1. Introduction
Nepal is a landlocked nation located in South Asia, and it is cited as having some of the most critical air quality challenges in the world, which can have detrimental human health impacts (Health Effects Institute 2020). Nepal also experiences an annual monsoon season, marked by periods of prolonged rainfall that significantly reduce average air pollution levels for four months (June–September) (Regmi et al. 2021). Given the nation’s air quality challenges, air pollution–related human health stressors, and annual monsoon season, it is an intriguing location for biometeorological research. Pokhara Metropolitan City (Pokhara), Nepal, was selected as the site of interest for this research project because of its large population (over 400 000 residents), ample air quality surveillance (three regulatory air quality monitors), and status as Nepal’s largest metropolitan city (464.24 km2) (Kathmandu Post 2017; Pokhara Metropolitan City 2020; Government of Nepal 2022). This study aimed to understand the complex biometeorological system, featuring air pollution and human health, in Pokhara Metropolitan City.
There were two key research questions addressed through this work. The first research question investigated whether monsoon seasons were a significant influencer of air pollution levels in Pokhara. Research conducted by Aryal et al. (2008) determined that Kathmandu, Nepal’s, air pollution levels were lowest during the monsoon season, and Regmi et al. (2021) determined Pokhara’s aerosol optical depth (a value describing the number of aerosols in a vertical column) to be lowest during monsoon seasons (Li et al. 2021). Therefore, this research question sought to learn whether the relationships between monsoon seasons and particulate matter levels during the air pollution sampling collection period (2017–20) were similar to previous studies’ findings. The hypothesis was that the monsoon seasons were significantly correlated with lower particulate matter levels due to wet deposition. In addition, it was hypothesized that the ranking of sites’ pollution levels changes with the presence of the monsoon. In other words, if the Department of Hydrology and Meteorology (DHM) monitor featured the highest levels of air pollution before the monsoon, the same would not be true during the monsoon. Exploring this hypothesis may explain whether the city’s highest pollution levels were located in the same region year-round or whether the maximum and minimum pollution levels changed locations during the monsoon season. This is important because if the location of the highest air pollution levels changes with the season, location-specific air pollution mitigation efforts may be less effective than behavior-specific mitigation efforts (i.e., limiting vehicle idling). This type of research question is best suited for settings with multiple air quality monitors spanning the city. At the time of the research, Pokhara was one of only two cities in Nepal that had such monitoring capabilities. The other city was Kathmandu, which may be considered for future research (Government of Nepal 2022).
The second research question aimed to understand the seasonality of air pollution, meteorological conditions, and human health within Pokhara. For this research, chronic obstructive pulmonary disease (COPD) was the human health condition of focus, given that exposure to air pollution is a risk factor for this irreversible disease (World Health Organization 2022). COPD was also important to consider because since at least 2018 and until the global pandemic of 2020, COPD was the reoccurring top cause of mortality at Pokhara Metropolitan City’s Western Regional Hospital (PWRH) (Pokhara Academy of Health Sciences Western Regional Hospital 2021, 2022). In 2019, COPD was the third most common cause of death worldwide, and a majority of these deaths occurred in lesser developed nations such as Nepal (World Health Organization 2022). It was hypothesized that COPD hospital admissions were strongly correlated with air pollution levels as a result of air pollutants’ ability to injure the respiratory system via oxidative stress and inflammation (Li et al. 2016). There is also evidence of a positive correlation between air pollution levels and COPD hospital admissions (Li et al. 2016). Similarly, it was hypothesized that COPD hospitalizations were more strongly correlated with the pollution levels from the Gandaki Boarding School (GBS) and DHM monitors because they are closer to the city center, where air pollutant levels are hypothesized to be the highest because of higher population levels (Pokharel and Khanal 2018).
