Assessing wastewater pollution from outdoor night markets and its impacts on river quality

A night market is an attractive place for international tourists to explore and a convenient location for local people to buy daily goods. Night markets are areas that should not be ignored due to their role as international and local tourist sites, especially in Asian cities. Both locals and tourists regard night markets as major recreational sites (Chou 2013; Kalnaovakul & Promsivapallop 2021) where people can experience unique cultural life activities and can enjoy local foods. Night markets can be located inside buildings; however, traditional night markets are outdoor markets, a type of street market, and these venues operate from late afternoon to midnight.

In Taiwan, night markets are vital culinary and shopping destinations (Chang & Hsieh 2006; Lee et al. 2008; Ackerman & Walker 2012; Chuang et al. 2014). However, there remain certain problems that must be improved, such as parking problems, traffic jams, physical environmental deterioration, and hygiene issues (Kuo et al. 2012; Sun et al. 2012; Lee & Pearce 2020). Regarding the environmental issues related to outdoor night markets, waste, waste gas (smoke), and wastewater should be considered. Waste, or litter, can be improved by installing more trash cans or using non-plastic bags or containers. Waste gas stemming from cooking contains large amounts of particles and reduce the ambient air quality. Many studies have addressed this issue. Zhao & Lin (2010) characterized the air quality in night markets and demonstrated that the levels of CO, CO₂, particulate matter with an aerodynamic diameter smaller or equal to 10 μm (PM₁₀), particulate matter with an aerodynamic diameter smaller or equal to 2.5 μm (PM_2.5), formaldehyde (HCHO), and polycyclic aromatic hydrocarbons (PAHs) at night markets were all higher during open hours than those during nonoperation hours. Amesho et al. (2021) assessed PM_2.5 in a night market and obtained similar results. Que et al. (2019) studied carbonyl compounds in cooking oil fumes in night markets and found that high emissions of carbonyl compounds could increase the cancer risk due to workplace exposure. Lee et al. (2019) assessed phthalate ester (PAE) pollution originating from night markets and observed significantly high PAE levels and high risks to people in these areas. Yusaf et al. (2010) examined air pollution stemming from different types of engine generators and suggested the use of liquefied petroleum gas to mitigate air pollution problems in Malaysian night markets. Similar to outdoor night markets, outdoor barbecue cooking might also cause air pollution (Song et al. 2018; Badyda et al. 2020)

Compared to air pollution, wastewater issues have rarely been examined. The discharged wastewater flows into street gutters and rainwater sewers. These facilities occur below the land surface and are generally overlooked. When vendors prepare and cook foods outside and when visitors consume foods outside, wastewater originating from cooking and dish washing activities is usually directly discharged into gutters without treatment. Raw wastewater produced by restaurants contains fat, oil, and grease (FOG), high biochemical oxygen demand (BOD) level, and large amount of total suspended solids (TSS), which can negatively impact the environment and wastewater collection systems (Tang et al. 2012; Gurd et al. 2019; Abomohra et al. 2020). FOG might block wastewater collection pipes and should be effectively removed during pretreatment (Tang et al. 2012; Williams et al. 2012; Yau et al. 2021). Wastewater resulting from night markets is similar to untreated restaurant wastewater; however, studies on night market wastewater are scarce, and there are no studies assessing the impacts of night market wastewater on receiving waterbodies. Wastewater might be minor and invisible, but it might contain high pollutant levels and might deteriorate the river quality. Similarly, Zhao et al. (2022) investigated parabens in road runoff and confirmed that the paraben level in road runoff exceeds that in effluents, and runoff might contaminate urban rivers. The air quality at night markets has been examined, as reported in the literature; however, night market wastewater has received nearly very little attention.

