Explore the vast range of titles available.

Drastic measures such as lockdown imposed in the countries worldwide to control the extent of COVID-19 have influenced environmental parameters substantially. The aim of the present study was to investigate the impact of lockdown on the air quality in Hyderabad city by comparing the pollutants concentration during lockdown and pre-lockdown periods. A comparative study was also made on the pollutant concentrations observed during the pre-lockdown (1st February – 23rd March 2020) and lockdown period (24th March - 30th April 2020) to those of the pollutants in the previous years (2018 and 2019). The Pearson correlation coefficient was employed to correlate the ozone (O 3 ) concentration with other pollutants. Carbon monoxide (CO), nitrogen oxides (NO X ) and O 3 were monitored along with meteorological parameters like temperature, relative humidity and solar radiation. It was observed that the O 3 concentration increased from 26 ppb (by volume) to 56.4 ppb during pre-lockdown and lockdown period, respectively, due to the decrease in CO and NO X concentration. The concentration of NO 2, NO and CO were also reduced during the lockdown period by 33.7%, 53.8% and 27.25%, respectively. To identify the statistical significance of the parameters, analysis of variance (ANOVA) was used. The present study provides new insights on the ambient air pollution in terms of the aforesaid parameters and could pave the way for regulatory authorities to implement control measures to curb the air pollution. Highlights • COVID-19 lockdown improved the air quality by the reduction in air pollutants leading to increased ozone concentration. • Pearson model was used to correlate the ozone (O 3 ) with oxides of nitrogen (NO X ), CO and other meteorological parameters. • NO X and carbon monoxide (CO) concentrations were reduced by 33.7 % and 27.25 %, respectively, due to COVID lockdown. Graphical abstract

To assess the surface ozone concentration (O 3 ), there is a need to establish relationship between air pollutants and meteorological parameters. The study was conducted on variation of air pollutants, viz. O 3 , nitrogen oxides (NO X  = NO 2  + NO) and carbon monoxide (CO) along with meteorological parameters like temperature (Temp), relative humidity (RH), solar radiation (SR) and wind speed (WS). The precursor gases were recorded in Hyderabad at Tata Institute of Fundamental Research-National Balloon Facility (TIFR-NBF; 17.47° N, 78.58° E). Correlation analysis is done on hourly averaged trace gases concentration and metrological data for the entire year 2016. O 3 is in negative correlation with NO X , CO and RH. NO X which is one of the precursor gases plays a major role in formation of O 3 by photo-chemical reaction (PCR). The increase in O 3 concentration is in proportion with the decrease in NO X concentration. O 3 correlated positively with Temp, SR and WS. Two sets of four models were constructed with multiple linear regression (MLR) representing the data for the three seasons (summer, winter and monsoon) and for the total year as well. The adjusted R 2 was determined and found to be in the range of 0.6 to 0.9 for the models using precursor gases and 0.9 by meteorological parameters. The models were validated by various performance indicators, viz. root mean square error (RMSE), mean absolute error (MAE) and mean biased error (MBE).

Development of innovative methodologies in environmental modelling using machine learning and artificial intelligence techniques for forecasting and prediction of surface ozone (O3), at ground level is needed to anticipate future climate projections and for effective air quality management. Several complexities like physical and chemical phenomena are involved in O3 formation from its precursors like nitrogen oxide (NO), carbon monoxide (CO), nitrogen dioxide (NO2) and oxides of nitrogen (NOx). It is important to understand and formulate nonlinear relationships in O3 formation using robust data-driven predictive models from time to time despite the existence of several O3 prediction models. This study focuses on the development of back-propagation artificial neural networks (BPANNs) for the prediction of seasonal and diurnal O3 concentrations using the data monitored at a specified location in Hyderabad, India (TIFR-NBF: Tata Institute of Fundamental Research-National Balloon Facility) during the period 2014–2016. The efficiency and performance of BPANNs in modelling with 80% of monitored data for training and 20% of data for validation of seasonal and diurnal ozone concentration and showed higher R2 (0.9999). The evaluation statistics of all the three seasonal models measured in terms of root mean square error (RMSE), mean absolute error (MAE) and mean square error (MSE) showed a better predictive ability against the monitored data fell within a 10% margin over response surface methodology (RSM) which indicate that it is much more confident and accurate in prediction of O3.

Development of innovative methodologies in environmental modelling using machine learning and artificial intelligence techniques for forecasting and prediction of surface ozone (O 3 ), at ground level is needed to anticipate future climate projections and for effective air quality management. Several complexities like physical and chemical phenomena are involved in O 3 formation from its precursors like nitrogen oxide (NO), carbon monoxide (CO), nitrogen dioxide (NO 2 ) and oxides of nitrogen (NO x ). It is important to understand and formulate nonlinear relationships in O 3 formation using robust data-driven predictive models from time to time despite the existence of several O 3 prediction models. This study focuses on the development of back-propagation artificial neural networks (BPANNs) for the prediction of seasonal and diurnal O 3 concentrations using the data monitored at a specified location in Hyderabad, India (TIFR-NBF: Tata Institute of Fundamental Research-National Balloon Facility) during the period 2014–2016. The efficiency and performance of BPANNs in modelling with 80% of monitored data for training and 20% of data for validation of seasonal and diurnal ozone concentration and showed higher R 2 (0.9999). The evaluation statistics of all the three seasonal models measured in terms of root mean square error (RMSE), mean absolute error (MAE) and mean square error (MSE) showed a better predictive ability against the monitored data fell within a 10% margin over response surface methodology (RSM) which indicate that it is much more confident and accurate in prediction of O 3 .