2. Materials and methods
This research was completed using an assortment of previously collected data sourced from government records and academic journals. The monthly meteorological values of accumulated rainfall and mean temperatures were retrieved from the Paudel (2020) article concerning Pokhara’s temperature and rainfall trends. However, the article lacked data for 2020. Basnet and Poudel (2021) provided accumulated rainfall values for June–September 2020 in Pokhara. Data for the remaining months of 2020 were not found in freely available, electronic government records. Although meteorological data are available for purchase from Nepal’s DHM, this research was conducted with freely accessible data to showcase what can be accomplished and discovered through open-access data. Pokhara’s three aforementioned air quality monitors are government-owned Grimm Aerosol Technik GmbH Electronic Dust Monitors (EDM) 180 models, and they measure air pollutants via laser scattering (Government of Nepal 2021a; GRIMM 2020). The monitors collected measurements every minute of operation, and the DHM’s officials reported daily averages in the publicly accessible datasets used for this research (Government of Nepal 2021a,b,c,d). The monitors’ PM2.5 and PM10 measurements were considered for this research. Particulate matter (PM) is a class of particles that are suspended in the air. Those with an aerodynamic diameter equal to or smaller than 2.5 μm are referred to as PM2.5, and particulate matter with an aerodynamic diameter greater than 2.5 and less than or equal to 10 μm are deemed PM10 (Health Effects Institute 2020). Pokhara’s air quality monitoring sites are referred to as the DHM, GBS, and Pokhara University (PU) (Government of Nepal 2021b,c,d, 2022). Figure 1 shows the spatial distribution of monitors, and Table 1 includes specific details for each monitor’s location. The starting and ending dates of Nepal’s monsoon seasons were retrieved from government records. Incidences of 2013–16 COPD hospital admissions to Pokhara’s Western Regional Hospital were sourced from Ghosh et al. (2017). The dataset included hospitalizations solely to this hospital, which serves patients from within the city and surrounding municipalities. Hospitalization rates were provided in monthly totals.
Google Earth Engine–produced map of the three monitoring sites (yellow markers) in Pokhara Metropolitan City (Gorelick et al. 2017). The map inset shows the location of Pokhara (yellow star) with respect to Nepal’s borders.
Citation: Weather, Climate, and Society 15, 2; 10.1175/WCAS-D-22-0098.1
Chart containing air quality monitoring sites’ names, location, city ward (i.e., district) number, and population (Government of Nepal 2021b,c,d; Pokharel and Khanal 2018; World Bank Nepal 2020; Gorelick et al. 2017; Gandaki Boarding School 2020; Pokhara University 2021).
Air quality data were collected from Nepal’s Department of Environment’s web page (Government of Nepal 2021b,c,d). Data analyses and visualizations were performed using RStudio software (version 4.2.1; RStudio Team 2020) and R packages listed in the online supplemental material. First, outliers were removed from each site’s 24-h-averaged data using the conservative interquartile range removal approach. This method removes outliers occurring beyond the 25th and 75th percentile of data (less than 2.5% of the entire data, which contain over 8700 PM measurements), and it was used to better account for the skewness of the data and limit the impact of localized events on the overall monthly means (The Pennsylvania State University 2022). Next, the air pollution measurements were sorted into monsoon and dry season categories. Official monsoon dates were retrieved from the DHM Climate Division’s report concerning monsoon onset and withdrawal dates, and this report provided historical monsoon dates from 1968 to 2020 (Government of Nepal 2020). At the time of retrieval, the date of the 2020 monsoon withdrawal was not contained in this report, and it was instead retrieved from a local news report citing a DHM bulletin declaring the withdrawal of the 2020 monsoon (Himalayan News Service 2020). For this research, days on or between the onset and withdrawal dates were classified as “monsoon” days, and days occurring outside of these periods were classified as “dry” days. Once the monsoon and dry seasons were classified, boxplots were constructed to understand the range of particulate matter levels experienced at each site during the monsoon and dry seasons. Then, a Spearman correlation analysis was performed on the data to determine correlations between the air pollution data and the monsoon’s presence. The daily PM measurements were used to calculate monthly averages of PM2.5 and PM10 for each site. Next, time series encompassing each site’s monthly averaged particulate matter measurements were generated. Because of unavailable monthly meteorological data for 2020 and the COPD data collection occurring prior to the air quality sampling period, a 12-month aggregation was performed on the data. This method allowed for the calculation of means for each data type throughout the 12 calendar months. Using these aggregated data, a Pearson correlation analysis was performed with the air pollution, meteorological, and COPD data. Last, a series of multiple regressions were performed to determine how accurately meteorological conditions and air pollution can predict COPD hospital admissions.
