaysia, two waste stabilization pond facilities and two mechanical-
biological treatment plants are in operation. The pond system this report will focus on is located south-
west of the UTM campus area (see Figure 6.1) and it is treating the wastewater from the western part of
the catchment area in UTM (see Appendix D). These WSP:s were built in 1985. The mechanical-biological
treatment plants were built later, as the university expanded.
Figure 6.1: Map over the UTM campus area with (1) The WSP studied in this report (2) another WSP within
UTM; (3) biological-mechanical treatment plant; (Another campus area and treatment plant is located in the
north-east, outside the map.)
26
The treatment system that this report will focus on consists of two parallel lines of facultative ponds
followed by maturation ponds (Figure 6.2). The available documents do not tell for how many persons
the ponds are dimensioned for, but according to the contractor for de-sludging operation, this pond
system was built to serve approximately a P.E. of around 8000. The pond system is a simple
construction, where the influent wastewater is divided in a chamber before it enters each treatment line
through pipes located under the water surface. The ponds have no screening and objects of many
different sizes have been seen floating in the pond. These objects have a tendency to clog the cannels
between the ponds. Since the water level of the receiving river is higher than the effluent level, a
pumping station is located between the pond and the river. The pumps are currently working
discontinuously, and during the time when the pumps are not pumping, the ponds accumulate
wastewater beyond the designed water levels. The technical data of the pond system can be found in
Appendix G.
Figure 6.2: Shape of pond system with facultative ponds (1) followed by maturation ponds (2).
6.3 Recipient
Introduction
The effluent from the WSP is released into a stream passing through the UTM area (Figure 6.3). The
stream transforms into two small lakes a few hundred meters downstream, where the main entrance to
UTM is located (see Figure 6.4). In these small lakes, canoeing activities take place which demands good
water quality. The part of water that origins from the WSP is large compared to the upstream river
water. Most of the water in the lakes is therefore considered mainly to be effluent water from the WSP.
27
Figure 6.3: The receiving river. Upstream (left) and downstream (right) of WSP.
Method
Composite samples were collected consisting of 3 samples upstream and 3 samples downstream the
WSP on the 14th of January 2010. The samples were analyzed for COD and TSS according to the
procedures described in chapter 7. The points where the three composite samples upstream and
downstream were collected can be seen in 6.4.
28
Figure 6.4: Map over recipient with upstream collection points (1) and downstream collection points (2)
Results and discussion
Figure 6.5: The water quality upstream and downstream of WSP on the 14th
of January 2010.
As expected, the COD and TSS levels are significantly higher downstream of the WSP (Figure 6.5). The
stream has not received any effluent from any kind of treatment facility before passing the WSP. The
initial COD may origin from pollution caused by living organisms such as fishes, birds and algae.
0
10
20
30
40
50
60
Upstream Downstream
mg/
l
COD
TSS
29
The high TSS increase downstream may be the result of the treated wastewater, but also from high
concentration of algae and bottom sediments stirred up in the stream.
Due to limitations in time and the cost for COD reagents, focus has been set on influent and effluent
water. These values are single analyses of composite samples and should only be considered to give an
indication of the changes of pollution levels in the river.
30
31
7. Sampling and analysis methods
Sampling and flow measurements The sampling procedure included filling plastic containers (each about 1 liter of volume) with sample
water from the influent chamber and one container with water from the effluent channel. The samples
were immediately transported to the laboratory where they were stored in a cooling room
(temperature between 10-13 °C).The following day the analyses of BOD and COD was started. TSS
analyses were usually carried out a few days after the sampling. The procedure for flow measurements
is described in Chapter 9.
COD COD analyses have been undertaken with Hach standard procedure for colorimetric determination. The
Hach reagents used were capable of “high range” (0-1500 mg COD/l). The digestions of the reagents
were done in a Hach DRB 200 for 2 hours at 150 °C (see Figure 7.1). The analysis has been done with
single samples.
The reading of the reagents was done in a Hach DR/4000 and a Hach DR/5000 (see Figure 7.1). The
average values from both machines were calculated.
Figure 7.1: Hach DRB 200 Digestor for reagents (left) and Hach DR/5000 for colorimetric
determination (right).
If the reading showed unexpected low or high values for a specific sample, a new analysis on that
sample was done the following day. If the new analysis gave the same result, the average of these values
was used. If another value, that was in the same range as the other samples from that measuring event,
was measured, that value was used instead.
