Talking points for “Coastal effects”

  1. Title page
  2. Objectives
  3. Topics
  4. Sea breeze forecasting – The sea-breeze will typically develop when the magnitude of the difference between the SST and the air temperature is greater than or equal to 6°F. Keep in mind climatology - what typically happens in a certain wind regime to the development of the sea breeze?
  5. Utility of GOES visible imagery during sea breeze events. Clear skies in the morning are a favorable ingredient for sea-breeze development. Watch for cumulus development along the sea-breeze front as well as the inland movement of this front. Thunderstorm development and evolution can be followed in the imagery to analyze intensification trends (i.e., larger overshooting top, favorable storm relative movement) as well as interactions with other surface boundaries (i.e., outflow boundaries, fronts, terrain induced boundaries etc).
  6. Sea breeze example – Gulf coast, 28 August 2004. Follow the principles discussed on the previous slide in this example.
  • Cloud coverage – skies are mostly clear early in the day.
  • Cumulus development along sea breeze front – correspond with land-water temperature difference threshold, and onshore flow development.
  • Where does the sea-breeze begin to move inland?
  • Highlight more intense thunderstorms and boundary interactions.
  1. Seabreeze interaction with trough – Long Island case
  2. Visible loop from 16:15 – 20:45 UTC 5 August 2005 centered over the OKX CWA. The key features are a cold front that extends from New York to Massachusetts and New Hampshire. Convection develops on the front early in New York. Southeast of the cold front we have a pre-frontal trough that goes from NYC northeastward to CT. Two sgements of the seabreeze intersect the pre-frontal trough, the first extends from west-central CT towards New Haven (around 17:30 UTC), the other is situated east-west along the north shore of Long Island. By 18:45 UTC the pre-frontal trough and the seabreeze over Long Island collide, resulting in thunderstorm development. Wind damage is reported at Plainview at 18:40 UTC. The storm dissipates but leaves an outflow boundary that moves eastward. New convection develops at the intersection of this outflow boundary and the seabreeze. Hail and wind damage is reported with this thunderstorm. Soon after that, the storm dissipates but the outflow boundary / seabreeze intersection further east forms a new thunderstorm which causes wind damage.This storm attained 55 dbZ at 38 kft before dissipation.
  3. 0.5 degree tilt reflectivity from KOKX radar 17:48 – 19:55 UTC 5 August 2005. Refer to talking points from previous slide. Outflow signature shows up well in the last storm (furthest east in Long Island).
  4. Sea-breeze case – 26 July 2007 – Charleston, SC CWA. Note SST analysis.
  5. Water Vapor imagery – 07:01 – 17:31 UTC 26 July 2007. A quasi-stationary trough is located over the Great Lakes region, this helps maintain west-southwest flow in a deep layer over the Charleston, SC CWA.
  6. Charleston, SC 12:00 UTC 26 July 2007 sounding. West-southwest flow over a deep layer, with light westerlies at the lowest levels. This would be light offshore flow for Charleston. Inversion is weak and low-level lapse rates are unstable (given the potential for erosion of the inversion by sea-breeze convergence).
  7. NAM12 Temperature (red contours, units of °F), CAPE (shaded, units of J kg-1) and winds (units of knots) from the 12:00 UTC 26 July run. 5 frames show the forecast out to 00:00 UTC 27 July. Recall the SST is between 82-83 °F. We should expect sea-breeze development when temperatures reach 88-89 °F (using the 6 °F land-water temperature difference “rule of thumb”). The afternoon forecast temperatures appear to be just below these values, however the NAM does show weak onshore flow in the CHS vicinity by 00:00 UTC. MOS temperatures (not shown) are also useful to look at.