Wastewater stemming from night markets differs from wastewater produced by traditional point pollution sources, which involves continuous emissions and fixed emission outfalls. This wastewater type also differs from that originating from nonpoint pollution sources, which is triggered by random rainfall or snow melt and exhibits diffuse properties. Wastewater resulting from night markets depends on the operation time and type. This type of water pollution source is unique and can be regarded as a new type of pollution source. Untreated wastewater might cause blockages in nearby gutters, leading to serious hygiene and health issues. Night market wastewater is also a pollution source affecting downstream rivers and can reduce the water quality. Therefore, a better understanding of wastewater originating from outdoor night markets is important to improve the local environment and to protect receiving waterbodies.

There are two questions to be answered in this study. The first question entails the characterization of the wastewater quality of night markets. A night market provides diverse services, including food products, drinks, clothes, accessories, and games. According to the different sales, the wastewater types and amounts can vary. A comprehensive survey of night markets is needed, and wastewater must be sampled and analyzed. The second objective aims to quantify the corresponding impacts on receiving waterbodies. This objective relies on a water quality model and comprehensive water pollution source investigation. With a complete water pollution source investigation, the contribution of night market wastewater can be revealed.

Taipei city is the capital city of Taiwan, where night markets function as important tourist sites for both international and domestic travelers. Night markets are also sites providing daily needs for local people. An official guidebook of Taipei night markets (2019) introduced 14 night markets and 3 themed markets widely known for their specific features and unique foods in Taipei city. In Taipei city, most night markets are outdoor market types and located along streets. Therefore, Taipei night markets are always named after roads, such as the Linjiang Street Tourist Night Market, Raohe Street Tourist Night Market, Liaoning Street Night Market, and Huaxi Street Tourist Night Market. Because night markets involve outdoor activities, environmental hygiene is an important issue. The generation of garbage, wastewater, waste gas, and noise should be considered. With an increasing number of trash cans, the litter issue can be resolved. Disposable tableware and plastic bags are limited in Taiwan, which helps to reduce litter. Waste gas should also be noted, and simple gas filter devices can usually be installed to collect and treat waste gas. Wastewater control is relatively rare. Wastewater discharged into roadside gutters might accelerate the proliferation of mosquitos and mice, which might cause health risks. This study particularly focused on wastewater issues among night markets.

The Keelung River runs through Taipei city. The drainage system in the urban area eventually connects with the Keelung River, a branch of the Tamsui River, and then flows into the sea. This indicates that rainwater drainage and possibly polluted urban runoff combined with untreated domestic wastewater can impact the water quality of the Keelung River and Tamsui River. Water quality protection of the Keelung River is one of the major policies of the Taipei city government.

This study was implemented in 3 steps. The first step encompassed a field investigation with interviews and questionnaires to collect basic data and wastewater treatment information on the considered night markets. The second step involved water sampling and analysis to characterize the water quality of night market wastewater. The third step aimed to assess the impacts on the water quality of the Keelung River. This step relied on the application of a water quality model, and the Water Quality Analysis Simulation Program (WASP) was used in this study.

A total of 10 night markets were investigated, and 801 vendors were interviewed. However, it should be noted that the vendors in a given night market are not always fixed. Some vendors operate on certain days, and several vendors might rent and share one space. Vendors take turns using the same night market space. Therefore, the 801 vendors interviewed might not represent the exact number of vendors at these night markets. Basic data, including the basic types, operation times, sold items, wastewater generation and treatment aspects, sources of water, and cleaning locations, were obtained through surveys. The questionnaire is provided in the Supplementary Material. A field investigation was implemented from mid-April to May 2021 and finished before implementation of the limitation policy in response to the COVID-19 pandemic.

