WSP Costs: Case Study in Sana'a, Yemen
WSP Costs: Case Study in Sana'a, Yemen
WSP Costs: Case Study in Sana'a, Yemen
A rigorous economic analysis was undertaken for a hypothetical large WSP system for the
city of Sana'a in the Yemen (Arthur, 1983) (Table 1). The overall cost of each of four
alternative treatment systems (WSPs, aerated lagoons, oxidation ditches and biological
filters) is given as a net present value (NPV), which is a figure that combines construction
costs with future operational costs discounted to their present value.
This least-cost analysis found that WSP costs are dictated by the cost of land, the discount
rate used and the value of the land at the end of the project life (taken in this study as 25
years). WSPs were found to be the least-cost solution for land prices in the range
US$50,000−150,000 per hectare, depending on the value of the discount rate used
(5−15%). These land prices refer, of course, only to the conditions of Arthur's case study.
However, they do indicate that WSPs are very competitive even at high land costs.
Arthur's study highlights an extremely important and universal aspect of WSP economics,
which is that their main capital item (land) is recoverable. Furthermore, since land is an
appreciating capital item, it can appear as such in company accounts. Including the end-
of-project value of the land dramatically alters the net present values in favour of WSPs −
their NPV decreases by nearly 90 percent (Table 2). Consideration of the end-of-project
value of land used for WSPs is not just a hypothetical or academic exercise − for example,
it has been extremely profitable in California, where WSP land in the city of Concord
increased in real terms by US$270,000 (1975 dollars) per ha during the 20-year period
1955−1975 (Oswald, 1976).
Table 1. The 25-year costs in millions of 1983 US dollars at a 12% discount rate of
alternative wastewater treatment systems for a population of 250.000 at a design
temperature of 20oC and for an effluent concentration of ≤10.000 faecal coliforms/100 ml.
Annual
Capital Net Present
System Operating Income*
Costs Value
Costs
Waste stabilization ponds 5.68 0.21 0.73 5.16
Aerated lagoons 6.98 1.28 0.73 7.53
Oxidation ditch 4.80 1.49 0.43 5.86
* From the sale of final effluent for crop irrigation and fish culture.
Arthur's study was done in 1983 and did not include treatment technologies such as up-
flow anaerobic sludge blanket (UASB) reactors and constructed wetlands. Also, the
effluent quality used (≤104 faecal coliforms/100 ml) would now be better taken as ≤1.000
faecal coliforms/100 ml, the WHO (1989) guideline for unrestricted irrigation (see Section
1.5).
Despite these criticisms, the basic cost-comparison methodology used by Arthur remains
the best available and its continued use is highly recommended to derive transparently
honest cost comparisons between competing treatment options for any given location.
Costa and Medri (2002) give the following functions for the construction costs of
WSPs in south Brazil:
(a) Cost of land (taken as the twice the pond area to allow for embankments and
access − hence the factor 2 in the equation):
CLi = 2PLAi
where CLi is the cost of the land for pond number i (US$); PL is the local cost of land
(US$/m2); and Ai is the volume of pond i (m2).
The costs are calculated for each pond in turn and the total cost for the system is then
determined by summing the costs of the individual ponds.
WSPs produce effluents of high microbiological quality that permit them to be used for crop
irrigation and/or the cultivation of fish and aquatic vegetables. WSP effluent reuse for
these purposes is described in detail in Section 10 of the India Manual6.
Agricultural reuse
Crop irrigation is divided into two broad categories: restricted crop irrigation, meaning
irrigation of all crops except salads and vegetables eaten uncooked; and unrestricted
irrigation which includes those crops). The World Health Organization has different
guidelines for the microbiological quality of treated wastewaters used for these two
categories of irrigation. These guidelines were originally published in 1989 (WHO, 1989)
and they are currently under revision (see Blumenthal et al., 2000). The revised
guidelines, due to be published in 2005, will be as follows:
a) Restricted irrigation
≤105 E. coli per 100 ml, and
≤1 human intestinal nematode egg per litre, reduced to ≤0.1 egg per litre when
children under the age of 15 are exposed (by working or playing in wastewater-
irrigated fields).
b) Unrestricted irrigation
≤1000 E. coli per 100 ml, and
≤1 human intestinal nematode egg per litre, reduced to ≤0.1 egg per litre when
children under the age of 15 are exposed locally by their field-worker parents bringing
home food crops eaten uncooked.