3. Results and discussion
The time series displayed in Fig. 2 show the monthly mean measurements collected from each site during the sampling period. DHM’s PM10 measurements were frequently the highest, followed by PU’s PM10 and DHM’s PM2.5 measurements. The GBS site’s PM2.5 and PM10 measurements were typically the lowest measurements during this period. The DHM site contradicted this trend during the 2018 monsoon period. There are gaps in the graph resulting mainly from no reported measurements during the month. Use of the interquartile outlier removal method also resulted in gaps for GBS’s PM2.5 and PM10 measurements from October through November 2020. All 24-h measurements during these months exceeded their upper-quartile limits (92 μg m−3 for PM2.5 and 97 μg m−3 for PM10) and were excluded from further calculations. While these values represent significant air pollution levels and warrant further scientific investigation, they were excluded from this research because of uncertainty about their accuracy.
RStudio-generated time series plot with lines showing mean PM measurements and monsoon seasons from 2017 through 2020.
Citation: Weather, Climate, and Society 15, 2; 10.1175/WCAS-D-22-0098.1
Table 2 shows that all particulate matter measurement variables were negatively correlated with the monsoon’s presence, and all correlation values were statistically significant (p values < 0.05). This confirmed the initial hypothesis that the monsoon was significantly and negatively correlated with particulate matter levels. However, the strength of correlations for DHM’s PM2.5 (correlation = −0.3013) and PM10 (correlation = −0.4061) measurements was lower than the other two sites’ correlation values. This likely implies that the monsoon’s presence and accompanying meteorological conditions may not have as much of an impact on localized air quality as anthropogenic emissions at the DHM site. This notion is further supported by the fact that the DHM site is located toward Pokhara’s city center, where there is a high-density population, and the GBS and PU sites are located farther from the city center and are in less densely populated regions of Pokhara (Pokharel and Khanal 2018). Therefore, their particulate matter levels may be less influenced by anthropogenic emissions than the DHM site. The differences in the monsoon seasons’ correlation to particulate matter levels show that weather characteristics such as wet deposition can lower pollution levels, but this ability is limited given the magnitude and the brief period of the monsoon season.
Spearman correlations between PM measurement levels and monsoon occurrence for the sampling period from 2017 through 2020. The p values were less than 0.0001 for all correlations, implying significance.
The boxplots in Fig. 3 show that air pollution levels during the dry season were greater than the pollution levels during the monsoon season. DHM experienced the highest median PM2.5 and PM10 values in both seasons, and GBS experienced the lowest median PM2.5 and PM10 values. These results show that proximity to the city center and population density may not lead to the highest air pollution levels; otherwise, GBS’s air pollution levels would be greater than PU’s levels. There are noticeably more remaining spikes in PM levels, following the outlier removal process, during the monsoon season measurements than the dry season measurements. This showcases that, even in the monsoon season, there are high spikes in air pollution levels that can increase health risks for citizens in the area. Table 3 quantifies the increase in pollution levels during the dry season. This increase is most noticeable for PU’s PM2.5 and PM10 measurements, which increased by 28.6844 and 37.7557 μg m−3, respectively, with the transition from monsoon to dry season. GBS experienced the lowest increase in PM2.5 and PM10 with increases of 19.9947 and 23.4194 μg m−3, respectively. As shown in Table 3, DHM experienced the highest PM2.5 and PM10 levels, and this was true across both the monsoon and dry seasons. Conversely, PU experienced the lowest PM2.5 and PM10 levels during monsoon season, while GBS experienced the lowest PM2.5 and PM10 levels during the dry season. This confirmed the original hypothesis, which stated that the ranking of air pollution means would change with respect to monsoon and dry seasons. This finding could imply that location-specific air quality mitigation efforts may not be as effective in reducing air pollution as behavioral mitigation methods such as reducing open waste burning (Choi et al. 2022). Results also show that during the monsoon period, disparities among PM10 levels decreased between sites, as the gap between the lowest and highest PM10 levels decreased from 19.9219 μg m−3 in the dry season to 9.8935 μg m−3 in the monsoon season. It is uncertain how much of this decrease can be attributed to meteorological conditions (i.e., precipitation and winds) or to human behavior change (i.e., less driving, biomass burning, etc.); however, this would make for interesting future research for air quality mitigation purposes. To better understand the intricacies of particulate matter levels during monsoon season and address the potential causes, higher temporal resolutions of particulate matter levels, rainfall, and temperature data, as well as wind and human activity data, would be essential.