32
BOD5 Influent samples from the 5-6 Nov were successfully analyzed by the incubator method. The effluent
samples from that day were analyzed by the manual method due to lack of available space in the
incubator. Samples from the other testing events failed due to technical problems, or are not used as
they are considered unreliable.
Incubator method
BOD analyzes were carried out using a BOD incubator (Hach, model 205) which is especially designed for
BOD tests. The sampling bottles are connected with a tube (for oxygen measurements) and put in a
compartment that keeps the temperature at 20 °C. A computer records the oxygen consumption over
time. The procedure was performed according to the product manual. One benefit of the BOD incubator
is that it allows reading of the oxygen consumption (BOD) continuously (instead of only a final
measurement after 5 days as the “manual” procedure described above). For calculations of the k-value
(Appendix E) readings were recorded with 0.5 days intervals (Appendix F).
One full testing event with the BOD incubator was successful. Other testing events failed due to
technical problems with the BOD incubator or lack of available storage space in the equipment.
Manual method
The “manual” measuring of BOD5 is done with the following procedure:
- Filing 300 ml airtight bottles with diluted wastewater (The purpose of diluting the samples was
to reach a DO above 7 mg O2 / liter, which may be necessary for a successful BOD reading after 5
days, as the DO must be above 2 mg O2 / liter at final reading. Dilution 1:1 with distilled water
was enough for this purpose).
- Measuring the initial DO level in the samples with a DO meter.
- Storing the bottles in a refrigerator at 20 °C for 5 days.
- Measuring the final DO with a DO meter.
The BOD5 was calculated as the decrease of DO per liter of water.
This method was considered as slightly unreliable as the reading of the DO meters was difficult. The DO
measured DO concentration was not stable over time, hence making it difficult to predict the final DO.
33
Total Suspended Solids
TSS analyses were made by filtering 100 ml of sampling through a microfilter (Whatman, GF/C glasfaser
microfilter). In order to force the sampling water through the filter, an air compressor was used to
create low pressure in the bottle receiving the filtered water (see Figure 7.2). The filters were measured
before and after filtering. To eliminate the weight contribution of water, the filters were dried in an
oven at around 110°C for one hour after filtering, but before final measuring.
Figure 7.2: Sample water was filled in the upper glass container. The air compressor (right in the picture)
created under pressure in the receiving container (containing filtered water in the picture).
34
35
8. Performance of the existing WSP
8.1 Method Effluent samples were collected every two hours at the pond outlet. Since the variation of effluent
quality does not change drastically, composite samples were created where each composite sample was
created from three effluent samples under a 6 hour period. This was done in order to reduce work and
costs in the laboratory. The analysis procedures are described in chapter 7.
8.2 Results and discussion
De-sludging conditions
The treatment efficiency is dependent on the sludge, since the sludge contains bacteria that degrade
pollutants in the wastewater. If the sludge has been removed recently, the system may operate
ineffectively due to the lack of bacteria. If however, the sludge has accumulated for a long period, the
volume of the sludge will reduce the HRT in the pond system, hence making the treatment less effective.
One can assume that the best treatment occurs somewhere in the middle between the de-sludging
operations.
The WSP (north treatment line) was de-sludged after the first measurements were done on the 5-6
November 2009. During de-sludging one treatment line at a time is closed down. The wastewater is
pumped out and the accumulated sludge from the bottom of the ponds is removed. The de-sludging
operation takes a few weeks to be completed and is done with 10 year intervals. On the 9-10 January
2010, the WSP had been in operation, after de-sludging, for about 2 weeks when the samples were
collected. The average effluent COD was 64 mg/l before and 58 mg/l after de-sludging (see Figure 8.1
and 8.2). The average BOD5 for 5-6 November was 39 mg/l BOD5 on average (Figure 8.1).
Since the samples were collected just before and after de-sludging, the results presented above may be
slightly higher than during most of the operation time. Zahari & Zain (1999) showed in their report an
average effluent of 40 mg/l COD (see Appendix A) which is approximately 30% better than the results in
this report. This could be due to the sludge conditions discussed above.
BOD5 and COD in effluent from WSP
The quality of the effluent water is relatively stable (see Figure 8.1 and 8.2). Variations may depend on
sunlight that changes the concentrations of algae in the surface layer. Another factor to consider is the
effluent pump that is working discontinuously.
