Next Article in Journal
The Widespread Use of Remote Sensing in Asbestos, Vegetation, Oil and Gas, and Geology Applications
Next Article in Special Issue
Seasonal and Diurnal Variability of Monoterpenes in the Eastern Mediterranean Atmosphere
Previous Article in Journal
Can Inner Mongolia Learn from Zhejiang’s Low-Carbon Policy?—Comparative Analysis Based on the EPS Model
Previous Article in Special Issue
An Experimental Method for Evaluating Ammonia Emission Rates of Bio-Curtain
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Atmosphere
Volume 14
Issue 1
10.3390/atmos14010171
atmosphere-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Emission Characteristics of Fine Particles in Relation to Precursor Gases in Agricultural Emission Sources: A Case Study of Dairy Barns

by
Hung-Soo Joo
1,
Sang-Woo Han
1,
Jin-Seok Han
1,* and
Pius M. Ndegwa
2,*
1
Department of Environmental Engineering, Anyang University, Anyang-si 14028, Gyeonggi-do, Republic of Korea
2
Department of Biosystems Engineering, Washington State University, Pullman, WA 99164, USA
*
Authors to whom correspondence should be addressed.
Atmosphere 2023, 14(1), 171; https://0-doi-org.brum.beds.ac.uk/10.3390/atmos14010171
Submission received: 30 November 2022 / Revised: 9 January 2023 / Accepted: 11 January 2023 / Published: 12 January 2023
(This article belongs to the Special Issue Ammonia Emission and Particulate Matter)
Browse Figures
Versions Notes

Abstract

:
Recently, precursor gases such as ammonia have sparked a growing interest in the secondary formation of particulate matter (PM). Most studies focus on urban areas and scientific data. Studies on precursor gases and PM emitted from agricultural sources are insufficient; thus, this paper presents a field monitoring study conducted from agricultural sources. To estimate the effect of precursor gases for PM2.5 from naturally ventilated dairy barns, correlation analyses were conducted using real-time monitoring data on the mass concentrations of PM2.5, NH3, SO2, NO2, and H2S and meteorological data. In addition to mass concentration, the emission and loading rates were used to closely analyze pollution status. The mass concentrations of PM2.5 and gaseous compounds did not correlate well, whereas the mass emission rates for PM2.5 and gaseous compounds (SO2, NH3, and NO2) correlated well because the unit of the emission rate reflected the ventilation factor. The correlation coefficients between PM2.5 and precursor gas emission rates ranged from 0.72 to 0.89 (R), with the SO2 emission rate exhibiting the highest correlation coefficient (R = 0.89). This correlation implies that SO2 from dairy farms is a dominant species among the gaseous precursors influencing the secondary formation of PM2.5; alternatively, SO2 and PM2.5 are produced from the same sources. The ambient PM2.5 loading rate and barn PM2.5 emission rate—estimated by multivariate linear regression using the gaseous independent variables NH3, SO2, and NO2—revealed high-correlation coefficients (0.60 and 0.92, respectively) with the measured data. At present, most studies investigating the precursor gases of PM in agricultural fields have focused on NH3; however, this study suggests that SO2 is a key factor in PM2.5 pollution. To elucidate the secondary formation of PM from precursor gases in agricultural sources, particulate ammonium, sulfate, nitrate, and chloride, which were not measured in this study, as well as oxidants and intermediates, should be considered in future research.

1. Introduction

Fine particles with an aerodynamic diameter of less than 2.5 microns (PM2.5) affect air pollution, climate change [1,2,3,4], visibility [1,2], and human health [3,4]. Fine particles are not only directly emitted from primary sources, such as automobiles, biomass burning, and various industrial combustion processes, but are also indirectly formed through gas-to-particle conversion in the atmosphere from gaseous compounds, such as ammonia (NH3), sulfur dioxide (SO2), and nitric oxides (NOx) [5,6,7], in the presence of oxidants via photochemical reactions [5,7]. Numerous studies have reported that ammonia influences secondary particle formation as a precursor of ammonium (NH4+) salts [2,3,7,8,9,10,11,12,13,14]. Some studies have demonstrated that sulfur dioxide and nitric oxides, i.e., precursors of sulfate (SO42−) and nitrate (NO3−) salts, are key factors in secondary formation rather than ammonia [10,15].
Agricultural sources, such as livestock building, manure management, and land application processes, not only significantly contribute to global ammonia emissions [8,9,16], but are also emission sources of precursor gases, such as SO2, nitric oxides (NOx), and volatile organic compounds (VOCs) [13]. In animal feeding operation (AFO) facilities, ammonia is mainly released from urea (CO(NH2)2), which is produced by the hydrolysis of urine by urease during microbial metabolism in feces [13,17]. Sulfur dioxide and NOx in AFOs are mainly produced from feces, i.e., from sulfur-containing protein decay and denitrification processes [13,18]. The formation of ammonium and sulfate salts in AFOs is positively correlated with solar radiation and temperature, and nitrate is correlated with solar radiation [19]. These findings imply that solar irradiance and temperature are important factors that drive photochemical reactions and convert gas precursors into fine particulates. Hydrogen sulfide (H2S), which is abundant in AFOs, plays an important role in the atmospheric budget of sulfur compounds, odor complaints, and influences the formation of aerosol sulfate via oxidation in the air [20]. Although many studies have been carried out on the role of precursor gases in secondary aerosol formation in the agricultural environment, the pathway of secondary formation of fine particles from those precursor gases is still not fully elucidated.
Major components of PM2.5 emitted from AFOs are composed of sulfate, nitrate, ammonium, and organic carbon (OC), followed by elemental carbon (EC), crustal elements, and heavy metals [11,13,15]. Inorganic PM2.5 mainly exists as ammonium sulfate and bisulfate and ammonium nitrate, respectively [8,9,15,21]. To clarify the formation mechanism of secondary inorganic PM2.5, Li et al., (2014) conducted correlation analyses between ammonium ions and sulfate or nitrate ions in PM2.5. Ammonium has high correlations with sulfate or nitrate [15,19]. The molar ratio of ammonium and sulfate ions was investigated to seek the major inorganic species in PM2.5 in AFOs [8,9,11,13]. Particulate ions, such as ammonium, sulfate, nitrate, and chloride, and gaseous NH3, emitted from livestock industries, are occasionally studied; however, monitoring studies of gaseous SO2 and NO2 emitted from livestock industries are rare. Recently, Wang et al., (2015) conducted a monitoring study in a Chinese layer house to verify the relationship between PM2.5 and gaseous precursors (NH3, NOx, and SO2); however, the correlation between those gases and PM2.5 concentrations was not significant [22]. Generally, the PM2.5 concentration should be highest during the daytime due to human and industrial activities, solar irradiance, temperature (secondary formation), etc. [23,24]. Unpredictably, many studies have shown that PM2.5 mass concentrations in the daytime are lower than that at night because the planetary boundary layer increases with increasing temperature during the daytime [1,25,26,27]. The use of distinguished units from PM concentration can be effective for explaining these inexplicable results. PM2.5 emission and loading rates, which include the ventilation rate, have been suggested as powerful units for interpreting more clearly the PM2.5 pollution status, as indicated in our previous study [23].
In this study, we carried out a correlation analysis using real-time monitoring data between PM2.5 and its precursor gases measured at naturally ventilated dairy barns. Data on mass concentrations, emission rates, and loading rates for PM2.5, NH3, SO2, NO2, and H2S inside and outside the barn were used in the correlation analysis to estimate key precursors for PM2.5 emissions at this dairy production site. In addition, correlations between PM2.5 and meteorological parameters, such as temperature, relative humidity, solar irradiance, and wind velocity, were also performed. Finally, we reproduced estimated barn PM2.5 emission rates and ambient PM2.5 loading rates using multi-linear regressions with respect to its precursor gases.