RStudio-generated boxplots of each monitoring site’s pollution measurements during the monsoon season and dry seasons following the removal of outliers via the interquartile range method.
Citation: Weather, Climate, and Society 15, 2; 10.1175/WCAS-D-22-0098.1
Seasonal mean PM values (μg m−3) and standard error for each site’s PM2.5 and PM10 measurements during the 2017–20 sampling period. The approximated shift in PM levels is also included.
Table 4 features Pearson correlation values among normally distributed air quality, meteorological conditions, and COPD hospital admissions data. There was a strong positive correlation between COPD hospital admissions and all sites’ air quality measurements (correlations = 0.7146–0.8944), and all p values were statistically significant (p values < 0.05). This confirms the initial hypothesis stating significant correlation would be present between air pollution levels and COPD hospital admissions. The strongest correlations were between COPD hospital admissions and GBS’s PM2.5 and PM10 values, and the weakest correlation occurred between hospital admission and DHM’s PM2.5 measurements. Therefore, the initial hypothesis stating that COPD admissions would have the strongest correlations between the DHM and GBS measurements is not completely valid. The hypothesis was made considering DHM’s and GBS’s proximity to the city center and population density, but DHM’s weaker correlation between its PM2.5 values and PWRH’s COPD admissions disrupts this theory. One reason for this weaker correlation may be that there are some unresearched, health-protective features in the DHM area (i.e., additional medical clinics) that reduce the need for citizens living with COPD to visit PWRH. This finding has prompted additional research questions concerning access to PWRH, additional medical resources available to Pokhara’s residents, and the built environment in Pokhara Metropolitan City (i.e., proximity of pollution sources to residential settings).
Pearson correlation values between 12-month-averaged air pollution, COPD hospital admissions, and meteorological data for Pokhara, Nepal.
In terms of meteorological conditions, mean low temperature (correlation = −0.7417) and high temperature (correlation = −0.7216) and rainfall accumulation (correlation = −0.6595) were strongly negatively correlated with COPD hospital admissions (p values < 0.05). The strong negative correlation to rainfall was understandable, given that precipitation can remove particulate matter from the air and prevent atmospheric loading (Regmi et al. 2021). However, the negative correlation to low temperatures was surprising, given that colder temperatures increase risks of COPD hospitalizations (Javorac et al. 2021; Yin et al. 2021). The research team that originally collected and analyzed the COPD data at Pokhara’s Western Regional Hospital also noted the relationship between COPD hospital admissions and cold temperatures (Ghosh et al. 2017). One possible reason for this negative correlation may be that cold temperatures prevented or deterred COPD patients from traveling to Pokhara’s Western Regional Hospital, although more information on human behavior is needed to confirm if this is the true reason for this correlation. To further investigate the relationship between air pollution, meteorological conditions, and COPD hospital admissions, greater temporal resolutions for coincident meteorological and COPD data would be useful. In addition, humidity data would be beneficial, considering that it is likely a protective factor for those with COPD (DeVries et al. 2016).
