4.3.1 Preliminary
climatology of cold air outbreaks of Wallops Island, VA
1. Buoy data
Buoy data
were examined by the UWash team. They used the following CAO (cold air
outbreak) signature: an increase in wind speed accompanied by an increase in sea-air
temperature difference [SST- Tair] when the wind was in an off-shore
direction, and a peak sustained wind speed of at least 10 m/s. The definition
of off-shore was broad since the coastline is so complex in the region. Two
buoys were used, one off Virginia Beach VA (#44014, at 36.58ºN, 74.83ºW), and
one off Rehoboth Beach DE (#44009, at 38.46ºN, 74.70ºW). The time series for
the first station is shown in Fig 1. At first glance CAOs can occur
anytime during the period shown (1 Oct - 1 Mar), and there is considerable
interannual variability.
Fig. 1: CAO occurrence at buoy 44014 (off Virginia Beach) between
1990-1999. The symbols are explained in the Table below. Data are missing
during the periods marked by a vertical purple line.
|
peak wind speed (m/s): |
Difference between sea and air
temperatures (K): |
||
|
<10-12 |
cyan |
0-4 |
circles |
|
12 to 14 |
blue |
4 to 8 |
squares |
|
14-16 |
green |
8 to 12 |
diamonds |
|
16-18 |
magenta |
12 to 16 |
triangles |
|
>18 |
red |
16-20 |
stars |
Data from buoy 44009 cover the period Oct 1994-Feb 1991, and those from 44014 cover Oct 1990-Feb 1999, so there is little overlap. Data are missing 20% (25%) of the time for buoy 4409 (44014) during the selected periods. Altogether 204 CAOs were selected during a period of 13 winters. Histograms of the duration, peak sustained wind speed, and maximum [SST- Tair] for these 204 events are shown in Fig 2. Data from both buoys are combined in these histograms because the CAO events because of the insignificant overlap period.
Fig 2. Duration, maximum sustained wind, and maximum [SST- Tair] for CAOs between 1984-’99. Selections from both buoys # 44014 and 44009 are merged.
The most likely duration (45 hrs) is quite long, therefore we may be able to examine cloud streets during various phases of a CAO. The distribution of the intensity parameters (maximum wind and temperature difference) is quite narrow, suggesting that is much more likely to find a ‘typical’ CAO (Vmax ~ 13 m/s, SST- Tair ~ 7 K) than a strong event. The corresponding histograms for one buoy (44014) are quite similar to the combined ones (Fig 2), suggesting that the 1991-’99 period was not very different from the ’84-’90 period.
The best month of the year to study CAOs appears to be December (Fig 3), both in terms of frequency and intensity, however other winter months are not much worse. CAOs are rather intense in February, but not as frequent. In October they tend to occur rather frequently, but they are less intense. Some inter-annual/decadal variation does occur: during 1991-’99 (buoy 44014) CAOs were less likely in October than during ’84-’90. During COWEX-I (Feb 1990), 4 CAOs were observed at buoy 44009, with a mean peak wind of 16 m/s, a mean [SST- Tair] of 8 K, and a total duration of 23 % of the month’s duration. In other words CAOs were more frequent and normally intense during Feb 1990. The CAO event of 17 Feb 1990 was rather typical in strength (Vmax = 13 m/s, SST- Tair =6.7 K) but short-lived.
Fig 3: Seasonal variation of frequency and intensity of CAOs, based on buoys 44014 and 44009. The blue line shows the average number of CAO events per month (times 10), the navy blue one shows what fraction of the time of a given month is occupied by CAOs. The yellow and pink lines display measures of CAO intensity.
Peak wind speed, air temperature deficit, and CAO duration are all positively correlated (Fig 4), but the correlations are rather weak. A strong-wind CAO may imply a large temperature deficit of air over water (R=0.37), but strong-wind events are about equally likely to be short-lived as long-lived (R=0.22).
Fig 4: Scatterplots of intensity variables (peak wind speed and air temperature deficit) and duration, for all 204 CAO events.
The relation with two intra-seasonal large-scale oscillations, the NAO and PNA oscillations, are shown in Fig 5. The NAO (North Atlantic oscillation) pattern concerns the intensity of the storm track in the central N. Atlantic and may be associated with CAOs off the mid-Atlantic coast of the USA. Positive values imply a stronger jet , implying that frontal disturbances affect western Europe more intensely/frequently. The PNA (Pacific North American) pattern affects weather from the northwestern Pacific to the Southeastern US. Positive values imply more frequent troughing of the upper-tropospheric flow over the southeastern US.
Fig 5: Variation of the NAO and the PNA indices during the winters of 1990-1999.
Positive values are red, negative blue.
2. GOES data
Geary Schwemmer's team is currently analyzing GOES IR and visible data.