The spatiotemporal variations of total column ozone concentration over Ethiopia

Abstract

We have studied spatiotemporal characteristics of ozone concentration over Ethiopia using data from the Ozone Mapper and Profiling Suite-Nadir Mapper (OMPS-NM). Daily total column ozone data of 108 observation points with spatial resolution 1° × 1° over the study area for the period of 2012–2019 have been analyzed. The spatial variations over the region have been studied by considering longitudinal and latitudinal bands separately through the sample mean difference among different bands using multi-comparison analysis of variance technique in order to identify the clusters in the region. For the temporal variability, we model the total column ozone timeseries observation as a sum of seasonal, trend, and temporally correlated noise components. We have found that the total column ozone concentration has a maximum value of 301DU during summer on August 18, 2013 and a minimum value of 216DU during winter on January 03, 2013 over the study period. The 95% confidence level of the overall mean of the total column ozone concentration during the study period was found to be (261.35 ± 2.38)DU. Our spatial data analysis revealed that the spatial distribution of ozone over Ethiopia can be classified into three clusters: southern cluster (4.5°N–8.5°N and 33.5°E–47.5°E), north-eastern cluster (9.5°N to 14.5°N and 41.5°E–47.5°E), and north-western cluster (9.5°N–14.5°N and 33.5°E–40.5°E). We have checked the coefficient of determination among bands in the same cluster to see if the concentration of ozone in one band can be explained by the concentration in another band for each cluster and confirmed the reliability of the classification. In order to capture temporal characteristics, we have computed the spectral periodogram for each cluster and obtained a power peak at frequency f = 0.002 768 Hz, which indicates that the ozone concentration over the region exhibits an annual cyclic behavior. A truncated Fourier series fit is used to determine the annual seasonal component. The non-parametric Mann–Kendall’s trend test with a 95% confidence level of significant has indicated a decreasing linear trend with a depletion rate of 0.77, 0.73, and 0.43 DU/yr over north-western, north-eastern, and southern clusters, respectively. The analysis of residuals for each cluster indicated that the residuals are normally distributed with no significant outliers, and the model explains 85%, 86%, and 79% of the variance in the north-western, north-eastern, and southern clusters, respectively, demonstrating the reliability of the model considered in this study.