Non-uniformity in outdoor CO 2 concentration in city of Copenhagen

. The accurate data of outdoor CO 2 concentration are important for the proper design of ventilation and thus for indoor air quality and energy use in buildings. Typical design practice is to assume outdoor CO 2 concentration to be 400 ppm. However, the outdoor CO 2 concentration may be different in different areas of cities. This paper presents preliminary results of long-term (one year) outdoor CO 2 concentration changes in four districts of Copenhagen (Denmark). The districts included downtown area and suburbs with different surroundings. Four buildings were selected for the measurements, one building in each district. Outdoor CO 2 concentration measurements were performed at two levels – ground level and top of the buildings. Special attention was paid to use accurate measuring instruments. The instruments were carefully calibrated before the measurements. The calibration of the instruments was checked periodically. In this paper, preliminary results from summer and autumn measurements are presented. The outdoor CO 2 concentration varied over the day and from day to day in the range between 340 and 450 ppm. The CO 2 concentration at the ground of the buildings was usually 10 to 40 ppm higher than that at the top level in autumn. At the buildings in the suburbs, during the working hours, the outdoor CO 2 concentration measured on the top level close to the intake duct was on average 408 ppm. At the building in the downtown area, that was on average 414 ppm. However, the outdoor CO 2 concentration varied depending on the building, level and time. During the working hours, the 75 percentiles of outdoor CO 2 concentration varied between 384 ppm and 442 ppm, which indicates that the required ventilation rate could be different over 10% depending on the building location site, measurement height and time. In order to ensure the required indoor limits of CO 2 concentration, CO 2 measurements must be performed close to the location of the outdoor air intake.


Introduction
The ventilation of spaces is intended primarily to provide occupants with clean air for breathing. The generated pollution (including CO 2 ) is diluted by supplying clean ventilation air, which in most of the cases is outdoor air surrounding the ventilated building. At present, indoor CO 2 concentration is used as a simple parameter to limit the level of pollution in rooms in the Danish Building Regulations 2018 (BR18) [1].
For designing the ventilation system, the outdoor CO 2 concentration is important. At constant number of occupants in a room, the ventilation rate of outdoor air required to reduce the indoor CO 2 level below the recommended maximum of 1000 ppm will depend on the CO 2 concentration in the outdoor air (Equation 1). q = G CO2 / ( C h,i -C h,o ) × 10 6 (1) Where, q is the required ventilation rate (m 3 /h); G CO2 is the CO 2 generation rate in the room at room conditions (m 3 /h); C h,i is the maximum allowed volume fraction of CO 2 in the indoor air (ppm); C h,o is the CO 2 volume fraction in the outdoor air (ppm).
BR18 suggests that the typical outdoor CO 2 concentrations are approximately 400-ppm in urban areas. However, George et al. [2], Idso et al. [3], Vogt et al. [4] and Kilkki et al. [5] have reported CO 2 levels higher than 500 ppm. Furthermore, the National Oceanic and Atmospheric Administration/ Earth System Research Laboratory (NOOA/ESRL) estimates that the recent average global CO 2 concentration is higher than 400 ppm [6]. Moreover, the outdoor CO 2 concentration is not constant; it is varying over the day and over the year [2][3][4][5]. In some cases, the variation reported over the same day was up to 60 ppm [4,7]. This variation is not considered in the present ventilation design practice, i.e. it is assumed constant outdoor CO 2 concentration of 400 ppm.
At present, there is no data on the outdoor CO 2 concentration in Copenhagen and its districts, as well as how it changes during the year seasons. The outdoor CO 2 concentration may be different in different areas of cities. Only in the most advanced demand control ventilation systems the outdoor CO 2 concentration is measured and used as one of the control parameters. This study aims to obtain reliable and accurate information on outdoor CO 2 concentration in Greater Copenhagen by conducting field measurements. This paper presents preliminary results of long-term (one year) outdoor CO 2 concentration changes in four districts of Copenhagen (Denmark).

Measuring sites
Four districts of Greater Copenhagen in Denmark, which included downtown area and suburbs with different surroundings, were selected to find the non-uniformity in outdoor CO 2 concentration. Four buildings were selected for the measurements, i.e. one building in each district. The location of the buildings is shown in Figure 1 and specified in Table 1.

