Processes influencing summertime ozone concentrations at a high Arctic site

Arctic surface ozone (O₃) is an important short lived climate forcer as it interacts with light both in the solar light and in the infrared region and thus it plays an important role during Arctic summer. Surface O₃ in the High Arctic exhibits substantial variability during summer months, driven by complex interactions between photochemistry, long-range transport, and boundary layer dynamics. Understanding the relative contributions of these processes and their long-term changes is critical for interpreting observed O₃ variability and projecting future changes under rapid Arctic climate warming. We present a comprehensive analysis of summertime (June-August) O₃ at a high Arctic monitoring station in North-east Greenland (Villum Research Station) spanning three decades (1995-2024), combining advanced statistical decomposition methods, back trajectory analysis, data-driven Bayesian modeling for entrainment detection, and chemical transport model simulations to quantify the major processes controlling surface O₃ concentrations and assess their temporal evolution. Long-term analysis using traditional (Mann-Kendall test and Sen’s slope) and STL (Seasonal-Trend decomposition using Loess) reveals complex temporal patterns in summertime O₃ and its baseline concentrations over the 30-year period, with substantial interannual variability. STL decomposes the time series into baseline, seasonal component, and residuals, enabling process-specific analysis. Back trajectory analysis comparing high versus low O₃ episodes identifies distinct source regions and transport pathways. Trajectories are categorized by surface type (land, snow, sea ice, ocean) and altitude (within versus above mixing layer). High O₃ episodes are predominantly associated with air masses from above the mixing layer and open ocean, whereas low O₃ periods show dominant patterns indicating sea ice and land sources. To quantify local boundary layer entrainment processes, we apply automated entrainment detection on the STL residuals, which isolate short-term variability after removing baseline and seasonal components. Entrainment events bring O₃-rich free tropospheric air to the surface, characterized by simultaneous O₃ increases, relative humidity (RH) decreases at 9m, and enhanced vertical RH gradients. A two-stage Bayesian inference approach is developed to first screens candidates using physical thresholds, following by probabilistically estimates event timing, magnitude, and persistence while accounting for measurement uncertainty.  We analyze temporal patterns in entrainment frequency and magnitude over the three decades to assess potential changes in boundary layer dynamics. To complement the observational analysis, we employ the Danish Eulerian Hemispheric Model (DEHM), a chemical transport model, to perform long-term O₃ simulations. The model quantifies stratospheric contributions to surface O₃ and enables evaluation of how well current chemical transport schemes capture the observed variability and process attribution identified through the statistical and trajectory analyses. This integrated approach provides robust process attribution and understanding by linking observed O₃ to air mass origin, transport characteristics, vertical mixing, and stratospheric inputs, demonstrating that Arctic O₃ variability results from complex interplay of hemispheric transport, local meteorology, and boundary layer dynamics.