36
Figure 8.1: BOD5 and COD in effluent from the northern treatment line on the 5-6 November 2009 (before
de-sludging).
Figure 8.2: COD in effluent from the northern treatment line on the 9-10 January 2010 (after de-sludging).
TSS in effluent
The TSS levels are higher and show a different pattern after the de-sludging was carried out. The
average TSS on the 5-6 November was 37 mg/l (Figure 8.3), but 22 mg/l on the 9-10 January (Figure 8.4).
The reason for lower TSS values after de-sludging may be caused by many reasons. One reason is the
fact that if the ponds had high levels of sludge, the volume of water in the ponds decreased, hence
creating less retention time and higher water flow, which disturbs the settling process. If the sludge
reaches a high level from the bottom of the pond, the distance between the top sludge layer and outlet
0
10
20
30
40
50
60
70
80
17-21.40 23-03 05-09 11-13
mg/
l
Time
COD
BOD5
0
10
20
30
40
50
60
70
80
12-16 18-22 00-04 06-10
CO
D (
mg/
l)
Time
37
channel gets narrower and more solid particles may get stirred up. The algae population may not have
developed entirely after the de-sludging, and since algae contribute to the effluent TSS values, it could
be another possible explanation.
For the composite value analyzed on the 9-10 Jan, between 00-04, it can be seen that it is significantly
lower than the other values from this date (Figure 8.4). One possible reason could be that a pump
station is located just after the outlet where the effluent samples were taken. The pumps are working
discontinuously and start when the water reaches a specific height. When the pumps start operating,
the effluent water flow increases drastically. This causes differences in the water quality and the
measured changes of effluent quality over time may therefore have been influenced by the unknown
pumping pattern.
Figure 8.3: Total Suspended Solids (TSS) in the effluent on the 5-6 November 2009.
0
5
10
15
20
25
30
35
40
45
50
17-21.40 23-03 05-09 11-13
TSS
(mg/
l)
Time
38
Figure 8.4: Total Suspended Solids (TSS) in the effluent on the 9-10 January 2010.
Average effluent values
The average weighted effluent from 5-6 Nov, 9-10 Jan and average values from Zahari & Zain (1999) give
the total average values seen in table 8.1. The values from the report by Zahari & Zain (1999) may
represent “better” sludge conditions in the pond, hence lowering the average level of pollutants (data is
available in Appendix A).
Table 8.1: Effluent average values.
BOD5 (mg/l) COD (mg/l) TSS (mg/l)
5-6 November 39 64 37 9-10 January 29 58 22 Zairi & Zain (1999) 16 40 13
Average 28 54 24
Algae in effluent
Algae are present in the effluent and are known to contribute to the effluent COD and BOD. The water
showed clear signs of microalgae in every effluent sample analyzed. In order to understand the
contribution of particles to the effluent COD, one test was carried out with filtered and unfiltered
composite effluent water from 9-10 January 2010 (see Figure 8.5).
0
5
10
15
20
25
30
35
40
45
50
12-16 18-22 00-04 06-10
TSS
(mg/
l)
Time
39
Figure 8.5: Unfiltered and filtered composite effluent water (from all 12 samples 9-10 Jan 2010).
The unfiltered effluent had a COD level of 57 mg/l and the same filtered effluent 27 mg/l. The difference
between unfiltered and filtered (suspended COD) made therefore up 30 mg/l COD. Since it has been
observed that the quantity of algae in the effluent is prominent, algae are assumed to contribute to a
major part of the 30 mg/l COD difference.
Reduction in pond system
In Tables 8.2 and 8.3 below the reduction of average COD and average TSS from the samplings made on
the 5-6 November, 2009 and on the 9-10 January, 2010 is shown.
Table 8.2: Influent (weighted COD average, see chapter 10.2) and Effluent COD values.
Date Influent COD (mg/l) Effluent COD (mg/l) Reduction
5-6 Nov, 2009 122 64 48%
9-10 Jan, 2010 113 58 49%
Table 8.3: Influent (weighted TSS average, see chapter 10.2) and Effluent TSS values.