Instruments
Four instruments from two manufactures were used; two identical instruments from each manufacturer. In Table 2 the instruments are identified as "Inst. T" and "Inst. G". The instruments used a nondispersive infrared method for the CO 2 measurements. The accuracy shown in Table  2 was calculated by considering the accuracy of the reference gas mixture as well as the effects of temperature, humidity, pressure and noises of output and logging on the long-stability of the measurements. The accuracy was calculated by assuming a reading value of 500 ppm CO 2 concentration. Both instruments compensate the measured data for temperature changes, and this compensation is based on built-in temperature sensors. During the measurements the relative humidity and the pressure changed in the ranges, respectively 24.  Special attention was paid to use accurate measuring instruments. The instruments were carefully calibrated in the laboratory at Technical University of Denmark at three reference CO 2 concentrations levels of 0 ppm, 500 ppm and 1001 ppm using Nitrogen mixed with CO 2 . Calibration equations were developed. The accuracy level of the gas mixture used was ± 0.5 %. During the calibration, Teflon tubes with the same length as during in the actual field measurements were used for air sampling. The calibration was conducted just before each measurement period.

Measurements
Outdoor CO 2 concentration measurements were performed at two levels -ground level and top of the buildings. Since, the target was to investigate the concentration of CO 2 in the outdoor air used for ventilation, the air sampling tube of one of the two instruments was placed very close to the intake duct of the ventilation system. The instruments T sampled air at the top level of the buildings, and the instruments G at the ground level. The sampling time was 1 minute.

Climate data
Data measured by a climate station located near the Building B (Table 1) was used as the representative data of the climate in Greater Copenhagen. The climate data were used to find the relation between the measured outdoor CO 2 concentration and outdoor temperature, wind direction and strength, etc. The climate station is located at the Department of Civil Engineering, Technical University of Denmark, at the roof of building 119 (55.79 N 12.53 W, elevation 8 m). The data can be downloaded from a website [8]. One-minute time step data was used in this study.
Specific humidity was calculated by Equations 2, 3 and 4 using the raw data of air pressure, relative humidity and air temperature. Equation 4 is known as the Tetens equation.
Where, φ s is the specific humidity (g/kg); ε is the molecular weight ratio of vapour to dry air (0.622); P is the air pressure (hPa); f is the vapour pressure (hPa); f s is the saturated vapour pressure (hPa); φ r is the relative humidity (%); T is the air temperature (°C).

Data processing
The calibration equations were used to obtain accurate outdoor CO 2 concentration based on the measurements. For the purpose of reducing random noise, the collected outdoor CO 2 concentration data (sampled each minute) and the data for the different meteorological parameters were used to calculate one hour averages. The average of wind speed was calculated by using scalar average method. Figure 2 shows one hour average of outdoor CO 2 concentration measured at Building A and B from 0:00 to 24:00 for one week (October 24 -October 31). Figure  3 shows the average CO 2 levels measured at Building C and D from 0:00 to 24:00 for one week (October 8 -October 15).

Outdoor CO 2 concentration
The outdoor CO 2 concentration measured at the top and ground level of each building, made a similar change with time, although no typical diurnal change can be seen. During these periods, the outdoor CO 2 concentration at the top tended to be around 10 to 40 ppm lower than that at the ground. The outdoor CO 2 concentration at the top varied in the range between 390 and 440 ppm and at the ground level it varied in the range between 400 and 450 ppm. Figure 4 shows one hour average of outdoor CO 2 concentration measured at the four buildings from 6:00 to 7:00 every day. Similarly, Figures 5, 6, 7 and 8 show the one hour average CO 2 concentrations for the periods

