ABSTRACT
The ventilated improved pit (VIP) latrine is a cheaper and convenient alternative to water-borne sanitation for many households in developing countries. To mimic the water closet toilet, some users have introduced prefabricated ceramic seats with covers, which are known to disrupt the odour control mechanism of the latrine. The objective of this study was to quantitatively establish the extent to which the use of seats with covers affects the ventilation rate in the VIP latrine and to explore whether an innovative seat with a partially-screened cover could significantly minimise the effect of seat covering. The ventilation rates in experimental latrines with various modifications of superstructure and user interface designs were monitored with an air flow meter. The study revealed that seat covering could lead to >87% reduction in the ventilation rate and allow the generation of <50% of the minimum ventilation rate required to achieve odour-free conditions in an otherwise standard VIP latrine. The study also established that lining part of the seat cover with an insect screen could increase the ventilation rate by more than three times and could attain the minimum ventilation rate when combined with the conventionally recommended superstructure design.
HIGHLIGHTS
Effects of fully-covered and novel partially-screened ceramic seats on ventilation rate in VIP latrines were assessed.
Fully-covered seats reduce ventilation rates in VIP latrines by 87%.
Partially-screened ceramic seat covers would control flies and odour in VIP latrines.
Using VIP latrine with windows in multiple sides of the superstructure and 100-mm vent pipe will lead to odour nuisance.
INTRODUCTION
Poor sanitation, hygiene, and access to safe drinking water are estimated to be the cause of 1.4 million deaths mostly in low- and middle-income countries (WHO/UNICEF JMP 2023). In such countries, the low coverage of sewerage systems places the burden of accessing improved sanitation facilities on individual households who have to resort to one form of non-sewered sanitation system or another. Most of these households tend to depend on dry sanitation systems because they are relatively cheaper as compared to wet sanitation technologies (Aburto-Medina et al. 2020; Scott 2020). Besides, some households, especially those that live in the outskirts of cities and peri-urban communities, who could otherwise afford wet sanitation technologies are constrained from adopting such technologies by some municipal-level infrastructural constraints. Such constraints include lack of access to regular piped water supply for the convenient operation of wet sanitation technologies and the absence of motorable access routes to individual houses by cesspit emptying trucks for the desludging of septic tanks when they are full (Parkinson et al. 2008; Adugna 2023).
For such households, the ventilated improved pit (VIP) latrine offers a more feasible and convenient alternative to water-borne sanitation. When properly constructed and operated, it is said to offer its users most of the health benefits of water-borne sanitation at a relatively cheaper cost (Kalbermatten et al. 1980; Ryan & Mara 1983b). In its conventional design, it has a mode of operation that employs natural ventilation mechanisms to draw out malodorous air from the pit into the atmosphere and, thus, control the nuisance of odour that is associated with the traditional pit latrine.
An illustration of the mode of operation of a VIP latrine (Source: Harvey et al. (2002)).
An illustration of the mode of operation of a VIP latrine (Source: Harvey et al. (2002)).
Some emerging VIP latrine user interface designs (Source: Author's field work).
The use of the seats with covers is likely to lead to odour nuisance in those toilets, which is a well-known barrier to consistent latrine usage and may encourage some prospective users to abandon the toilets and resort to open defaecation (Obeng et al. 2015). Even though seat covering in VIP latrines is generally known to reduce the ventilation rate through the vent pipe, there is no data seen in scientific literature that demonstrates the extent or magnitude of its effect on the ventilation rate as compared to an uncovered seat or squat hole. Such data are needed to make a case for producers of the ceramic pots with covers to rethink the design of the prefabricated seat covers meant for use in VIP latrines. Similarly, such data are needed to educate VIP latrine owners and prospective ones on the need to avoid seat covering.
Furthermore, there is the need to explore potential user interface innovations that would address the needs of prospective owners who are keen on aesthetics. A potential innovation could be the redesigning of the seat cover to replace a portion of the cover with an insect screen so that it would allow entry of air into the pit. Nevertheless, such a design would still compromise the movement of air into the pit due to head losses across the screen. Hence its potential in allowing adequate ventilation in the vent pipe needs to be investigated.
To address the above gaps, the objective of this paper is to assess the extent to which the use of ceramic seats with covers affects the ventilation rate in the VIP latrine as compared to the conventional guideline of leaving the seat or squat hole uncovered, and to explore the potential of a partially-screened seat cover in minimising the effect of seat covering to allow adequate ventilation in the VIP latrine.