Faecal coliforms may be substituted for E. coli. The intestinal nematodes are Ascaris
lumbricoides (the human roundworm), Trichuris trichiura (the human whipworm) and
Ancylostoma duodenale and Necator americanus (the human hookworms). Details of
these nematodes can be found in Sanitation and Disease7.
6
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/india/IPDMc10.pdf
7
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/watsan/sandis/sandis.html
These calculations are detailed in Section 4 of the India Manual. They are briefly described
in Box 3:
where R is the percentage egg removal in a single pond, and θ is the retention time in
the pond (days).
The equation is applied first to the anaerobic pond, then to the facultative pond, to
calculate the number of eggs per litre of the facultative pond effluent, as follows:
Efac = Erw (1 − ran) (1 − rfac)
where Efac and Erw are the number of eggs per litre of facultative pond effluent and the
raw wastewater, respectively, and r = R/100 with the subscripts 'an' and 'fac' referring
to the anaerobic and facultative ponds.
If Efac is >1 (or >0.1 if children under 15 are exposed), then the facultative pond
effluent requires further treatment in a maturation pond (usually a single 3-day
maturation pond is sufficient, but this must always be checked).
Removal of E. coli
where Nfac and Nrw are the number of E. coli per 100 ml of facultative pond effluent
and the raw wastewater, respectively; and kB(T) is the value of the first-order rate
constant for E. coli removal at T °C (day−1), given by:
kB(T) = 2.6 (1.19) T−20
If Nfac is >105 per 100 ml, then further treatment in one or more maturation ponds is
necessary.
The fish most commonly grown in wastewater-fed fishponds are carp and tilapia (see
Figure 10). Details of wastewater-fed fishponds in metropolitan Kolkata, India are given in
Section 2 of the India Manual8.
The revised WHO microbiological quality guidelines for aquacultural reuse will be:
≤104 E. coli per 100 ml of fishpond water (or aquatic vegetable pond water), and
zero detectable human trematode eggs per litre of treated wastewater.
Faecal coliforms may be substituted for E. coli. The human trematodes are Schistosoma
spp. (human blood flukes) Clonorchis sinensis (oriental liver fluke) and Fasciolopsis buski
(giant intestinal fluke). Details of these trematodes are given in Sanitation and Disease9.
Wastewater-fed fishponds are designed to receive facultative pond effluent on the basis of
a total nitrogen loading on the fishpond of 4 kg N/ha d. Checks are then made to
determine whether the E. coli count in the fishpond is ≤104 per 100 ml and whether the free
ammonia (NH3) concentration in the fishpond is ≤0.5 mg N/l (higher concentrations are
toxic to fish).
Figure 10. Harvesting carp from one of the wastewater-fed fishponds in Kolkata, India.
8
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/india/IPDMc2.pdf
9
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/watsan/sandis/sandis.html
Biogas (methane and carbon dioxide) can be profitably recovered from anaerobic ponds if
they are covered with a floating plastic membrane. This is done at the Western Treatment
Plant in Melbourne, Australia, where the three large WSP systems each receive a
wastewater flow of 120,000 m3/day. Half of the anaerobic section of the first pond is
covered (Figure 11) and the biogas collected is used to generate 6,000 kW of electricity 8–
16 hours per day, 365 days per year, which is worth about US$ 1 million per year (DeGarie
et al., 2000). The cover has three layers: a high-tensile UV-resistant geo-membrane for
biogas recovery at the top; a 12.5 mm polyfoam insulation and flotation layer in the middle;
welded to a base layer of high-density polyethylene. The cover measures 171 × 200 m (an
area of 3.4 ha).
So, biogas recovery from anaerobic ponds is obviously feasible at large WSP sites, but it
has also been done at smaller sites − for example, at Arad (population 22,000) in the
Negev desert, Israel (Shelef and Azov, 2000). Biogas recovery is especially feasible if
high-rate anaerobic ponds are used (see Peña, 2002).
Figure 11. Covered anaerobic pond at the Western Treatment Plant in Melbourne,
Australia.
WSP start-up
Before commissioning a WSP system, any vegetation growing in the empty ponds must be
removed. The facultative ponds and maturation ponds are commissioned before the
anaerobic ponds so as to avoid odour release when the anaerobic pond effluent
discharges into empty facultative ponds. The facultative ponds and maturation ponds
should ideally be filled initially with fresh surface water or groundwater to permit the
development of the required algal and heterotrophic bacterial populations. If freshwater is
not available, then the facultative pond can be filled with raw wastewater and allowed to
rest in batch mode for 3−4 weeks to allow the microbial populations to develop. Some
odour release may be expected during this period.
