27 Potential Impact of a Low Emission Zone on Street-Level Air …
173
tal resolution to predict spatial patterns and in specific coordinates to compare with
observations.
To evaluate predicted hourly NO 2 concentrations we use observations from an
experimental campaign conducted in April and May 2013 in Barcelona [2]. During
the campaign, mobile laboratories placed at the parking lane of several street segments measured air quality parameters at 3 m height. For this study, data gathered
every 30 min at Valencia Street No. 445 is shown. This street presents a marked
canyon pattern where building height to street width ratio is approximately 1.
As an initial step towards developing a methodology to assess the potential impact
of the LEZ on NO 2 concentration levels, in this work we analyze the sensitivity of
CALIOPE-Urban to potential reductions of NO x emissions that could be caused by
a policy to improve air quality in the city without taking into account the LEZ at
this stage. In the analysis presented here, we only use a mesoscale model run that
incorporates the full emissions in the region and we force the street-scale model to
use as input explicit emissions for each street segment with a reduction of 15%. This
is a preliminary methodology to address the question of the effectiveness of such
measure. In a future work, the reductions will be considered both in the mesoscale
and urban scale models.
27.3 Preliminary Results
We start by assessing the ability of the model to represent spatial and temporal
variability of NO 2 concentrations in Barcelona, then we analyze the sensitivity to a
potential reduction of NO x emissions. Figure 27.1 shows the spatial detail provided
by street-scale and mesoscale combined solutions over Eixample district for 8 UTC
9 April 2013. The increased resolution of street-scale makes possible to visualize
NO 2 street gradients.
To visually analyze hourly NO 2 temporal variation of the models and to evaluate
the sensitivity of the system to changes in NO x emissions we use observations from
the highly trafficked experimental campaign site of Valencia street, which has an
average daily traffic of 32,500 vehicles and represents the typical street canyon pattern
with an average of 20 m building height and 20 m street width, during the period
4–8 April 2013. This period and site are chosen because they represent the main
NO 2 concentration patterns found during the experimental campaign in the most
extended geometry of the city (i.e. Eixample district). Figure 27.2 shows observed
NO 2 concentration levels (black) compared to CALIOPE (blue), CALIOPE-Urban
(red) and for the street-scale model using reduced NO x emissions of 15% (green).
From Fig. 27.2 we can see a high difference in performance between mesoscale (blue)
and street-scale (red) models. Overall mesoscale is systematically under-predicting
the concentration during the central hours of the day and is not able to capture
the peaks. In contrast, street-scale represents more precisely the measured diurnal
variation, giving estimates on weekdays in accordance with observations levels and
a better reproduction of peak values. For example, on the 8th at 6 am (UTC) a
173
tal resolution to predict spatial patterns and in specific coordinates to compare with
observations.
To evaluate predicted hourly NO 2 concentrations we use observations from an
experimental campaign conducted in April and May 2013 in Barcelona [2]. During
the campaign, mobile laboratories placed at the parking lane of several street segments measured air quality parameters at 3 m height. For this study, data gathered
every 30 min at Valencia Street No. 445 is shown. This street presents a marked
canyon pattern where building height to street width ratio is approximately 1.
As an initial step towards developing a methodology to assess the potential impact
of the LEZ on NO 2 concentration levels, in this work we analyze the sensitivity of
CALIOPE-Urban to potential reductions of NO x emissions that could be caused by
a policy to improve air quality in the city without taking into account the LEZ at
this stage. In the analysis presented here, we only use a mesoscale model run that
incorporates the full emissions in the region and we force the street-scale model to
use as input explicit emissions for each street segment with a reduction of 15%. This
is a preliminary methodology to address the question of the effectiveness of such
measure. In a future work, the reductions will be considered both in the mesoscale
and urban scale models.
27.3 Preliminary Results
We start by assessing the ability of the model to represent spatial and temporal
variability of NO 2 concentrations in Barcelona, then we analyze the sensitivity to a
potential reduction of NO x emissions. Figure 27.1 shows the spatial detail provided
by street-scale and mesoscale combined solutions over Eixample district for 8 UTC
9 April 2013. The increased resolution of street-scale makes possible to visualize
NO 2 street gradients.
To visually analyze hourly NO 2 temporal variation of the models and to evaluate
the sensitivity of the system to changes in NO x emissions we use observations from
the highly trafficked experimental campaign site of Valencia street, which has an
average daily traffic of 32,500 vehicles and represents the typical street canyon pattern
with an average of 20 m building height and 20 m street width, during the period
4–8 April 2013. This period and site are chosen because they represent the main
NO 2 concentration patterns found during the experimental campaign in the most
extended geometry of the city (i.e. Eixample district). Figure 27.2 shows observed
NO 2 concentration levels (black) compared to CALIOPE (blue), CALIOPE-Urban
(red) and for the street-scale model using reduced NO x emissions of 15% (green).
From Fig. 27.2 we can see a high difference in performance between mesoscale (blue)
and street-scale (red) models. Overall mesoscale is systematically under-predicting
the concentration during the central hours of the day and is not able to capture
the peaks. In contrast, street-scale represents more precisely the measured diurnal
variation, giving estimates on weekdays in accordance with observations levels and
a better reproduction of peak values. For example, on the 8th at 6 am (UTC) a