Assessment of Landfill Odorous Gas Effect on Surrounding Environment

Landfill odorous gas emission has been a serious environmental problem, especially for the residents or passersby on the road near the landfill. In this paper, in situ monitoring and the numerical CALPUFF model were adopted to analyze the odor nuisance problem caused bymunicipal solid waste (MSW) degradation in a large landfill.,e static chamber technique was used tomeasure the odorous gas emission rate on the working area, temporary cover area, and final cover area of the landfill during Dec. 2016 and Apr. 2018.,e results showed that the emission rate of H2S on the working area varied from 0.003mg/m/min to 0.98mg/m/min, and it was positively correlated with the ambient temperature.,e emission of H2S varied between 0.125 kg/h and 1.09 kg/h on the working area, and it varied between 9.2×10 kg/h and 6.8×10 kg/h on the temporary cover when considering the impact of the holes in a high-density polyethylene (HDPE) membrane, and it was negligible on the final cover. ,e contribution rates of H2S emission in the whole landfill were 90.79%∼98.59% and 0.0008%∼0.52% for the working area and the temporary cover area, respectively. ,e numerical simulation showed that wind velocity and gas emission rates were the critical factors that affect odor dispersion. To limit the H2S-influenced area within the landfill site, proper engineering measures should be taken to ensure the H2S emission rate of lower than 15% of its original value.


Introduction
With the increasing of urbanization, a large number of transportation infrastructures with tremendous traffic and human flow (e.g., highways and tunnels) appear around the municipal solid waste (MSW) landfill. e degradation of MSW would generate less than 1% v/v nonmethane compounds, which are considered to be one of the primary sources of odor emissions [1]. e formation of these odorous gases originated from the secondary product during the waste degradation or strip of the mixed materials existed in the MSW. Compared with the MSW in developed European and American countries, the kitchen waste accounts for 40%∼85% (wet basis) of the MSW in China, and the initial water content varies from 30% to 70% (wet basis) [2]. us, a greater variety of odorous gases could be measured around the landfill in China [3][4][5]. e health of the residents and passersby in and around the landfill site will be significantly influenced by the harmful odorous gases. e odor-related nuisance has become the major cause of complaints to the local authorities [6]. e high concentration of H 2 S could be measured in all the landfills monitoring at home and abroad, ranging from 0.004 ppm to 0.64 ppm [7][8][9]. Besides, the olfactory threshold of H 2 S is only 0.41 ppb, which caused the H 2 S chosen as the representative odor gas of landfills [5]. e gas migration in the air directly related to the local meteorological conditions [10]. Some specific weather conditions, including low mixing height or low wind speed could severely affect the odor dispersion process. Landfills are often located in complex terrain areas. e meteorological conditions are challenging to predict, which leads to the complexity of odor prediction. To assess the individual's perception of the odorous material, the quantification of compounds concentration, which is perceived by receptors, could be the most effective way [11]. Capelli et al. indicated that dispersion models had become a valuable tool to evaluate odor pollution [12]. Some air dispersion models, such as AERMOD and CALPUFF, recommended by the National Standard (HJ2.2-2008) could analyze the gas migration in the range of tens of meters to tens of kilometers [13]. e existing research studies prefer to use the CAL-PUFF to assess the gas migration due to the limitations of the AERMOD model, including the incapability of handling the calm wind condition and steady-state assumption [14].
is paper mainly focused on the impact of the air pollution caused by the MSW degradation on the residents or passersby on the road near the landfill. In situ monitoring and the CALPUFF model were used to evaluate the long-term H 2 S emission rate and the key factors which may affect the gas diffusion in the air. A proper method should be adopted to reduce the H 2 S fugitive emission in the working area. e result could be used to improve landfill operations and management.

