GOAL: To develop stratified satellite based cloud frequency climatologies to aid in forecasting the timing and extent of sea breeze convective development under various synoptic wind regimes.
|
BACKGROUND: Ken Gould, NWS Tallahassee, started developing sea-breeze satellite composites on RAMSDIS for the summer months of June, July, and August in 1995. The initial results have shown immediate applicability to convective forecasts, nowcasts, and terminal forecasts, especially when integrated with the mesoscale thermodynamic and lightning composites. Initially, the analysis was qualitative, focusing on averaged visible imagery. The analysis is now quantitative, focusing on visible and infrared cloud frequency composites. Examples are shown here of the original visible average satellite composites, the new visible cloud frequency composites and select infrared cloud frequency composites. A few of the corresponding lightning composites derived from ground based sensors are also shown for comparison.. Over time, these satellite composites will form climatologies. |
METHODS
1) Regime Designation: During June, July and August, each day is designated with a particular synoptic
flow regime based on the 12 UTC 1000-700 hPa Mean Layer Vector Wind (MLVW). The MLVW was calculated using
PC-Gridds Interactive Display and Diagnostic System (PC-GRIDDS) software for 1996-98 and Advanced Weather Information
Processing System (AWIPS) during 1999-2000, using both the Eta and NGM model output. The designated
regimes are shown in Table 1.
|
Table 1. |
Tallahassee Summer Sea-breeze Regimes |
|
|---|---|---|
|
Regime |
Description |
|
|
1 |
Light and variable or light SE |
|
|
2 |
Light to moderate (3 to 10 kts) E to NE |
|
|
3 |
Strong (> 10 kts) E to NE |
|
|
4 |
Light to moderate (3 to 10 kts) W to SW |
|
|
5 |
Strong (> 10 kts) W to SW |
|
|
6 |
Moderate (6 to 10 kts) SE to S |
|
|
7 |
Strong (> 10 kts) SE to S |
|
|
8 |
Light to moderate (3 to 10 kts) N to NW |
|
|
9 |
Strong (> 10 kts) N to NW |
2) Collection, Archival , and Processing of GOES-8
Visible Imagery:
The first year of imagery was collected at 2.2 km resolution. Imagery for the following years were collected
at 1 km resolution. Only the results from the 1 km data set are shown here (ie from 1996 - 2000).
Visible imagery from 1315-2215 UTC were archived at hourly intervals at CIRA.
The imagery were adjusted for albedo variations as a function of the changing solar directions throughout
the day.
The imagery were checked for navigation errors.
Background images were determined from compositing the minimum brightness counts (darkest pixels) of various
images.
Using the background image as a reference, cloud/no-cloud images were created.
10.7 um Imagery:
Imagery were collected at full resolution (4 km) and were archived at hourly intervals from 0015-2315 UTC.
Since the end of the convective day did not correspond to 2315 UTC, the 10.7 µm imagery for a particular regime
day were analyzed from 0715 UTC of the current day to 0615 UTC of the next day (1115 PM EDT to 1015 PM EDT).
In order to compare with the visible cloud frequency composites, the threshold temperature of 283 K was chosen to distinguish between cloudy and clear "background" regions.
Pixels equal to or colder than this value were designated as cloudy.
A threshold of 235 K was used to determine the frequency and location of deep convection.
The imagery were then compiled into the appropriate synoptic regime using programs developed with McIDAS
software at CIRA.
In the example imagery loops presented below, the number of images that were used to create a particular hourly
composite is listed as the sample size (n=) on the lower portion of the figure. This number may vary from hour to
hour due to missing imagery.
