Today, we have inexpensive sensors to detect
gases and particles in indoor and outdoor environments. For example, some models of cars now
come equipped with low-cost sensors in the ventilation system to detect pollutant levels in the
incoming air. This information is useful for controlling the ventilation system and reducing exposure when the vehicle is driving on polluted streets
in tight traffic. When high pollutant levels are
detected, the car shifts to recirculation, reducing
pollution exposure for the driver and passengers.
Low-cost sensors are becoming more specific,
accurate, and useful, and with their increasing
utility in new applications, the demand for effective low-cost sensors is increasing [12].
The development of a new generation of small,
portable, low-cost sensors and sensor nodes has
been rapid in recent years, although they are typically less accurate than conventional instruments.
However, high precision is not always important,
as in the example of ventilation in cars where the
sensor only needs to detect whether pollution
levels are high or low. There is also a value in
the speed and density of monitoring promised by
low-cost sensors, potentially providing high temporal and spatial resolution, respectively.
As sensors become smaller, more accurate, and
cheaper, they are finding use in an increasing
number of applications, beyond the traditional
areas of mining and industry. Low-cost sensors
are unlikely to replace conventional instruments
for accurate measurements of well-mixed gases in
the atmosphere such as CO 2 (with the exception of
variable CO 2 levels in buildings) but could be
particularly useful for measurements of shortlived pollutants with sources inside cities, such
as NO 2, PM, and O 3 . In addition to poor time
resolution, current pollution monitoring networks
are expensive (maintenance, staff, instrument
cost, and space), and cities have too few monitoring sites to create an accurate map of a city.
A detailed time-dependent pollution map would
be extremely useful in pollution control efforts, to
pinpoint sources and exposure hot spots and help
evaluate remediation strategies. For instance, in
urban areas in the USA, there are only ~2–5
sensors per 1000 km
2 or per one million people
[13]. A network containing low-cost sensor nodes
could have orders of magnitude more sites with
greatly improved time resolution, and so provide a
pollution map with far higher temporal and spatial
resolution. This map could be combined with a
map of population density to visualize pollution
exposure throughout the day. Such a sensor network could thereby be a great improvement on
today’s measurements and estimates of pollution
exposure. Anchoring this network to reference
sensors would also aid in calibration and improve
accuracy. Accurate, widespread, low-cost sensor
networks are just starting to become a reality and
there is massive potential for their widespread use
[14, 15].
Low-cost sensors promise to be a vital tool for
monitoring both the personal exposure of individuals and the pollution observed at specific locations. These locations could be emissions sources
(livestock barns, construction sites, factories,
pipelines, etc.) or along trafficked routes
(highways, shipping lanes, etc.), or polluted
areas (ports and industrial parks, neighborhoods,
etc.) Previously, passive sampling methods have
been used with some success for personal exposure measurements, but this approach significantly lowers the temporal resolution of the data
Low-Cost Sensors for Indoor and Outdoor Pollution,
Fig. 1 A typical canary bird
Low-Cost Sensors for Indoor and Outdoor Pollution
425
gases and particles in indoor and outdoor environments. For example, some models of cars now
come equipped with low-cost sensors in the ventilation system to detect pollutant levels in the
incoming air. This information is useful for controlling the ventilation system and reducing exposure when the vehicle is driving on polluted streets
in tight traffic. When high pollutant levels are
detected, the car shifts to recirculation, reducing
pollution exposure for the driver and passengers.
Low-cost sensors are becoming more specific,
accurate, and useful, and with their increasing
utility in new applications, the demand for effective low-cost sensors is increasing [12].
The development of a new generation of small,
portable, low-cost sensors and sensor nodes has
been rapid in recent years, although they are typically less accurate than conventional instruments.
However, high precision is not always important,
as in the example of ventilation in cars where the
sensor only needs to detect whether pollution
levels are high or low. There is also a value in
the speed and density of monitoring promised by
low-cost sensors, potentially providing high temporal and spatial resolution, respectively.
As sensors become smaller, more accurate, and
cheaper, they are finding use in an increasing
number of applications, beyond the traditional
areas of mining and industry. Low-cost sensors
are unlikely to replace conventional instruments
for accurate measurements of well-mixed gases in
the atmosphere such as CO 2 (with the exception of
variable CO 2 levels in buildings) but could be
particularly useful for measurements of shortlived pollutants with sources inside cities, such
as NO 2, PM, and O 3 . In addition to poor time
resolution, current pollution monitoring networks
are expensive (maintenance, staff, instrument
cost, and space), and cities have too few monitoring sites to create an accurate map of a city.
A detailed time-dependent pollution map would
be extremely useful in pollution control efforts, to
pinpoint sources and exposure hot spots and help
evaluate remediation strategies. For instance, in
urban areas in the USA, there are only ~2–5
sensors per 1000 km
2 or per one million people
[13]. A network containing low-cost sensor nodes
could have orders of magnitude more sites with
greatly improved time resolution, and so provide a
pollution map with far higher temporal and spatial
resolution. This map could be combined with a
map of population density to visualize pollution
exposure throughout the day. Such a sensor network could thereby be a great improvement on
today’s measurements and estimates of pollution
exposure. Anchoring this network to reference
sensors would also aid in calibration and improve
accuracy. Accurate, widespread, low-cost sensor
networks are just starting to become a reality and
there is massive potential for their widespread use
[14, 15].
Low-cost sensors promise to be a vital tool for
monitoring both the personal exposure of individuals and the pollution observed at specific locations. These locations could be emissions sources
(livestock barns, construction sites, factories,
pipelines, etc.) or along trafficked routes
(highways, shipping lanes, etc.), or polluted
areas (ports and industrial parks, neighborhoods,
etc.) Previously, passive sampling methods have
been used with some success for personal exposure measurements, but this approach significantly lowers the temporal resolution of the data
Low-Cost Sensors for Indoor and Outdoor Pollution,
Fig. 1 A typical canary bird
Low-Cost Sensors for Indoor and Outdoor Pollution
425