Once the facultative ponds and maturation ponds have been commissioned, the anaerobic
ponds are filled with raw wastewater and, if possible, inoculated with active biomass
(sludge seed) from another anaerobic bioreactor. The anaerobic ponds are then loaded
gradually up to their design load over a period of 2−4 weeks (the time depends on whether
the anaerobic pond was inoculated with an active sludge seed or not). The pH of the
anaerobic pond has to be maintained at around 7−7.5 during the start-up to allow for the
methanogenic archaeal populations to develop. If the pH falls below 7 during this period,
lime should be added to correct it.
Routine maintenance
Once the ponds have started functioning in steady state, routine maintenance is minimal
but essential for good operation. Full details are given in Section 6 of the India Manual10.
The main routine maintenance activities are:
As a rough guide one full-time operator is required at WSPs receiving wastewater flows up
to about 1,000 m3/d, two operators for wastewaters flows up to about 2,500 m3/d and pro
rata for higher flows (Arthur, 1983). A foreman/supervisor is required at sites treating more
10
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/india/IPDMc6.pdf
All WSP operators should receive adequate training so that they understand what they
have to do and how to do it correctly. If, for example, the pond operators have not been
told to remove scum from facultative ponds and maturation ponds, they will not know that it
should be removed. As a result, scum can cover a substantial part of the pond, algal
photosynthesis becomes impossible, and the pond turns anoxic (Figure 12).
Anaerobic ponds need to be desludged when they are around one-third full of sludge. This
occurs every 2−5 years, but it is operationally better to remove some sludge every year (as
a task to be done every February, for example, has a better chance of being done on time
than one which has to be done every few years). The sludge removed from anaerobic
ponds can be dewatered on sludge drying beds (Figure 13). Facultative ponds store any
sludge for their design life, which is a significant operational advantage.
When the travel time in the sewers is long (more than a day), the wastewater arriving at
the WSP site may be highly septic, and that can cause odour from the preliminary
treatment works.
The quality of the final effluent should be regularly determined at all WSP sites. For large
systems this should be done monthly, and for small systems serving just a few thousand
people at least every three months. Samples should be analysed for those parameters for
which the effluent standards have been set by the local environmental regulator (usually
BOD5, suspended solids, pH, possibly also ammonia; and, if the effluent is reused in
agriculture, E. coli or faecal coliforms and helminth eggs).
WSPs are essentially a simple technology, but some common problems tend to occur
regularly. These are generally the results of mistakes made during design, construction
and operation.
WSP design is often poor as too many designers do not understand the microbiological
processes occurring in them. They may think that, because ponds are "simple", it is not
strictly necessary to follow the correct design procedures. Either they do not know what
11
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/india/IPDMc7.pdf
The result of this regrettable ignorance is that many WSPs have been "designed" with
inappropriate BOD5 loadings. Inappropriately high loadings lead to odour and pond failure;
inappropriately low loadings, especially on anaerobic ponds, lead to under-performance
and overall costs are increased as the land area used is greater than necessary. Inlets
and outlets are often incorrectly located (Figures 14 and 15) − indeed pond hydrodynamics
is rarely considered. In some cases, there is no provision for preliminary treatment
(screening and grit removal), which adversely affects the ponds because of too much scum
and a higher rate of sludge accumulation. Pond failure or poor performance is also caused
by inadequate attention being given to geotechnical aspects during the physical design of
WSPs (Bernhard and Kirchgessner, 1987; Mantilla et al., 2002).
Outlet
Inlet
Figure 14. An example of poor inlet and outlet arrangements in an anaerobic pond. They
are located in adjacent corners of the pond rather than in diagonally opposite corners,
leading to hydraulic short-circuiting.
Lloyd et al. (2003) studied 14 WSP systems in Mexico: all produced poor quality effluents.
The reasons for under-performance included gross under-design, adverse environmental
conditions, a very high degree of hydraulic short-circuiting, and very poor operation and
maintenance. The main adverse environmental conditions were the large diurnal
variations in temperature in winter (from −4°C to +30°C) and very high wind speeds (peaks
of more than 8 m/s), both of which were major factors in the excessive hydraulic short-
circuiting. In one pond the dead space was 80 percent of the pond volume).
Another common problem is that, despite the loading rates being correctly selected by the
designer, based on reasonable values of the key design parameters, such as population,
per caput wastewater flow and BOD5 contribution, the actual influent loads are different.