  8. NAM12 QPF (red contours, units in inches), moisture convergence (shaded, units of g kg-1 12 hr-1) and winds from the 12:00 UTC 26 July run. 5 frames show the forecast out to 00:00 UTC 27 July. The NAM only has weak convergence along the coastline by the late afternoon hours. The model does forecast precipitation over portions of the CWA (not just along the coastline), this may contribute to temperatures being too cool in the model, thus the land-water temperature difference magnitude would be smaller. Next, we’ll look at mid-day observations to see if this trend is observed.
  9. Visible imagery from 13:45 – 17:31 UTC with METARs as overlay. The forecast for precipitation between 15:00 – 18:00 UTC from the NAM 12:00 UTC run is not taking place. Visible imagery shows partly cloudy skies over South Carolina, allowing heating to take place. The observed temperatures are higher than what was forecast from the 12:00 UTC NAM run. By 17:00 UTC we see tempertures along the coast of South Carolina in the 87-91 °F range (note: land-water temperature differences now exceed 6 °F from southeast South Carolina to Georgia coastal region). We also see a trend in the observations towards an onshore wind. An offshore convergence line shows up in the visible imagery.
  10. NAM12 Temperature (red contours), CAPE (shaded) and winds from the 18:00 UTC 26 July run. 3 frames show the forecast out to 00:00 UTC 27 July. The NAM now looks much more favorable for sea-breeze development compared to the 12:00 ITC run. Surface temperatures are higher, which is helping the development of onshore flow (stronger and over a larger area compared to the 12:00 UTC run). The CAPE forecast by 00:00 UTC shows values around 3000 J/kg in the Charleston vicinity, this suggests that any thunderstorm that develops along the sea-breeze will have ample instability.
  11. NAM12 QPF (red contours), moisture convergence (shaded) and winds from the 18:00 UTC 26 July run. 3 frames show the forecast out to 00:00 UTC 27 July. The NAM has strong convergence along the coastline by 21:00 UTC, note how much more favorable this solution looks for sea-breeze development compared to the 12:00 UTC NAM run. By 00:00 UTC the NAM does have QPF in the Charleston vicinity, we’ll check the observational trends soon to see if this will verify.
  12. Visible imagery from 17:45 – 23:01 UTC with METARs as overlay. Thunderstorm development occurs along the sea-breeze as it moves inland. It appears that there are boundary interactions between the sea-breeze and outflow boundary from convection further inland in southeast South Carolina. These boundary interactions appear to play a role in terms of thunderstorm evolution in southeast South Carolina for a period of time. Convergence zone boundaries setup with inland lakes over South Carolina also interact with the sea-breeze boundary, aiding thunderstorm development. Lake boundaries interact with the sea-breeze and outflow boundaries as well.
  13. KCLX 0.5 degree reflectivity from 17:58 – 23:07 26 July 2007 (METARs available in overlay). There are multiple interactions that occur between the sea-breeze and various outflow boundaries. One of the noteworthy interactions results in thunderstorm development just east of the KCLX radar site between 19:22 – 19:41 UTC. This thunderstorm produces an outflow boundary to its south where additional thunderstorms develop south of KCLX around 20:45 UTC. The new thunderstorm continues southeast towards Hilton Head Island which produced the following storm reports:
    .TIME...... EVENT...... CITY LOCATION...... LAT.LON
    .DATE...... MAG...... COUNTYLOCATION..ST.. ...SOURCE.
    ..REMARKS..
    0555 PM TSTM WND GST BLUFFTON 32.24N 80.86W
    07/26/2007 E45 MPH BEAUFORT SC PUBLIC
    PUBLIC REPORT OF STRONG ROTATION IN CLOUDS WITH POSSIBLE
    FUNNEL FORMING. VERY GUSTY WINDS THAT TORE PURSE FROM
    HAND WHILE WALKING TO HOUSE.
    0558 PM HAIL 2 E BLUFFTON 32.24N 80.83W
    07/26/2007 E0.25 INCH BEAUFORT SC TRAINED SPOTTER
    ON BRIDGE GOING TO HILTON HEAD ON HWY 278.
    (2200 UTC)0600 PM FUNNEL CLOUD 4 W HILTON HEAD ISLAND 32.22N 80.82W
    07/26/2007 BEAUFORT SC TRAINED SPOTTER
    SPOTTER DRIVING ACROSS HILTON HEAD BRIDGE ON HWY 278
    NOTICED A FUNNEL CLOUD OVER THE CALIBOGUE SOUND. FUNNEL
    WAS EXTENDING 1/4 OF THE WAY TO THE SURFACE BUT THERE WAS
    NO INDICATION OF VORTEX AT THE SURFACE.