Site Description.
e investigated MSW landfill is located in the northern part of Hangzhou City (120°12′E, 30°23′N). It is the first standardized valley-type sanitary landfill in China. e area of the landfill is 700,000 m 2 . It was designed to contain 22 million m 3 of MSW. e site does not accept medical, industrial, and hazardous waste. e daily amount of landfilled MSW increased to 6655 t/d since 2017 [15]. e total surface area of the landfill is 438,000 m 2 and comprises three parts: working area, temporary cover area, and final cover area. 1.5 mm thick high-density polyethylene (HDPE) was used as the only temporary cover material. Gas wells were installed to collect landfill gas (LFG) for electricity generation. e size of the working area is approximately 6000 m 2 .
ere are many municipal structures scattered around the landfill within a radius of 5 km. Figure 1, sampling points were distributed on the landfill working area, temporary cover area, and final cover area. Twelve sampling points for each campaign were randomly arranged on the working area in each campaign following US EPA requirements. e monitoring above the temporary cover focused on the gas emission from the leakage point. e static chamber and 10 ml syringe were used to collect landfill gas from selected locations. e chamber used in this study consists of a glass cylinder and a steel pedestal. e chamber is 550 mm in height and 500 mm in diameter. Once the chamber was placed on the sampling location, the groove in the steel pedestal was filled with water to keep the chamber sealed. All the valves were then shut off to allow the gas to accumulate into the chamber. e gas was sampled for 1 hour every 4 hours. A 1-L empty aluminum bag was connected to the chamber to equilibrate inner pressure. e gas emission rate was determined by equations (1) and (2) [16]:

Gas Sampling and Analysis. As shown in
where V is the volume of the chamber (m 3 ); C is the H 2 S concentration (mg/m 3 ); S is the area of the chamber base (m 2 ); j′ is the gas flux (mg/m 2 /h); t is the time (min); ΔC is the gas concentration difference between time t and time t + Δt. e sulfur-containing gas analysis was carried out by using a gas chromatograph with a flame photometric detector (FPD, HC-3, Hubei, China shafts Technology Co., Ltd.). FPD conditions were as follows: nitrogen was used as the carrier gas at the pressure of 0.1 MPa; the pressure of hydrogen and oxygen was adjusted to 0.04 MPa to form a hydrogen-rich flame; the temperature of the column and detector was held at 125°C and 100°C, respectively; 1 mL sample from anaerobic tubes was directly injected into the injection port during the measurement; the detection limit is 0.05 ppm for hydrogen sulfide.

e CALPUFF Model.
e CALPUFF model was approved by the US Environmental Protection Agency (EPA) as a useful tool to determine atmospheric dispersion, especially in areas with pollution sources such as landfills. It combines three major components: CALMET, CALPUFF, and CALPOST. A detailed description of the CALPUFF model could be found in a study by Scire et al. [17].

Meteorological Data.
Due to the difficulty of the upper-air meteorological data acquisition in the selected area, the meteorological data simulated by the mesoscale meteorological model (MM5) were used as the initial guess wind field of the CALMET meteorological model. e MM5 model's domains for the processing of the meteorological parameters in the study area consisted of a main domain (Domain 1) with a resolution of 1 km per point and a nested domain (Domain 2) with a resolution of 1 km per point. e integration step was 9 s.
Using the MM5 as the initial guess wind field has been proven to be trustful in different research studies [18]. e validity of calculated meteorological data was verified by comparing them with the measured data from the meteorological station in the landfill ( Figure 2).

Odor
reshold Regulations. e odor annoyance criterions of different countries are also rather diverse [19]. Mostly are based on the 98th percentile method proposed by the Lombardia region guidelines [20]. In odor diffusion modeling, very short averaging times (e.g., a few minutes) must be used [21]. Odors can be smelt intermittently for short periods. In the CALPUFF model, the averaged 1-h odor concentrations are converted to shorter averaging periods based on the power-law equation of the form: where C s is the required shorter time average concentration; C h is the hourly average concentration calculated by the CALPUFF model; T h is the 1-h period expressed in minutes; T s is the short-term period in minutes; exponent (p): the CALPUFF manual advises the use of 1/5th power relationship. Nagata and Takeuchi showed that the olfactory threshold of H 2 S was 0.41 ppb (0. 62 μg/m 3 ) [22]. erefore, the H 2 S 1-h average limited concentration was 0.34 μg/m 3 . Figure 3 illustrates this study's diagram for CALPUFF operation. e spatial scale for analysis is 10 km × 10 km, and the mesh accuracy is 100 m.

Operation of CALPUFF.
e map used in this study was a Universal Transverse Mercator (UTM) Reference Ellipsoid WGS-84, Global Coverage, UTM Zone 11 with coordinates 30.38°N and 120.21°E. e mesh used was 10 km × 10 km and 11 vertical levels for meteorology. e topography and land use were defined using the preprocessors "Terrel" and "CTGPROC."