2. Experimental Methods

2.1. Monitoring Site

This study was conducted in a pair of naturally ventilated free-stall dairy barns (Mabton, WA, USA). Barn 1 (183 m × 30 m) housed approximately 650 cows and barn 2 (214 m × 39 m) housed approximately 850 cows, respectively. Barns 1 and 2 had 4 and 6 lines for free stalls, respectively, and the width of each stall was 1.32 m. The dairy farms had two vertical curtain systems at the north and south walls, and all curtains were fully open in the summer season (2.7 m maximum opening). The openings in the east and west walls and roof were 4.27, 4.27, and 1.6 m, respectively. Each cow was milked 3 times a day at the milking parlor and spent approximately 4 h outside the barn. Excreted manure in the barn was flushed every six hours via flushing water from the lagoon. Additional details and information on the study site were provided in a previous study [28,29,30,31]. Potential emission sources of PM and gaseous compounds in these dairy buildings—manure decay, animal and machine activities, feeds, soil beddings, dusts from outside, secondary aerosol formation, etc.—were thought to be derived from the animals themselves.

2.2. Measurement Instruments

As shown in Figure 1, three measurement points—two points inside the two barns and one point outside the two barns—were operated in real-time for the monitoring of PM2.5, NH3, SO2, NO2, and H2S. To monitor the corresponding ventilation data, 16 three-dimensional ultrasonic anemometers (Model RM Young 81000, RM Young Company, Traverse City, MI, USA) were installed at the doors of each barn. Gas sampling ports for the monitoring of NH3, SO2, NO2, and H2S were located at the same points with the anemometers. To monitor the meteorological parameters, a pyranometer (Model LI-200SL, LiCOR, Lincoln, NE, USA), a relative humidity–temperature probe (NOVUS Model RHT-WM, NOVUS Electronics, Canoas, Brazil) and a wind anemometer (Model 03002VM Wind Sentry, RM Young Company, MI, USA) were installed on the weather tower on the roof of barn 1. Continuous gas sampling was carried out using a gas sampling system (GSS), which consisted of 10 solenoids placed in the on-farm instrument shelter (OFIS). The ammonia concentration was measured using a photoacoustic IR multigas monitor (Model 1412, Innova AirTech Instruments, Ballerup, Denmark, 0.2 ppm of LDL); sulfur dioxide and hydrogen sulfide concentrations were measured using a fluorescence-based analyzer (Model 450i, Thermo Fisher Scientific, Waltham, MA, USA, 1 ppb of LDL); and the NO2 concentration was measured using a chemiluminescence NH3 analyzer (TEC Model 17C, Thermo Environmental Instruments, Franklin, MA, USA) composed of a combination NH3 converter and an NO-NO2-NOX analyzer (1 ppb of LDL). The NH3 concentrations measured by this instrument were compared with the concentration measured by the Innova Model 1412, which was used as an official analyzer for NH3 in the National Air Emission Monitoring Study (NAEMS) project [28]. The barn PM2.5 mass concentrations were monitored using two TEOMs (tapered element oscillating microbalances, Model 1400a, Rupprecht & Patashnick, Albany, NY, USA, 0.6 µg/m3 of LDL) inside the barn, while ambient PM2.5 concentrations were monitored using a beta gauge (Model FH62C14, Thermo Electron Corporation, Waltham, MA, USA, 0.6 µg/m3 of LDL). Periodical maintenance, such as a zero-span check and multi-point calibration, was conducted according to the manufacturers’ recommendations during the monitoring period [28,29].

2.3. Sampling and Data Acquisition

The average and standard deviations of PM2.5, NH3, SO2, NO2, H2S, and meteorological parameters during the measurement period are presented in Table 1. The weather conditions of the monitoring site were hot and dry in the summer and cold and wet in the winter seasons [30,31]. Previous studies showed that a higher amount of air pollutants was usually emitted in the summer season [23,30,31]. Because the measurement PM2.5 was conducted between 9 July and 21 July 2009, the presented data were limited to 13 days in this report. Emission and loading rates were calculated by multiplying each concentration and barn ventilation rate or ambient wind speed, respectively. Continuous measurements of PM2.5 were taken at two points in the barns and at one point on the outside of the barn (ambient air), while meteorological parameters were continuously recorded at the weather tower located on the rooftop of barn 1. Measurements of gases were performed on a sequential rotation at 10 sampling locations (SLs) in the two barns (5 locations for each barn: SLs 1, 3, 5, 7, 9 for barn 1 and SLs 2, 4, 6, 8, 10 for barn 2) and were conducted for 10 min at each sampling location. It took 100 min to measure all 10 sampling locations in the two barns [29]. A desktop computer located in OFIS acquired real-time measurements data using an AirDAC software [28]. Once a day, the collected data were saved on the site computer and transmitted to a data scientist’s desktop computer. Site design, sampling schedule, sampling layout, data acquisition, periodical maintenance, etc., were conducted under the protocols provided by the NAEMS project [28,29,30,31].

