Multitemporal Analysis of the Influence of PM10 on Human Mortality According to Urban Land Cover
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Site
2.2. Information Collection
2.3. Information Analysis
3. Results and Discussion
3.1. Spatiotemporal Analysis of PM10
3.2. Spatiotemporal Variation of IDM
3.3. Relationship between Land Cover, PM10, and IDM
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Karanasiou, A.; Alastuey, A.; Amato, F.; Renzi, M.; Stafoggia, M.; Tobias, A.; Reche, C.; Forastiere, F.; Gumy, S.; Mudu, P.; et al. Short-Term Health Effects from Outdoor Exposure to Biomass Burning Emissions: A Review. Sci. Total Environ. 2021, 781, 146739. [Google Scholar] [CrossRef] [PubMed]
- Sun, X.; Luo, X.-S.; Xu, J.; Zhao, Z.; Chen, Y.; Wu, L.; Chen, Q.; Zhang, D. Spatio-Temporal Variations and Factors of a Provincial PM2.5 Pollution in Eastern China during 2013–2017 by Geostatistics. Sci. Rep. 2019, 9, 3613. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Soto, D.F.P.; Mejía, C.A.Z.; Miranda, J.P.R. Evaluación de la calidad del aire mediante un laboratorio móvil: Puente Aranda (Bogotá D.C., Colombia). Rev. Fac. Ing. Univ. Antioq. 2014, 71, 153–166. [Google Scholar]
- Blanco-Becerra, L.C.; Miranda-Soberanis, V.; Hernández-Cadena, L.; Barraza-Villarreal, A.; Junger, W.; Hurtado-Díaz, M.; Romieu, I. Effect of Particulate Matter Less than 10µm (PM10) on Mortality in Bogota, Colombia: A Time-Series Analysis, 1998–2006. Salud Pública México 2014, 56, 363–370. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Whyand, T.; Hurst, J.R.; Beckles, M.; Caplin, M.E. Pollution and Respiratory Disease: Can Diet or Supplements Help? A Review. Respir. Res. 2018, 19, 79. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Ning, G.; Wang, S.; Ma, M.; Ni, C.; Shang, Z.; Wang, J.; Li, J. Characteristics of Air Pollution in Different Zones of Sichuan Basin, China. Sci. Total Environ. 2018, 612, 975–984. [Google Scholar] [CrossRef]
- Yang, W.; Jiang, X. Evaluating the Influence of Land Use and Land Cover Change on Fine Particulate Matter. Sci. Rep. 2021, 11, 17612. [Google Scholar] [CrossRef]
- Liang, L.; Gong, P. Urban and Air Pollution: A Multi-City Study of Long-Term Effects of Urban Landscape Patterns on Air Quality Trends. Sci. Rep. 2020, 10, 18618. [Google Scholar] [CrossRef]
- East, J.; Montealegre, J.S.; Pachon, J.E.; Garcia-Menendez, F. Air Quality Modeling to Inform Pollution Mitigation Strategies in a Latin American Megacity. Sci. Total Environ. 2021, 776, 145894. [Google Scholar] [CrossRef]
- Zafra, C.; Suárez, J.; Pachón, J.E. Public Health Considerations for PM10 in a High-Pollution Megacity: Influences of Atmospheric Condition and Land Coverage. Atmosphere 2021, 12, 118. [Google Scholar] [CrossRef]
- Santovito, A.; Gendusa, C.; Cervella, P.; Traversi, D. In Vitro Genomic Damage Induced by Urban Fine Particulate Matter on Human Lymphocytes. Sci. Rep. 2020, 10, 8853. [Google Scholar] [CrossRef] [PubMed]
- Liu, C.; Henderson, B.H.; Wang, D.; Yang, X.; Peng, Z.-R. A Land Use Regression Application into Assessing Spatial Variation of Intra-Urban Fine Particulate Matter (PM2.5) and Nitrogen Dioxide (NO2) Concentrations in City of Shanghai, China. Sci. Total Environ. 2016, 565, 607–615. [Google Scholar] [CrossRef] [PubMed]
- Yang, H.; Chen, W.; Liang, Z. Impact of Land Use on PM2.5 Pollution in a Representative City of Middle China. Int. J. Environ. Res. Public Health 2017, 14, 462. [Google Scholar] [CrossRef]