Night market wastewater sampling required approval from vendors and should not interrupt their business activities. A total of 28 wastewater samples were collected, and samples were collected on weekdays and weekends. Wastewater was collected during and after operation. Wastewater during operation represents cooking- and dish washing-derived wastewater. The water samples collected after operation indicated washing- and equipment cleaning-derived wastewater. Water was sampled at 2 night markets. It was not necessary to collect wastewater at all night markets because the stall types at Taiwanese night markets are similar. For example, typical local food stalls can be found at each night market, and it is expected that the wastewater quality based on the same food type is similar. Among the 28 water samples, 15 water samples were obtained from the Raohe Street Tourist Night Market, and 13 water samples originated from the Liaoning Street Night Market. One of the 28 water samples was not directly acquired from a vendor. This sample was obtained from the gutter to determine the local water quality.

The sampled water was analyzed to determine dissolved oxygen (DO), pH, conductivity, temperature, suspended solids (SS), BOD, ammonia nitrogen (NH₃-N), total phosphorous (TP), alkyl-benzene sulfonate (ABS), and grease levels. The water samples were sent to an approved inspection company, and the analysis process followed official methods.

The WASP was developed by the US Environmental Protection Agency (EPA) and released for free use. The WASP can simulate the water quality in rivers, reservoirs or estuaries. The WASP includes various dissolution and absorption mechanisms, including pollutant discharge, advection and diffusion processes in hydraulics. The WASP can assess traditional water quality components, such as dissolved oxygen, organic compounds, nutrients, and toxic compounds, including heavy metals. Due to its advantages, the WASP is suitable for most water body types, most water quality items, and different fluid processes, and this model can be coupled with external models. In regard to its limitations, the WASP cannot manage particular variables and processes, such as flood plains, and cannot automatically perform calibration. Also, the WASP requires a large amount of hydrodynamic data.

The newest version is WASP 8.2, which contains advanced eutrophication and toxicant modules (Wool et al. 2020). WASP inputs include segment data, boundary conditions, parameters, and coefficients for various computational mechanisms, and initial conditions for each system of the model. The WASP is a widely used water quality model and has been applied in Taiwan, such as the Keelung River and Tamsui River (Taipei City 2018, 2021). Based on successful experiences, the WASP was adopted in this study. Model details can be found on the US EPA website, including software download instructions and documentations (https://www.epa.gov/ceam/water-quality-analysis-simulation-program-wasp).

Because all vendors occurred outdoors, wastewater was directly discharged into road gutters. According to the field investigation, water use and emissions at the considered night markets could be classified as operational water use and clean water use after operation. Operational water use and emissions were further classified into batched, continuous, or zero-use categories. If the surveyed vendors used a water barrel as their water source, wastewater might be emitted as batch-type wastewater. If vending stalls with wash basins and tap water systems comprised the source, wastewater was regarded as a continuous emission source. If the vendors exhibited no obvious water use source and no emissions, they were considered to produce zero emissions during operation. However, all food service vendors used water to clean their stalls and sites after operation. Wastewater emissions were investigated to evaluate the wastewater amount. The wastewater emission results are summarized in Table 2. It was determined that 45.7% of the 801 vendors produced batch-type emissions and 39.3% of the vendors produced continuous emissions. Fifteen percent of the vendors did not emit wastewater during operation. Table 2 also indicates that the operation types notably differed among the different night markets. For example, the Raohe Street Tourist Night Market mostly produced batch-type emissions and very few continuous emissions. In contrast, the Liaoning Street Night Market and Dalong Street Night Market only produced continuous emissions because all stalls at these night markets operated their own wash basins. Although all stalls were cleaned after operation, one particular condition was considered, e.g., some stalls might be moved back to storehouses to be cleaned and thus might not be cleaned at the night markets. Under this condition, cleaning wastewater was not emitted onsite but was emitted at other locations. Eighty-four of the 801 vendors did not clean their stalls on site but rather conducted cleaning activities in storehouses, and 77 of these vendors occurred at the Raohe Street Tourist Night Market. This suggests that most of the vendors at the night markets tended to clean their stalls on site after operation, and the amount of cleaning wastewater should be considered.