2.4. Data Processing

In this study, the averages of 30 min of data obtained from the raw data collected every minute were used for analysis in this paper. For example, the 30-min average data for gaseous compounds meant that the data comprised an average of two measurements in barn 1 and one measurement in barn 2 (SLs 1, 2, and 3). Then, the next 30-min data comprised an average of one measurement in barn 1 and two measurements in barn 2 (SLs 4, 5, and 6). This process was repeated with subsequent 30-min data. The 30-min average data of precursor gases and PM2.5 for the two barns were integrated and merged by considering the weighted averages of the barn ventilation rates. Outliers were deemed to be any data lying outside of the interquartile range (1.5 × IQR) for each dataset and were not used in subsequent data analyses [31,32,33]. The 30-min means (48 datasets per day) for the entire 13 days of the measurement campaign were used in diurnal pattern, correlation, and linear regression analyses and principal component analysis (PCA). Multivariate linear regression and PCA were performed using IBM SPSS Statistics version 20 software.

3. Result and Discussion

3.1. Summary of Measurement Data

Table 1 shows the average concentrations and emission rates of PM2.5 and precursor gases, as well as the average values of meteorological parameters during the monitoring period. The PM2.5 concentrations at ambient temperature and in the barns were approximately 31 and 70 μg/m3, respectively, which were much higher than the U.S. daily PM2.5 standard [34]. The PM2.5 emission rates of barn 1 (emission factor (EF): 0.17 g/cow/h) were slightly higher than those of barn 2 (EF: 0.14 g/cow/h). The emission rates of NH3, H2S, SO2, and NO2 were 29 g/min (EF: 2.32 g/cow/h), 1.0 g/min (EF: 0.08 g/cow/h), 0.2 g/min (EF: 0.016 g/cow/h), and 3.7 g/min (EF: 0.30 g/cow/h), respectively.
The average wind speed, relative humidity, and temperature were 2.2 m/s, 32%, and 27.6 °C, respectively. Zou et al., (2020) reported that the emission rates from naturally ventilated dairy barns in North China were approximately 1.52 g/AU/h (1 to 8 ppm) for ammonia, 0.001 g/AU/h for hydrogen sulfide, and 0.002 g/AU/h for sulfur dioxide, respectively [35]. In another study from Canada, the emission rates from a dairy barn were approximately 26.5 to 97.5 mg/min (2.1 to 7.8 mg/AU/h) for PM2.5 and 0.45 to 1.78 g/AU/h (1.9 to 3.1 ppm) for ammonia, respectively [36]. Ammonia emissions reported in dairy barns in previous studies and our study were similar, whereas the emission levels of PM2.5, sulfur dioxide, and hydrogen sulfide varied. These differences are probably due to differences in farm management, monitoring methods, feed, surrounding environment, and other factors.

3.2. Diurnal Patterns of Fine Particles and Gaseous Pollutant Concentrations

Figure 2 shows the diurnal profiles of average PM2.5, SO2, NH3, and NO2 concentrations in the two barns. The PM2.5 concentrations ranged from 55 to 75 μg/m3 and were usually higher in the morning. The SO2 concentrations ranged from 1.3 to 2.2 ppb and showed lower levels at 06:00 to 09:00 in the morning, at 13:00 to 14:30 in the afternoon, and at 22:00 to 24:00 in the nighttime. The NH3 concentrations ranged from 0.7 to 1.4 ppm and were higher during the nighttime than during the daytime. The NO2 concentrations ranged from 0.015 to 0.075 ppm and were higher in the morning than in the afternoon, similar to the PM2.5 concentration pattern. The diurnal variations of the PM2.5 and SO2 concentrations were not as significant as those of the NH3 and NO2 concentrations. The pattern of the NH3 concentrations was inversely proportional to that of the NO2 concentrations, which most likely meant that the emission source or formation mechanisms of the two gases were not the same or they were emitted at different times. Bjerg et al., (2012) reported that NH3 emissions were higher in the morning (before noon) because animals urinated more at this time [37]. NH3 emissions were lower at night because of reduced feeding and defecation [35]. Other studies also showed that ammonia concentrations during the daytime were lower than those during the nighttime. This pattern was also frequently observed in urban areas and can be explained by the increased planetary boundary layer (PBL, atmospheric mixing layer of air pollutants) caused by surface radiation during the daytime [1,26,27]. In general, wind fields affect concentration because air pollutants emitted from various sources are diluted and mixed into ambient air. Therefore, the interpretation of pollution status is sometimes very difficult and can yield unusual results (a limitation of concentration unit).
Figure 3 shows the correlations between PM2.5 concentrations and SO2, NH3, and NO2 concentrations, respectively. Although NH3, SO2, and NO2 are precursor gases for the formation of secondary aerosols [2,5,7,8,9,14,20], the correlations between PM2.5 and the concentrations of these gaseous compounds were not statistically significant. Wang et al., (2015) reported that correlations between PM2.5 and NOx, SO2, and NH3 concentrations were statistically significant, finding the greatest correlation between PM2.5 and NOx [22]. As shown in Figure 3, no correlations were found between PM2.5 and precursor gases, suggesting that further analysis of the field monitoring data using other units, which is more in-depth than the concentration unit, should be conducted.

