The net carbon footprint of green stormwater infrastructure; bioretention basins

Sustainable stormwater management systems have evolved to address the adverse environmental impacts of increased runoff, which is a consequence of urbanisation. Along with an increase in urbanisation, the sources of contaminants are increasing. Rain washes off surfaces such as roofs, roads, and car parks and adds a high pollution load to stormwater systems. To address this, there exist various types of vegetated stormwater infrastructure, such as bioretention basins, green roofs, rain gardens, vegetated swales, and stormwater ponds. Bioretention basins are a prominent type of vegetated stormwater infrastructure which has been designed for volume reduction and water quality improvement. Despite the attention paid to the environmental impacts and pollutant removal performance of bioretention basins, there is little understanding of the carbon (C) dynamics of bioretention soil. One of the most valuable added ecosystem benefits of these systems is C storage and sequestration within the soil media, while soil organic C also plays an essential role in the removal of nitrogen (N) from stormwater in bioretention basins. This thesis investigates the whole life cycle carbon footprint of vegetated stormwater infrastructure and provides field-based measurements of C and total nitrogen (TN) storage in bioretention soil and the associated greenhouse gas (GHG) fluxes from these basins. The soils’ C and N dynamics, denitrification potential and the GHG fluxes from bioretention basins were investigated in relation to a variety of factors, such as the site’s age, soil texture and nutrient concentrations. The primary objectives of this research were: a) to review and analyse the net carbon footprint of green stormwater infrastructure; b) to measure the C sequestration potential of bioretention basins; c) to investigate the performance of bioretention basins in accumulation of TN and denitrification potential; and d) to investigate the total GHG fluxes from bioretention basins. The thesis is prepared in six chapters including published and unpublished papers. Chapter Two systematically reviews and analyses the life cycle carbon footprint and C sequestration potential of five types of vegetated stormwater infrastructure. The results show the significant potential of C sequestration in mitigating the life cycle carbon footprint in rain gardens, bioretention basins, green roofs, vegetated swales and stormwater ponds. The mitigation potential over a 30-year lifetime was estimated as >100%, 70%, 68%, 45% and 8%, for each of these systems respectively. In Chapter Three, the spatial, vertical and temporal variation of C were explored in 25 bioretention basins in subtropical climatic conditions in south-east Queensland, Australia. A thirteen-year soil chronosequence method was employed to measure the C sequestration potential of bioretention basins. The results showed that there was a strong influence of age on the soil C density throughout the top 20 cm of the bioretention soil only, with a sequestration rate of 0.31 kg C m-2 yr-1 over 13 years of operation. The accumulation rate of C in the bioretention soil was very high in the top 5 cm of soil, while in the lower soil depths, C accumulated at a more gradual rate. This chapter argued that future design considerations for bioretention basins should consider an addition of a C source layer at the bottom of the soil filter media, an increase in the saturation level in the soil profile and/or an increase in the hydraulic conductivity. This would potentially promote the pollutant removal performance of bioretention basins through denitrification by providing C in a deeper soil profile where anaerobic conditions are more likely to occur. In Chapter Four, the variation of TN in the bioretention soil was investigated spatially, vertically and temporally. The values of stable isotopes of C and N was investigated in both bioretention plants and below ground soil media. In addition, the nutrient removal performance of bioretention basins was investigated through assessment of the potential denitrification of a subset of seven bioretention basins. The TN accumulation in soil strongly correlated with the soil C content. C source tracing showed that the bioretention plants (C3 plants) are the major source of C in the bioretention soil. However, the existence of external sources of C with high 𝜹�13C value was noted, which possibly occurred through C amendment of the soil via maintenance work. The sites’ ages were shown to significantly positively influence both the C and the TN accumulation in bioretention soil. The soil TN and 𝜹�15N levels were very low in the bioretention basins which were three years of age or younger. In addition, the NOx concentration significantly increased with the TN in the soil. The ages of the sites showed a strong positive correlation with the denitrification potential of bioretention basins. The young bioretention basins showed low TN content, 𝜹�15N and denitrification potential. It is argued that biotic uptake through bioretention plants could be the major fate of nitrate in young bioretention basins, as opposed to their low denitrification potential. Chapter Five investigated the fluxes of N2O, CH4 and CO2 in two bioretention basins and their embankment area. The two bioretention basins had contrasting hydraulic conductivities of 28 mm hr-1 and 312 mm hr-1. The static chambers method was used to measure the GHG over 34 days with a total rainfall of 147 mm. The measurements were taken before and after each rainfall event. The influence of the Carex appressa plant on the GHG fluxes was also examined in the fast draining bioretention basin. N2O fluxes positively correlated with precipitation, although the results showed that N2O was a negligible source of GHG emissions in bioretention basins. The fast draining basin and the embankment area were both sinks of CH4 but were large sources of CO2 fluxes. The slow draining basin had large fluxes of CH4, while its CO2 fluxes were much lower than those in the fast draining basin. CO2 was the major source of GHG fluxes from bioretention basins, and it was noted that a balance between the CO2 and CH4 emissions can minimise the total GHG emissions. The presence of the Carex appressa plant increased the magnitude of the sink of CH4 and the source of N2O fluxes. It was concluded that adopting a slow draining design for bioretention basins can minimise the total GHG emissions while potentially achieving treatment objectives for the basins. Chapter Six of this thesis presented the GHG fluxes as part of the overall net carbon footprint of bioretention basins. A conceptual framework was developed for the net carbon footprint of bioretention basins, including the life cycle carbon footprint, C sequestration and GHG fluxes. This thesis makes a contribution to theory, method, design and maintenance of bioretention basins. This study was the first survey on the abundance of soil C and TN and values of stable isotopes of C and N in bioretention basins, and some design recommendations have resulted from this research. The outcomes of the thesis can contribute to the body of knowledge and provide designers and policymakers with further considerations of C implications in the technical design characteristics and future operation strategies of bioretention basins.