Mitigating landfill gas emissions: the effectiveness of HDPE covers in reducing atmospheric concentrations of H₂S, NH₃, and PM₁₀

The WMC Marišćina is located near the town of Marčelji, north of Rijeka. It consists of a mechanical-biological treatment (MBT) plant, a wastewater treatment plant, and a bioreactor landfill and receives residual municipal solid waste to produce refuse-derived fuel (RDF).

Upon arrival, the waste is stored in a reception bunker and preshredded to a size of 20 cm. This pre-shredded material, consisting mainly of mixed municipal waste, is then loaded into bioreactors (Rottebox^®, Herhof, Solms, Germany), where it is subjected to forced aeration to dry the waste. This bio-drying process prepares the waste for subsequent mechanical treatment and refining.

The mechanical treatment begins with passing the waste through a 20 mm vibrating sieve to separate the fines, which are primarily composed of organic material. The fines are then transported to a bioreactor landfill, where they serve as a substrate for landfill gas production, currently flared but planned to be used for energy recovery. The remaining material is then passed through a magnetic separator to extract ferrous metals, followed by an eddy current separator to remove non-ferrous metals.

After metal separation, the material is transported to an air classifier, which removes the heavy fraction, and then to a nearinfrared (near-IR) optical separator. The optical separator is used to eliminate polyvinyl chloride (PVC) from the waste stream to reduce chlorine content in RDF. The residues from the mechanical processing are landfilled, while the recovered metals and RDF are sent to other facilities for material and energy recovery. The flow of residual waste through the various processing stages at WMC Marišćina is illustrated in Figure 1.

Figure 1

Material flow of residual waste in the mechanical-biological treatment plant at the WMC Marišćina, Croatia; I – import flow; E – export flow

Implementation of improved gas management strategies (2019–2021)

In response to the operational inefficiencies and environmental concerns (foul odours) noted in 2018, the landfill management team implemented several modifications to the bioreactor operational procedures. The first step involves covering the fine fraction with a low-density polyethylene (LDPE) cover on top of a 10 cm layer of soil at the end of each working day. The second step involves covering most of the landfill area with HDPE membranes welded together to prevent landfill gas escape. To further reduce landfill gas emissions, the landfill working area has been reduced to a maximum of 3,000 m². This allows for a better control of landfill gas emissions and provides more efficient gas collection and extraction. For this purpose, a gas collector has been deployed, connecting all individual vents. A central blower is then used to create negative pressure within the gas collection system and the landfill body and minimise the venting of the landfill gas into the atmosphere. The excess gas is currently being burnt in a process known as flaring, which involves controlled combustion of gas that cannot be captured or utilised. This method, while effective in reducing immediate safety and environmental risks posed by venting raw gas, results in the release of carbon dioxide and other emissions. Having recognised the potential for more sustainable energy use, the WMC plans to replace flaring with an electric energy generator that will convert landfill gas into electricity. A schematic of the current venting system is given in Figure 2.

Figure 2

Active venting system schematic at the HDPE cover WMC Marišćina landfill. W1 – gas well 1; W2 – gas well 2

Table 1 shows seasonal mean and median levels of H₂S, NH₃, and PM₁₀ measured near the WMC landfill in 2018 (baseline) and 2021 (after the HDPE cover was set in place). The decrease in H₂S and PM₁₀ levels between the two monitoring years suggests that the HDPE cover effectively reduced their emissions throughout the year.

In contrast, NH₃ medians soared in the spring of 2021 and continued to be higher than respective seasonal medians in 2018. This increase in NH₃ levels and variability in 2021 may point to other emission sources in the area and calls for further investigation.

Changes in pollutant levels near the WMC between 2018 and 2021 are even clearer with time series analysis presented in Figure 4. The 2018 H₂S levels exhibited greater variability than those measured in 2021, whereas variability in NH₃ and PM₁₀ was less pronounced over the studied years.

Figure 4

Time-trends of 24-hour averaged concentrations of H₂S, NH₃, and PM₁₀ in 2018 (blue) and 2021 (orange). Peak air quality values are indicated by red (2018) and green (2021) crosses

Our trend analysis (Figure 5) shows a general upward trend for H₂S in 2018, which became stable following the installation of the HDPE cover. NH₃ had a rising trend in 2018 but a distinct decline in 2021. The PM₁₀ trends, in contrast, remained relatively stable across both years.

Figure 5

Linear trends in H₂S, NH₃, and PM₁₀ concentrations (red line) measured near the WMC Marišćina with individual data points plotted separately for each year (blue crosses)

Figure 6 histograms and kernel density estimates (KDE) highlight further the differences in atmospheric concentrations of H₂S, NH₃, and PM₁₀ between 2018 and 2021. In 2018, the distributions of all three pollutants were more spread out with heavier tails, indicating higher variability and a higher prevalence of peak atmospheric concentrations of pollutants. This variability is particularly pronounced with PM₁₀, suggesting multiple contributing sources or varying emission intensities throughout the year. One such source could be the asphalt plant mentioned earlier. In contrast, the 2021 data exhibit more peaked distributions with narrower spreads, indicating more consistent pollutant levels and fewer extreme events.

