Study of spatial and temporal variation of atmospheric optical parameters and their relation with PM 2.5 concentration over Europe using GIS technologies

Abstract

The purpose of this study was to examine the use of remote sensing aerosol data as an estimator of ground level fine particulate matter concentration (PM 2.5).

In order to examine this possible relation, daily MODIS Aerosol Optical Depth (AOD) data were used, collected for an entire year. The analysis involved manipulation of pollution and meteorological data, such as the PM 2.5 concentration which resulted from a regional photochemical model and meteorological parameters like wind speed (WS), planetary boundary layer height (PBL) and relative humidity (RH), which in turn resulted from the application of a prognostic meteorological model for the whole of Europe. Statistical regression analysis was performed for the aforementioned data in several locations of big urban agglomerations all over Europe, where the problem of particulate matter air pollution is higher, as well as its impact on man and the environment. Furthermore, the relation of AOD with PM 2.5 and meteorological parameters was also examined using PM 2.5 measurements of two operational air pollution stations located in Attica, Greece.

The study confirmed a conclusion reached by other relevant studies, that the relationship between AOD and PM 2.5 is highly variable for different regions and for different time scales (Engel-Cox, 2004; Hu et al, 2013). A strong correlation of AOD – PM 2.5 was established for winter and autumn in most locations. During spring and especially summer the regression models did not produce good results for most of the places that were applied.

The study also confirmed that the use of meteorological data can improve the PM 2.5 to AOD correlation. AOD or AOD/PBL was the most dominant factor in the regression analysis only in 40 % of the cases with good results. In 60 % of cases, one of the meteorological factors (RH, WS or PBL) was the most important factor in the regression equation.

Abstract (Greek, Modern (1453-))

Σκοπός της συγκεκριμένης εργασίας είναι να εξετάσει την πιθανή σχέση ανάμεσα στα σωματίδια που μετρούνται με μεθόδους τηλεπισκόπισης και τα αιωρούμενα σωματίδια της κατώτερης ατμόσφαιρας (PM 2.5). Για το σκοπό αυτό χρησιμοποιείται το οπτικό βάθος (Aerosol Optical Depth) του MODIS σε ημερήσια βάση και για ένα ολόκληρο έτος. Για τις συγκεντρώσεις των σωματιδίων χρησιμοποιούνται τα αποτελέσματα ενός φωτοχημικού μοντέλου. Η περιοχή αναφοράς είναι η Ευρώπη. Στην εξεταζόμενη σχέση λαμβάνονται επιπλέον υπόψη και μετεωρολογικές παράμετροι όπως η υγρασία, η ταχύτητα ανέμου και το ύψος ανάμιξης όπως αυτά έχουν προκύψει από την εφαρμογή ενός προγνωστικού μετεωρολογικού μοντέλου. Η στατιστική ανάλυση πραγματοποιείται σε επιλεγμένα σημεία που αφορούν μεγάλα αστικά κέντρα της Ευρώπης, όπου το πρόβλημα τη σωματιδιακής αέριας ρύπανσης είναι μεγαλύτερο όπως και οι επιπτώσεις της στον άνθρωπο και στο περιβάλλον. Η σχέση αυτών των παραμέτρων εξετάζεται επίσης με βάση τα δεδομένα μετρήσεων δύο σταθμών ατμοσφαιρικής ρύπανσης στην Αττική.

Τα αποτελέσματα της εργασίας επιβεβαίωσαν ότι, η συσχέτιση της συγκέντρωσης των αιωρούμενων σωματιδίων με το οπτικό βάθος που προκύπτει από τις δορυφορικές παρατηρήσεις, είναι ιδιαίτερα μεταβλητή για διαφορετικές περιοχές και χρονικές κλίμακες. Επομένως η σχέση μεταξύ αυτών των παραμέτρων μπορεί να εξεταστεί μόνο τοπικά και λαμβάνοντας υπόψη την εποχική μεταβλητότητα.

Η εργασία επιβεβαίωσε επίσης ότι η χρήση μετεωρολογικών παραμέτρων μπορεί να βελτιώσει τη συσχέτιση μεταξύ PM 2.5 και AOD. Αυτό προκύπτει από το γεγονός ότι, στις περιπτώσεις που εξετάστηκαν, μόνο στο 40% αυτών το AOD ήταν η πιο σημαντική παράμετρος συσχέτισης, ενώ στο υπόλοιπο 60% ως πλέον σημαντική παράμετρος εμφανίζεται μία εκ των μετεωρολογικών παραμέτρων όπως η υγρασία, η ταχύτητα ανέμου και το ύψος ανάμιξης.

Popular Abstract

Air pollution is one of the most important environmental problems, since it is the cause of diseases or even deaths in humans, it damages other living organisms like animals and plants, and generally it is harmful for the natural or even the built environment. In order to monitor air pollution, most European countries have set up air quality monitoring networks to assess ambient air quality in respect to several pollutants like Ozone (O3), Carbon Monoxide (CO), Nitrogen Oxides (NOx), and Particulate matter (PM). However, according to the European Environmental Agency, since setting up a monitoring station is costly, other techniques, such as satellite imagery and air quality models, should be used in order to acquire deeper knowledge of Europe’s air quality.

In this context, the use of remote sensing data could be an excellent alternative solution, since, its global coverage, has the potential to provide air quality information for the entire European territory. Nevertheless, since the various substances that the European policy requires monitoring are not measured directly using the satellite instruments, a relation between them should be initially established. This study aims to investigate this method as an alternative for air quality assessment / monitoring. For that purpose, remote sensing data related to the quantity of the aerosols in the atmosphere were used. These data were compared to fine particulate matter (PM 2.5) concentration in several European cities all over Europe, taking into account also the meteorological conditions, in order to establish a possible relation between them.

The results of this analysis showed that air pollution is a very localized problem and thus any relation between satellite data and air pollution levels should be examined separately in each location. Also, since during each year the factors that are related to air pollution, like emission and meteorology, are very different, the relation should also be examined seasonally. Having that in mind, a strong correlation of remote sensing data and PM 2.5 concentration was found for winter and autumn in most locations. During spring and especially summer the results were not good for most of the places. The use of meteorological parameters like wind speed and relative humidity can significantly improve our ability to estimate air pollution levels using remote sensing data. It is foreseen, that in the future, improvements in satellite remote sensing of the atmosphere, as well as the use of higher spatial and temporal resolution data, will lead to more accurate predictions of PM 2.5 based on satellite observations.

• author: Symeonidis, Panagiotis
• supervisor: Thomas Holst
• organization: Dept of Physical Geography and Ecosystem Science
• course: GISM01 20171
• year: 2017
• type: H2 - Master's Degree (Two Years)
• subject: Earth and Environmental Sciences
• keywords: aerosols, air quality, satellite remote sensing, GIS, geography, Geographical Information Systems, particulate matter, Aerosol Optical Depth, MODIS
• publication/series: Master Thesis in Geographical Information Science
• report number: 66
• language: English
• id: 8902398
• date added to LUP: 2017-02-06 10:57:46
• date last changed: 2017-02-06 10:57:46