METHODOLOGY
The study location
The study was conducted on the campus of the Cape Coast Technical University, which is located within the northern part of the Cape Coast Metropolitan Area (CCMA). The CCMA is the administrative capital of the Central Region of Ghana. The city is located within latitude 5°07′–5°20′N and longitude 1°11′–1°41′W (CCMA 2014). Cape Coast is a coastal city, bounded at the south by the Gulf of Guinea and covers a land area of 122 km2 (Ghana Districts 2006). According to Ghana's population and housing census of 2021, the population of Cape Coast is 189,925 persons. The average daily temperature and wind speed are, respectively, 29 °C and 1.68 m/s (Weatherandclimate.com 2023). The rainfall pattern is bi-modal with the major season spanning from April to July and the minor season from September to November (Ghana Meteorological Agency 2023). The total annual rainfall ranges between 750 and 1,000 mm (CCMA 2014).
According to the Ghana Statistical Service, up to 30% of households in Cape Coast have access to at least basic sanitation (Ghana Statistical Service 2022). The current census reports do not contain data on the various sanitation technologies available in Cape Coast. However, the data are aggregated for the various regions. From the regional level data, 1.7% of households in Central region practice open defaecation, 23.0% use public toilets, 0.05% are connected to sewer systems and the remaining households depend on on-site sanitation technologies for excreta collection and disposal. The main on-site sanitation technologies are water closet or pour flush connected to septic tanks (22.7%), pit latrines (20.3%), and VIP/KVIP latrines (19.3%). The number of households using VIP/KVIP latrine has seen a significant growth from 11.8% in 2010 (Ghana Statistical Service 2013) to 19.3% in 2021 (Ghana Statistical Service 2022). In terms of actual numbers, two and a half times more people now use VIP/KVIP than in 2010. Currently, about 20% of all households and 27.5% of households with toilet facilities in urban areas of the Central Region use private (individual or shared) VIP/KVIP latrines (Ghana Statistical Service 2022).
Research design
The study was designed as an experimental comparative study to assess how the emerging use of industrially-produced ceramic seats with covers affects the ventilation rate in a VIP latrine as compared to an innovative design in which part of the seat is replaced with an insect screen which allows passage of air from the latrine cubicle into the pit while restraining flies from leaving the pit via the seat. These user interface designs were also compared to the conventionally recommended option of leaving the seat or squat hole entirely uncovered. The four user interface designs were assessed in two types of VIP latrine superstructure designs representing the standard design based on conventional guidelines and a popular design commonly encountered in Ghana, which are described in detail in the following section.
Description of experimental setup
The experimental setup involved two types of VIP latrine superstructure design modifications in each of which a set of four user interface designs were tested. The two design modifications were made in the same base structure and monitored one after the other. Hence, aside from the specific modifications in the superstructure design described below, other design parameters such as those relating to the pit were identical. For each superstructure design, the four user interfaces were tested one after the other. The base experimental latrine was built on the campus of the Cape Coast Technical University in Cape Coast.
Description of experimental latrine superstructures
The two VIP superstructure designs used for the study and the basis for their selection are described below:
Superstructure type 1 (ST1): a near-standard VIP latrine superstructure built in conformity with conventional design guidelines except that the window was fitted with an insect screen. It was fitted with a 150-mm vent pipe and had a window provided on only the windward side. The installation of an insect screen in the window is discouraged by conventional design guidelines. However, this was done due to its increasing popularity among latrine owners and artisans as observed in Ghana for the practical purpose it serves by preventing reptiles and rodents from entering the toilet (Obeng et al. 2015, 2019a). Besides, it has been established that, when used in a VIP latrine fitted with a 150-mm vent pipe, its effect can be conveniently overlooked (Obeng et al. 2019c).
Superstructure type 2 (ST2): a worst-case scenario superstructure design fitted with a 100-mm vent pipe and windows on each side and fitted with insect screens. Nearly all VIP latrines (93.2%) surveyed in Ghana were built with 100-mm vent pipes due to the high cost of the 150 mm PVC pipe; 25% had windows provided on multiple sides of the superstructure (Obeng et al. 2024a). This paper sought to investigate how the four user interface designs would perform in this popular but compromised superstructure design, which is expected to reduce the ventilation rate as a result of the use of a smaller vent pipe and escape of air through the extra windows (Ryan & Mara 1983b; Mara 1984).
Plan, elevation, and section of experimental VIP latrines. (a) Plan, (b) side elevation, and (c) section.