- Anenberg, S.C.; Achakulwisut, P.; Brauer, M.; Moran, D.; Apte, J.S.; Henze, D.K. Particulate Matter-Attributable Mortality and Relationships with Carbon Dioxide in 250 Urban Areas Worldwide. Sci. Rep. 2019, 9, 11552. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Manojkumar, N.; Srimuruganandam, B. Health Effects of Particulate Matter in Major Indian Cities. Int. J. Environ. Health Res. 2021, 31, 258–270. [Google Scholar] [CrossRef] [PubMed]
- Leikauf, G.D.; Kim, S.-H.; Jang, A.-S. Mechanisms of Ultrafine Particle-Induced Respiratory Health Effects. Exp. Mol. Med. 2020, 52, 329–337. [Google Scholar] [CrossRef] [PubMed]
- Guo, P.; Su, Y.; Wan, W.; Liu, W.; Zhang, H.; Sun, X.; Ouyang, Z.; Wang, X. Urban Plant Diversity in Relation to Land Use Types in Built-up Areas of Beijing. Chin. Geogr. Sci. 2018, 28, 100–110. [Google Scholar] [CrossRef] [Green Version]
- Rodríguez-Villamizar, L.A.; Rojas-Roa, N.Y.; Fernández-Niño, J.A. Short-Term Joint Effects of Ambient Air Pollutants on Emergency Department Visits for Respiratory and Circulatory Diseases in Colombia, 2011–2014. Environ. Pollut. 2019, 248, 380–387. [Google Scholar] [CrossRef]
- Franco, J.F.; Gidhagen, L.; Morales, R.; Behrentz, E. Towards a Better Understanding of Urban Air Quality Management Capabilities in Latin America. Environ. Sci. Policy 2019, 102, 43–53. [Google Scholar] [CrossRef]
- Aguiar-Gil, D.; Gómez-Peláez, L.M.; Álvarez-Jaramillo, T.; Correa-Ochoa, M.A.; Saldarriaga-Molina, J.C. Evaluating the Impact of PM2.5 Atmospheric Pollution on Population Mortality in an Urbanized Valley in the American Tropics. Atmos. Environ. 2020, 224, 117343. [Google Scholar] [CrossRef]
- Sefair, J.A.; Espinosa, M.; Behrentz, E.; Medaglia, A.L. Optimization Model for Urban Air Quality Policy Design: A Case Study in Latin America. Comput. Environ. Urban Syst. 2019, 78, 101385. [Google Scholar] [CrossRef]
- Mendieta, E. Medellín and Bogotá: The Global Cities of the Other Globalization. City 2011, 15, 167–180. [Google Scholar] [CrossRef]
- SDA Informe Anual calidad del aire Bogotá 2018 » Observatorio Ambiental de Bogotá. Available online: https://oab.ambientebogota.gov.co/?post_type=dlm_download&p=13003 (accessed on 28 October 2022).
- U.S. EPA Title 40 of the CFR—Protection of Environment. Available online: https://www.ecfr.gov/current/title-40 (accessed on 28 October 2022).
- World Health Organization. Occupational and Environmental Health Team Guías de Calidad del aire de la OMS Relativas al Material Particulado, el Ozono, el Dióxido de Nitrógeno y el Dióxido de Azufre: Actualización Mundial 2005; Organización Mundial de la Salud: Geneva, Switzerland, 2006. [Google Scholar]
- Shakya, A.K.; Ramola, A.; Kandwal, A.; Prakash, R. Comparison of Supervised Classification Techniques with Alos Palsar Sensor for Roorkee Region of Uttarakhand, India. ISPRS-Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 2018, 425, 693–701. [Google Scholar] [CrossRef] [Green Version]
- Zafra, C.; Ángel, Y.; Torres, E. ARIMA Analysis of the Effect of Land Surface Coverage on PM10 Concentrations in a High-Altitude Megacity. Atmos. Pollut. Res. 2017, 8, 660–668. [Google Scholar] [CrossRef]
- DANE Proyecciones de Población. Available online: https://www.dane.gov.co/index.php/estadisticas-por-tema/demografia-y-poblacion/proyecciones-de-poblacion (accessed on 28 October 2022).
- MAVDT Manual de Operación de Sistemas de Vigilancia de la Calidad del Aire » Observatorio Ambiental de Bogotá. Available online: https://oab.ambientebogota.gov.co/?post_type=dlm_download&p=3768 (accessed on 28 October 2022).