As a final step in this research, a series of multivariate regressions were performed to determine how well COPD hospital admissions could be predicted using particulate matter levels and meteorological data. Because of asynchronous datasets, all data were aggregated across the 12 calendar months. This method was beneficial for examining the relationship between air pollution, meteorological conditions, and COPD hospital admissions across a generalized year. The first multivariate regression included both PM2.5 and PM10 data, the second regression excluded PM10 data, and the third regression excluded PM2.5 data. The multivariate regression including PM2.5 and PM10 aggregations was statistically insignificant (p value > 0.05), so it is not featured in Table 5. The remaining regression results were statistically significant, and both regressions yielded fairly confident results (R2 > 0.85 and p values < 0.05). Model validation was performed via the Jarque–Bera test using the fBasics R package (version 3042.89.2) on the residuals of the regressions that computed the normal distribution of residuals (Wuertz et al. 2022). Because of the high correlation between mean low and high temperatures (correlation = 0.9781), multivariate regressions were also performed by excluding first the low temperatures and then the high temperatures. However, these multivariate regressions yielded lower R2 values than those that included both mean low and high temperatures. This likely means that both mean high and low temperatures help better characterize the biometeorological system, as opposed to a singular temperature value. This complements previous research findings that both high and low temperatures can exacerbate COPD symptoms (Almagro et al. 2015; Javorac et al. 2021; de Miguel-Díez et al. 2019).
Results from multiple regressions in which COPD hospital admissions were predicted using PM2.5, PM10, temperature, and rainfall data. For PM2.5 regression results, the R2 was 0.8623, with a p value of 0.0451. For PM10 regression results, the R2 was 0.8829, with a p value of 0.0311.
A key finding from the multivariate regressions was that PM10 values were better predictors of COPD hospital admissions than PM2.5. This was surprising given that PM2.5 is cited as having more significant health impacts due to its ability to travel deeper into the respiratory system (Pope and Dockery 2006). This may be the result of differences in the time lapse between exposure to PM2.5 or PM10 and subsequent COPD exacerbation, but more research is needed to support this theory. From the multivariate regressions, it is also shown that DHM’s PM2.5 and PM10 values had negative coefficients as predictor variables for COPD hospital admissions. All other pollution measurements had positive coefficients. This was surprising; however, a previous study noted an inverse relationship between PM2.5 and COPD exacerbations (DeVries et al. 2016). One possible reason that DHM’s values were inversely related to COPD hospital admissions may be that exposure measurements occurred after the period of COPD hospitalization data. Gaining access to COPD data coincident with the air quality monitoring sampling period (2017–20) may result in a different predicted relationship between DHM’s air pollution levels and Pokhara’s Western Regional Hospital’s COPD admissions.
The findings from this research are promising, but they are not without limitations. In terms of meteorological measurements, a lack of humidity, wind direction, and wind speed data limited analyses. Humidity data would have been used to examine the relationship with COPD hospitalizations given its protective properties (DeVries et al. 2016). Wind direction and speed data would also have been useful for this research because these data would have allowed for more in-depth analyses of air pollution transport, which influences air pollution exposure. These variables were not provided in accessible datasets, but future research on this topic will benefit from acquiring and using these data. Additionally, limited data availability caused a reliance on monthly mean values of temperatures and precipitation for data analyses. A monthly analysis is not ideal for studying the relationship between PM exposure and COPD hospitalizations because the hospitalizations stem from acute exposure. Instead, daily or weekly hospitalizations would have been more beneficial as they would allow for epidemiologic time series analyses. This resolution of COPD hospitalization data was not publicly available during this research, and the available data used during this research did not occur simultaneously with the meteorological or air pollution measurements. Because of this shortcoming, analyses are limited to a seasonality as opposed to a time series or generalized additive model analyses. Despite this limitation, the conclusions of this research are promising, as it appears air pollution and meteorological measurements may be useful indicators of COPD hospitalizations. To achieve more definitive predictions for COPD hospitalizations in Pokhara Metropolitan City, it is essential to secure more comprehensive data for these three factors.