3.3. Diurnal Profiles of Fine Particles and Gaseous Compound Concentrations, Emission Rates, and Loading Rates

The diurnal profiles of PM2.5 concentration and emission and loading rates in the barns and in ambient air (outside the barn) are shown in Figure 4. In our previous study, the PM2.5 emission and loading rates were suggested as parameters that could be used to effectively interpret or determine the PM2.5 pollution status [23]. The diurnal variations in the PM2.5 emission rates and ambient PM2.5 loading rates in the barns were more significant than the variations in concentration data (Figure 2), which were higher during the daytime than the nighttime. The average concentrations inside or outside the barns showed similar levels throughout the day (60–75 and 30–40 μg/m3, respectively), whereas barn emission rates and ambient loading rates were widely distributed (1–4 g/min and 2–6 μg/m3/min, respectively), with the highest peaks at 15:00 in the barns and 18:00 outside the barns. The units of the emission and loading rates were calculated by multiplying mass concentration and ventilation rate or wind velocity. At the monitoring site, wind velocities during the daytime had much higher values than those during the nighttime (data not shown).
The diurnal patterns of precursor gas emission rates from the barns are shown in Figure 5. In contrast to the patterns in Figure 2, the SO2, NH3, and NO2 emission rates were significantly higher during the daytime than the nighttime, similar to the PM2.5 emission rate profile (Figure 4 upper), but the H2S emission rates did not exhibit a similar pattern with other precursor gases. The NH3 emission rate increased during the daytime, which was consistent with previous research [15,35,37,38]. When we identify the emission and pollution status, the use of emission or loading rates, which includes wind fields, can be more effective than the use of a concentration unit.

3.4. Correlation Analyses between Fine Particles and Gaseous Compounds

Figure 6 shows the correlations between the PM2.5 emission rates and gaseous emission rates from the barns. NH3, SO2, and NO2 emission rates correlated well with the PM2.5 emission rates (R = 0.72–0.89) and the p values were significant (p < 0.05), with the SO2 emission rates having the highest correlation (R = 0.89). The correlation between H2S and PM2.5 emission rates was not significant (p > 0.05). Li et al., (2014) reported that aerosol (PM2.5) mass concentration correlated well with sulfate (R = 0.70) and H2SO4, whereas gaseous NH3, nitrate, and chloride salts did not correlate well with aerosol mass concentrations in the livestock facility. In their model simulation, the reduction in sulfate and H2SO4 significantly influenced the reduction in PM, rather than the reduction in NH3 [10,15]. These findings suggested that both gaseous and particulate sulfur compounds would be a dominant species for PM pollution in livestock facilities. In many studies, ammonia has been recognized as the key precursor gas of secondary PM formation in agricultural sources. This finding suggested that SO2 emissions can also play an important role in PM pollution in agricultural areas.
Figure 7 shows the correlation between barn PM2.5 emission rates and meteorological parameters. Solar irradiance, wind velocity, and temperature were directly proportional to the barn PM2.5 emission rates (significant p values < 0.05) with solar irradiance showing the highest correlation (R = 0.87). Numerous studies reported that solar radiation strongly influences the formation of secondary aerosols because solar radiation mediates the photochemical reactions between precursor gases with O3, OH, and H2O2, as well as NO in the atmosphere; thus, an increase in solar radiation increases the formation of secondary particulate matter [5,39,40,41]. However, the PM2.5 emission rates from the barns were inversely proportional to relative humidity. In general, temperature is inversely correlated with relative humidity, whereas temperature is positively correlated with wind speed [1]. Zou et al., (2020) reported that SO2 and NH3 emissions from Chinese dairy buildings were increased with increasing temperature and ventilation rates during the daytime [35]. The higher PM2.5 emission rate during the daytime might be caused by the increasing SO2 and NH3 emissions in this study. This is because animal and machine activities, manure excretion, manure decay, and the resuspension of fine particles from soil beddings might increase during the daytime. Secondary aerosol formation by gaseous precursors could also progress during the daytime.
Figure 8 shows the principal component analysis (PCA) using 13 variables to evaluate the relationship between PM2.5 emission or loading rates and gas emission rates or meteorological parameters. Although PCA results do not prove causation, many scientists employ PCA to assess associations when performing a correlation analysis involving multiple variables [42,43,44]. Relative humidity (BarnRH and AmbRH) and barn H2S emission rates (BarnH2S) were separated by component 1 and 2 with negative loadings. The PM2.5 emission rate (BarnPM2.5) and loading rate (AmbPM2.5) were grouped with barn SO2 emission rate (BarnSO2), temperature (BarnT and AmbT), ambient solar irradiance (AmbSolar), ambient wind velocity (AmbWV), barn NH3 emission rate (BarnNH3), barn NO2 emission rates (BarnNO2), and barn animal activity (BarnAct). BarnPM2.5 was closely clustered with BarnSO2, BarnT, AmbT, AmbSolar, and AmbPM2.5, while AmbPM2.5 was clustered with AmbWV, BarnSO2, and BarnNH3. Barn SO2 emission rate was highly correlated with barn PM2.5 and ambient PM2.5, which coincided with the correlation shown in Figure 7. The ambient PM2.5 loading rate was close to the emissions of SO2 and NH3, indicating that ammonium sulfate is likely to be a dominant secondary PM in ambient air, based on this PCA result. Both particulate ion salts (ammonium, sulfate, nitrate, and chloride) and intermediates (HNO3, HONO, and H2SO4) should be measured to identify this secondary PM formation in further studies.

3.5. Multivariate Linear Regressions between PM2.5 and Its Precursor Gases

To elucidate the relationship between PM2.5 and its precursor gases, multivariate linear regression was conducted, with the dependent variable of PM2.5 emission or loading rates and three independent variables (NH3, SO2, and NO2 emission rates), as shown in Equation (1). Table 2 and Table 3 show the results for barn PM2.5 emission rate and ambient PM2.5 loading rate, respectively. Both regressions showed significant p values and regression coefficients (R = 0.93 for the barn PM2.5 emission rate and R = 0.67 for the ambient PM2.5 loading rate, respectively). The slopes (B-values) between the PM2.5 emission rate and NH3, SO2, and NO2 emission rates were positive, whereas the B-values between PM2.5 and NO2 emission rates were negative in the regression for ambient PM2.5 loading rate. In both regressions, the T-value (coefficient divided by standard error) for the SO2 emission rate was highest amongst the three gases, implying that the barn SO2 emission rate had the most significant influence on PM2.5 emissions.
y (PM2.5) = A·x1 (NH3) + B·x2 (SO2) + C·x3 (NO2) + Intercept
The estimated ambient PM2.5 loading rates and barn PM2.5 emission rates were reproduced using an empirical linear regression equation (Table 2 and Table 3) and compared with the measured data (Figure 9). The p-values in both regressions were significant (less than 0.05). The estimated ambient PM2.5 loading rates underestimated the direct measurements (slope = 0.40 and R = 0.60). However, the estimated barn PM2.5 emission rates were comparable to the direct measurements (slope = 0.85 and R = 0.92). The latter regression was more significant for the estimations of PM2.5 emissions from measurements of SO2, NH3, and NO2 in dairy barns because the gaseous compounds were measured only in the barn (not in the ambient site) in this study. In addition, regarding the reduction in PM2.5 emissions from the livestock facility, these regressions clearly revealed that the emission of these precursor gases (especially SO2 emissions) must be controlled or managed.

