| Surface-based temperature inversions(SBIs)are low-level inversion layers directly influenced by surface,which limit the diffusion of substance,energy and vapor in the near-surface layer due to the weak turbulence and low wind speed of the SBIs.They had a great impact on the environment and human activities by accumulating of pollutants in the boundary layer and influencing the formation of clouds,fog,and the sea ice concentration.In short,understanding the long-term changes in SBIs can help us better understand their impacts on air quality,human health,and some of the potential impacts on climate change and investigating surface-based inversions from a global perspective can filled a gap in the current research on SBIs.This study used radiosonde data from 493 radiosonde stations worldwide from the Integrated Global Radiosonde Archive(IGRA)during 1989-2019 to investigate the variations in surfacebased inversions(inversion frequency,inversion difference,and inversion depth)from a global perspective.Firstly,the interannual variations and spatial distributions of the global SBIs were calculated,followed by the analysis of the diurnal variations of SBIs.Second is the comparisons of SBIs characteristics in different regions.Finally,various meteorological factors affecting the long-term variations of the SBIs have been analyzed and their contributions to the variations of the SBIs have been quantified.The main results are as follows:(1)During 1989-2019,SBI frequency showed an increasing trend and SBI depth showed a decreasing trend,but there was no significant increasing or decreasing trend of SBI difference.However,the spatial distributions of frequency,difference,and depth did not have consistent trends and not displayed the same trends as those of the global average.Generally speaking,SBI frequency,difference,and depth increased with increasing latitude and values of the three parameters of SBIs were smaller over the ocean than over land.(2)The correlation analysis of the interannual variations of the three parameters of SBIs showed that SBI frequency were significantly negatively correlated with SBI depth and significantly positively correlated with SBI difference while the correlation between SBI depth and difference was very weak.(3)The frequency of nighttime SBIs was almost twice that of daytime SBIs,but the difference and depth of nighttime and daytime SBIs were very close.The SBI frequency,difference,and depth showed obvious diurnal variations.(4)The SBI difference and depth interannual variations in the Northern Hemisphere seemed similar to those in the Southern Hemisphere.The SBI frequency,difference,and depth in both hemispheres showed clear annual cycles.The values of frequency,difference,and depth were greater in the Northern Hemisphere than in the Southern Hemisphere and the same as over land than the ocean.What’s more,the variations in SBI over the ocean were not as drastic as those over land in both the Southern and Northern Hemispheres.The latitudinal distribution of SBIs was characteristic: The interannual and annual variations of SBIs were usually greater at higher latitudes than at lower latitudes.(5)The correlation coefficient between SBI frequency and surface temperature was 0.59;the correlations between SBI difference and meteorological factors were very small and did not pass the significance test;the correlation coefficients between the SBI depth and surface temperature,total column water,and LTS were-0.82,-0.39 and 0.48,respectively.SBI frequency,difference,and depth in most of the stations displayed positively correlation with LTS and negatively correlation with surface temperature.The results showed that significant effects of surface temperature and LTS on long-term variations in global SBI frequency and their contribution to the long-term variations of SBI frequency was 71.0% and 29.0%,respectively;significant effects of surface temperature,surface net thermal radiation,and total column water on long-term variations in global SBI depth and their contribution to the long-term variations of SBI depth was 58.7%,15.2%,and 26.1%,respectively. |