The area around Bluffton often times report funnel clouds / waterspouts due to the favorable shape of the coastline. Finally, note the triple point of 2 outflow boundaries with the sea-breeze in Georgia produce thunderstorms as it advects westward between 22-23 UTC.

  1. Sea fog – Excluding synoptic scale influences, sea fog will increase in speed (due to growth) as it moves from relatively warm to colder water. Conversely, sea fog speed will generally decrease (due to front-end dissipation) as it moves from relatively cold to warm water.

Movement of sea fog:

Direction: Very close to surface wind direction, except where fog motion is affected by landfall

Speed: The maximum speed of sea fog moving from warm to colder water is about twice that of the observed wind speed.

  1. Topographic map – highlight the coastline of Maine.
  2. NAM Surface temperature, winds and MSLP from 12:00 UTC July 25, 2006. 3 frames show the forecast out to 00:00 UTC 26 July. The region of interest is just east of the Massachusetts coastline. Southerly flow around 15-20 knots is forecast in this region.
  3. GOES 5-day running mean SST analysis (degrees C) for the region the Atlantic seaboard. Knowledge of the SST distribution is essential for successful fog forecasting. Our region of interest in this case is off the coast of New England. Note the SST’s decrease from roughly 20°C east of Massachusetts to about 16°C off the coast of Maine.
  4. GOES-12 visible loop from 11:15 – 21:45 UTC 25 July 2006. Note the rapidly moving region of fog that originates east of Cape Cod (eastern Massachusetts) and advects north towards the coast of Maine. The fog bank appears to be increasing in speed (due to growth) because it is moving towards a region of colder SST’s. Once the fog intersects the Maine shoreline, frictional convergence causes it to spread out parallel to the shoreline. The fog may become quite deep right at the shoreline but it only advects inland a short distance.
  5. Visible imagery – 20 July 2004 – western Great Lakes. This example illustrates that sea (“lake”) fog and the sea (“lake”) breeze also exist over the Great Lakes. In this example, we see both, including convection being inhibited along the lake-breeze boundary as the air mass over the lake is more stable. We do observe some convection that initiates along various outflow boundaries inland over Wisconsin and Michigan. The convection in northern Wisconsin is inhibited as it approaches the more stable air mass along the lake-breeze boundary. Also notice the fog bank appears to slow down as it interacts with the coastline over Minnesota and Wisconsin.
  6. SST analysis from 20 July 2004 over the Great Lakes. The fog bank appears to slow down (due to front end dissipation) because it’s moving towards warmer SST’s.

27 to 29

•Especially when synoptic scale forcing is weak, low level boundaries and their interactions are central to (severe) thunderstorm development

•Strong mid to upper level synoptic forcing can induce or strengthen low level boundary interactions

•Boundaries can induce low level miso/meso-cyclone formation (as well as tornadoes) when horizontal vorticity along the cool side of a temperature boundary (e.g. sea breeze front) is tilted and stretched by strong updrafts

•Low level mesoscale convergence lines play a major role in determining where/when storms will form. What appear to sometimes be “random” occurrences of TSs, actually result from:

•A convergence line related kinematic feature

•Collision of convergence lines

•Local intensification of existing clouds when a boundary passes underneath

•A boundary encountering a local unstable air mass

•The interaction of a boundary with hills or mountains

•Study by J. Wilson and W. Schreiber concluded that over the high plains, 80% of TSs with cores >= 30dBZ initiated because of convergence line boundaries and 95% of storms with cores >= 60dBZ were associated with boundary interactions

30. Title. San Diego March 11, 2006

31. Cold Season Severe Wx in West Coast (CA)

•Upper Level Trof w/500mb heights falling below 560 dam and located offshore of central CA

•Relatively tight 500mb height gradient giving at least 40m height falls over a 24 hr. period

•Left front quad of Low Level Jet nosing into the region with a core of at least 25-30kts

•850mb winds into region should be at least 15-20kts with a tight gradient present

•Strong Upper level jet moving over region around trof base

•Low Level airmass should be relatively moist and unstable

32. Soundings: 12Z and Modified for early afternoon. Toggle between the two and note differences.

33. NAM 500 mb Ht and Vort. 4 frames.

34. NAM 850mb Low Level Jet. 2 frames.

35. SFC Analysis (optional)

36. San Diego Area Terrain

37. Topographically induced/enhanced helicity:

•With west to southwesterly flow into the region, blocking effects due to terrain features will turn the flow northwesterly (southeasterly) – artificially “backing” the flow in the lowest levels and therefore artificially enhancing the vertical veering through the boundary layer. This in turn increases the local helicity values as well as the depth of the local boundary layer.

38 to 40. Topography, Flow, Analysis and Convergence areas for 1800Z, 1900Z, and 2000Z.

