Particulate matter variability in Kathmandu based on in-situ measurements, remote sensing, and reanalysis data

Kathmandu has one of the highest particulate matter air pollution levels in the world. However, few direct measurement data are available for long-term analyses, limiting policy interventions and public health advisories. Remote sensing-based data sets provide an alternative approach to address this issue. In this paper, we present an approach to analyze and understand the diurnal, seasonal, annual, and multi-annual variability of pollution levels based on in situ measurements of particulate matter (PM2.5), remote sensing data based on Moderate Resolution Imaging Spectroradiometer (MODIS) Aqua and Terra Aerosol Optical Depth (AOD), and aerosol mass concentration retrievals, as well as AOD and PM2.5 data from Copernicus Atmosphere Monitoring Service (CAMS) and Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA-2) reanalysis data sets. The analysis of the direct measurements revealed distinct annual patterns, characterized by winter maxima and summer minima. With the exception of the summer monsoon season, public health guidelines are frequently exceeded significantly throughout the year, particularly in winter. The analysis furthermore pointed toward distinct daily patterns with primary maxima in the mornings, secondary maxima in the late evenings, and minima in the afternoons. The annual pattern of AOD derived from the MODIS data is markedly different from that. Due to the coarse spatial resolution and the fact that MODIS AOD is a column integrated property, it does not reflect the small scale phenomenon of the Kathmandu urban pollution pattern but instead shows a maximum in the spring. The same pattern was observed with the CAMS and MERRA-2 reanalysis AOD data, even though MERRA-2 captures pollution levels during the summer monsoon season very well. The CAMS reanalysis PM2.5 data are generally well-aligned with the near-surface measurement data, even though they overestimate the daily and monthly maxima and do not capture the morning maxima in the diurnal course. Nevertheless, CAMS PM2.5 data can be adjusted via linear regression to reasonably mirror the measurements. It shows that PM2.5 concentrations in Kathmandu have increased significantly in the past decades (almost 2 mu g/m3 annually), mainly after the monsoon season from September to February. Our results indicate that around 85% of all winter days in Kathmandu are categorized as "unhealthy" according to the Air Quality Index (AQI).