4. Conclusions

To estimate the key precursors of PM2.5 emissions, the concentrations and emission rates of NH3, SO2, and NO2, as well as meteorological parameters, were analyzed for correlations with the PM2.5 concentrations, emission rates, and loading rates at the dairy site. No statistically significant differences between concentration data were observed, whereas NH3, SO2, and NO2 emission rates correlated well with the PM2.5 emission rates in the correlation analysis, with correlation coefficients ranging from 0.72 to 0.89 in the PCA. SO2 (R = 0.89) was the most significant species among three precursor gases influencing PM2.5 emissions in the livestock facility. This result suggests that the management of sulfur-containing compounds (SO2 and H2S) in agricultural sources is important from the standpoint of PM emission control and odor control. Solar irradiance, wind speed, and temperature were positively correlated with the PM2.5 emission rates. Solar irradiance showed the highest correlation coefficient (R = 0.87), whereas relative humidity was negatively correlated with PM2.5 emissions. In the multivariate linear regressions, barn the PM2.5 emission rates and ambient PM2.5 loading rates could be reproduced using the precursor gases SO2, NO2, and NH3, and the correlation coefficient of the estimated barn PM2.5 emission rates correlated well with the measured barn PM2.5 emission rates (R = 0.92).

Author Contributions

Conceptualization, P.M.N., H.-S.J. and J.-S.H., Data curation, S.-W.H.; Correlation analysis, H.-S.J. and J.-S.H.; Supervision, P.M.N.; Writing—original draft, H.-S.J. and P.M.N.; Writing—editing, H.-S.J. and J.-S.H. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Cooperative Research Program for Agriculture Science & Technology Development (Project No. PJ017075), the Rural Development Administration of Korea, the FRIEND (Fine Particle Research Initiative in East Asia Considering National Differences) Project through the NRF and Ministry of Science and ICT (No. 2020M3G1A1114999), the Korea Ministry of Environment as R&D Project for Management of Atmosphere environment Project (No. 2021003410002), and the Graduate School of Particulate Matter Specialization of Anyang University.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

Nomenclature

AFOsAnimal feeding operations
AmbPM2.5Ambient PM2.5 loading rate
AmbRHAmbient relative humidity
AmbSolarAmbient solar irradiation
AmbTAmbient temperature
AmbWVAmbient wind velocity
AUAnimal unit
BCoefficients for independent variables
BarnActAnimal activity in barn
BarnNH3NH3 emission rate from barn
BarnNO2NO2 emission rate from barn
BarnPM2.5PM2.5 emission rate from barn
BarnRHRelative humidity in barn
BarnSO2SO2 emission rate from barn
BarnTTemperature in barn
CO(NH2)2Urea
DACData acquisition and control
ECElemental carbon
g/cow/hGram per cow per hour (cow·h)
GSSGas sampling system
H2SHydrogen sulfide
H2SO4Sulfuric acid
HNO3Nitric acid
HONONitrous acid
IQRInterquartile range
LDLLower detection limit
μg/m3/minMicrogram per m3 minute (m3·minute)
NH3Ammonia
NH4+Ammonium ion
NO3−Nitrate ion
NOxNitric oxides
OCOrganic carbon
OFISOn-farm instrument shelter
PBLPlanetary boundary layer
PCAPrincipal component analysis
PMParticulate matter
PM2.5Fine particles with an aerodynamic diameter of less than 2.5 micron
SDStandard deviation
SLsSampling locations
SO2Sulfur dioxide
SO42−Sulfate ion
TCoefficient divided by standard error
VOCsVolatile organic compounds