- Berger, V.W.; Zhou, Y. Kolmogorov–Smirnov Test: Overview. In Wiley StatsRef: Statistics Reference Online; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2014; ISBN 978-1-118-44511-2. [Google Scholar]
- Rebekić, A.; Lončarić, Z.; Petrović, S.; Marić, S. Pearson’s or Spearman’s correlation coefficient—Which one to use? Poljoprivreda 2015, 21, 47–54. [Google Scholar] [CrossRef]
- Conagin, A.; Barbin, D.; Demétrio, C.G.B. Modifications for the Tukey Test Procedure and Evaluation of the Power and Efficiency of Multiple Comparison Procedures. Sci. Agric. 2008, 65, 428–432. [Google Scholar] [CrossRef] [Green Version]
- Montealegre, B.J.E. Técnicas Estadísticas Aplicadas en el Manejo de Datos Hidrológicos y Meteorológicos; Instituto Colombiano de Hidrología, Meteorología y Adecuación de Tierras: Bogotá, Colombia, 1990. [Google Scholar]
- Carrera-Villacrés, D.V.; Guevara-García, P.V.; Tamayo-Bacacela, L.C.; Balarezo-Aguilar, A.L.; Narváez-Rivera, C.A.; Morocho-López, D.R. Relleno de Series Anuales de Datos Meteorológicos Mediante Métodos Estadísticos En La Zona Costera e Interandina Del Ecuador, y Cálculo de La Precipitación Media. Idesia 2016, 34, 81–90. [Google Scholar] [CrossRef] [Green Version]
- MADS Derecho Del Bienestar Familiar [RESOLUCION_MINAMBIENTEDS_2254_2017]. Available online: https://www.icbf.gov.co/cargues/avance/docs/resolucion_minambienteds_2254_2017.htm (accessed on 28 October 2022).
- Xu, T.; Liu, Y.; Tang, L.; Liu, C. Improvement of Kriging Interpolation with Learning Kernel in Environmental Variables Study. Int. J. Prod. Res. 2022, 60, 1284–1297. [Google Scholar] [CrossRef]
- Stafoggia, M.; Bellander, T.; Bucci, S.; Davoli, M.; de Hoogh, K.; de’ Donato, F.; Gariazzo, C.; Lyapustin, A.; Michelozzi, P.; Renzi, M.; et al. Estimation of Daily PM10 and PM2.5 Concentrations in Italy, 2013–2015, Using a Spatiotemporal Land-Use Random-Forest Model. Environ. Int. 2019, 124, 170–179. [Google Scholar] [CrossRef]
- Yuan, Z.; Yang, Y. Combining Linear Regression Models. J. Am. Stat. Assoc. 2005, 100, 1202–1214. [Google Scholar] [CrossRef]
- Franceschi, F.; Cobo, M.; Figueredo, M. Discovering Relationships and Forecasting PM10 and PM2.5 Concentrations in Bogotá, Colombia, Using Artificial Neural Networks, Principal Component Analysis, and k-Means Clustering. Atmos. Pollut. Res. 2018, 9, 912–922. [Google Scholar] [CrossRef]
- SDM Encuesta de Movilidad. Available online: https://www.movilidadbogota.gov.co/web/encuesta_de_movilidad (accessed on 28 October 2022).
- Ramírez, O.; Sánchez de la Campa, A.M.; Amato, F.; Catacolí, R.A.; Rojas, N.Y.; de la Rosa, J. Chemical Composition and Source Apportionment of PM10 at an Urban Background Site in a High–Altitude Latin American Megacity (Bogota, Colombia). Environ. Pollut. 2018, 233, 142–155. [Google Scholar] [CrossRef] [PubMed]
- MinTrabajo Leyes Desde 1992—Vigencia Expresa y Control de Constitucionalidad [CODIGO_SUSTANTIVO_TRABAJO]. Available online: http://www.secretariasenado.gov.co/senado/basedoc/codigo_sustantivo_trabajo.html (accessed on 28 October 2022).