Results and Discussion
3.1. e Emission Rate of Odorous Gas. All dispersion models need reasonably accurate emission data to produce reliable results. e H 2 S emission rate on the working surface and temporary area during the monitoring period is shown in Figure 4. It is estimated that 20% of the landfill gas released directly from the landfill working area [23]. For the monitoring of H 2 S in this site, the statistical data shown in Figure 4(a) indicated that the H 2 S flux rate varied from 0.003 mg/m 2 /min to 0.98 mg/m 2 /min, equivalent to the annual H 2 S emission varied between 1.57 × 10 4 g/a and 515 × 10 4 g/a, which was comparable to the results measured in South Korean landfills (0∼13.6 × 10 4 g/a) [24]. e arithmetic mean value of H 2 S flux rate increased with the ambient temperature, reaching its peak value of 0.2502 mg/ m 2 /min in June 2017. e maximum flux rate of 0.98 mg/m 2 / min was also monitored at a single point in June 2017. is rule has also been verified by other scholars. Kim indicated that the flux of sulfur-contained organic compounds would peak in summer and changed with seasons [24]. Ji reported the H 2 S concentration peaked between July and September, which caused by the higher water content of MSW in summer. e field monitoring results found that the variation tendency of H 2 S flux was consistent with that of moisture content [25].

e Analysis of the Critical Factors Affecting H 2 S Diffusion.
Set a fixed point at the height of 10 meters, which was 1.7 km to the northwest from the landfill center to be analyzed. e initial ambient temperature was 21°C, the ambient air pressure was 101.3 kPa, and the wind speed was 5.2 m/s. Several factors that may influence the gas diffusion in the air were analyzed by the CALPUFF model.

Wind Velocity.
e wind velocity degree was categorized by wind speed at 10 meters high off the open ground. e average wind speed varies with the altitude which is usually described by the logarithmic law or exponential law [26]. e regulation named "Load Code for the Design of Building Structure" (GB 50009-2001) adopts the exponential law [27]: where Z is the height to the ground; α is surface roughness; V 10 is the average wind velocity at 10 meters above the ground. e wind velocity at 10 meters above the ground was revised in accordance with the wind velocity level 1∼7. e variation of H 2 S concentration with wind velocity is shown in Figure 5(a). Wind velocity played a very significant role in gas diffusion.
e concentration appeared to significantly increase at initial and then decrease in the process of wind velocity variation from calm to level 3.

Air Pressure.
Considering the amplitude of the atmospheric pressure fluctuation [28], the pressure variation was set as 0, ±3 kPa, and ±6 kPa. e variation of H 2 S concentration with atmospheric pressure is shown in Figure 5(b). e result of the gas concentration at the fixed point indicated that amplitude of concentration change caused by the air pressure variation was 0.3%. e pollutant concentration is negatively correlated with the increase in atmospheric pressure. e stronger turbulence caused by the higher atmospheric pressure will increase the regional wind speed and therefore reduce the pollutant concentration. Sadowska-Rociek et al. indicated that the variation of odorous gas concentration was directly related to air pressure [29]. e results of field monitoring showed that gas emission would increase during the low pressure in one day [30,31].

Air Temperature.
e variation of air temperature would affect the H 2 S migration rates directly. Set the air temperature to 0°C, 20°C, and 35°C; the variation of H 2 S concentration with air temperature is shown in Figure 5(c). e results indicated that the H 2 S concentration was changed positively with the air temperature variation, and the amplitude Advances in Civil Engineering of concentration change was 0.6%. e rise of temperature would reduce the atmospheric stability which may weaken the pollutant vertical convection in the air. Qiang et al. reported that the concentration of odorous gas in spring and summer was significantly higher than that in autumn and winter [32].

Emission
Rates. e gas emission rate would be affected by several factors such as the composition, fill age, saturation, and mechanical movement of MSW. e measured H 2 S emission rate is set to F 0 (0.125 kg/h); the variation of H 2 S concentration with emission rate is shown in Figure 5(d). e emission rate varied in the scope of 0.5 F 0 to 1.5 F 0 , and the calculated results indicated that the H 2 S concentration was linearly positively correlated with the variation of gas emission rate in the origin. e equation of the CALPUFF model also proved the linear positive correlation between the pollutant flux and concentration.

