Geography Reference
In-Depth Information
Despite these shortcomings, we can say that the dosimetry data, if collected with
sufficient accuracy, can be used to construct maps of the spatial distribution of noise
or even be combined with data from conventional measurements.
Cartographic Processing of Voluntarily Measured
Environmental Noise Data
Currently there is not too much scientific literature that would address cartographic
visualization of environmental noise data. Technical standards (old ISO 1996-
2:
1987
) and best practice literature (European Commission
2006
) present only
overview maps with contours, or choropleth maps. The area method prevails in
large scale and detailed maps. Virtually all maps give information only on the
average noise in the area, but not (except for maps at a very large scale) on the
effects of noise pollution on humans (this information must be deduced), and are
usually result of calculations. Such maps can be constructed from measurements if
there is a sufficiently dense network of sensors which produces representative data.
The number of required measurements, however, is directly proportional to size of
the reference area and the variability of noise sources. Therefore, in most cases of
using volunteered data collected by mobile phones, choropleth or contour methods
are not applicable, because not enough of the space is sampled.
Authors of the first attempts at environmental noise measurement using mobile
phones (D
Hondt et al.
2013
; Kanjo
2010
;N¨st
2013
) divided the area of interest
into regular cells (usually square). The measured point noise data are then assigned
to overlaying cell. If multiple points fall within one cell, their values are simply
averaged.
Older maps, which depict traffic noise measurement and calculation, use the
method of linear symbols. The noise level of surface transport at a standardized
distance from source (e.g. railways or roads), or in some cases the noise level at the
facades (e.g. Senatsverwaltung f¨r Gesundheit, Umwelt und Verbracherschutz
2008
) is typically what is represented in these maps. In both cases it is necessary
generalize measured values into cells (or lines).
'
Generalization of Regions with Presupposed Homogeneous
Noise Distributions
The basic visualization parameters of a volunteered noise map, formed from a large
volume of values, measured at various positions, are (1) shape, (2) size and
(3) arrangement of visualization cells. In most common cases, the cells are square
and are arranged to a square grid because processing of that configuration is the
most simple, but other shapes and arrangements, e.g. hexagonal, may be more