Date Influent TSS (mg/l) Effluent TSS (mg/l) Reduction
5-6 Nov, 2009 63 37 41%
9-10 Jan, 2010 47 22 53%
0
10
20
30
40
50
60
Unfiltered effluent
Filtered effluent
mg/
l
40
41
9. Wastewater flow
9.1 Survey of wastewater producers at UTM campus
Method
Construction drawings over the area were used to identify the sewage pipes connected to the WSP. The
chief engineer at UTM helped to identify the outer borders of the catchment. The amounts of students
and staff living and studying/working within the catchment were identified by consulting housing and
university offices. Some assumptions have been made where no information could be found about the
number of people living and working in buildings connected to the treatment pond (for more
information of the estimation of P.E., see Appendix B).
The number of students and staff working within the faculties at UTM were given a P.E. of 0.2 instead of
1 (Guidelines for developers) as they do not consume as much water at the faculties as they would have
done at home, thus lowering the PE value.
Outside UTM campus there are private family houses where the sewerpipes are connected to the
studied treatment pond. The number of people living in each household is not exactly known so it was
estimated that each household consists of 5 family members.
Other buildings connected to the treatment pond where no information of the number of residents
could be given were the UTM Main Office and the Mosque. The number of staff in the UTM Main Office
was assumed to be 500 and the number of people visiting the Mosque could be up to 3000 people. The
P.E. for the people working and visiting the buildings was set to 0.2 in both cases (Guidelines for
developers).
The water consumption for UTM’s students is based on a value collected from a report by Katimon and
Demun (2004). The report concludes that the average water consumption for UTM is 260
litres/(person*day). This value is higher than the average value of 208 liter/(person*day) for the rest of
Malaysia (Ithnin, 2007).
The reason for the higher consumption according to our proposition could be:
1. Students and staff at UTM have a more water consuming behavior, such as more showering and more
use of laundry facilities.
2. Leakage of fresh water from delivery pipes.
A third proposition by Katimon and Demun (2004) for the higher water consumption at UTM is:
3. Water intensive faculties at UTM, such as the Environmental Engineering Laboratory and Marine
Engineering Laboratory, consume large quantities of water.
42
When estimating the water consumption behavior of UTM the value from Katimon and Demun (2004) of
260 litres/(person*day) is initially used. The water consumption and waste water production is normally
of equal amount if no irrigation occurs. At UTM it was observed that washing machines, cooking
facilities and students’ laundry in some cases were not connected to the sewer network so the final
waste water production behavior of UTM will be lower than 260 litres/(person*day).
Results and discussion
The pond receives wastewater from around 10.500 persons according to the survey (see Appendix B).
No factories or other similar activities are present within the catchment. Hence, the wastewater is pure
domestic.
If the value of 260 litres/(person*day) from Katimon and Demun (2004) is used the wastewater flow into
the pond can be calculated:
(Eq. 9.1)
However, all the freshwater consumed will not end up as wastewater due to losses. For example, it has
been observed that the public dinner facilities do not drain their kitchen wastewater into the sewer
system, but into the storm water channels. These facilities support the majority of the students with
food since students usually do not have access to kitchens.
It has also been observed that washing machines for the students’ laundry in some cases are connected
to the storm water drain. Washing machines generally consumes much water and therefore some losses
also from laundry take place.
According to the household water consumption in Sweden around 20 percent of the total water
consumption is used for cooking and laundry. If the same value is applied to Malaysian circumstances
the consumed fresh water going into the sewer system at UTM will be:
(Eq. 9.2)
or if calculating water consumption per capita at UTM:
(Eq. 9.3)
The value of will be used in Chapter 9 when estimating the infiltration of
water into the sewer pipes.
43
9.2 Flow measurements
Method
The inlet water velocity has been measured during three 24h-events. Water speed was measured with a
current meter (model SWOFFER 2100) in one of the two inlet concrete chambers (see Figure 9.1 for
point of measurement) every two hours. The propeller was placed so that it measured the water
velocity in the centre of the inlet pipe.
Figure 9.1. The flow measurements were made in a concrete chamber illustrated from above in the drawing.
Before the wastewater is sent to the two different treatment lines, the waterflow is split in the water divider
chamber. The pipes seen in the figure are placed beneath the surface and have a diameter of 18 inches (45.7
cm) each.