Plan, elevation, and section of experimental VIP latrines. (a) Plan, (b) side elevation, and (c) section.
Description of the user interface designs
Pictures of the four user interface designs assessed (Source: Author's field work). (a) USH, (b) UCS, (c) PCS, and (d) FCS.
Pictures of the four user interface designs assessed (Source: Author's field work). (a) USH, (b) UCS, (c) PCS, and (d) FCS.
Monitoring of experimental variables/parameters
The experimental procedure for measuring the ventilation rate followed the procedure outlined by Ryan & Mara (1983a) and used by Obeng et al. (2019b, 2019c, 2022). For each experimental setup, data were logged at a minute interval for 10 continuous minutes and repeated at hourly intervals for 12 h a day (6:00 am–6:00 pm). Each setup was monitored for a single day. However, the taking of 10 data points within each hour across a 12-h interval ensured adequate data over a wide variation of weather conditions. Nevertheless, the inability of the study to repeat the setups for multiple days is recognised as a weakness of this study.
External wind speed, temperature and relative humidity were measured using the Logia® 5-in-1 Wireless Weather Station (Model: LOWSB510PB, manufactured by Logia Weather Stations, Incorporated, New Jersey, USA). The weather station was mounted at a height nearest to or equal to the height of the top of the vent pipe as recommended by Ryan & Mara (1983a) and used by Obeng et al. (2019b, 2019c, 2022). Data were logged at the device's minimum logging interval of 5 min.
Data analysis
Data organisation and statistical analysis
The data were inputted into Microsoft Excel 2016 and imported into SPSS software version 23 for analysis. All data were tested for normality with the Shapiro–Wilk test at 95% significance level and were found to be non-normally distributed. Non-parametric statistical methods were used in the data analysis. The Kruskal–Wallis test was used to test for variance among groups while multiple comparisons of all design setups were done using the Dunn's post hoc test. As non-parametric analyses are based on the medians rather than means, the medians and interquartile ranges of the data have been presented. Nevertheless, the means and standard deviations have also been presented to facilitate easy comparison of the results with those of other studies and technical guidelines that are normally based on means of the various variables.
Dealing with the effect of variation in the elements of weather
A key challenge in studies of this nature is the difficulty of ensuring that the ventilation rates in the different experimental setups were measured under similar weather conditions. A typical solution is to site the different setups at the same location and monitor them simultaneously. However, where many different design modifications are intended to be compared as in this study, such an approach becomes prohibitively expensive due to the need to build a toilet for each design modification and ensure availability of equipment and personnel for monitoring all setups simultaneously. Besides, there is the technical challenge of ensuring that none of the toilets would obstruct air flow to others at any time as the direction of the local winds keep changing, and that the movement of the winds into the different toilets would be equal. The availability of technically similar equipment with the same levels of sensibility could also be another challenge.
The above model was reported to have an adjusted coefficient of multiple determination, R2, of 91.44%, and a predicted R2 of 91.22%. This means it would explain 91.22% of changes in the ventilation rate when a new set of data are used to predict an unknown ventilation rate and could be expected to minimise the effect of weather variability in such studies. In relation to the key elements of weather, the model can be interpreted that (if all other factors are held constant):
a unit (1 m/s) change in wind speed would lead to 28.7% change in the ventilation rate;
a unit (1%) change in relative humidity would lead to 1% change in the ventilation rate;
a unit (1 °C) change in external temperature will lead to 2.8% change in the ventilation rate
The current study used the factors established in this model to adjust or standardise the nominal ventilation rates recorded in each of the experimental setups. This was done by adopting the overall averages of the weather conditions recorded at the study site as an imaginary common weather condition under which the setups were compared. The difference between the hourly averages of the elements of weather during the monitoring of each setup and the overall averages were used to compute the effect of changes in each of the elements of weather and, subsequently, the standardised ventilation rates. Thus, the standardised ventilation rates are those that are expected to be measured in the various setups if they were monitored under the average wind speed, relative humidity and external wind speed. The averages of these elements of weather recorded at the study site over the period of the data collection were 2.07 m/s, 78.18%, and 28.94 °C.
RESULTS AND DISCUSSION
Weather conditions at the study location
Table 1 presents data on the key elements of weather (temperature, relative humidity, and wind speed) during the monitoring period of each of the experimental setups and the overall averages for the period of the study. Because the data were found to be non-normally distributed, the median and interquartile ranges of each weather element have also been presented alongside the means and standard deviations. Also presented in the table are the results of the analysis of variance for each element of weather among the four user interfaces using the Kruskal–Wallis test. The data have been presented separately for the two types of superstructure designs.