References

  1. Zhao, X.; Zhang, X.; Xu, X.; Xu, J.; Meng, W.; Pu, W. Seasonal and diurnal variations of ambient PM2.5 concentration in urban and rural environments in Beijing. Atmos. Environ. 2009, 43, 2893–2900. [Google Scholar] [CrossRef]
  2. Wang-Li, L. Insights to the formation of secondary inorganic PM2.5: Current knowledge and future needs. Int. J. Agric. Biol. Eng. 2015, 8, 1–13. [Google Scholar] [CrossRef]
  3. Pozzer, A.; Tsimpidi, A.P.; Karydis, V.A.; de Meij, A.; Lelieveld, J. Impact of agricultural emission reductions on fine-particulate matter and public health. Atmos. Chem. Phys. 2017, 17, 12813–12826. [Google Scholar] [CrossRef] [Green Version]
  4. Foy, B.D.; Schauer, J.J. Changes in speciated PM2.5 concentrations in Fresno, California, due to NOx reductions and variations in diurnal emission profiles by day of week. Sci. Anth. 2019, 7, 45. [Google Scholar] [CrossRef] [Green Version]
  5. Hodan, W.M.; Barnard, W.R. Evaluating the Contribution of PM2.5 Precursor Gases and Re-Entrained Road Emissions to Mobile Source PM2.5 Particulate Matter Emissions; MACTEC: Research Triangle Park, NC, USA, 2004. Available online: http://www3.epa.gov/ttnchie1/conference/ei13/mobile/hodan.pdf (accessed on 11 December 2017).
  6. Choi, S.Y.; Park, S.W.; Byun, Y.J.; Han, Y.J. Characteristics of locally occurring high PM2.5 concentration episodes in a small city in South Korea. Atmosphere 2021, 12, 86. [Google Scholar] [CrossRef]
  7. Wang, S.; Xing, J.; Jang, C.; Zhu, Y.; Fu, J.S.; Hao, J. Impact assessment of ammonia emissions on inorganic aerosols in East China using response surface modelling. Environ. Sci. Technol. 2011, 45, 9293–9300. [Google Scholar] [CrossRef]
  8. Hristov, A.N. Technical note: Contribution of ammonia emitted from livestock to atmospheric fine particulate matter (PM2.5) in the United States. J. Dairy Sci. 2011, 94, 3130–3136. [Google Scholar] [CrossRef]
  9. Hristov, A.N.; Hanigan, M.; Cole, A.; Todd, R.; McAllister, T.A.; Ndegwa, P.M.; Rotz, A. Ammonia emissions from dairy farms and beef feedlots. Can. J. Anim. Sci. 2011, 91, 1–35. [Google Scholar] [CrossRef] [Green Version]
  10. Li, Q.F.; Wang-Li, L.; Shah, S.B.; Jayanty, R.K.M.; Bloomfield, P. Ammonia concentrations and modeling of inorganic particulate matter in the vicinity of an egg production facility in Southeastern USA. Environ. Sci. Pollut. Res. 2014, 21, 4675–4685. [Google Scholar] [CrossRef]
  11. Liang, D.; Ma, X.; Zhang, J.; Liu, Z.; Wu, J.; Feng, Y.; Zhang, Y. Chemical analysis of particulate matter in the harvest period in an agricultural region of eastern China. Aerosol Air Qual. Res. 2017, 17, 2381–2389. [Google Scholar] [CrossRef]
  12. Ye, Z.; Guo, X.; Cheng, L.; Cheng, S.; Chen, D.; Wang, W.; Liu, B. Reducing PM2.5 and secondary inorganic aerosols by agricultural ammonia emission mitigation within the Beijing-Tianjin-Hebei region, China. Atmos. Environ. 2019, 219, 116989. [Google Scholar] [CrossRef]
  13. Dai, P.; Shen, D.; Tang, Q.; Huang, K.; Li, C. PM2.5 from a broiler breeding production system: The characteristics and microbial community analysis. Environ. Pollut. 2020, 256, 113368. [Google Scholar] [CrossRef]
  14. Meng, Z.; Wu, L.; Xu, X.; Xu, W.; Zhang, R.; Jia, X.; Liang, L.; Miao, Y.; Cheng, H.; Xie, Y.; et al. Changes in ammonia and its effects on PM2.5 chemical property in three winter seasons in Beijing, China. Sci. Total Environ. 2020, 749, 142208. [Google Scholar] [CrossRef]
  15. Cheng, B.; Wang-Li, L. Responses of Secondary Inorganic PM2.5 to precursor gases in an ammonia abundant area in North Carolina. Aerosol Air Qual. Res. 2019, 19, 1126–1138. [Google Scholar] [CrossRef]
  16. Aneja, V.P.; Schlesinger, W.H.; Erisman, J.W. Effects of agriculture upon the air quality and climate: Research, policy, and regulations. Environ. Sci. Technol. 2009, 43, 4234–4240. [Google Scholar] [CrossRef] [Green Version]
  17. Hao, C.; Pan, Y.; Zhang, Z.; Zeng, Y. Kinetic determination of urease activity in fresh pig feces and slurry and the effect on ammonia production at different conditions. Sustainability 2019, 11, 6396. [Google Scholar] [CrossRef] [Green Version]
  18. Joo, H.S.; Hirai, M.; Shoda, M. Piggery wastewater treatment using Alcaligenes faecalis strain No. 4 with heterotrophic nitrification and aerobic denitrification. Water Res. 2006, 40, 3029–3036. [Google Scholar] [CrossRef]
  19. Li, Q.F.; Wang-Li, L.; Liu, Z.; Jayanty, R.K.M.; Shah, S.B.; Bloomfield, P. Major ionic compositions of fine particulate matter in an animal feeding operation facility and its vicinity. J. Air Waste Manag. Assoc. 2014, 64, 1279–1287. [Google Scholar] [CrossRef] [Green Version]
  20. Feilberg, A.; Hansen, M.J.; Liu, D.; Nyord, T. Contribution of livestock H2S to total sulfur emissions in a region with intensive animal production. Nat. Commun. 2017, 8, 1069. [Google Scholar] [CrossRef] [Green Version]
  21. USEPA. Air Quality Criteria for Particulate Matter; EPA (United States Environmental Protection Agency): Washington, DC, USA, 2004; Volume I.
  22. Wang, Y.; Niu, B.; Ni, J.Q.; Xue, W.; Zhu, Z.; Li, X.; Zou, G. New insights into concentrations, sources and transformations of NH3, NOx, SO2 and PM at a commercial manure-belt layer house. Environ. Pollut. 2020, 262, 114355. [Google Scholar] [CrossRef]
  23. Joo, H.S.; Park, K.; Lee, K.; Ndegwa, P.M. Mass concentration coupled with mass loading rate for evaluating PM2.5 pollution status in the atmosphere: A case study based on dairy barns. Environ. Pollut. 2015, 207, 374–380. [Google Scholar] [CrossRef] [PubMed]
  24. Chen, S.; Cui, K.; Yu, Y.T.; Chao, H.R.; Hsu, Y.C.; Lu, I.C.; Arcega, R.D.; Tsai, M.H.; Lin, S.L.; Chao, W.C.; et al. A big data analysis of PM2.5 and PM10 from low cost air quality sensors near traffic areas. Aerosol Air Qual. Res. 2019, 19, 1721–1733. [Google Scholar] [CrossRef] [Green Version]
