Assessment of Spatiotemporal Variability and Rising Trends of Surface Ozone Over India

Abstract With rising anthropogenic activities, surface ozone levels have increased across different parts of the world including India. Studies have shown that surface ozone shows distinct characteristics across India, however these studies are mostly based on isolated locations. Any comprehensive and spatiotemporally consistent study on surface ozone is lacking thus far. Keeping these facts in mind, we utilize ground-based observations and reanalysis datasets to investigate the surface ozone variability, seasonality, and linkages with meteorology over India. A validation exercise shows that the Copernicus Atmosphere Monitoring Service Reanalysis (CAMSRA) reasonably compares against the ground-based observations. Results show that the CAMSRA ozone is in good agreement with the observations across India, where it shows better correlations (r>0.7) over southern regions and relatively lesser (> 0.5) over northern and eastern regions indicating larger variability and spread over these regions. We further quantify this agreement in terms of range, mean absolute error (MAE), and root mean square error (RMSE). Time series analysis shows that the CAMSRA captures seasonal variations irrespective of location. Spatial distribution of surface ozone shows higher (lower) concentrations of about 40-60 ppb (15-20 ppb) during pre-monsoon (monsoon) months over a broad region covering northern and western parts, and peninsular India. A prominent increase during May is noted over the northern region especially over Indo-Gangetic Plains (IGP). These seasonal variations are linked to solar radiation (SR), temperature, low-level circulation, and boundary layer height (BLH). Furthermore, Principal Component Analysis (PCA) is performed to understand the dominant patterns of spatiotemporal variability for different seasons. It is seen that the first (second) mode shows a high percentage variance explained ranging between 30-50% (10-20%). The time series of PCA-1 mode indicates an overall increasing trend across India with a notable increase over south and central India. The second mode indicates prominent variability over the IGP (southern India) in the pre-monsoon (post-monsoon) season, which shows significant interannual variability. During the monsoon season, an interesting dipole pattern is seen which closely resembles the active and break spell patterns of the Indian summer monsoon. Overall, the spatiotemporal variations in surface ozone are closely tied to meteorology while the rising trends indicate the potential role of increasing precursors across India.