- Martilli, A.; Sanchez, B.; Rasilla, D.; Pappaccogli, G.; Allende, F.; Martin, F.; Román-Cascón, C.; Yagüe, C.; Fernandez, F. Simulating the Meteorology during Persistent Wintertime Thermal Inversions over Urban Areas. The Case of Madrid. Atmos. Res. 2021, 263, 105789. [Google Scholar] [CrossRef]
- Gharibzadeh, M.; Saadat Abadi, A.R. Estimation of Surface Particulate Matter (PM2.5 and PM10) Mass Concentration by Multivariable Linear and Nonlinear Models Using Remote Sensing Data and Meteorological Variables over Ahvaz, Iran. Atmos. Environ. X 2022, 14, 100167. [Google Scholar] [CrossRef]
- Zhang, B.; Jiao, L.; Xu, G.; Zhao, S.; Tang, X.; Zhou, Y.; Gong, C. Influences of Wind and Precipitation on Different-Sized Particulate Matter Concentrations (PM2.5, PM10, PM2.5–10). Meteorol. Atmos. Phys. 2018, 130, 383–392. [Google Scholar] [CrossRef]
- Li, W.; Pei, L.; Li, A.; Luo, K.; Cao, Y.; Li, R.; Xu, Q. Spatial Variation in the Effects of Air Pollution on Cardiovascular Mortality in Beijing, China. Environ. Sci. Pollut. Res. Int. 2019, 26, 2501–2511. [Google Scholar] [CrossRef]
- Cohen, A.J.; Brauer, M.; Burnett, R.; Anderson, H.R.; Frostad, J.; Estep, K.; Balakrishnan, K.; Brunekreef, B.; Dandona, L.; Dandona, R.; et al. Estimates and 25-Year Trends of the Global Burden of Disease Attributable to Ambient Air Pollution: An Analysis of Data from the Global Burden of Diseases Study 2015. Lancet 2017, 389, 1907–1918. [Google Scholar] [CrossRef] [Green Version]
- Fazlzadeh, M.; Rostami, R.; Yousefian, F.; Yunesian, M.; Janjani, H. Long Term Exposure to Ambient Air Particulate Matter and Mortality Effects in Megacity of Tehran, Iran: 2012–2017. Particuology 2021, 58, 139–146. [Google Scholar] [CrossRef]
- Zhang, W.; Ma, R.; Wang, Y.; Jiang, N.; Zhang, Y.; Li, T. The Relationship between Particulate Matter and Lung Function of Children: A Systematic Review and Meta-Analysis. Environ. Pollut. 2022, 309, 119735. [Google Scholar] [CrossRef]
- IDEAM, Informes del Estado de la Calidad del Aire en Colombia. Available online: http://www.ideam.gov.co/web/contaminacion-y-calidad-ambiental/informes-del-estado-de-la-calidad-del-aire-en-colombia?p_p_id=110_INSTANCE_3uZc3mUViyRu&p_p_lifecycle=0&p_p_state=normal&p_p_mode=view&p_p_col_id=column-1&p_p_col_count=1&_110_INSTANCE_3uZc3mUViyRu_struts_action=%2Fdocument_library_display%2Fview_file_entry&_110_INSTANCE_3uZc3mUViyRu_fileEntryId=68521975 (accessed on 28 October 2022).
- Trinh, T.T.; Trinh, T.T.; Le, T.T.; Nguyen, T.D.H.; Tu, B.M. Temperature Inversion and Air Pollution Relationship, and Its Effects on Human Health in Hanoi City, Vietnam. Environ. Geochem Health 2019, 41, 929–937. [Google Scholar] [CrossRef]
- Gautam, S.; Brema, J. Spatio-Temporal Variation in the Concentration of Atmospheric Particulate Matter: A Study in Fourth Largest Urban Agglomeration in India. Environ. Technol. Innov. 2020, 17, 100546. [Google Scholar] [CrossRef]
- Xu, W.; Wu, Q.; Liu, X.; Tang, A.; Dore, A.J.; Heal, M.R. Characteristics of Ammonia, Acid Gases, and PM2.5 for Three Typical Land-Use Types in the North China Plain. Environ. Sci. Pollut. Res. 2016, 23, 1158–1172. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Salata, S.; Ronchi, S.; Arcidiacono, A. Mapping Air Filtering in Urban Areas. A Land Use Regression Model for Ecosystem Services Assessment in Planning. Ecosyst. Serv. 2017, 28, 341–350. [Google Scholar] [CrossRef]
- Riondato, E.; Pilla, F.; Sarkar Basu, A.; Basu, B. Investigating the Effect of Trees on Urban Quality in Dublin by Combining Air Monitoring with I-Tree Eco Model. Sustain. Cities Soc. 2020, 61, 102356. [Google Scholar] [CrossRef]