The diameter of the two inlet pipes is 18 inches (45.7 cm). The water velocity was measured at a point
where the pipe ends and releases the water into the inlet chamber. If one assumes that the measured
water velocity is the average water velocity in the pipe, the flow can be calculated by equation 9.4:
Q (Flow of wastewater) (m3/s) = Area of pipe (m2) * Water velocity (m/s) (Eq. 9.4)
Calculations of the flow are done on two hours basis. This means that the flow measured at one event is
assumed to be valid for two hours (3600 seconds). Since the measurements on the 5-6 November are
not complete (9 measurements instead of 12 due to technical problem), the first three flow values from
17:30 to 22:00 on the 5th of November is estimated with “designed” values. The designed values are
designed similar to the flow measurements made on the 18-19 November where rain occurred in the
afternoon around the same time as on 5-6 November (see Figure 9.5 on page 48).
44
In the case 5-6 November the following equations below are used:
(Eq. 9.5)
(rain-peak-flow at 14:15 is excluded) (Eq. 9.6)
In the case 18-19 November the following equations below are used:
(Eq. 9.7)
(rain-peak-flow at 16:15 is excluded) (Eq. 9.8)
In the equations, Q is the measured flow values from each sample day. In the data from 5-6 November
the total water flow is also calculated without the rain-peak- flow at 14:15 with Eq. 9.6. The same
procedure is made with the data from 18-19 November where the total water flow is calculated without
the rain-peak- flow at 16:15 with Eq. 9.7. This is made in order to estimate the amount of infiltrating
water into the sewer pipes (see Chapter 10).
Results
Measurements 5-6 Nov 2009
“Thursday-Friday under school period”
The first measurements were made from the 5th of November to the 6th of November. This was the week
before the examination week. During this period no lectures at the university were held and most
students were assumed to stay at home at campus. The sampling period started at 17:30 on the 5th of
November but unfortunately there were no flow measurements made from the first three sample
occasions that day due to technical problems. The first flow measurement started at 23:00 which can be
seen in Figure 9.2 below. The flow decreased at night and increased slightly in the morning. From 13:00
to 16:00 a rainstorm started which can be seen in Figure 9.2 in the precipitation data from UTM’s own
rain gauge. In connection to the rainstorm there was a significant increase of the flow.
45
Figure 9.2: Above: The water velocity in the inlet chamber for the northern treatment line of the WSP system
during 5-6 November. Below: Precipitation data at site (UTM weather station data)
Figure 9.5: The inlet wastewater flow to the WSP system. The designed flow is made to fill the gap between
the measurements during 5-6 Nov.
* On the 5-6 of November and 9-10 of January, the total flow is assumed to be twice the flow in one
inlet, since there are two inlets with equal amounts of water.
* On the 18-19 of November, there was only one line and one chamber in use. However, some leakage
to the closed line occurred except at the last measurement at 10 a.m., when workers managed to stop
the leakage. The leakage was measured to 9 l/s and is considered when calculating the flow from 18-19
November.
Total flow
The total inflow for 5-6 November has been calculated with Eq. 9.5 where the peak-flow during the rain
is included and Eq. 9.6 where the peak-flow during the rain is excluded. The total inflow for 18-19
November has been calculated with Eq.9.7 where the peak-flow during the rain is included and Eq. 9.8
where the peak-flow during the rain is excluded. The total inflow for 9-10 January has been calculated
with Eq. 9.5 and since there was no precipitation during this measuring period no rain peaks have been
considered. Average inflow quantity to the pond each day will be based on Qtot with rain from 5-6
November, 18-19 November and Qtot without rain from 9-10 January. The results can be seen in Table
9.1.
0
0.05
0.1
0.15
0.2
Wat
er
Flo
w (
m3
/s)
designed flow 5-6 Nov5 Nov (Thursday)
6 Nov (Friday)
18-19 Nov (holiday)
9-10 Jan (weekend)
49
Table 9.1: Calculated total influent waterflow into the pondsystem for one day. *Partly using designed values.
** Rainfall peaks excluded (on the 5-6 November designed values are used).
Date
Qtot
With rain
(m3/day)
Qtot
Without rain
(m3/day)
Q max
(m3/s)
Q min
(m3/s)
Q Average
With rain
(m3/s)
Q Average
Without rain
(m3/s)
5-6 Nov
18-19 Nov
9-10 Jan
6346*
7527
No rainfall
5428**
6617**
5526
0.190
0.203
0.085
0.039
0.059
0.049
0.073*
0.087
No rainfall
0.058**
0.077**
0.064
Discussion
It is obvious that the rain influences the wastewater flow significantly. It can be seen on the velocity
measurements that soon after the rainfall occur, the flow speed increases rapidly. As the rainfall ends,
the flow tends to go down quickly. On the 18-19 of November, most students left UTM for holiday.
