An analogue dynamical model for forecasting fog‐induced visibility: validation over Delhi

ABSTRACT Accurate forecasts of fog and visibility are important for many applications; while prolonged fog can adversely affect many crops, even a short duration of dense fog can lead to disruption of air and highway traffic. The genesis and dynamics of fog are a result of many processes; accurate forecasting of fog thus continues to be a challenge. A forecast model of the occurrence of fog, measured in terms of visibility, is presented. The model is formulated as an analogue model; thus the merit of the model is primarily based on its validation against observation. Two forecasts using two sets of meteorological fields are considered: one as the benchmark forecasts with visibility calculated from observed meteorological fields and the other based on meteorological forecasts from an atmospheric mesoscale model (Weather Research and Forecasting). While the benchmark (perfect) forecasts from observed meteorological fields provide the potential skill of the model, the mesoscale forecasts provide an assessment of realizable skill in an operational setting. The validation was carried out against hourly visibility data recorded at Indira Gandhi International Airport over Delhi during the winter months (December and January) for the period 2009–2012. Error statistics show that the analogue fog model can capture a significant part of the observed variability of fog. The forecasts have more success in forecasting intense (visibility 4 h) fog events. The model provides a useful forecasting tool, as shown by measures such as average error, number of false warnings and the number of misses.