  25. Monks, P.; ApSimon, H.; Carruthers, D.; Carslaw, D.; Derwent, D.; Harrison, R.; Laxen, D.; Stedman, J. Fine Particulate Matter (PM2.5) in the United Kingdom, Department for Environment, Food and Rural Affairs. 2012. Available online: https://uk-air.defra.gov.uk/assets/documents/reports/cat11/1212141150_AQEG_Fine_Particulate_Matter_in_the_UK.pdf (accessed on 20 August 2020).
  26. Zhang, Y.L.; Cao, F. Fine particulate matter (PM2.5) in China at a city level. Sci. Rep. 2015, 5, 14884. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  27. Lee, J.; Hong, J.W.; Lee, K.; Hong, J.; Velasco, E.; Lim, Y.J.; Lee, J.B.; Nam, K.; Park, J. Ceilometer monitoring of boundary-layer height and its application in evaluating the dilution effect on air pollution. Bound-Layer Meteorol. 2019, 172, 435–455. [Google Scholar] [CrossRef] [Green Version]
  28. Ni, J.Q.; Heber, A.J. An on-site computer system for comprehensive agricultural air quality research. Comput. Electron. Agric. 2010, 71, 38–49. [Google Scholar] [CrossRef]
  29. Joo, H.S.; Ndegwa, P.M.; Heber, A.J.; Ni, J.Q.; Bogan, B.W.; Cortus, E.L.; Ramirez-Dorronsoro, J.C. A direct method of measuring gaseous emissions from naturally ventilated dairy barns. Atmos. Environ. 2014, 86, 176–186. [Google Scholar] [CrossRef]
  30. Joo, H.S.; Ndegwa, P.M.; Heber, A.J.; Ni, J.Q.; Bogan, B.W.; Ramirez-Dorronsoro, J.C.; Cortus, E.L. Particulate matter dynamics in naturally ventilated freestall dairy barns. Atmos. Environ. 2013, 69, 182–190. [Google Scholar] [CrossRef]
  31. Joo, H.S.; Ndegwa, P.M.; Heber, A.J.; Ni, J.Q.; Bogan, B.W.; Ramirez-Dorronsoro, J.C.; Cortus, E.L. Greenhouse gas emissions from naturally ventilated freestall dairy barns. Atmos. Environ. 2015, 102, 384–392. [Google Scholar] [CrossRef]
  32. Li, L.; Wu, J.; Ghosh, J.K.; Ritz, B. Estimating spatiotemporal variability of ambient air pollutant concentrations with a hierarchical model. Atmos. Environ. 2013, 71, 54–63. [Google Scholar] [CrossRef] [Green Version]
  33. Sancho, J.; Martínez, J.; Pastor, J.J.; Taboada, J.; Piñeiro, J.I.; García-Nieto, P.J. New methodology to determine air quality in urban areas based on runs rules for functional data. Atmos. Environ. 2014, 83, 185–192. [Google Scholar] [CrossRef]
  34. USEPA. Revised Air Quality Standards for Particle Pollution and Updates to the Air Quality Index (AQI); Office of Air Quality Planning and Standards: Research Triangle Park, NC, USA, 2012; EPA 454/R99-010.
  35. Zou, B.; Shi, Z.; Du, S. Gases emissions estimation and analysis by using carbon dioxide balance method in natural-ventilated dairy cow barns. Int. J. Agric. Biol. Eng. 2020, 13, 41–47. [Google Scholar] [CrossRef]
  36. Mali, D.J. Measurement of Ammonia, Methane and Particulate Matter Emissions from a Dairy Barn; The University of Guelph: Guelph, OA, Canada, 2013. [Google Scholar]
  37. Bjerg, B.; Zhang, G.; Madsen, J.; Rom, H.B. Methane emission from naturally ventilated livestock buildings can be determined from gas concentration measurements. Environ. Monit. Assess. 2012, 184, 5989–6000. [Google Scholar] [CrossRef]
  38. Manning, M.I.; Martin, R.V.; Hasenkopf, C.; Flasher, J.; Li, C. Diurnal patterns in global fine particulate matter concentration. Environ. Sci. Technol. 2018, 5, 687–691. [Google Scholar] [CrossRef]
  39. Hsu, Y.C.; Lai, M.H.; Wang, W.C.; Chiang, H.L.; Shieh, Z.X. Characteristics of water-soluble ionic species in fine (PM2.5) and coarse particulate matter (PM10–2.5) in Kaohsiung, Southern Taiwan. J. Air Waste Manag. Assoc. 2008, 58, 1579–1589. [Google Scholar] [CrossRef]
  40. Yang, K.; Kong, L.; Tong, S.; Shen, J.; Chen, L.; Jin, S.; Wang, C.; Sha, F.; Wang, L. Double High-Level Ozone and PM2.5 co-pollution episodes in Shanghai, China: Pollution characteristics and significant role of daytime HONO. Atmosphere 2021, 12, 557. [Google Scholar] [CrossRef]
  41. Wang, Y.; Chen, J.; Wang, Q.; Qin, Q.; Ye, J.; Han, Y.; Li, L.; Zhen, W.; Zhi, Q.; Zhang, Y.; et al. Increased secondary aerosol contribution and possible processing on polluted winter days in China. Environ. Pollut. 2019, 127, 78–84. [Google Scholar] [CrossRef]
  42. Parsons, K.J.; Cooper, W.J.; Albertson, R.C. Limits of principal components analysis for producing a common trait space: Implications for inferring selection, contingency, and chance in evolution. PLoS ONE 2009, 4, e7957. [Google Scholar] [CrossRef] [Green Version]
  43. Lenz, M.; Müller, F.J.; Zenke, M.; Schuppert, A. Principal components analysis and the reported low intrinsic dimensionality of gene expression microarray data. Sci. Rep. 2016, 6, 25696. [Google Scholar] [CrossRef]
  44. Bloomer, C.; Rehm, G. Using Principal Component Analysis to Find Correlation and Patterns at Diamond Light Source. In Proceedings of the IPAC, Dresden, Germany, 15–20 June 2014. [Google Scholar]
Atmosphere 14 00171 g001 550
Figure 1. Description of monitoring site (Joo et al., (2015a)) [23].
Figure 1. Description of monitoring site (Joo et al., (2015a)) [23].
Atmosphere 14 00171 g001
Atmosphere 14 00171 g002 550
Figure 2. Diurnal profiles of the PM2.5 and sulfur dioxide concentrations (upper) and ammonia and nitrogen dioxide concentrations (bottom) in the barns.
Figure 2. Diurnal profiles of the PM2.5 and sulfur dioxide concentrations (upper) and ammonia and nitrogen dioxide concentrations (bottom) in the barns.
Atmosphere 14 00171 g002
Atmosphere 14 00171 g003 550
Figure 3. Correlations between PM2.5 mass concentrations and gaseous concentrations in the barns. (p-values = 0.160 for SO2, 0.161 for NO2, 0.814 for NH3, and 0.781 for H2S, respectively).
