Technical note: An interactive dashboard to facilitate quality control of in-situ atmospheric composition measurements

Abstract. In-situ measurements of trace gases are crucial for monitoring changes in the atmosphere's composition and understanding the underlying processes that drive them. For over three decades, the Global Atmosphere Watch (GAW) programme of the World Meteorological Organization (WMO) has coordinated a network of surface monitoring stations and facilities with the goal of providing high-quality atmospheric composition measurements worldwide. One of the critical challenges towards this goal is the spatially unbalanced availability of high-quality time series, and the lack of near-realtime quality control (QC) procedures that would allow the prompt detection of unreliable data. Here, we describe an interactive dashboard designed for GAW station operators, but which may be of much wider use, that is able to flag anomalous values in near-realtime or historical data. The dashboard combines three distinct algorithms that identify anomalous measurements: (i) an outlier detection based on the Subsequence Local Outlier Factor (Sub-LOF) method, (ii) a comparison with numerical forecasts coupled with a machine learning model, and (iii) a Seasonal Autoregressive Integrated Moving Average (SARIMA) regression model. The application, called GAW-QC, can process measurements of methane (CH₄), carbon monoxide (CO), carbon dioxide (CO₂), and ozone (O₃) at hourly resolution, offering multiple statistical and visual aids to help users to identify problematic data. By enhancing QC capabilities, GAW-QC contributes to the GAW programme's goal of providing reliable atmospheric measurements worldwide.