After the field investigation and interviews, the current wastewater treatment measures are summarized in Table 3. Except for the Ningxia Tourist Night Market, the grease trap was used as a common night market wastewater treatment method. However, a fixed water basin is necessary for grease trap installation, and this method is unsuitable for itinerant individual stalls. To control wastewater originating from night markets, the government encouraged vendors to install grease trap equipment. The results indicated that more than half of the generated wastewater could not be treated by grease traps. Most vendors at the surveyed night markets did not operate fixed stalls but individual movable stalls. Installing a grease trap in these stalls is almost impossible. In the current state of Taipei night markets, stalls with water basins will be encouraged to install a grease trap. Wastewater not originating from water basins can be directly discharged without any treatment.

A total of 27 water samples were collected from vendor stalls, and 1 sample was collected from the gutter. Of the 27 samples, 14 samples were collected on a weekday night, and 13 samples were obtained on a weekend night. Because the water samples were collected from stalls and vendor agreement was needed, wastewater was randomly sampled without consideration of a certain type of food service or a fixed sampling time. As a result, the random water quality could represent realistic conditions. However, two of the 27 water samples contained waste oil originating from stalls, and extremely high oil, SS, and BOD concentrations could skew the average wastewater quality. Therefore, these 2 extreme data points were not included in the discussion. Even though the above 2 extreme samples were removed, the variability of the water quality remained high among the remaining 25 samples. This result reflects the properties of night market wastewater: during and after operation, water used for cooking, washing, or cleaning purposes yields emissions of different water quality levels. Water use is affected by the food service type and access to water sources. The pH of night market wastewater ranged from 3.4 to 8.9, the conductivity ranged from 446 to 5,520 μmho/cm, the SS concentration ranged from 50 to 14,700 mg/L, the BOD concentration ranged from 167 to 21,300 mg/L, the NH₃-N concentration ranged from 0.71 to 28.70 mg/L, the TP concentration ranged from 0.42 to 33.20 mg/L, the ABS concentration ranged from 0.1 to 1,290 mg/L, and the grease concentration ranged from 5.3 to 704 mg/L. The median and average values are listed in Table 4. Results for the water sampled from the nearby gutter are also listed in Table 4. The concentration was much lower than that in direct wastewater, indicating that wastewater entering gutters might be diluted by other unknown water. The possible water sources of gutters are complex and might include untreated domestic wastewater, residual rainwater, or seepage groundwater.

Water quantity evaluation was divided into batch-type and continuous emissions. Batch-type emissions originated from the use of water barrels as a water source, and continuous emissions originated from the use of tap water. The volume of a water barrel is 15 L, and the surveyed vendors usually refilled their barrels one time during operation, so 0.03 m³/day water was needed during operation. The average water use time needed for cleaning after operation was 45 min, and the tap water flow ranged from 10 to 12.5 L/min. Therefore, the average water amount after operation was 0.56 m³/day. The total stall water emissions of the batch type reached 0.59 m³/day. The water amount for the continuous emission type was based on water fee records. At the Liaoning Street Night Market, independent water fees were employed, which are used for night market operations. We assumed all used water became wastewater, and the stall water emissions of the continuous type reached 0.747 m³/day. At the 10 night markets, there were 340 batch-type emissions and 451 continuous emissions, and the total wastewater reached 520.1 m³/day. The water use amount was related to the number of consumers. With increasing number of consumers, water use might increase. The presented values could be regarded as an average state because water fees were derived from two-month records and the influence of the number of consumers was smoothened. The water use associated with batch-type emissions was only slightly affected by consumers. Vendors were too busy to refill their water barrels during operation, and the use of cleaning water after operation was needed. Therefore, the water use associated with batch-type emissions was insensitive to the number of consumers.

Based on the sampled wastewater, the wastewater quality was highly variable. This finding is related to the sale items and water use behaviour patterns of the venders. The distribution of night market wastewater had high variance, but the average or the median value might represent an overall status. The derived unit pollution load of one stall in this study represents average conditions. The derived unit pollution load of one stall represents average conditions. It was suggested that this coefficient could be used to assess the pollution loads of the entire night market but not the loads of a particular stall. This process might result in stall load over- or underestimation. When the number of stalls and unit pollution loads are known, the potential pollutant emissions of a given night market can quickly be calculated. The unit pollution load coefficient could be useful as a preliminary assessment to evaluate the possible pollution generated at night markets.

The WASP model was used in water quality assessment of the Keelung River. First, the inputs into the simulated river segments were clarified. Along the simulated river, there occurred a total of 23 inputs along the right and left sides of the river. These 23 inputs included 22 drainage outfalls and 1 outfall of a municipal wastewater treatment plant (MWWTP). The MWWTP treated 200,000 m³ sewage per day and contributed to the largest pollution inputs in this river. Therefore, the downstream water quality from the MWWTP, which is approximately 8 km, was influenced. Increasing BOD and NH₃-N concentrations were observed. These drainage outfalls drain stormwater under wet weather conditions but still contain water under dry weather conditions. Water sources under dry weather conditions were unidentified. Water might originate from septic tanks, car washing, market wastewater, unconnected domestic wastewater, or groundwater seepage. However, the quality and quantity of drainage outfalls on dry days were very diverse. Only the emissions of the MWWTP were associated with continuous records. Without considering the emissions of the MWWTP, the pollution loads originating from the 22 drainage outfalls ranged from 4,814 to 9,112 kg/day for BOD and 1,965–2,658 kg/day for NH₃-N. Details of the 23 inputs are listed in the supplementary file, Table S1.

To assess the impacts of night market wastewater on the Keelung River, two scenarios were designed. The first scenario considered that wastewater from the night markets along the Keelung River was comprehensively collected and treated and was not discharged into the river. Under this scenario, the total loads of these 6 night markets were 1,136.4 kg/day for BOD, 2.0 kg/day for NH₃-N, and 490.5 kg/day for SS. The second scenario considered not only night market wastewater but also other unknown drainage pollution. Unknown drainage pollution refers to wastewater except that at the estimated point and from nonpoint pollution sources. Scenario 2 aimed to show that drainage pollution is an important pollution source in this area and that the current pollution sources were not identified well. Scenario 1 is night market wastewater, and scenario 2 is drainage wastewater pollution, including night market wastewater. The removal pollution loads in the scenarios are listed in Table 5. It can be noted that based on the field investigation, nitrogen compounds were less abundant in night market wastewater, so the removal degree of NH₃-N in scenario 1 was very low. However, the removal degree of BOD in scenario 1 was significant.

Night markets are popular sites, but night market wastewater is usually not suitably managed. This might occur because wastewater originating from night markets does not constitute a traditional point or nonpoint pollution source, so it can be easily neglected. Through field investigation and water quality analysis in this study, the unit loads of food service stalls were generated. The wastewater amount of each stall was approximately 0.75 m³/day, which is similar to the water use of 3 people per day (assuming one person uses 250 L/day). However, the BOD load of each stall reached 2,509 g/day, which is higher than the emissions of 50 people (assuming the BOD concentration in raw domestic wastewater is 180 mg/L). Wastewater resulting from night markets contained high levels of BOD, TP, and grease but a low level of N. Although the wastewater amount was not large, the pollution loads should be noted.

In the case study, there occurred 22 drainage outfalls along the Keelung River, and market wastewater contributed 18.5% to BOD but only 1.3% to NH₃-N under dry weather conditions. Night market wastewater treatment could reduce the BOD concentration in the Keelung River by approximately 10%, but no improvement in NH₃-N and DO could be realized. This study confirmed that night market wastewater is a potential pollution source of BOD. This study characterized night market wastewater and evaluated the unit pollution loads of food stalls, which could be used for preliminary assessments of other night markets in Asian cities.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.