Treated sewage sludge, commonly referred to as biosolids, is the organic by-product of urban waste water treatment. When appropriate treatment is applied, it may be reused as an agricultural fertiliser. Despite this benefit, there are several issues associated with the reuse of municipal sewage sludge in agriculture. While many of these are issues of perception, there is considerable concern over the presence of metals, nutrients, pathogens, pharmaceutical and personal care products (PPCPs), and other endocrine-disrupting and synthetic compounds in biosolids, which may cause environmental and human health problems.
The main aims of this research were to (1) quantify the range of concentrations of metals and two of the most abundant PPCPs in the world, the antimicrobials triclosan (TCS) and triclocarban (TCC), in biosolids from a range of wastewater treatment plants (WWTPs) in the Republic of Ireland (2) undertake a field-scale experiment to assess losses of nutrients (nitrogen and phosphorus), metals, TCS and TCC, and microbial matter (total and faecal coliforms) following successive rainfall events on grassland onto which biosolids had been applied, and to compare the results with another commonly spread organic fertiliser, dairy cattle slurry (DCS) (3) to measure the uptake of metals by ryegrass for a period of time after the application of biosolids (4) conduct a risk assessment of potential hazards of human health concern based on the experimental data.
Three types of biosolids commonly used in Ireland were examined as part of this study: anaerobically digested, lime stabilised (LS) and thermally dried (TD). Biosolids and DCS were surface applied in accordance with the legislation in Ireland. A rainfall simulator was used to generate surface runoff over three successive events (24 hr, 48 hr and 360 hr) after a single application.
The metals in the biosolids from the WWTPs examined were below the maximum allowable concentrations of metals for use in agriculture in the European Union (EU). Some priority metals such as antimony and tin, which are potentially harmful to human health, were identified in some of the samples analysed. As these parameters are not currently regulated, this means that a number of toxic metals, which are up to 40 times higher than their baseline concentrations in soils, are being applied to land without regulation. In the WWTPs examined, concentrations of TCS and TCC were 0.61 and 0.08 µg g-1, which were below the concentrations for these parameters measured in other countries. Similar to the findings for metals, the possibility exists that these potentially harmful, unregulated contaminants, for which no international standards currently exist for recycling in agriculture, may accumulate in the soil upon repeated application.
When losses of nutrients, metals, and indicator microorganisms arising from biosolid-amended plots were compared with slurry treatments, biosolids did not pose a greater risk in terms of losses along the surface runoff pathway. The concentrations of TCS and TCC in surface runoff were also mainly below the limits of detection (90 ng L-1 for TCS, 6 ng L-1 for TCC). Furthermore, there was no significant difference in metal bioaccumulation of the ryegrass between plots that received biosolids and those that did not cover the study duration.
A literature review identified contaminants of concern based on relevant risk factors, persistence, bioaccumulation and toxicity (PBT). The contaminants identified were persistent organic contaminants (POPs), pharmaceuticals and PPCPs. A suite of 16 contaminants identified in the literature were further analysed in a risk ranking model to include health based risk endpoints. A probabilistic model was constructed in Excel 2010 (incorporating @Risk 6.0) to estimate human exposure to organic contaminants that are contained within biosolids destined for land application. Nonylphenols ranked the highest across all environmental compartments. The use of these contaminants is heavily restricted in the EU; however, because of their persistence, bioaccumulation and toxicity of these compounds in the environment remains a concern. Triclosan and TCS also ranked high, and may be considered as a potentially greater risk, as their use is not restricted and they are known to cause adverse health effects.
An exposure assessment model was further developed for both metals and E. coli. The model considered exposure to metals and E. coli through surface water abstracted for drinking taking account of surface runoff, dilution and water treatment effects. The likelihood of illness arising from exposure and the severity of the resulting illness was evaluated. Different dose-response relationships were characterised for the different pollutants with reference Lifetime Average Daily Dose (LADD) and Hazard quotient (HQ) used for metals, whilst a worst-case negative exponential dose-response model was used for E. coli. Of all the scenarios considered (three biosolids treatments), and with regards to the LADD, results showed that mean copper exposure concentrations for children were highest in all three rainfall events (mean values 2.07 × 10-2, 2.07 × 10-2 and 1.18 × 10-2 µg kg-1 bw d-1) corresponding to the LS treatment. This was followed by adult copper exposure concentrations (mean value 1.80 ×10-2, 1.31 × 10-3 and 9.21 × 10-3 µg kg-1 bw d-1, for all three rainfall events). The results for the hazard quotient showed that, of all the scenarios considered, the metal copper and the biosolid treatment LS had the highest HQ for children for all three rainfall events with mean child HQ values 5.59 × 10-4, 4.09 × 10-4 and 3.18 × 10-4 respectively, followed by mean adult HQ values of 4.87 × 10-4, 3.54 × 10-4 and 2.49 × 10-4, respectively. However, these were still below the threshold value of risk (HQ < 0.01, no existing risk).
The results for viable E. coli consumed show that one of the sludges examined (an anaerobically digested sludge originating from a WWTP in the UK (ADUK)) was highest for the first and second rainfall events with mean exposure values 5.20 × 10-1 MPN/100 ml and 2.34 ×10-1, MPN/100 ml, respectively. The results for the probability of illness for healthy and immunocompromised populations showed that the biosolid treatment ADUK (first and second rainfall event) among immunocompromised populations (Ic) had the greatest probability of illness/d with mean probability values of 3.68×10-3 and 2.1 × 10-3 illness/d, respectively. The results indicate that the risk of illness was negligible for healthy individuals; however, care is required with immunocompromised individuals where the annual risk was greater than the threshold risk of illness (10-4) as set by the USEPA.
The overall conclusion from this study is that although, in general, land applied biosolids pose no greater threat to water quality than dairy cattle slurry and cattle exclusion times from biosolids-amended fields may be overly strict (within the context of current exclusion criteria), a matter of concern is that unlegislated metals and PPCPs, found to be present in biosolids originating from a selection of WWTPs examined in this study, may be inadvertently applied to land. With multiple applications over several years, these may build up in the soil and may enter the food chain, raising concerns over the continued application of biosolids to land in Ireland.