Approximately one half to two thirds of all students were not at the University during these
measurements. Despite this, both Qmax and Qmin were higher during the holiday than during the school
period. This could be explained by that the monsoon period started just after the measurements on the
5-6 November, but before the 18th of November. More or less every day received heavy rainfalls which
could have raised the groundwater level, thus creating more infiltration into the sewer system (see
chapter 10).
50
51
10. Infiltration High infiltration of groundwater into the sewer pipes is a common problem. Sewers may leak
wastewater out of the pipes or infiltrate groundwater into the sewer pipes, depending on the
surrounding soil water content. It has been observed during our measurements that there is a flow of
wastewater even in late night or early morning time even during vacation periods. The wastewater flow
drastically increases during rain events, and the wastewater strength is weak (diluted). This is clear
evidence of infiltration into the pipes. The extra infiltrating water is mostly unwanted since it demands
bigger tanks, or ponds, and to some degree will make the treatment less effective. However, one benefit
is that since the water is already diluted, it may be easier to meet the requirements for discharging the
water.
Three different approaches have been used to find the amounts of infiltrating water.
10.1 Calculations Method 1: Estimation of domestic water consumption If one assumes that the average contribution of wastewater per capita is 208 l/day, and the number of
people is known to be 10.500, the infiltration may be calculated (see Table 10.1).
Table 10.1: Estimation of infiltrating water, taking total water flow and assumed domestic water
consumption into consideration. **Assumes all inhabitants present. *Assumes only 1/3 of students and staff
present due to holiday.
Date Measured total
inlet water during
24h (m3)
Assumed
Domestic water
consumption (m3)
Infiltration (Total
– Domestic)
(m3)
Amount of
infiltration in
pipes
5 – 6 Nov 6346 2184** 4162 66%
18 – 19 Nov 7527 728* 6799 90%
9 – 10 Jan 5526 2184** 3342 60%
Method 2: Calculating base flow and rain peaks
The lowest flow measured during night times is considered as base flow. At this time (lowest flow
usually occurred between 04:00 a.m. and 06:00 a.m.) the water is considered to be purely infiltrated
water. The contribution from rainfall has been excluded according to Eq. 9.5 and Eq. 9.6 respectively.
52
Table 10.2: Calculations of infiltration by using baseflow and peak flow.
Date Measured total
inlet water during
24h (m3)
Total inlet water
without base-flow
and rain peak
(m3)
Infiltration
(Base
flow+Rainpeak)
(m3)
Amount of
infiltration in
pipes
5 – 6 Nov 6346 2028 4318 68%
18 – 19 Nov 7527 1559 5968 79%
9 – 10 Jan 5526 1275 4251 77%
Method 3: Calculating the COD dilution in wastewater
The load per capita is assumed to be 120 g COD per day (Kemira, 2004). With assumed fresh water
consumption it is possible to estimate the theoretical concentration of COD in the wastewater.
(eq. 10.1)
Since the concentration of COD in the measured samples is lower than the theoretical value, one
explanation may be infiltrating water. This model assumes that the infiltrating water is clean and does
not contribute with any pollution.
Table 10.3: Amount of infiltrating water through theoretical dilution of COD concentration.
Date Average
weighted COD in
inlet (mg/l)
Theoretical COD
(mg/l)
Amounts of
infiltration in pipes
5 – 6 Nov 122 577 79%
18 – 19 Nov 112 577 81%
9 – 10 Jan 113 577 80%
10.2 Discussion and conclusion The methods in chapter 10.1 show that infiltration and possible misconnections with storm water
channels contributes with 60-90% of the total incoming water to the WSP. Since the storm water
channels are built to receive all the storm water, no additional water should enter the sewage network.
If the infiltrated water could be reduced, the HRT in the WSP would increase, hence improving the
treatment efficiency. An unavoidable effect of reducing the infiltration is the increase in concentrations
of pollutants in the wastewater. These both aspects must be considered against each other.
53
11. Influent wastewater quality
11.1 Method The influent water was collected every two hours during the measurement periods. Samples were
withdrawn from the inlet chamber and immediately taken to the laboratory and stored in a refrigerator.