Figure 3. Correlations between PM2.5 mass concentrations and gaseous concentrations in the barns. (p-values = 0.160 for SO2, 0.161 for NO2, 0.814 for NH3, and 0.781 for H2S, respectively).
Atmosphere 14 00171 g003
Atmosphere 14 00171 g004 550
Figure 4. Diurnal profiles of PM2.5 mass emission rates and concentrations in the barns (upper) and in the ambient air (bottom).
Figure 4. Diurnal profiles of PM2.5 mass emission rates and concentrations in the barns (upper) and in the ambient air (bottom).
Atmosphere 14 00171 g004
Atmosphere 14 00171 g005 550
Figure 5. Diurnal patterns of hydrogen sulfide and sulfur dioxide emission rates (upper) and ammonia and nitrogen dioxide emission rates in the barns (bottom).
Figure 5. Diurnal patterns of hydrogen sulfide and sulfur dioxide emission rates (upper) and ammonia and nitrogen dioxide emission rates in the barns (bottom).
Atmosphere 14 00171 g005
Atmosphere 14 00171 g006 550
Figure 6. Correlations between PM2.5 mass emission rates and gaseous air pollutant emission rates at the barns (p-values < 0.001 for NH3, NO2, and SO2, and p = 0.071 for H2S, respectively).
Figure 6. Correlations between PM2.5 mass emission rates and gaseous air pollutant emission rates at the barns (p-values < 0.001 for NH3, NO2, and SO2, and p = 0.071 for H2S, respectively).
Atmosphere 14 00171 g006
Atmosphere 14 00171 g007 550
Figure 7. Correlations between PM2.5 mass emission rates and meteorological parameters in the barns (p-values for all regressions < 0.001).
Figure 7. Correlations between PM2.5 mass emission rates and meteorological parameters in the barns (p-values for all regressions < 0.001).
Atmosphere 14 00171 g007
Atmosphere 14 00171 g008 550
Figure 8. Principal component analysis (PCA): results rotated by Varimax to capture factors influencing PM2.5 emission in 24 h. Components 1 and 2 explain 67.1% and 10.9% of the variance, respectively.
Figure 8. Principal component analysis (PCA): results rotated by Varimax to capture factors influencing PM2.5 emission in 24 h. Components 1 and 2 explain 67.1% and 10.9% of the variance, respectively.
Atmosphere 14 00171 g008
Atmosphere 14 00171 g009 550
Figure 9. Correlations between direct PM2.5 measurements and estimated PM2.5 (from the multivariate linear regressions): PM2.5 loading (left, p < 0.001) and emission rates (right, p < 0.001).
Figure 9. Correlations between direct PM2.5 measurements and estimated PM2.5 (from the multivariate linear regressions): PM2.5 loading (left, p < 0.001) and emission rates (right, p < 0.001).
Atmosphere 14 00171 g009
Table
Table 1. Average of PM2.5, gaseous compounds and meteorological parameters during the monitoring periods.
Table 1. Average of PM2.5, gaseous compounds and meteorological parameters during the monitoring periods.
Measurement Period9 July~21 July 2009
PM2.5Ambient PM2.5Barn 1 PM2.5Barn 2 PM2.5
Concentration (μg/m3)Loading rate
(mg/m3/min) *		Concentration (μg/m3)Emission rate (g/min)Concentration (μg/m3)Emission rate
(g/min)
Mean30.814.0372.772.0867.851.79
SD8.542.267.711.6211.641.30
Barn NH3Barn H2SBarn SO2
Gaseous compoundsConcentration (ppm)Emission rate
(g/min)		Concentration (ppb)Emission rate
(g/min)		Concentration (ppb)Emission rate
(g/min)
Mean1.0129.1017.691.031.850.21
SD0.3518.936.010.720.920.17
Barn NO2Ambient meteorological parameters
Gaseous compounds and othersConcentration (ppm)Emission rate
(g/min)		Wind Velocity (m/s)Solar irradiance
(W/m2)		Relative humidity (%)Temperature
(°C)
Mean0.043.662.21354.5931.5827.55
SD0.022.601.1832.246.773.26
* Ambient loading rate = Concentration (μg/m3) × Air exchange rate (frequency/min) = μg/m3/min, air exchange rate means the exchanged frequency of air in an imaginary control volume box (1 m3) during 1 min.
Table
Table 2. Multivariate linear regressions between barn PM2.5 emission rate (dependent variable) and its precursor gas emission rates from the barns.
Table 2. Multivariate linear regressions between barn PM2.5 emission rate (dependent variable) and its precursor gas emission rates from the barns.
RegressionIndependent VariablesB (Coefficients for Independent Variables)T (Coefficient
/Standard Error)
R = 0.933
p value < 0.001		Intercept−0.088−0.468
NH30.0252.727
SO24.9424.755
NO20.1663.516
Table
Table 3. Multivariate linear regressions between ambient PM2.5 loading rate (dependent variable) and PM2.5 precursor gas emission rates from the barns.
Table 3. Multivariate linear regressions between ambient PM2.5 loading rate (dependent variable) and PM2.5 precursor gas emission rates from the barns.
RegressionIndependent VariablesB (Coefficients for Independent Variables)T (Coefficient
/Standard Error)
R = 0.666
p value < 0.001		Intercept1.3692.354
NH30.0411.460
SO27.7332.410
NO2−0.062−0.424
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Joo, H.-S.; Han, S.-W.; Han, J.-S.; Ndegwa, P.M. Emission Characteristics of Fine Particles in Relation to Precursor Gases in Agricultural Emission Sources: A Case Study of Dairy Barns. Atmosphere 2023, 14, 171. https://0-doi-org.brum.beds.ac.uk/10.3390/atmos14010171

AMA Style

Joo H-S, Han S-W, Han J-S, Ndegwa PM. Emission Characteristics of Fine Particles in Relation to Precursor Gases in Agricultural Emission Sources: A Case Study of Dairy Barns. Atmosphere. 2023; 14(1):171. https://0-doi-org.brum.beds.ac.uk/10.3390/atmos14010171

Chicago/Turabian Style

Joo, Hung-Soo, Sang-Woo Han, Jin-Seok Han, and Pius M. Ndegwa. 2023. "Emission Characteristics of Fine Particles in Relation to Precursor Gases in Agricultural Emission Sources: A Case Study of Dairy Barns" Atmosphere 14, no. 1: 171. https://0-doi-org.brum.beds.ac.uk/10.3390/atmos14010171

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop