Hurricane Matthew: heavy rainfall and flooding across the Southeast and Mid-Atlantic US

October 9th, 2016 |

MIMIC Total Precipitable Water product, from 06 October/04 UTC to 08 October/16 UTC [click to play MP4 animation]

MIMIC Total Precipitable Water product, from 06 October/04 UTC to 08 October/16 UTC [click to play MP4 animation]

Copious amounts of moisture associated with Hurricane Matthew resulted in heavy rainfall (map | text list) and widespread flooding across the Southeast and Mid-Atlantic US during the 07 October to 09 October 2016 period. Hourly images of the MIMIC Total Precipitable Water (TPW) product (above; also available as a 22 Mbyte animated GIF) showed the high TPW values that spread from Florida to the Mid-Atlantic states; all-time record high TPW values were measured via rawinsonde at Jacksonville, Florida and Charleston, South Carolina, with a record high value for the month of October at Newport/Cape Hatteras, North Carolina (Tweet). For more details, see the Weather Underground blog.

Track of Matthew, from 28 September at 12 UTC to 09 October at 18 UTC

Track of Matthew, from 28 September at 12 UTC to 09 October at 18 UTC

Matthew set numerous records for intensity, longevity, and landfall (summary) — an animation of hourly GOES-13 Water Vapor (6.5 µm) images covering the 11-day period from 12 UTC on 28 September to 12 UTC on 09 October is shown below (also available as a large 113 Mbyte animated GIF). The CIMSS Tropical Cyclones site posted GOES-13 Visible (0.63 µm) animations from the individual days of 03 October, 04 October, 05 October, 06 October, 07 October, and 08 October.

GOES-13 Water Vapor (6.5 µm) images [click to play MP4 animation]

GOES-13 Water Vapor (6.5 µm) images [click to play MP4 animation]

The combination of high winds and flooding led to widespread power outages, with over 2 million homes and businesses without power. A comparison of nighttime Suomi NPP VIIRS Day/Night Band (0.7 µm) images from 28 September (before Mattthew arrived) and 09/10 October (after the passage of Matthew) showed a notable reduction in the glow of city lights in areas with no power (below; images courtesy of William Straka, SSEC). Note that the presence of patchy clouds on all 3 images tended to diffuse or even obscure the appearance of city lights below, depending on the thickness of the cloud layer(s).

Suomi NPP VIIRS Day/Night Band (0.7 µm) images on 28 September, 09 October and 10 October [click to enlarge]

Suomi NPP VIIRS Day/Night Band (0.7 µm) images on 28 September, 09 October and 10 October [click to enlarge]

As clouds cleared in the wake of Hurricane Matthew on 09 October, a Suomi NPP VIIRS true-color Red/Green/Blue (RGB) image at 1859 UTC, viewed using RealEarth (below), revealed patterns of turbidity in the offshore waters of the Atlantic Ocean from Florida to North Carolina; this increased turbidity was a result of high amounts of particles suspended in the water due to a combination of mixing from prolonged high winds and runoff from inland flooding.

Suomi NPP VIIRS true-color image [click to enlarge]

Suomi NPP VIIRS true-color image [click to enlarge]

About 8 hours later, a Terra MODIS Sea Surface Temperature (SST) product image at 0243 UTC on 10 October (below) showed a large eddy of warm Gulf Stream water (with a maximum SST value of 85.2º F, darker red color enhancement) surrounding a pocket of cooler water (with a minimum SST value of 78.5º F, darker blue color enhancement) off the coast of South Carolina.

Terra MODIS Sea Surface Temperature product [click to enlarge]

Terra MODIS Sea Surface Temperature product [click to enlarge]

The VIIRS Instrument on Suomi NPP provides data that are used in a River Flood Product (discussed previously on this blog). The product uses three reflective bands (I01, I02, and I03 at 0.64 µm, 0.86 µm and 1.61 µm, respectively) and the infrared window band I05 at 11.45 µm. The image below (courtesy of Sanmei Li at George Mason University) identifies many flooded regions over North Carolina. In particular, the flooding near Goldsboro and Lumberton is identified.

snppviirs_floodmap_south_north_carolina_usa_11oct_2016_18_17

JPSS River Flood product produced with Suomi NPP data, 1817 UTC on 11 October 2016 (Click to enlarge)

A sequence of 1 pre-Matthew (06 September) and 3 post-Matthew (09, 10 and 12 October) Terra/Aqua MODIS false-color RGB images from the SSEC MODIS Today site (below) also helped to highlight areas of flooding (darker shades of blue, especially notable along river valleys) that resulted from the heavy rainfall.

Terra and Aqua MODIS false-color images, from 06 September and 09, 10 and 12 October 2016 [click to enlarge]

Terra and Aqua MODIS false-color images, from 06 September and 09, 10 and 12 October 2016 [click to enlarge]

Hurricane Matthew makes landfall in western Haiti, then eastern Cuba

October 4th, 2016 |

GOES-13 Visible (0.63 µm) Imagery, 1045-1245 UTC on 4 October 2016 (Click to enlarge)

Hurricane Matthew has made landfall in western Haiti. The rocking animation (click here for a straight animation) above shows the cloud-filled eye of the storm crossing the Tiburon Peninsula. The storm’s center is forecast to remain largely over water as it moves through the Windward Passage between Cuba and Hispaniola.

A closer look using a 2-panel comparison of GOES-13 Visible (0.63  µm) and Infrared Window (10.7 µm) images, below, shows the deteriorating satellite presentation following interaction with the topography of the islands. The GOES-13 satellite was in Rapid Scan Operations (RSO) mode, providing images as frequently as every 5-7 minutes.

GOES-13 0.63 µm Visible (left) and 10.7 µm Infrared Window (right) images [Click to play animation]

GOES-13 0.63 µm Visible (left) and 10.7 µm Infrared Window (right) images [Click to play animation]

NOAA-18 overflew the region around 1130 UTC while the eye was on land, and the toggle below shows Visible (0.64 µm) and Infrared Window Channel (10.8 µm) imagery from 1130 UTC. The cloud-filled eye is distinct in the infrared image at that time, but a sequence of POES AVHRR Infrared (12.0 µm) images showed the rapid deterioration shortly after landfall (as was seen in the GOES-13 images above).

NOAA-18 AVHRR Visible (0.64 µm) and Infrared (10.8 µm) Imagery, 1130 UTC on 4 October 2016 (Click to enlarge)

A toggle between 1215 UTC GOES-13 Infrared Window (10.7 µm) and 1217 UTC DMSP-18 SSMIS Microwave (85 GHz) images from the CIMSS Tropical Cyclones site, below, revealed that a well-defined eye was still evident in the microwave data.

GOES-13 Infrared Window (10.7 µm) and DMSP-18 SSMIS Microwave (85 GHz) images [Click to enlarge]

GOES-13 Infrared Window (10.7 µm) and DMSP-18 SSMIS Microwave (85 GHz) images [Click to enlarge]

Aqua overflew Matthew shortly after 1800 UTC on 4 October, and the toggle below shows the 1-km visible (0.65 µm) and the 1-km ‘Cirrus Channel’ (1.38 µm). The Cirrus Channel detects radiation at a wavelength where very strong absorption by water vapor is occurring; only high clouds are detected with this channel, and the toggle between the Cirrus Channel and the Visible nicely outlines the cirrus canopy of the storm. The Advanced Baseline Imager (ABI) on GOES-R also includes a Cirrus Channel.

Aqua MODIS Visible (0.65 µm) and

Aqua MODIS Visible (0.65 µm) and “Cirrus Channel” (1.38 µm) at 1832 UTC on 4 October 2016 [Click to enlarge]

Meanwhile, to the northeast of Matthew, in the tropical Atlantic, Tropical Storm Nicole has formed. The animation of visible imagery from GOES-13, below, shows a sheared storm; the low-level circulation is west of the deepest convection. It’s unlikely that Nicole will intensify much under such sheared conditions. Cirrus outflow from Matthew is evident at the south and west of Nicole.

GOES-13 Visible (0.63 um) images [click to play animation]

GOES-13 Visible (0.63 um) images [click to play animation]

ASCAT on METOP-A sampled both storms in its morning overpass over the western Atlantic, as shown below. The maximum scatterometer-derived wind speeds were 60 knots with Matthew and 40 knots for Julia.

GOES-13 Visible (0.63 µm) image, with Metop-AASCAT winds [Click to enlarge]

GOES-13 Visible (0.63 µm) image, with Metop-AASCAT winds [Click to enlarge]

Late in the day on 04 October, Category 4 Hurricane Mathew made a second landfall along the far eastern tip of Cuba. As seen in the image toggle below, in spite of a ragged appearance on GOES-13  Infrared Window (10.7 µm) imagery, a distinct eye was still seen using DMSP-18 SSMIS Microwave (85 GHz) data.

