Spatio-temporal analysis of Aerosol Optical Thickness using Visible Infrared Imaging Radiometer Suite Suomi-National Polar-orbiting Partnership at different aggregation levels

Abstract

Air pollution has considerably increased in south Asia because of industrialization, urbanization and burning of agricultural residue. Agricultural Plains and northern region of Pakistan and India have been severely affected by dense fog episodes. Hence, we mapped Aerosol Optical Thickness (AOT) and its sources at regional level. The objective of this study was to investigate the trend analysis of AOT, Enhanced Vegetation Index (EVI) and fire events over the plains and northern region of Pakistan and India. To achieve this, data from the VIIRS S-NPP satellite for AOT, EVI and fire events was downloaded. It was transformed to annual, monthly and fog time period (October to February) aggregation level. Lag-1 autocorrelation coefficient (r) were calculated at each aggregation level to determine the autocorrelated time series trend. EVI had the greatest number of autocorrelated time series (29/52) as compared to AOT (10/60) and fire events (13/54). These autocorrelated time series were replaced by per-whitened series and Mann-Kendall trend were calculated. It was observed that AOT had a significantly increasing trend for monthly aggregation over Bahawalpur (p-value = 0.273, ZMK = 1.905). While for for-time period Bahawalpur (p-value = 0.175, ZMK = 1.755), Multan (p-value = 0.133, ZMK = 1.943) and for annul aggregation level Multan (p-value = 0.072, ZMK = 3.113), Jhang (p-value = 0.2, ZMK = −2.3), Chakwal (p-value = 0.2, ZMK =2.075). Similarly, Fire events have a significant trend over Jhang (p-value = 0.196, ZMK = 2.2), Chakwal (p-value = 0.2, ZMK = 1.936), Bahawalnagar (p-value = 0.2, ZMK = 2.18) for monthly aggregation level. Whereas for fog-time Chakwal (p-value = 0.2, ZMK = 2.14) and for annul aggregation level New Delhi (p-value = 0.072, ZMK = 1.342), Bahawalpur (p-value = 0.007, ZMK = −80.48), Shimla (p-value = 0.230, ZMK = 0.895), Aligarh (p-value = 0.072, ZMK = 1.342), Multan (p-value = 0.23, ZMK = 0.895), Chakwal (p-value = 0.133, ZMK = 1.118), Bahawalnagar (p-value = 0.072, ZMK = −53.657). However, EVI have significant trend for monthly aggregation over Bahawalnagar (p-value = 0.139, ZMK = 3.961). Moreover, for fog-time period Multan (p-value = 0.2, ZMK = 1.661), Patiala (p-value = 0.078, ZMK = 2.548) and Sirsa (p-value = 0.096, ZMK = 2.409). While for annual aggregation level Shimla (p-value = 0.2, ZMK = 1.767) and Chakwal (p-value = 0.2, ZMK = -18.106) had the significant trend. AOT increased from May to July and October to February. Whereas, EVI has an increasing trend from January to March and July to October. It was noticed that AOT were high over Lahore and New Delhi. While Chamba, Gurdaspur and Shimla had the lowest AOT. Whereas, Lahore and Patiala had the highest number of fire events at all aggregation levels and lowest EVI were recorded in Bahawalnagar, Chakwal and Churu.