[ppm]
A  Fig. 4. One hour average of outdoor CO 2 concentration from 6:00 till 7:00 am every day.     [    Figures 4, 5, 6, 7 and 8 indicate that the outdoor CO 2 concentration was different during each day even during the same time period of the days at the same measured location. During the period from July till September, the outdoor CO 2 concentration at the top of Building B tended to be around 50 ppm higher than that at the ground. On the other hand, in October and November, the outdoor CO 2 concentration at the top tended to be around 10 to 40 ppm lower than that at the ground. In the latter half of September, the outdoor CO 2 concentration at the ground of Building A was relatively high compared to concentration measured at Building B, at both levels. The concentration measurements at the ground of Building A might have been affected by human activities near the building. The average of the outdoor CO 2 concentration in October and November seems to be higher than that in the period from July till September. Table 2 shows the statistics of one hour average of outdoor CO 2 concentration measured at the top of the buildings. Similarly, Table 3 shows the statistics of the average CO 2 levels measured at the ground of the buildings.
Differences in the CO 2 levels were observed between the top and ground levels of each building. Thus, in order to perform proper calculation of the amount of supplied outdoor ventilation air that will ensure the required indoor limits of CO 2 concentration, CO 2 measurements must be performed close to the outdoor air intake.
The average CO 2 concentration at noon tended to be lower than that in the morning and evening. This might be caused by photosynthesis of plants. The possibility will be analysed in our future study with the data in winter.
The maximum CO 2 concentration at the top tended to be lower than that at the ground. The difference between 75 percentile and 25 percentile at the top varied at the range only between 6 ppm to 18 ppm which was within the accuracy of the instruments. On the other hand, the corresponding difference at the ground varied between 14 ppm to 54 ppm. These results indicate that the outdoor CO 2 concentration at the ground might have larger change and could be higher than that at the top. We can conclude that it is better to place the intake of supply air duct for the ventilation as close as possible to the building top.  Fig. 11. Specific humidity vs outdoor CO 2 concentration at the ground of Building B. met) is assumed 19 L per hour [9]. The results in Table 4 show that a change by 50 ppm in the outdoor CO 2 concentration causes a change by approximately 10% in the required ventilation rate.
During the design of building ventilation, it is not efficient to adopt the maximum value of the outdoor CO 2 concentration because the high required ventilation rate will lead to high energy consumption of the fan, heating and cooling. But, to reduce the time that the indoor CO2 concentration exceeds the maximum allowed level, in calculations it might be better to consider not the average of outdoor CO2 concentration but the 75percentile of that as the assumed outdoor CO 2 concentration.
During the working hours from 8:00 to 18:00, the value of 75 percentiles in Tables 2 and 3 varied between 384 ppm (at 9:00 at the ground of Building B) and 442 ppm (at 9:00 at the ground of Building C), which indicates that the required ventilation rate could be different over 10% depending on the site, height and time. Table 5 shows the correlation coefficient between different climate parameters and outdoor CO 2 concentration.

Correlation with climate parameters
The wind speed was correlated negatively with the outdoor CO 2 concentration, especially at the top of the buildings as shown in Figure 9 and 10. The wind might mix the atmosphere and balance its CO 2 concentration. This tendency will be stronger in the area of higher elevation than that of lower.
The specific humidity was correlated negatively with the outdoor CO 2 concentration measured at the ground of Building B as shown in Figure 11. This supports the result reported by Marrero at al [10]. They analysed the data obtained at a building surrounded by residential houses with gardens and parks and concluded that the outdoor CO 2 concentration in dry days becomes higher compared to wet days. The high humidity during wet days boosts the photosynthesis of plants thus absorbing more CO 2 . On the other hand, as shown in Table 5, the specific humidity seemed to be correlated positively with the outdoor CO 2 concentration measured at the top of Building C, but as shown in Figure 12, the specific humidity in this building varied in small range, between 5 g/kg and 9 g/kg. This means the correlation coefficient in this case is not that indicative.

Conclusions
Long-term (one year) measurements of outdoor CO 2 concentration at the ground and the top level of four buildings in Greater Copenhagen (Denmark) have been planned to find the CO 2 variation in different areas of the city. In this paper, preliminary results from summer and autumn measurements were presented.
At buildings in suburbs, over the working hours from 8:00 to 18:00, the outdoor CO 2 concentration measured on the top level close to the intake duct was on average 408 ppm. At the building in downtown area, the corresponding concentration was on average 414 ppm. The CO2 concentration measured at the ground of the buildings was usually 10 ppm to 40 ppm higher than that at the top level in autumn. The differences were within the uncertainty of the measurements. However, the outdoor CO2 concentration variations over the day and from day to day were greater than the uncertainty of the measurements: in the range between 341 ppm at the lowest of 25 percentiles and 450 ppm at the highest of 75 percentiles. During the working hours, the value of 75 percentiles of outdoor CO 2 concentration varied between 384 ppm and 442 ppm, which indicates that the required ventilation rate could be different over 10% depending on the building location site, measurement height and time. The outdoor CO 2 concentration could be affected by the climate parameters of wind speed and humidity. Therefore, in order to perform proper calculation of the supplied ventilation rate that will ensure the required indoor limits of CO2 concentration, CO2 measurements must be performed close to the outdoor air intake.
The measurements of outdoor CO 2 concentration and analyses of results will be continued in 2019.