 |
One of the factors affecting the development and progression of the sea-breeze front is the shape of the coastline. Bays and inlets tend to be good examples of concave coastlines with respect to the water. The sea-breeze front moving inland off the bays without synoptic influence tends to diverge. Where the shape of the coastline is convex with respect to the water, such as the region where Apalachicola is at the point, the sea-breeze front moving inland will converge and convective development is more likely. This region shows up as a preferred area of convective development in many of the regimes with variations on the inland penetration, east-west location, and strength of convective development depending on the background synoptic flow. Regimes 1, 2, and 4 also have loops of lightning flash density composites, being developed at Tallahassee, which complement the satellite climatology efforts. Look for the publication on the 1996-1999 results in Weather and Forecasting (end of the year 2001) |
RESULTS for VISIBLE CLOUD FREQUENCY, VISIBLE AVERAGE, and 10.7 um CLOUD FREQUENCY by TEMPERATURE THRESHOLD (both 283 K and 235 K)
    |
Regime 1 (Light and variable or light SE): For a ‘typical' summer, this is the most common regime type. It is characterized
by the dominant "Bermuda High", which gives the Florida peninsula southeasterly flow. Because of the
westward termination of the ridge axis, the Florida panhandle is left in a col point and experiences light and
variable or light SE flow. Loop: Lightning Flash Density Composites, Regime 1: Light and variable |
    |
Regime 2 (Light to moderate E to NE): This is a fairly uncommon regime type during a typical summer. It is characterized
by a high pressure system of continental origin located to the north of the Tallahassee region. This produces a
general easterly flow (3 to 10 kts) over the region, and because this type of high originates over land, convection
is usually sparse on this kind of day. The summers 1996 - 1999 were atypical (1996 - building El Niño, 1997
- strong El Niño, 1998-1999 rapid onset and influence of La Niña) and this turned out to be a frequently occurring regime. Loop: Lightning Flash Density Composites, Regime 2: Light to moderate E to NE |
    |
Regime 3 (Strong E to NE): This has the same direction of flow as regime 2 , but with stronger winds (>10 kts). |
    |
Regime 4 (Light to moderate W to SW): This is a fairly common type of regime flow resulting when the "Bermuda
High" is shifted further to the south. This sets up a general west-southwest flow (3-10 kts) over the Florida
panhandle. Convective development is likely on this type of day. Loop: Lightning Flash Density Composites, Regime 4: Light to moderate W to SW |
    |
Regime 5 (Strong W to SW): This has the same direction of flow as regime 4 , but with stronger winds (>10 kts). |
    |
Regime 6 (Moderate (SE to S): This is similar to a Regime 1 day with the "Bermuda High" shifted to the W. |
    |
Regime 8 (Light to moderate N to NW): This regime usually occurs after passage of a cold front or trough and the winds shift to the N-NW (3-10 kts). Although the flow is offshore, keeping the sea breeze near the coast (along the panhandle) later in the day, the opposing flow also allows for a more intense sea breeze front. If the thermodynamics are favorable, there is a severe weather potential later in the day. |
    |
Regime 9 (Strong N to NW): This has the same direction of flow as regime 8 , but with stronger winds (>10 kts). |
Currently, the regime cloud frequency results are being used extensively in aviation and public forecasting to supplement existing information. In short-term forecasts, it is being used to 'fine tune' convective initiation and the timing of frontal passages. In zone forecasts, cloud frequency information is being used subjectively to produce more accurate and detailed probability of precipitation as well as severe weather or flood potential. In marine forecasts, it has provided more insight into the occurrence of land breeze convection and the sea fog/stratus potential. In aviation forecasts, the cloud frequency results have also been used subjectively to give better information on ceilings, timing of convection, and convective coverage en route.
In the future, the visible and infrared cloud frequency composites will be combined with lightning, precipitation, and radar data. With the combined information, we hope to address questions pertaining to frequency and location of rain in association with each of the regimes.
References:
Gould, K. J., and H. E. Fuelberg, 1996: The use of GOES-8 imagery and RAMSDIS to develop a sea breeze climatology over the Florida Panhandle. Preprints, Eighth Conf. on Satellite Meteorology and Oceanography, Atlanta, GA, Amer. Meteor. Soc., 100-104.
Camp, J. P., Watson, A. I., Fuelberg, H. E., 1998: The diurnal distribution of lightning over North Florida and its relation to the prevailing low-level flow. Weather and Forecasting, 13: 729-739.
Connell, B. H., abd K. J. Gould, 2000: GOES-8 Visible cloud frequency composites of the convectively active sea breeze under stratified synoptic flow over the Florida Panhandle. Preprints, Tenth Conference on Satellite Meteorology and Oceanography, Long Beach, CA, Amer. Meteor. Soc., 438-441.
Molenar, D. A., K. J. Schrab, and J. F. W. Purdom, 2000: RAMSDIS contributions to NOAA Satellite Data Utilization. Bull. Amer. Meteor. Soc., 81, 1019-1030.
Connell, B. H., K. J. Gould, and J. F. W. Purdom, 2001: High resolution GOES-8 visible and infrared cloud frequency composites over Northern Florida during the Summers 1996-1999. Look for it in Weather and Forecasting near the end of the year.
Updated: August 2001
