인천공항의 안개 유형별 분석 및 예측연구

Abstract

ICA(Incheon Airport) was awarded Best Airport Worldwide in 2005, while having been settled down as a hub airport for North-East Asia. Nevertheless it recorded 11,685 cancels or delays out of 541,125 total in air traffic operations for the recent four years. Among 11,685 cancels or delays, cancels which were caused by weather, were 1,694 (14%), primarily which caused by fogs were 968 (58%). For the purpose of improving the capability of fog forecast, we analysis fog climatology with the data collected from 2001 to 2004 and developed the AWSRPS(Aviation Weather Short range Prediction System). In addition to analysis two neighboring weather stations, observed on Gimpo Airport located in inland and on ICA, Incheon City in seashore, respectively. Fog cases during the study period was characterized by occurring causes into radiation, advection, stream and front fogs according to the classification criteria being employed at operational service in Aviation Meteorological Office. For fog classification, surface weather data collected from neighboring regions that may significantly influence fog occurrence at ICA were analyzed, while upper-layer data from Baengnyeongdo and Osan, lighthouse data from Yeonpyeongdo and Sonmido, sea-surface buoy data from Deokjeokdo, respectively. Analysis on the occurring characteristics of fogs using observations reveals that, in both ICA and Incheon Weather Station, most frequent fogs took place in summer (97 days - 36% and 64 days - 39%, respectively), least frequent in Fall (34 days - 13%, 17 days - 10%, respectively). On the contrary, at Gimpo Airport for 71 days (36%) in Fall, while 34 days (17%) in spring fog took place on average. This indicates that Incheon was significantly influenced by sea fog, while Gimpo by surface inversion of inland in Autumn. Each frequency of fogs occurred at ICA was in the order of 162 Advection fog, 143 Frontal fog, 71 Radiation fog, and 36 Steam fog, respectively, being commonly influenced by South-westerly advection in summer, while precipitation falling into a weakened inversion in summer and shallow cold pool in winter. In this case of radiation fog, 32% was occurred from October to November, mostly between 0400∼0700 KST of 37 events (52%). For advection fog it occurred in July with 40 events (25%), while most frequently between 0400∼0700 KST of 52 events (32%). steam fog took place from November through March, most frequently in February with 14 events (39%), while frontal fog was formed mostly from January through July with 126 times (85%), primarily in the morning with 101 times (71%). Under saturation deficit, with no significant correlation with any cause for fogs, below 1.5℃ fog was mostly formed as many as 375 times (91%). In order to analyze the fogs at ICA in terms of main cause, numerical validation was conducted under the Short-range Aerodrome Forecasting System. NWP(Numerical Weather Prediction) model can simulate synoptic background similar to field observations in fog forecasting, but significant differences in simulating pressure system flow speed, wind and humidity at local scale. Radiation fog (Case 1) can be formed in general under pressure pattern with saddle type or weak pressure gradient and surface inversion, while negative value at surface but positive value at upper layer in Richardson number with about 200 m height. Advection fog (Case 2) can be characterized when South-westerly wind was dominant and surface inversion took place with 7∼8 kts South-westerly, 5∼7 of Richardson number and below 100 m height. Front fog (Case 3) occurred commonly when front or low pressure passes through Korea with 7∼8 kts southerly wind, 5∼7 of Richardson number, and below 100 m height. At ICA when the humidity was about 98% with 1∼4 of turbulence, the cloud water (fog) condensed on the surface was correlated with fog events. For more accurate fog forecasts in Korea in the future, we need not only more dense observation network, but also radiometers at each aerodrome station to analyze observation radiosonde. Furthermore, great effort should be made to develop NWP models that can reflect complex nature of Aerodrome's topographic features.