Improving surface-based precipitation phase determination through air mass boundary identification

Most hydrological models apply one empirical formula based on surface air temperature for precipitation phase determination. This approach is flawed as surface precipitation phase results from energy exchanges between falling precipitation and air in the lower atmosphere. Different lower atmospheric conditions cause different precipitation phase probabilities for near-freezing temperatures. Often directly measured lower atmospheric conditions are not available for remote areas. However, meteorological observations occurring before/after similar air mass boundaries have similar atmospheric conditions that vary from most other observations. Therefore, hydrological models can indirectly account for lower atmospheric conditions. Twenty years of manual observations from eight United States weather stations were used to compare misclassified precipitation proportions when analyzing (a) all precipitation observations together and (b) identified cold air mass boundary observations (CAB) and non-CAB observations separately. The CAB observations were identified by a rapid surface air temperature decrease. A two-surface air temperature threshold method with one threshold all snow (T_S°C) and one all rain (T_R°C) having a linear snow fraction decrease between the thresholds was used. The T_S (0 °C), and T_R (4 °C) values for CAB were 1 °C warmer than for non-CAB (−1 °C, 3 °C). Analyzing CAB and non-CAB separately reduced misclassified precipitation 23%, from 7.0 to 5.4%.