- Wu, J.; Wang, Y.; Qiu, S.; Peng, J. Using the Modified I-Tree Eco Model to Quantify Air Pollution Removal by Urban Vegetation. Sci. Total Environ. 2019, 688, 673–683. [Google Scholar] [CrossRef] [PubMed]
Station | CAR | KEN | GUA | SUB | TUN | USQ | |
---|---|---|---|---|---|---|---|
Characteristics | |||||||
Coordinates | 4°39′30.5″ N 74°5′2.28″ W | 4°37′30.2″ N 74°9′40.8″ W | 4°47′1.52″ N 74°2′39.1″ W | 4°45′40.5″ N 74° 5′36.5″ W | 4°34′34.4″ N 74°7′51.4″ W | 4°42′37.3″ N 74°1′49.5″ W | |
Elevation (masl) | 2577 | 2580 | 2580 | 2571 | 2589 | 2570 | |
Annual mean PM10 (µg/m3) | 28 | 50 | 28 | 46 | 38 | 39 | |
Annual mean rainfall (mm) | 932 | 1280 | 796 | 454 | 544 | 905 | |
Hourly mean temperature (°C) | S/R a | 15.2 | 14.5 | 14.4 | S/R a | S/R a | |
Mean wind speed (m/s) | 1.19 | 2.30 | 1.00 | 1.39 | 1.19 | 1.60 | |
Locality | Barrios Unidos | Kennedy | Suba | Suba | Tunjuelito | Usaquén | |
Land classification | Urban | Urban | Urban/Rural | Urban/Rural | Urban | Urban/Rural | |
Urban land use b | AUI/CS/D/I/R/SP | AUI/CS/D/I/R/SP | AAC/AUI/CS/D/I/R/SP | AAC/AUI/CS/D/I/R/SP | AUI/CS/D/I/R/SP | AAC/AUI/CS/D/R/SP | |
Rural land use c | N/A | N/A | SAP/ME/AC | SAP/ME/AC | N/A | SAP/SE | |
Population density (inhabitants/ha) | 112 | 265 | 115 | 115 | 173 | 82.2 | |
Population/locality | 133,126 | 1,019,748 | 1,152,387 | 1,152,387 | 171,632 | 535,693 | |
Area/locality (ha) | 1189 | 3855 | 10,048 | 10,048 | 990 | 6514 |
Station | Radius (m) | Land Cover (%) | |||||||
---|---|---|---|---|---|---|---|---|---|
Impermeable | Vegetation | Water Bodies | Bare Land | ||||||
2008 a | 2018 | 2008 | 2018 | 2008 | 2018 | 2008 | 2018 | ||
CAR | 800 | 46.9 | 63.5 | 50.8 | 34.4 | 2.28 | 2.18 | 0.00 | 0.00 |
1600 | 63.3 | 74.1 | 34.7 | 24.1 | 2.00 | 1.76 | 0.00 | 0.00 | |
KEN | 800 | 87.3 | 90.5 | 9.6 | 10.4 | 0.58 | 0.60 | 2.48 | 0.49 |
1600 | 91.7 | 95.7 | 6.4 | 4.02 | 0.14 | 0.16 | 1.71 | 0.13 | |
GUA | 800 | 18.6 | 44.8 | 64.2 | 48.4 | 0.22 | 0.20 | 16.9 | 6.70 |
1600 | 20.1 | 39.3 | 59.7 | 62.5 | 0.33 | 0.13 | 19.9 | 8.97 | |
SUB | 800 | 46.1 | 48.7 | 46.1 | 41.1 | 0.10 | 0.32 | 7.65 | 9.83 |
1600 | 53.8 | 60.4 | 40.9 | 35.9 | 0.14 | 0.2 | 5.14 | 3.6 | |
TUN | 800 | 78.9 | 87.1 | 9.27 | 17.53 | 0.00 | 0.00 | 11.9 | 3.59 |
1600 | 80.3 | 83.3 | 12.7 | 17.9 | 0.04 | 0.20 | 6.92 | 2.73 | |
USQ | 800 | 61.4 | 55.8 | 38.6 | 44.2 | 0.00 | 0.00 | 0.00 | 0.00 |
1600 | 56.3 | 49.1 | 43.2 | 50.2 | 0.53 | 0.57 | 0.00 | 0.00 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ochoa-Alvarado, L.M.; Zafra-Mejía, C.A.; Rondón-Quintana, H.A. Multitemporal Analysis of the Influence of PM10 on Human Mortality According to Urban Land Cover. Atmosphere 2022, 13, 1949. https://doi.org/10.3390/atmos13121949
Ochoa-Alvarado LM, Zafra-Mejía CA, Rondón-Quintana HA. Multitemporal Analysis of the Influence of PM10 on Human Mortality According to Urban Land Cover. Atmosphere. 2022; 13(12):1949. https://doi.org/10.3390/atmos13121949
Chicago/Turabian StyleOchoa-Alvarado, Laura Marcela, Carlos Alfonso Zafra-Mejía, and Hugo Alexander Rondón-Quintana. 2022. "Multitemporal Analysis of the Influence of PM10 on Human Mortality According to Urban Land Cover" Atmosphere 13, no. 12: 1949. https://doi.org/10.3390/atmos13121949
APA StyleOchoa-Alvarado, L. M., Zafra-Mejía, C. A., & Rondón-Quintana, H. A. (2022). Multitemporal Analysis of the Influence of PM10 on Human Mortality According to Urban Land Cover. Atmosphere, 13(12), 1949. https://doi.org/10.3390/atmos13121949