4. Conclusions
From investigating this biometeorological system in Pokhara Metropolitan City, the relationships between localized air pollution, meteorological conditions, and COPD hospital admissions were better understood. All sites’ air pollution levels decreased during the monsoon season, and this decrease was most pronounced in PU’s PM10 measurements and least pronounced in DHM’s PM2.5 measurements. In general terms, DHM experienced the highest levels of PM2.5 and PM10, whereas GBS experienced the lowest levels. However, PU’s air pollution levels were greater than GBS’s levels during the dry season. This knowledge could lead to further investigation of health burden disparities and inform air pollution mitigation efforts that account for seasonal behaviors. In terms of health trends, COPD hospital admissions were positively correlated with all air pollution datasets. Conversely, COPD hospital admissions were negatively associated with mean high and low temperatures and accumulated rainfall. This highlights the benefits that meteorology can provide for settings experiencing high air pollution levels. Multivariate regressions concluded that these meteorological factors may be useful for estimating COPD hospitalizations throughout the calendar year. Results showed that using both mean low and high temperatures helped best predict COPD hospital admissions; however, using both PM2.5 and PM10 levels did not yield statistically significant results. In addition, using PM10 values was more accurate than using PM2.5 values. A shortcoming of the regression models was limited data availability, and this was overcome through a 12-month aggregation of data. However, more comprehensive and coincident air pollution, meteorological, and COPD data may allow for more robust predictions. Future research may be conducted to understand how inversion layers and Pokhara’s urban heat island influence air pollution levels and health risks. As this work progresses, the researcher intends to work with collaborators to further investigate biometeorological systems and assess solutions that may address pollution exposure and protect human health in Pokhara Metropolitan City and beyond.
Acknowledgments.
The author is a 2021 Robert Wood Johnson Foundation Health Policy Research Scholar and received funding from this organization.
Data availability statement.
Air quality data analyzed in this study were sourced from existing datasets and were openly available at locations cited in the reference section. Should these locations no longer be accessible, details on the data and how to request access are available from Nepal’s Department of Environment (info@doenv.gov.np). The monsoon occurrence data were also publicly available, at the location cited in the reference section, during the data-collection phase of this research. This resource may no longer be accessible, but details about the data and how to request access may be available from Nepal’s Department of Hydrology and Meteorology (info@dhm.gov.np). The data for COPD hospitalizations at Pokhara’s Western Regional Hospital are included in Ghosh et al. (2017). The data for monthly meteorological means are included in Paudel (2020) and Basnet and Poudel (2021).
REFERENCES
Almagro, P., C. Hernandez, P. Martinez-Cambor, R. Tresserras, and J. Escarrabill, 2015: Seasonality, ambient temperatures and hospitalizations for acute exacerbation of COPD: A population-based study in a metropolitan area. Int. J. Chronic Obstruct. Pulm. Dis., 10, 899–908, https://doi.org/10.2147/COPD.S75710.
Aryal, R. K., B.-K. Lee, R. Karki, A. Gurung, J. Kandasamy, B. K. Pathak, S. Sharma, and N. Giri, 2008: Seasonal PM10 dynamics in Kathmandu Valley. Atmos. Environ., 42, 8623–8633, https://doi.org/10.1016/j.atmosenv.2008.08.016.
Basnet, K., and K. Poudel, 2021: Developing a formula to estimate precipitation at ungauged location and analysis of rainfall pattern: Case study for Rastra Bank Chowk, Pokhara, Nepal. Himalayan J. Appl. Sci. Eng., 2 (2), 1–13, https://doi.org/10.3126/hijase.v2i2.43240.
Choi, E., N. Shrestha, and T. R. Bhandari, 2022: Open waste burning contrary to other air pollution-related perceptions and practices in Pokhara, Nepal. Arch. Environ. Occup. Health, 77, 721–733, https://doi.org/10.1080/19338244.2021.2004985.
de Miguel-Díez, J., J. Hernández-Vázquez, A. López-de-Andrés, A. Álvaro-Meca, V. Hernández-Barrera, and R. Jiménez-García, 2019: Analysis of environmental risk factors for chronic obstructive pulmonary disease exacerbation: A case-crossover study (2004-2013). PLOS ONE, 14, e0217143, https://doi.org/10.1371/journal.pone.0217143.