Surface Parameters.
For the surface parameters of one landfill, the heat flux caused by human activity could be ignored; theoretical analysis and experiment proved that Bowen ratio and surface roughness are two sensitive  parameters affecting surface gas concentration [33]. However, the geographical environment around the landfill site would keep constant. erefore, the surface parameters would be regarded as constant values that would not be analyzed separately when analyzing the dynamic factors.
From the parametric analysis, it can be concluded that each parameter has a certain effect on the H 2 S diffusion in the air; however, the influence of air temperature and air pressure could be ignored. e variation of gas emission rate was positively correlated with the change of gas concentration in fixed points. In addition, the variation of wind velocity was positively correlated with the H 2 S concentration within a specific wind level range. is would imply that the high gas emission coupled with appropriate wind speed could cause a serious odorous problem in a certain range.
us, it can be concluded that wind level and emission rate are the two key factors affecting H 2 S diffusion in the air.

Assessment of Odor Influence by CALPUFF.
Massive complaints against the odorous problems were investigated during 2016. e spatial distribution of complain points is shown in Figure 3. e complaints were concentrated in August, September, and October. No. 11 and no. 12 points occupied the major proportion of all the complaints, with 37 cases and 6 cases, respectively. e no. 11 point was selected for further analysis. e detailed complaints which contain the specific date and time are listed in Table 1.
According to the 16-wind direction map, the no. 11 point was located 3.5 km south-southwest (SSW) to the landfill. e malodorous material released from the landfill could be directly blown to the no. 11 point when the wind direction was north-northeast (NNE), northeast (NE), and eastnortheast (ENE). e CALPUFF model was used to analyze the percentage of wind that blew directly to the no. 11 point; the results are shown in Table 2. e percentage of wind that blew directly to no. 11 reached 27.9%, 7.86%, 20.9%, 58.32%, and 47.3%, respectively, during the 5 periods with massive complaints. e frequency of wind direction was consistent with the complaint occurrence. e H 2 S flux at the working area was assumed to be kept constant at 0.125 kg/h. According to the H 2 S distribution calculated by CALPUFF (Figure 6(a)), the area within 5 km of the landfill would be interfered by the odorous problem. e averaged H 2 S concentration at the no. 11 point reached 9.1 μg/m 3 , which is 15 times greater than the olfactory threshold. erefore, proper engineering measures such as controlling the exposed area or increasing the gas collection should be taken to ensure the emission rate of H 2 S decreases to 15% of its original value. Cai et al. reviewed nine LFG mitigation measures, including refinement process for MSW landfilling (RPL), LFG collection and flaring (LCF), and LFG collection and power generation (LCP) and found that the implementation of these methods could lead to the malodorous emission reduction of 90∼95% theoretically [34].    Odors from the Tianziling landfill occurred in the midnight recently.

Conclusions
e odorous gas H 2 S was monitored and analyzed in a large-scale landfill in Hangzhou, China. e CALPUFF model was adopted to evaluate the odor influence to the surrounding residential areas. e main conclusions are drawn as follows: (1) e emission rate of H 2 S varied from 0.003 mg/m 2 /min to 0.98 mg/m 2 /min on the working area of the landfill, and it was positively correlated with the ambient temperature. e emission of H 2 S varied between 0.125 kg/h and 1.09 kg/h on the working area, and it varied between 9.2 × 10 −6 kg/h and 6.8 × 10 −4 kg/h on the temporary cover when considering the impact of the holes in the HDPE membrane, and it was negligible on the final cover. e contribution rates of H 2 S emission in the whole landfill were 90.79%∼98.59% and 0.0008%∼0.52% for the working area and the temporary cover area, respectively.
(2) e varied wind direction and speed are the key reasons of large-scale complaints against the odorous problems. e degree of coincidence between the point orientation and the wind direction reached 27.9%, 7.86%, 20.9%, 58.32%, 21.5%, and 47.3%, respectively, during the six periods with high complaints occurrence. During that time, the prevailing wind speed varied from 1.6 m/s to 3.3 m/s; the increasing of wind speed within this range would promote the higher H 2 S concentration.
(3) Gas flux variation is another critical factor to affect the odor dispersion. Proper engineering measures should be taken to ensure the emission rate of H 2 S lower than 15% of its original value, such as increasing the gas collection rate.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request Disclosure A part of the findings in this paper has been published in the Proceedings of 8th International Congress on Environmental Geotechnics (Volume 2).

Conflicts of Interest
e authors declare that they have no conflicts of interest regarding the publication of this paper. Advances in Civil Engineering 9