Analysis of prevailing weather conditions during the monitoring of the various experimental setups
Experimental set up description . | External temperature (°C) . | Relative humidity (%) . | External wind speed (m/s) . | |||||||
---|---|---|---|---|---|---|---|---|---|---|
Superstructure design . | User interface design . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . |
Type 1: Single window with insect screen and 150-mm vent pipe | Uncovered squat hole | 28.98 (3.41) | 30.80 (7.58) | 119.233 (0.000)** | 76.75 (12.32) | 72. 01 (26.98) | 94.082 (0.000)** | 1.72 (1.29) | 2.33 (2.79) | 75.060 (0.000)** |
Uncovered ceramic seat | 28.74 (3.46) | 30.33 (6.47) | 74.25 (7.07) | 72.01 (13.47) | 2.54 (1.81) | 3.49 (3.74) | ||||
Partially-screened ceramic seat | 29.71 (2.71) | 30.31 (6.58) | 77.57 (10.60) | 73.00 (25.02) | 2.08 (1.18) | 2.39 (2.30) | ||||
Fully-covered ceramic seat | 28.53 (2.85) | 29.60 (5.90) | 77.54 (10.04) | 77.80 (23.99) | 1.78 (1.32) | 1.29 (2.20) | ||||
Type 2: Multi-window with insect screen and 100-mm vent pipe | Uncovered squat hole | 29.57 (1.52) | 29.97 (1.97) | 29.695 (0.000)** | 81.32 (5.99) | 80.67 (9.33) | 4.173 (0.243) | 2.59 (1.01) | 2.69 (1.73) | 31.351 (0.000)** |
Uncovered ceramic seat | 28.04 (1.61) | 27.68 (1.98) | 80.69 (7.40) | 83.00 (4.42) | 1.75 (0.83) | 1.67 (1.60) | ||||
Partially-screened ceramic seat | 30.28 (2.90) | 31.48 (4.37) | 72.72 (9.31) | 71.00 (14.42) | 2.50 (1.33) | 2.64 (1.93) | ||||
Fully-covered ceramic seat | 27.66 (2.33) | 28.73 (4.47) | 84.65 (7.30) | 82.00 (15.67) | 1.59 (0.98) | 1.82 (1.82) | ||||
Overall | 28.94 (2.80) | 29.68 (4.06) | 78.18 (9.66) | 78.67 (15.00) | 2.07 (1.31) | 2.34 (2.18) |
Experimental set up description . | External temperature (°C) . | Relative humidity (%) . | External wind speed (m/s) . | |||||||
---|---|---|---|---|---|---|---|---|---|---|
Superstructure design . | User interface design . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . |
Type 1: Single window with insect screen and 150-mm vent pipe | Uncovered squat hole | 28.98 (3.41) | 30.80 (7.58) | 119.233 (0.000)** | 76.75 (12.32) | 72. 01 (26.98) | 94.082 (0.000)** | 1.72 (1.29) | 2.33 (2.79) | 75.060 (0.000)** |
Uncovered ceramic seat | 28.74 (3.46) | 30.33 (6.47) | 74.25 (7.07) | 72.01 (13.47) | 2.54 (1.81) | 3.49 (3.74) | ||||
Partially-screened ceramic seat | 29.71 (2.71) | 30.31 (6.58) | 77.57 (10.60) | 73.00 (25.02) | 2.08 (1.18) | 2.39 (2.30) | ||||
Fully-covered ceramic seat | 28.53 (2.85) | 29.60 (5.90) | 77.54 (10.04) | 77.80 (23.99) | 1.78 (1.32) | 1.29 (2.20) | ||||
Type 2: Multi-window with insect screen and 100-mm vent pipe | Uncovered squat hole | 29.57 (1.52) | 29.97 (1.97) | 29.695 (0.000)** | 81.32 (5.99) | 80.67 (9.33) | 4.173 (0.243) | 2.59 (1.01) | 2.69 (1.73) | 31.351 (0.000)** |
Uncovered ceramic seat | 28.04 (1.61) | 27.68 (1.98) | 80.69 (7.40) | 83.00 (4.42) | 1.75 (0.83) | 1.67 (1.60) | ||||
Partially-screened ceramic seat | 30.28 (2.90) | 31.48 (4.37) | 72.72 (9.31) | 71.00 (14.42) | 2.50 (1.33) | 2.64 (1.93) | ||||
Fully-covered ceramic seat | 27.66 (2.33) | 28.73 (4.47) | 84.65 (7.30) | 82.00 (15.67) | 1.59 (0.98) | 1.82 (1.82) | ||||
Overall | 28.94 (2.80) | 29.68 (4.06) | 78.18 (9.66) | 78.67 (15.00) | 2.07 (1.31) | 2.34 (2.18) |
**Significant at a 1% confidence level.
SD, standard deviation; IQR, interquartile range.
