Odour management at landfills.

Abstract

Landfill sites that are situated near residential zones pose significant management challenges. One of the key issues concerns odour emissions; they can be a major cause of public opposition to the existence of landfills. Mounting opposition can result in costly and premature closure of waste management facilities. The Bisasar Road landfill in Durban is one of the largest general waste landfills in South Africa, and is surrounded by residential areas on three sides. Odour emissions are therefore of primary concern. The situation at Bisasar provided an opportunity to involve the surrounding residents as odour "receptors" in order to undertake research that would provide a better understanding of odour emissions and how best to manage them. Effective management of odour emissions requires a model to accurately predict the occurrence and extent of the problem. For effective modelling and prediction of odour at a landfill, it becomes necessary to characterize the odour sources in terms of their odour emission rates and also to establish the population's response to the odour over a wide range of concentrations. This thesis describes the application of numerical dispersion modelling, coupled with interactive community involvement, in order to indirectly estimate odour emission rates, where the community acts as an odour assessment panel. A reanalysis of data captured from a 2002 community survey was performed in order to investigate the odour emission rate from the open waste piles - transfer station and working face - at the Bisasar road Landfill site. The odour emission rate from these similar odour sources was found to lie in the range 102-103 OU/m2/s. For the duration of the survey, the working face contributed 82% to the odour perceived at the receptors and the Transfer Station contributed 18% to the overall offsite odour. The 2002 survey required receptors to fill out a weekly diary of odour observations, but a new community survey in 2004 used direct telephonic communication with receptors to establish the odour impact. Residents were contacted telephonically at their homes on a regular basis to ascertain whether they perceived an odour at the time of call. This data, in conjunction with the concentration predictions form the odour prediction model, allowed for an emission rate for the open waste piles to be inferred through backward dispersion. The emission rate from an open waste pile was estimated at being 250 OU/m2/s. The individual response of each receptor as well as the combined response for the panel to a range of odour concentrations was assessed in terms of probability of detection and perceived odour intensity. The odour emission rate, inferred through backward dispersion using the 2004 community survey data, was used to establish the impact of the odour from the open waste piles on the surrounding community through long term forward dispersion calculations. This provided a scientifically defensible methodology for the specification of buffer zones around sources that emit nuisance odours.