diode laser (TDL) or quantum cascade laser (QCL), although these instruments are
very expensive. For the relaxed eddy accumulation method, the reactive N species in
the upward air and downward air are sampled separately into two containers, and the
mean concentrations of reactive N species in the containers measured in the laboratory under ideal conditions (Businger and Oncley 1990). The flux equation is
f ¼ bσ w
À
c þ À c À
Á
ð3:3Þ
where c þ À c À is the mean concentration difference in the upward wind and
downward wind, σ w is the standard deviation of vertical wind velocity and b is an
empirical constant. Though this method does not need fast-responding instruments,
an effective sampling system to separate the upward and downward moving air is
necessary (Businger and Oncley 1990).
The time-averaged gradient method was developed for long-term deposition flux
measurements at relatively low cost because the gradient, eddy covariance and
relaxed eddy accumulation methods are expensive and labour-intensive (Erisman
et al. 2001; Fowler et al. 2001). It measures the averaged vertical concentration
gradient of N r species as well as the same meteorological parameters as those
measured in the gradient method. Although low-cost, large errors will occur when
the atmosphere is stable (Erisman et al. 2001) because the real deposition flux is very
small, the concentration gradient is large, which can cause an erroneously high dry
deposition to be estimated. To solve this problem, the conditional time-averaged
gradient method was developed, which avoids the continuous sampling of the timeaveraged gradient method and only measures the concentration gradient when
atmospheric conditions meet the requirements of the gradient method (Famulari
et al. 2010). Thus, no measurement is made when the wind velocity is low, the air is
stable and the fetch is not enough long.
3.2.1.2 Estimating Dry Deposition Using the Inferential Method
Using micrometeorological methods to measure dry deposition usually requires
rapid response instruments and homogeneous and even underlying surfaces, which
makes setting up the site costly in terms of money and labour. This limits the
application of these methods. As a substitute for micrometeorological methods, the
inferential method was proposed in which the dry deposition flux (F) is the product
of the concentration of N species at the measuring height (C Z ) and the deposition
velocity (V d ):
F ¼ ÀC Z V d
ð3:4Þ
The direct measurement of V d is difficult, and thus the V d is usually calculated
using well-tested models based on empirical parameters and meteorological data.
When calculating the V d of reactive N gases, the function of the reciprocal of the sum
44
X. Liu et al.
very expensive. For the relaxed eddy accumulation method, the reactive N species in
the upward air and downward air are sampled separately into two containers, and the
mean concentrations of reactive N species in the containers measured in the laboratory under ideal conditions (Businger and Oncley 1990). The flux equation is
f ¼ bσ w
À
c þ À c À
Á
ð3:3Þ
where c þ À c À is the mean concentration difference in the upward wind and
downward wind, σ w is the standard deviation of vertical wind velocity and b is an
empirical constant. Though this method does not need fast-responding instruments,
an effective sampling system to separate the upward and downward moving air is
necessary (Businger and Oncley 1990).
The time-averaged gradient method was developed for long-term deposition flux
measurements at relatively low cost because the gradient, eddy covariance and
relaxed eddy accumulation methods are expensive and labour-intensive (Erisman
et al. 2001; Fowler et al. 2001). It measures the averaged vertical concentration
gradient of N r species as well as the same meteorological parameters as those
measured in the gradient method. Although low-cost, large errors will occur when
the atmosphere is stable (Erisman et al. 2001) because the real deposition flux is very
small, the concentration gradient is large, which can cause an erroneously high dry
deposition to be estimated. To solve this problem, the conditional time-averaged
gradient method was developed, which avoids the continuous sampling of the timeaveraged gradient method and only measures the concentration gradient when
atmospheric conditions meet the requirements of the gradient method (Famulari
et al. 2010). Thus, no measurement is made when the wind velocity is low, the air is
stable and the fetch is not enough long.
3.2.1.2 Estimating Dry Deposition Using the Inferential Method
Using micrometeorological methods to measure dry deposition usually requires
rapid response instruments and homogeneous and even underlying surfaces, which
makes setting up the site costly in terms of money and labour. This limits the
application of these methods. As a substitute for micrometeorological methods, the
inferential method was proposed in which the dry deposition flux (F) is the product
of the concentration of N species at the measuring height (C Z ) and the deposition
velocity (V d ):
F ¼ ÀC Z V d
ð3:4Þ
The direct measurement of V d is difficult, and thus the V d is usually calculated
using well-tested models based on empirical parameters and meteorological data.
When calculating the V d of reactive N gases, the function of the reciprocal of the sum
44
X. Liu et al.