It can be seen from the table that the weather is tropical, with overall averages of temperature, relative humidity and wind speed during the period of the study being 28.94 °C, 78.18%, and 2.07 m/s, respectively. The temperature and wind speed are comparable to those under which Obeng et al. (2019b, c) conducted similar studies in Prampram, Ghana, where the averages of the two elements of weather were, respectively, 30.40 °C and 2.10 m/s. With respect to relative humidity, Obeng et al. (2019b, c) recorded a lower figure (63.5%) during their study in Prampram.
With the exception of relative humidity under ST2, the variance of each element of weather across the monitoring periods of the four user interface designs in both types of superstructures were statistically significant. This implies that, without standardising the results to a common set of weather conditions, the differences of the ventilation rates in the four user interface designs could be influenced by the effect of weather rather than the effect of the user interface design. For instance, the wind speed, which is known to be the most influential element of weather (Obeng et al. 2019b) had an average of 2.54 m/s during the monitoring of the UCS under superstructure 1 as compared to 1.72 m/s in the USH. With a unit (1 m/s) change in wind speed having been found to lead to 28.7% change in the ventilation rate, if all other factors are held constant, the difference of 0.82 m/s could account for as much as 23.5% of the difference in the nominal ventilation rates recorded in the two setups if all other factors are held constant. This explains why the ventilation rates were standardised or adjusted to the overall average weather conditions using the factors established by Obeng et al. (2019b) in order to eliminate or minimise the effect of weather variations in the comparison of the ventilation rates in the various user interface designs.
Comparison of ventilation rates among the user interface designs
As seen in Table 2 the analyses of variance of both nominal and standardised ventilation rates showed highly significant differences among the levels recorded among the different user interface designs. This was the case for the user interface designs tested in both types of superstructure designs, with Kruskal–Wallis, H > 270 and p = 0.000 in all cases of comparisons. Post hoc pairwise comparisons using the Dunn's test shown in Table 3 reveal that only the standardised ventilation rates in the two uncovered designs are similar. All other pairs of user interface designs compared in both types of superstructures are significantly different (p = 0.000).
Comparison of ventilation rates measured in the various experimental setups
Experimental set up description . | Nominal ventilation rate (m3/h) . | Standardised ventilation rate (m3/h) . | |||||
---|---|---|---|---|---|---|---|
Superstructure design . | User interface design . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . |
Type 1: Single window with insect screen and 150-mm vent pipe | Uncovered squat hole | 52.52 (28.55) | 57.26 (52.17) | 272.567 (0.000)** | 50.29 (22.73) | 51.58 (35.57) | 311.445 (0.000)** |
Uncovered ceramic seat | 65.70 (42.86) | 72.21 (79.68) | 40.26 (26.09) | 35.26 (42.17) | |||
Partially-screened ceramic seat | 13.60 (6.64) | 13.36 (7.95) | 12.41 (6.70) | 10.47 (7.02) | |||
Fully-covered ceramic seat | 4.90 (5.08) | 4.45 (8.27) | 4.11 (4.87) | 3.49 (6.80) | |||
Type 2: Multi-window with insect screen and 100-mm vent pipe | Uncovered squat hole | 23.47 (10.13) | 23.47 (13.29) | 345.960 (0.000)** | 16.30 (5.72) | 16.19 (8.03) | 326.523 (0.000)** |
Uncovered ceramic seat | 14.4 (6.34) | 12.72 (8.20) | 14.21 (5.11) | 13.19 (6.35) | |||
Partially-screened ceramic seat | 8.59 (5.66) | 6.76 (5.94) | 7.69 (5.91) | 6.94 (5.24) | |||
Fully-covered ceramic seat | 2.05 (0.75) | 1.98 (1.13) | 2.14 (0.58) | 2.11 (0.75) |