DeVries, R., D. Kriebel, and S. Sama, 2016: Low level air pollution and exacerbation of existing COPD: A case crossover analysis. Environ. Health, 15, 98, https://doi.org/10.1186/s12940-016-0179-z.
Gandaki Boarding School, 2020: Welcome to Gandaki Boarding School. Gandaki Boarding School, accessed 18 May 2022, https://gbs.edu.np/#.
Ghosh, V., and Coauthors, 2017: Study on epidemiology of chronic obstructive pulmonary disease (COPD) at Western Regional Hospital, Pokhara. J. Gandaki Med. Coll.-Nepal, 9, 65–69, https://doi.org/10.3126/jgmcn.v9i2.17873.
Gorelick, N., M. Hancher, M. Dixon, S. Ilyushchenko, D. Thau, and R. Moore, 2017: Google Earth Engine (version 7.3.6.9285). Google, https://earthengine.google.com/.
Government of Nepal, 2020: Monsoon onset and withdrawal date information. Ministry of Energy, Water Resources and Irrigation Department of Hydrology and Meteorology Climate Division (Climate Analysis Section) Rep., 3 pp.,https://www.dhm.gov.np/uploads/dhm/climateService/monsoon_onset_n_withdrawal_English_6_June_20221.pdf.
Government of Nepal, 2021a: Air quality status of Nepal 2016-2020: Main report. Department of Environment Rep., 9 pp., https://doenv.gov.np/progressfiles/Main-report-60120-1660561765.pdf.
Government of Nepal, 2021b: Air quality status of Nepal 2016-2020, volume VII: DHM Pokhara air quality monitoring station. Ministry of Forest and Environment Department of Environment Rep., 124 pp., https://doenv.gov.np/progressfiles/8_DHM-Pokhara-42992-1660561643.pdf.
Government of Nepal, 2021c: Air quality status of Nepal 2016-2020, volume IX: Gandaki Boarding School Pokhara air quality monitoring station. Ministry of Forest and Environment Department of Environment Rep., 100 pp., https://doenv.gov.np/progressfiles/9_GBS-Pokhara-53954-1660561631.pdf.
Government of Nepal, 2021d: Air quality status of Nepal 2016-2020, volume X: Pokhara University, Pokhara air quality monitoring station. Ministry of Forest and Environment Department of Environment Rep., 109 pp., https://doenv.gov.np/progressfiles/10_PU-Pokhara-33144-1660561617.pdf.
Government of Nepal, 2022: Air quality monitoring. Ministry of Forests and Environment Department of Environment Air Quality Monitoring, accessed 20 May 2022, https://pollution.gov.np/portal/#/home?_k=zndq5z.
GRIMM AEROSOL TECHNIK, 2020: EDM180 environmental dust monitor for approved PM measurements. GRIMM Datasheet, 2 pp., https://www.grimm-aerosol.com/fileadmin/files/grimm-aerosol/3%20Products/Environmental%20Dust%20Monitoring/Approved%20PM%20Monitor/EDM180_The_Proven/Product%20PDFs/Datasheet_EDM180_ENG_2020.pdf.
Health Effects Institute, 2020: State of global air 2020: A special report. Health Effects Institute Rep., 28 pp., https://www.stateofglobalair.org/sites/default/files/documents/2020-10/soga-2020-report-10-26_0.pdf.
Himalayan News Service, 2020: Monsoon finally withdraws after a delay of 23 days. Himalayan, 18 October, https://thehimalayantimes.com/nepal/monsoon-finally-withdraws-after-a-delay-of-23-days.
Javorac, J., M. Jevtić, D. Živanović, M. Ilić, S. Bijelović, and N. Dragić, 2021: What are the effects of meteorological factors on exacerbations of chronic obstructive pulmonary disease? Atmosphere, 12, 442, https://doi.org/10.3390/atmos12040442.
Kathmandu Post, 2017: Pokhara Lekhnath becomes largest metropolitan city. Kathmandu Post, 13 March, https://kathmandupost.com/national/2017/03/13/pokhara-lekhnath-becomes-largest-metropolitan-city.
