Appendix 6. Odour Buffer Assessment [As Published] · 132007.2399

                                                                                               [29]
              Odour Buffer Assessment
              StarCo Developments


Odour intensity observations have been used in line with the method described by Ormerod et al.
(2002). The method permits approximations of odour concentration to be derived from a series of
odour intensity observations made in the field, following an adaptation of standard German
methods VDI 3882 and VDI 3940.

VDI 3940 describes a procedure for logging odour in the field, involving noting the odour intensity
every 10 seconds over a 10 minute period. The method is employed over a grid of points on the
ground at various times throughout a year, in order to determine the number of ‘odour hours’,
which form the basis of determining acceptability. An odour hour occurs when the specific odour in
question is clearly recognisable for at least 10% of the sample period, i.e., at least 6 of the 60
observations over a 10-minute period. Over a year, the odour is considered to be unacceptable if
odour hours occur for more than 10% of the time, i.e., more than 876 hours in a year. Where
observations are not possible to prove this, a modelling procedure is also allowed, and it involves
the routine application of a factor of 10 to modelled 1-hour odour concentrations to account for
odour fluctuations (Frechen, 1997). Effectively, the odour hour approach equates to a nose-
response-time averaged odour concentration equivalent to the intensity of weak or distinct odour
on the VDI scale, occurring at a frequency of at least 1% of the time over the whole year. (A
clearly recognisable odour is assumed to equate to either weak or distinct on the VDI scale).

Observations of odour intensity in the field for this study are only loosely based on the German
approach because the objectives are not identical. Observations are made at various locations
downwind of the odour source. However, rather than defining ‘odour hours’, the procedure is
based on using the observed odour intensity measurements as a basis for estimating mean and
peak odour concentrations and peak-to-mean ratios, and for relating the field observations to
odour complaints and other reports. The observations also provide a basis for ground-truthing of
model simulations of odour concentration.

Most observing periods are of 30 minutes or greater in order to obtain a more representative
sample of the fluctuations in odour intensity than is required in the German odour hour approach.

The odour intensity observations are made every 10 seconds (within a margin of error estimated to
be typically ±10%), but the actual sampling of odour occurs only during inhalation, which typically
has a duration of 2-3 seconds. Thus, the odour data represent 2- to 3-second average samples
once every 10 seconds.

Analysis of odour intensity field observations

An important aim of the odour intensity observations is to provide data from the field that might
assist in odour dispersion model evaluation. The method of analysing the data for this purpose is
described in Ormerod et al. (2002). Briefly, it involves use of a Monte Carlo simulation method
based on the initial observations, to convert odour intensity to odour concentration and to generate
a cluster of estimates of the odour concentration averaged over the sampling period. The cluster of
results reflects the uncertainty involved in the conversion from intensity to concentration. A total of
100 Monte Carlo simulations are made for each 30-minute field observation..

The results of the analysis are plotted as time series and as scatter plots showing the results of the
mean vs. peak concentrations and of the mean vs. peak-to-mean ratios for each of the 100
simulations. These scatter plots visually display the uncertainty inherent in the estimated
concentration data.

Plotting of the field results as time series provides a convenient format to illustrate the fluctuations
of odour over time. Plots have been prepared for the original odour intensity time series and for



1898a report final.doc                                                      PAE Job Number 1898a