Experimental set up description . | Nominal ventilation rate (m3/h) . | Standardised ventilation rate (m3/h) . | |||||
---|---|---|---|---|---|---|---|
Superstructure design . | User interface design . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . | Mean (SD) . | Median (IQR) . | Kruskal–Wallis, H (p-value) . |
Type 1: Single window with insect screen and 150-mm vent pipe | Uncovered squat hole | 52.52 (28.55) | 57.26 (52.17) | 272.567 (0.000)** | 50.29 (22.73) | 51.58 (35.57) | 311.445 (0.000)** |
Uncovered ceramic seat | 65.70 (42.86) | 72.21 (79.68) | 40.26 (26.09) | 35.26 (42.17) | |||
Partially-screened ceramic seat | 13.60 (6.64) | 13.36 (7.95) | 12.41 (6.70) | 10.47 (7.02) | |||
Fully-covered ceramic seat | 4.90 (5.08) | 4.45 (8.27) | 4.11 (4.87) | 3.49 (6.80) | |||
Type 2: Multi-window with insect screen and 100-mm vent pipe | Uncovered squat hole | 23.47 (10.13) | 23.47 (13.29) | 345.960 (0.000)** | 16.30 (5.72) | 16.19 (8.03) | 326.523 (0.000)** |
Uncovered ceramic seat | 14.4 (6.34) | 12.72 (8.20) | 14.21 (5.11) | 13.19 (6.35) | |||
Partially-screened ceramic seat | 8.59 (5.66) | 6.76 (5.94) | 7.69 (5.91) | 6.94 (5.24) | |||
Fully-covered ceramic seat | 2.05 (0.75) | 1.98 (1.13) | 2.14 (0.58) | 2.11 (0.75) |
**Significant at a 1% confidence level.
SD, standard deviation; IQR, interquartile range.
Post hoc comparisons of standardised ventilation rates among the user interface designs
Superstructure description . | Pair of user interfaces compared . | Dunn's Z-score . | Adjusted p-value . |
---|---|---|---|
Type 1 (Single window with insect screen and 150-mm vent pipe | Uncovered squat hole seat versus uncovered ceramic seat | 44.158 | 0.06 |
Uncovered squat hole versus partially-screened ceramic seat | 174.866 | 0.000** | |
Uncovered squat hole versus fully-covered ceramic seat | 277.515 | 0.000** | |
Uncovered ceramic seat versus partially-screened ceramic seat | 130.707 | 0.000** | |
Uncovered ceramic seat versus fully-covered ceramic seat | 233.357 | 0.000** | |
Partially-screened ceramic seat versus fully-covered ceramic seat | 102.650 | 0.000** | |
Type 2 (Multi-window with insect screen 100-mm vent pipe) | Uncovered squat hole seat versus uncovered ceramic seat | 31.193 | 0.485 |
Uncovered squat hole versus partially-screened ceramic seat | 161.177 | 0.000** | |
Uncovered squat hole versus fully-covered ceramic seat | 287.620 | 0.000** | |
Uncovered ceramic seat versus partially-screened ceramic seat | 129.987 | 0.000** | |
Uncovered ceramic seat versus fully-covered ceramic seat | 256.427 | 0.000** | |
Partially-screened ceramic seat versus fully-covered ceramic seat | 126.443 | 0.000** |
Superstructure description . | Pair of user interfaces compared . | Dunn's Z-score . | Adjusted p-value . |
---|---|---|---|
Type 1 (Single window with insect screen and 150-mm vent pipe | Uncovered squat hole seat versus uncovered ceramic seat | 44.158 | 0.06 |
Uncovered squat hole versus partially-screened ceramic seat | 174.866 | 0.000** | |
Uncovered squat hole versus fully-covered ceramic seat | 277.515 | 0.000** | |
Uncovered ceramic seat versus partially-screened ceramic seat | 130.707 | 0.000** | |
Uncovered ceramic seat versus fully-covered ceramic seat | 233.357 | 0.000** | |
Partially-screened ceramic seat versus fully-covered ceramic seat | 102.650 | 0.000** | |
Type 2 (Multi-window with insect screen 100-mm vent pipe) | Uncovered squat hole seat versus uncovered ceramic seat | 31.193 | 0.485 |
Uncovered squat hole versus partially-screened ceramic seat | 161.177 | 0.000** | |
Uncovered squat hole versus fully-covered ceramic seat | 287.620 | 0.000** | |
Uncovered ceramic seat versus partially-screened ceramic seat | 129.987 | 0.000** | |
Uncovered ceramic seat versus fully-covered ceramic seat | 256.427 | 0.000** | |
Partially-screened ceramic seat versus fully-covered ceramic seat | 126.443 | 0.000** |