Li, J., S. Sun, R. Tang, H. Qiu, Q. Huang, T. G. Mason, and L. Tian, 2016: Major air pollutants and risk of COPD exacerbations: A systematic review and meta-analysis. Int. J. Chronic Obstruct. Pulm. Dis., 11, 3079–3091, https://doi.org/10.2147/COPD.S122282.
Li, J., X. Ge, Q. He, and A. Abbas, 2021: Aerosol optical depth (AOD): Spatial and temporal variations and association with meteorological covariates in Taklimakan Desert, China. PeerJ, 9, e10542, https://doi.org/10.7717/peerj.10542.
Paudel, U., 2020: Trends of temperature and rainfall in Pokhara. Prithvi Acad. J., 3, 22–32, https://doi.org/10.3126/paj.v3i0.29556.
Pokhara Academy of Health Sciences Western Regional Hospital, 2021: Annual report F/Y 2076/077. Department of Health Services Rep., 473 pp., https://dohs.gov.np/wp-content/uploads/2021/07/DoHS-Annual-Report-FY-2076-77-for-website.pdf.
Pokhara Academy of Health Sciences Western Regional Hospital, 2022: Annual report F/Y 2077/078. Department of Health Services Rep., 572 pp., https://dohs.gov.np/wp-content/uploads/2022/07/DoHS-Annual-Report-FY-2077-78-date-5-July-2022-2022_FINAL.pdf.
Pokhara Metropolitan City, 2020: Brief introduction. Pokhara Metropolitan City Doc., 12 pp.
Pokhara University, 2021: An official site of Pokhara University. Pokhara University, accessed 20 May 2022, https://pu.edu.np/.
Pokharel, R. P., and N. R. Khanal, 2018: Open space: Typology and distribution in Pokhara Lekhnath Metropolitan City. Geogr. J. Nepal, 11, 25–44, https://doi.org/10.3126/gjn.v11i0.19547.
Pope, C. A., III, and D. W. Dockery, 2006: Health effects of fine particulate air pollution: Lines that connect. J. Air Waste Manage. Assoc., 56, 709–742, https://doi.org/10.1080/10473289.2006.10464485.
Regmi, J., K. N. Poudyal, A. Pokhrel, M. Gyawali, A. Barinelli, and R. Aryal, 2021: Analysis of aerosol optical depth and Angstrom exponents over an AERONET site at Pokhara, Nepal. BIBECHANA, 18, 118–127, https://doi.org/10.3126/bibechana.v18i1.29448.
RStudio Team, 2020: RStudio: Integrated development for R, version 4.2.1. RStudio, PBC, http://www.rstudio.org/.
The Pennsylvania State University, 2022: 3.2—Identifying outliers: IQR method: Stat 200. PennState: Statistics online courses, accessed 27 December 2022, https://online.stat.psu.edu/stat200/lesson/3/3.2#:∼:text=We%20can%20use%20the%20IQR,add%20this%20value%20to%20Q3.
World Bank Nepal, 2020: Environment and social impact assessment and environment and social management plan of improvement of Talchowk-BegnasRoad-P23.i. World Bank Nepal Doc., 228 pp., https://documents1.worldbank.org/curated/zh/103131573190265382/pdf/Environmental-and-Social-Impact-Assessment-and-Environment-and-Social-Management-Plan-of-Improvement-of-Talchowk-Begnas-Road-Pokhara-Metropolitan-City.pdf.
World Health Organization, 2022: Chronic obstructive pulmonary disease (COPD). WHO, accessed 1 September 2022, https://www.who.int/news-room/fact-sheets/detail/chronic-obstructive-pulmonary-disease-(copd).
Wuertz, D., T. Setz, Y. Chalabi, M. Maechler and C. Team, 2022: fBasics: Rmetrics—Markets and basic statistics, version 3042.89.2, R package. https://CRAN.R-project.org/package=fBasics.
Yin, X., and Coauthors, 2021: Impacts of Indian summer monsoon and stratospheric intrusion on air pollutants in the inland Tibetan Plateau. Geosci. Front., 12, 101255, https://doi.org/10.1016/j.gsf.2021.101255.
