Assessing Temperature-Dependent Bias in Offline Filter-Based Nitrate Measurements: Insights from Multi-Year PM2.5 and PM10 Comparisons with Monitor for Aerosols and Gases in Ambient Air (MARGA)

Nitrate is a major component of atmospheric particulate matter (PM), and its accurate quantification is essential for air quality assessment, policy evaluation, and atmospheric modeling. Offline filter-based methods are widely used due to their cost-effectiveness and simplicity, but their reliability for nitrate is compromised by volatilization and other sampling artifacts. Here, we compared offline filter measurements against an online monitor for aerosols and gases in ambient air (MARGA) system, used as a more reliable operational reference, for PM2.5 and PM10 nitrate at an urban and a suburban site in Hong Kong over multi-year periods by comparing daily filter-based data with daily averages derived from hourly MARGA measurements. PM2.5 sulfate and ammonium showed excellent agreement between methods (slopes ≈1, R2 ≥0.87), whereas nitrate exhibited greater scatter and a substantial, systematic downward bias in the filter-based data. Temperature emerged as the most prominent first-order predictor of this bias. Using an empirical threshold of 23 °C, derived from the observed intermethod relationship and interpreted as an operational divider rather than a universal thermodynamic threshold, we found that filter-based PM2.5 nitrate was notably underestimated at higher temperatures and less correlated with MARGA, implying substantially greater uncertainty in empirical correction. At lower temperatures, underestimation persisted, but the filter and MARGA measurements were more strongly correlated, supporting more constrained correction. Filter-based PM10 nitrate showed better agreement with MARGA than PM2.5 nitrate under both temperature regimes, although the relationships remained temperature-dependent. These results support the use of temperature-stratified, data set-specific evaluation of filter-based nitrate measurements and motivate further studies to establish more broadly applicable correction approaches.