**Significant at a 1% confidence level.
As expected, Table 2 shows that the designs with squat hole or ceramic seat without cover showed remarkably higher levels of ventilation than the designs with the partially-screened and fully-covered seats for both the nominal and standardised ventilation rates. This confirms the importance of easy access of external air into the pit to displace warm, malodourous air from the pit as emphasised by pioneering studies into the design of the VIP latrine such as Mara (1984). The role of the easy movement of air into the pit also explains why the ventilation rates in the design with partially-screened seat cover led to higher levels of ventilation rates as compared to that with fully-covered seats in both types of superstructures and for both nominal and standardised ventilation rates. The sharp drop in the ventilation rate in the ceramic seat with partially-screened cover as compared to the UCS results from the loss of air pressure across the screen as the air moves from the superstructure into the pit (Mara 1984).
It is noted that the trend of the standardised ventilation rates among the user interface designs is not much different from that of the nominal rates. This suggests that the effect of partial or full covering of the seat on the ventilation rate could be stronger than that of weather variability within the range observed at the study site. Comparing the mean standardised ventilation rates in the design with FCS and that with UCS, it can be deduced that the full covering of the ceramic seat led to 92% reduction in the ventilation rate for ST1 and 87% in ST2. A similar comparison between the partially-covered and UCS shows a reduction of 75 and 53% in superstructure types 1 and 2, respectively. On the other hand, an innovation to introduce partially-screened seat covers to replace full covering of the ceramic seats would lead to 3 and 3.6 times increase in the ventilation rates for superstructure types 1 and 2, respectively.
Comparison of the ventilation rates with conventional benchmarks
Comparison of nominal and standardised ventilation rates with conventional benchmarks.
Comparison of nominal and standardised ventilation rates with conventional benchmarks.
In ST1, the ceramic seat with a partially-screened cover was able to attain the minimum ventilation rate of 10 m3/h at which the VIP latrine is found to attain odourless conditions (Ryan & Mara 1983b; Mara 1984). However, it may not be able to guarantee adequate factor of safety against odour development when some other unfavourable design choices such as provision of windows in more than one side of the superstructure are introduced. On the other hand, the design with a FCS could only attain less than half (4.11 m3/h) of the minimum ventilation rate or less than a quarter of the recommended rate. This means that latrines with a covered ceramic seat are very likely to generate odour in the cubicle when a prospective user opens the cover for use. Unfortunately, no study was identified that reported the ventilation rate in a VIP latrine with a partially-covered and fully-covered seat or squat hole to which the results of this study could be compared.
In ST2 (frequently encountered in a survey conducted in Ghana), which represents a potentially worst-case scenario in which a 100-mm vent pipe is used and windows provided in more than one side of the superstructure, the USH and ceramic seat attained the minimum but failed to reach the recommended ventilation rate. As in superstructure 1, the ventilation rates measured in this study are comparable to those obtained by Obeng et al. (2019c), which reported a mean nominal ventilation rate of 17.63 m3/h for a latrine with an USH as compared to 23.47 and 14.40 m3/h, respectively, recorded in the USH and UCS, respectively, in this study. Even though the USH in this study recorded a relatively higher mean nominal ventilation rate, that setup had the highest prevailing wind speed in this study. Upon standardisation to the average weather conditions, the setup recorded a mean standardised ventilation rate of 16.30 m3/h. Thus, in principle, both studies found the USH or ceramic seat used in a superstructure with multiple windows fitted with insect screens and a 100-mm vent pipe to deliver the minimum ventilation rate of 10 m3/h but fails to attain the recommended rate of 20 m3/h at an average local wind speed of 2.1 m/s.