Figure 6

Distribution of H₂S, NH₃, and PM₁₀ 24-hour averaged levels near the WMC Marišćina and their kernel density estimates

Comparison of atmospheric pollutant levels between 2018 and 2021 and their correlations

The Mann-Whitney U test reveals statistically significant differences (p<0.05) in the levels of all three pollutants between 2018 and 2021 (Figure 7). Figure 8 shows even more clearly that H₂S and PM₁₀ levels dropped, whereas NH₃ levels increased after the HDPE cover was set up, suggesting possible other sources of ammonia emissions in the vicinity of the bioreactor landfill, shifts in waste composition, operational changes, or potential accumulation of ammonia beneath the HDPE cover. Figure 9 shows the relationship between NH₃, H₂S, and PM₁₀ levels obtained with the Spearman’s rank correlation analysis for 2018 and 2021. In 2018, before the landfill was covered with an HDPE membrane, H₂S, NH₃, and PM₁₀ had a moderate positive correlation, which suggests that emissions of these pollutants were likely influenced by a shared underlying process, such as waste decomposition and the lack of containment, leading to the concurrent release of gases and particulates into the air. In 2021, after the landfill was covered with an HDPE membrane, H₂S showed a weaker correlation with both NH₃ and PM₁₀ compared to 2018. This change may indicate that the HDPE membrane disrupted the simultaneous release of these pollutants by containing H₂S more effectively than PM₁₀ or NH₃and could reflect independent emission mechanisms. Overall, the reduced correlations in 2021 suggest that HDPE was effective in modifying the interaction between emission processes, leading to a more controlled and segmented release of pollutants.

Figure 7

Comparison of 2018 and 2021 H₂S, NH₃, and PM₁₀ levels measured near the WMC Marišćina. The central line within each box represents medians, while the edges of the box indicate the interquartile range (IQR). Whiskers extend to the minimum and maximum values within 1.5 times of the IQR, and any outliers are plotted as individual points

Figure 8

Changes in H₂S, NH₃, and PM₁₀ measured near the WMC Marišćina from 2018 to 2021

Figure 9

Spearman’s rank correlation matrix illustrating the relationships between H₂S, NH₃, and PM₁₀ levels measured in ambient air near the WMC Marišćina for the years 2018 and 2021. The colour scale represents the strength and direction of the correlations, with values ranging from -1 (perfect negative correlation) to 1 (perfect positive correlation)

Our findings confirm the efficacy of HDPE covers in significantly reducing emissions of H₂S and PM₁₀ from the WMC landfill. These results are consistent with previous research, which has demonstrated that HDPE covers can effectively contain and reduce emissions of landfill gases (14). The increase in NH₃ levels in ambient air following the installation of HDPE cover is somewhat unexpected, however, and suggests that while HDPE covers are effective for certain pollutants, they may alter the landfill’s microenvironment and gas dynamics in such way that increases NH₃ emission. The reason for this is that NH₃ is typically a product organic waste decomposition under anaerobic conditions (15) and that the HDPE cover may have diminished or cut off oxygen supply and accelerated the anaerobic breakdown of nitrogenous materials into ammonia. Another reason could be that the type and composition of waste delivered to the landfill have changed between 2018 and 2021 in favour of nitrogen-rich waste materials, such as food waste.

Yet another reason for the observed increase in NH₃ may be other sources in the vicinity of the landfill, such as the asphalt plant north of the WMC. Although ammonia is not typically a by-product of standard asphalt production, emissions may arise from the use of nitrogen-containing chemical additives, such as those in warm mix asphalt technologies, or from incomplete combustion of fuels with nitrogen compounds in asphalt plants, or from the use of nitrogen-rich reclaimed or recycled materials, or from the use of ammonium-based dust suppressants.

Whatever the reason, our findings highlight the complexity of landfill gas management and the need for comprehensive monitoring systems that can detect and address unexpected shifts in emission profiles (16, 17).

Another important thing to consider when assessing air quality data is wind direction. Our analysis revealed that peak pollutant concentrations were observed when the wind was blowing from the northwest (NW), that is, directly from the WMC to the monitoring station. Peak concentrations of pollutant levels recorded during these episodes may reflect local atmospheric transport dynamics. Accounting for such directional influences is essential for accurately identifying the sources of ambient pollution and for developing targeted mitigation strategies.

In conclusion, while HDPE covers do offer a promising solution for reducing certain landfill emissions, their implementation must be part of a broader, integrated approach to landfill management. This approach should include comprehensive environmental monitoring programmes, consider all potential emissions, and employ adaptive management strategies to ensure that interventions lead to a net positive environmental and health outcome.

A novel strategy in that respect could be the deployment of advanced monitoring technologies such as drones and sensor networks, which could also help identify the sources of increased NH₃ emissions. In addition, future research should focus on understanding the mechanisms which are driving the increase in NH₃ emissions and on developing strategies to mitigate them. This includes exploring potential interactions between different landfill gases and conditions under which they are produced and released. Future research should also investigate long-term effects of HDPE cover on landfill emissions and how they could contribute to broader environmental goals, such as reducing greenhouse gas emissions and improving air quality.