The first sampling started 5th November at 17:30 and continued every second hour until 14:15 the next
day. The second sampling started at 12:30 a.m. on the 18th of November and continued every second
hour until 10:00 a.m. the next day. The third sampling started at 12:00 a.m. on the 9th of January and
continued every second hour to 10:00 a.m. the 10th of January.
From all these samplings days COD, BOD5 and TSS were analyzed according to the methods described in
chapter 7. From the samples 5-6 November k-values were analyzed. Since the measurements during 5-6
November lacks flow data between 17:30 and 22:00, a designed imaginary flow was created and used in
order to get complete measuring data and to calculate weighted parameters (see Figure 9.5). For more
information about how the imaginary designed curve was created see Chapter 9.
11.2 Results and discussion
BOD5 and COD The inlet COD concentrations show a tendency to follow the typical pattern where the concentration
rises when people use water consuming facilities at most, usually in the morning and late afternoon.
When water flow increases due to rainfall, the COD values peaks and reaches levels much higher than
during normal flow. At late night or early morning (04:00-06:00) all three sampling periods reach their
lowest values (12 mg/l – 33 mg/l) (Figure 11.1). The pattern of COD concentrations is lower during the
holiday period, which is explained by the lower load due to absence of students in combination with
infiltration. According to Mara (2003), a COD concentration of less than 400 mg/l is considered as “weak
strength”. This water is therefore placed into this category.
From the first sampling period 5-6 November, BOD5 from all 12 measurements were analyzed. For all 12
measurements that period, COD was also analyzed. This was done in order to get a ratio between BOD5
and COD for the wastewater entering the pond. In Figure 11.2 below COD and BOD5-values have been
plotted.
54
Figure 11.1: The inlet COD during three 24-hours measurements.
Figure 11.2: BOD5 and COD values in inlet to the WSP during 5-6 November 2009.
George, I., Crop, P & Servais, P. (2002) Fecal coliform removal in wastewater treatment plants studied by
plate counts and enzymatic methods. Water Research 36 (10), 2607-2617
Hammer, Mark J. (1986) Water and Wastewater Technology. 2nd edition. Toronto, Canada: John Wiley &
Sons, Inc.
Ithnin, H.B (2007) Rate of water use in Malaysian homes. Department of Geography. Faculty of Arts and
Social Sciences. University of Malaya. Kuala Lumpur, Malaysia.
Jacangelo, Joseph J. & Trussell, R. Rhodes (2002) International Report Water and Wastewater Disinfection: Trends, Issues and Practices. Available at: http://74.125.155.132/scholar?q=cache:RN3HcKfLwDMJ:scholar.google.com/+disinfection+of+wastewater+in+Malaysia&hl=sv&as_sdt=2000&as_vis=1 [Accessed at: 15 April 2010]
Jansen, J. (n.d) Danish design practice for nitrogen removal. Published by: Water Quality Institute,
Hørsholm, Denmark.
Katimon, A. & Demun, A.S (2004) Water use trend at Universiti Teknologi Malaysia: Application of Arima
model. Jurnal Teknologi, 41(B) Dis. 2004: 47–56.
Available at: http://eprints.utm.my/1936/1/JURNALDIS%5BB4%5D41.pdf
[Accessed at: 25 February 2010]
Kemira (2003) About Water Treatment. Helsingborg, Sweden: Kemira Kemwater.
Liew, R. (2008) Country: Malaysian Water & Waste Water. U.S. Department of Commerce.
Malaysia Water Association (1994) Design Guideline for Water Supply System. Kuala Lumpur, Malaysia:
Percetakan Nasional Malaysia Berhad.
Mara, D. (2003) Domestic Wastewater Treatment in Developing Countries. London, U.K: Earthscan.
Masterton District Council (2009) Review of Pathogen Reduction In Waste Stabilisation Ponds.
Available at: http://resources.mstn.govt.nz/sewerageupgrade/nz1report.pdf
[Accessed at: 5 Mars 2010]
Maynard, H. E., Ouki, S. K. & Williams, S.C. (1999) Tertiary lagoons: a review of removal mechanisms and
performance. Water Research 33 (1), 1-13
McKinney, Ross. E (1968) Overloaded Oxidation Ponds: Two Case Studies. Water Pollution Control
Federation. Vol. 40, No. 1, pp. 49-56. Published by: Water Environment Federation.
Richmond, S. et al., 2007. Malaysia, Singapore & Brunei. 10th Edition. Hong Kong, China: Lonely Planet