In this type of superstructure, the ceramic seat with a partially-screened cover failed to attain the minimum ventilation rate. It recorded 7.69 m3/h, representing only about three-quarters of the minimum required to attain odour-free conditions. This means that the innovation to replace fully-covered ceramic seats with those having partially-screened covers would only be able to control odour if that user interface design is used in a standard superstructure design that is fitted with a 150-mm vent pipe and a window provided in only one side of the superstructure as conventionally recommended. Obeng et al. (2019a) found the cost of the 150-mm vent pipe to be three times that of the 100 mm, with the actual difference in price (the equivalent of 10.66 United States dollars) being 5.4 times Ghana's daily minimum wage at the time. Considering the high cost of the 150-mm vent pipe, further research is required to explore other design modifications to increase the ventilation rate in a superstructure with a 100-mm vent pipe and a ceramic seat with a partially-screen cover to, at least, the minimum rate of 10 m3/h. First, a screen with bigger apertures than the one used in this study but still capable of restraining flies from leaving the pit should be tested to see whether the potential decrease in the head loss across the screen could boost the ventilation rate above the minimum threshold. Secondly, the study should be repeated with the additional windows in superstructure 1 sealed to leave only one window in the windward side to assess whether the anticipated increase in air pressure in the cubicle (Mara 1984) could increase the ventilation rate beyond the minimum threshold. With Obeng et al. (2019b) establishing that the provision of windows in all sides of the superstructure as done in superstructure 1 leading to a 32% drop in the ventilation rate, if all other factors are held constant, this modification could raise the ventilation rate recorded in this setup from 7.69 to 10.2 m3/h.
Spectacularly, the use of the FCS in ST2 could attain only about 20% of the minimum ventilation rate (2.14 m3/h). Thus, the practice of installing a ceramic seat with a full plastic cover in VIP latrines with 100-mm vent pipes and windows or other openings provided in multiple sides of the superstructure would adversely compromise the odour control mechanism of the latrine and potentially make it end up nearly functioning as an ordinary simple pit latrine. In Ghana where nearly all VIP latrines that were surveyed (93.2%) were found to be fitted with 100-mm vent pipe (Obeng et al. 2024a), there is the need for collaboration among relevant stakeholders to guide the use of prefabricated ceramic seats in VIP latrines. It is recommended that academic and relevant state institutions engage the producers and importers of ceramic seats with the findings of scientific studies such as this one and additional ones as recommended above to innovatively introduce partially-screened seat covers that would satisfy the aesthetic requirements of prospective users while minimising the adverse effect of seat covering on the odour control function of the VIP latrine. Furthermore, there is the need for local artisans and prospective owners to be educated on the type of superstructure design that would allow such innovative seat covers to attain, at least, the minimum ventilation rate required for odour control.
CONCLUSIONS
Even though the practice of VIP seat covering has been known to have a negative effect on the ventilation rate, the results of this study reveal the extent to which the practice affects the ventilation rate through the vent pipe. This study has provided further insight into the effect of the covering of a VIP latrine seat by establishing that the practice could reduce the ventilation rate by 87% or more depending on the design of the superstructure. Combined with the use of a 100-mm PVC vent pipe and provision of windows in multiple sides of the superstructure, the ventilation rate could drop to barely 20% of the minimum required to attain odour-free conditions. Even when a superstructure with a vent pipe size of 150 mm is used and a window provided in only one side of the superstructure as conventionally recommended, the covering of the seat would not allow up to 50% of the minimum ventilation rate. Thus, using a ceramic seat with a seat cover which is left covered when the latrine is not in use is likely to lead to the generation of odour nuisance in the latrine cubicle when a prospective user lifts the cover. The study also identifies the replacement of part of the solid cover with an insect screen as a potential innovation for offering the user the aesthetics of a ceramic seat with a cover as used in the water closet with a minimised effect on the ventilation rate. This innovation could lead to more than three times the ventilation rate in a covered seat and attain the minimum ventilation rate when combined with the conventionally recommended superstructure design. However, when used in a superstructure fitted with a 100-mm vent pipe and having windows provided in multiple sides as commonly practised in Ghana, the innovation would not be able to attain the minimum ventilation rate. There is the need for research and state institutions to engage producers of prefabricated ceramic seats intended for use in VIP latrines on how to introduce such innovations to make their products compatible with the VIP latrine's mode of operation. There is also the need to educate latrine builders and prospective owners on how to design the latrine superstructure if such innovations are to be adopted.
AUTHOR CONTRIBUTIONS
P.A.O. contributed to writing – original draft, visualisation, methodology, investigation, formal analysis, conceptualisation. E.A.D. contributed to writing – review and editing, supervision, resources, project administration, funding acquisition, conceptualisation. E.A. contributed to writing – review and editing, supervision, project administration, formal analysis, conceptualisation. S.O.-K. contributed to writing – review and editing, supervision, resources, project administration, funding acquisition, conceptualisation. E.A. contributed to writing – review and editing, supervision, resources, project administration, funding acquisition, conceptualisation. P.A.O. contributed to writing – review and editing, visualisation, validation, supervision, resources, project administration, methodology, formal analysis, conceptualisation.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.