The actual loading at the start may be much lower than the design value used, leading to
critical underloading in the anaerobic ponds; or it may increase at a greater rate than
predicted in the design, so leading to early critical overloading in the anaerobic and
facultative ponds.
Inappropriate changes made during construction can also adversely affect pond
performance. For example, it has been found in Colombia that contractors sometimes
decide to change the pond length-to-breadth ratio and/or increase the depth, or fail to
install the lining material correctly. They do this in the misguided belief that WSPs are just
holes in the ground and everything will be fine as long as the pond volumes are more or
less correct. They get away with it because their site work is not adequately supervised by
the designer or an independent civil/environmental engineer.
Figure 16. An example of a WSP system suffering from total maintenance neglect.
Training
Professional staff involved in WSP projects include design and construction engineers,
engineers responsible for operation of the WSPs once commissioned, and chemical and
microbiological laboratory managers and analysts. Financial analysts and sociologists
may also be involved at the pre-design and design stages. It is now becoming more
common for the local community to be involved at these stages; among other things, local
residents may need reassurance that WSPs will not cause odour problems, etc. All the
professionals require appropriate training. Ideally, this is best done at a university or
technical college, but that requires specialist lecturers who teach up-to-date curricula, and
this is not so common in many universities and colleges. Few local professional
institutions, such as national environmental engineering associations, take continuous
professional development (CPD) seriously, yet CPD is necessary to keep professionals up
to date in their specialism.
Sections 1.1−1.8 have dealt with conventional free-surface WSP systems. Some other
types of pond systems merit discussion:
WSTR were developed in Israel to store the effluent from a WSP system during the non-
irrigation period, so that the whole year's treated wastewater could be used in the irrigation
season, permitting a much greater area to be irrigated and more crops produced (Juanicó
and Shelef, 1991). WSTR are especially suitable in arid and semi-arid areas where the
value of treated wastewater for irrigation is high. Current practice is to treat the wastewater
in an anaerobic pond and discharge the effluent into a single 5−20 m deep WSTR with a
retention time equal to the length of the non-irrigation season (Figure 17a). This is perfectly
satisfactory if only restricted irrigation is practised.
If unrestricted irrigation is intended, three or four sequential batch-fed WSTR are required
to achieve the WHO guideline for unrestricted irrigation (Mara and Pearson, 1992) (Figure
17b). These sequential batch-fed WSTR are operated on a cycle of fill, rest and use, such
that the E. coli count in a reservoir at the end of its rest phase (i.e., just before it is used for
irrigation) is less than 1000 per 100 ml.
Hybrid WSPs−WSTR systems provide treated wastewater for both restricted and
unrestricted irrigation (Mara and Pearson, 1999) (Figure 17c).
Nightsoil ponds are not very common, as bucket latrines are increasingly being
replaced by more appropriate sanitation technologies. Nightsoil treatment ponds
(anaerobic ponds + facultative ponds) are designed in the normal way (Box 1) for a
BOD5 contribution of about 20 g/person day. Suitable arrangements have to be made
for the nightsoil tankers to discharge their loads into the anaerobic ponds. Tanker
wash water is also discharged into the anaerobic ponds.
Lessons learnt
Experience from around the world has shown that WSPs are very often the most cost-
effective wastewater treatment method. However, both the process design and the
physical design of WSPs have to be carried out very carefully by competent design
engineers since WSPs are more than just holes in the ground.
Effluents from WSPs can achieve stringent discharge standards that make them highly
suitable for reuse in agriculture and/or aquaculture. This puts WSPs at the forefront of
wastewater reclamation technologies.
Despite more than 50 years of continued research into different aspects of WSPs, there is
still much work to do on topics such as hydrodynamic improvement, ecological modelling
Research perspectives
Recent research has shown that it should be possible to improve the performance of all
types of WSPs (i.e. anaerobic ponds, facultative ponds and maturation ponds). For
example, the recent development of high-rate anaerobic ponds has shown that it is
possible to reduce the retention time to 12 hours, yet still achieve average BOD5 removals
of 70 percent at 25 ºC (Peña, 2002). Future research should investigate high-rate
facultative ponds and maturation ponds to treat the effluents from both high-rate anaerobic
ponds and UASBs.
The following are case studies the reader may find useful since they are from different
regions in the world where WSPs have been successfully implemented.
12
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/med/emedwsp.pdf
13
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/med/wspna.pdf
14
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/med/wspmed.pdf
15
http://www.leeds.ac.uk/civil/ceri/water/tphe/publicat/pdm/india/IPDMc2.pdf