41. Visible Loop: 15Z Mar 11 to 145Z Mar 12.

42. Reflectivity 1836Z to 2119Z. Show storm in question and “dead area” behind hill/mountain blocking radar. (“right mover.”)

43. 4 Panel Reflectivity 1.3 deg to 10 deg…from 1940Z to 2010Z.

44. 4 Panel SRM 0.5 deg to 4.3 deg…from 1940Z to 2010Z.

45. SRM 6.4 deg…from 1940 to 2010Z.

46. Photo of storm moving east/southeast of Escondido.

47. Title to San Diego Sept. 22, 2007 case

48. Soundings 12Z and mid-afternoon modified. (toggle)

49. NAM 500mb ht and vort.

50. Hand/subjective analysis of flow and convergence zones from 14Z to 19Z.

51. IR image loop from 14Z to 20Z. Temp difference on either side of line that formed convection was between 1.5 deg and 2.0 deg C. (12.5-13.0 deg W of line…14.5 deg E of line).

52. Hybrid Visible/IR overlay – to identify cloudy/cloud-free areas and boundary detection.

53. Visible image loop from 1700Z to 1945Z.

54. Sfc Analysis.

55And 56 - Waterspouts.

* Tend to form under convective clouds deeper that 3km (although, not always the case)

•Begin in a region of horizontal wind shear (boundary). These wind shear boundaries can be pre-existing from terrain induced convergent zones or can actually form from convection itself between the updrafts and downdrafts.

•The intensity of a convective vortex (waterspout in this case) depends only on the thermodynamics of the convective “heat engine.” (Renno and Bluestein)

•Can be interpreted as the difference in temp and water vapor content between updrafts and downdrafts

•Waterspouts are most likely to form in regions where the warmest and most moist updrafts lie adjacent to the coldest and driest downdrafts

•This is why waterspouts are most often observed near boundaries (discontinuities) that separate relatively warm waters/air from cold waters/air

57. Waterspout Schematic. : Vortex intensity depends on the difference in temperature and water vapor content of the air at large radius from and at the center of the vortex.

58. 4 panel reflectivity 0.5 deg to 3.4 deg. Showing first elevated convection along convergence line showing first.

59. 4 panel reflectivity 0.5 to 3.1 deg continued…after change from VCP 11 to VCP 12.

60. 4 panel SRM 0.5 deg to 3.1 deg.

61. Spectrum Width…0.5 deg. From 1845Z to 1924Z.

62. Photo – Waterspoouts.

63. Title March 24, 2006 HA case.

64. Sounding 12Z and modified for later in day.

65. Hawaii Severe Weather facts

•Most (90%) severe weather occurs during the “winter” season – October thru April.

•Most deep convection occurs when perturbations aloft help to eradicate the trade wind inversion.

•A typical deep convection pattern has a mid to upper level low tilted and located to the northwest of the islands…giving strong divergent southwesterly flow over the islands.

66. GFS(75) 500mb height forecast

67. GFS height/temp/ and short wave analysis/forecast.

68. GFS sfc analysis/forecast

69. RSM model sfc wind forecast and convergence zone analysis – talk about flow convergene zone and how formed/difference to trade wind/inversion convergence.

70. IR imagery from 1345Z to 1700Z.

71. IR imagery from 1715Z to 1930Z.

72. Visible imagery from 1645Z to 2030Z.

73. 4 panel: Composite, 0.5 reflectivity, 1.5 deg reflectivity, and VIL.

74. Hail photo.

75. Best Approach for locating LL boundaries

•1997 Koch and Ray paper produced the following results:

•Hand meso-analysis identified between 50 and 60 percent of the convective boundaries

•Satellite imagery analysis identified just over 80% of the convective boundaries

•Radar analysis identified near 85% of the convective boundaries

•When using all three analysis techniques together, 0ver 95% of all the convective boundaries were detected

76. LL boundaries continued…

•When using all three techniques together, up to a four hour lead time was gained for identifying a significant weather event and where it would most likely occur

•Hand meso-analysis – good for determining the “nature” of the boundaries (temp diff, moisture diff, stability discontinuities, airflow, convergence/divergence

•Satellite analysis – good for long range identification of boundaries and tracking/timing movement. Obscuration can be a problem

•Radar analysis – good for detecting both air density or air flow boundaries…even in clear air. Not good for low level boundaries > about 50nm from radar

77. Contact Info.