GOES-13 Infrared Window (10.7 µm) and DMSP-18 SSMIS Microwave (85 GHz) images [Click to enlarge]

GOES-13 Infrared Window (10.7 µm) and DMSP-18 SSMIS Microwave (85 GHz) images [Click to enlarge]

Hurricane Matthew

September 30th, 2016 |

GOES-13 Visible (0.63 µm) imagery, 1115 UTC on 30 September 2016 [Click to enlarge]

GOES-13 Visible (0.63 µm) imagery, 1115 UTC on 30 September 2016″

Early morning visible imagery over Matthew, above, from GOES-13, shows a circular storm with many overshooting tops and no apparent eye. However, Microwave imagery, below, from GCOM at about 0620 UTC, shows an eye structure beneath the clouds. (Information on AMSR-2 is here; Imagery was produced using Polar2Grid, part of CSPP, the Community Satellite Processing Package). Matthew is forecast to turn north and move over the Greater Antilles, threatening Jamaica, Hispaniola and Cuba from Sunday to early Tuesday. More information is available at the National Hurricane Center website.

GCOM AMRS-2 Brightness Temperatures at 36.5 and 89.0 GHz, 0620 UTC on 30 September 2016 [Click to enlarge]

GCOM AMRS-2 Brightness Temperatures at 36.5 and 89.0 GHz, 0620 UTC on 30 September 2016 [Click to enlarge]”

One benefit of Polar2Grid is that it puts different satellite imagery on the same grid, and therefore an animation of Polar data can be produced. The 10.8 µm window channel animation below has imagery from the AVHRR on METOP-A (1349 UTC) and METOP-B (1441 UTC) (Imagery from NOAA-18 (1030 UTC) was projected onto a different grid). A Visible image toggle using data from METOP-A and METOP-B is here.

10.8 µm brightness temperatures from AVHRR on METOP-A (1349 UTC) and METOP-B (1441 UTC) on 30 September 2016 [Click to enlarge]

10.8 µm brightness temperatures from AVHRR on METOP-A (1349 UTC) and METOP-B (1441 UTC) on 30 September 2016 [Click to enlarge]”


===== Update, 1800 UTC =====
Visible imagery from GOES-13 (below) shows the development of an eye. Matthew continues its motion to the west-southwest in between Colombia and Hispaniola.

GOES-13 Visible 0.63 µm Imagery, 1345-1715 UTC on 30 September 2016 [Click to enlarge]

GOES-13 Visible 0.63 µm Imagery, 1345-1715 UTC on 30 September 2016 [Click to enlarge]”

===== Update, 22 UTC 01 October =====

GOES-13 Infrared Window (10.7 µm) images [click to play animation]

GOES-13 Infrared Window (10.7 µm) images [click to play animation]

According to the NHC, Hurricane Matthew reached Category 5 intensity around 00 UTC on 01 October (public advisorydiscussion). An animation of GOES-13 Infrared Window (10.7 µm) images from 09 UTC on 30 September to 18 UTC on 01 October, above, showed the evolution of cold cloud-top brightness temperatures surrounding the small eye during the period of rapid intensification (SATCON)  on 30 September – 01 October.

Interestingly, the center of what had since been downgraded to a Category 4 Matthew did a circular loop during the daylight hours of 01 October, as seen in a 2-panel comparison of GOES-13 Visible (0.63 µm) and Infrared Window (10.7 µm) images, below.

GOES-13 0.63 µm Visible (left) and 10.7 µm Infrared Window (right) images [click to play animation]

GOES-13 0.63 µm Visible (left) and 10.7 µm Infrared Window (right) images [click to play animation]

One important point about the location of Matthew:


It is because of this southerly location that the storm was not adequately sampled within the CONUS scan sector (which was being provided with imagery as often as every 5-7 minutes, since GOES-13 was in Rapid Scan Operations mode) — so imagery of Matthew was only available every 30 minutes from the Northern Hemisphere scan sector. Once GOES-R becomes operational, a full disk scan can be performed as freqently as once every 5 minutes, which would provide much better sampling for an important tropical cyclone such as Matthew.

Tropical Storm Matthew in the Windward Islands

September 28th, 2016 |

Note: Matthew was upgraded to a Hurricane at 1800 UTC on 29 September. See the National Hurricane Center Website for the latest information.

GOES-13 Visible (0.63 µm) and shortwave Infrared (3.9 µm) at 6-hour intervals, 23-28 September 2016 [Click to animate]

GOES-13 Visible (0.63 µm) and shortwave Infrared (3.9 µm) at 6-hour intervals, 23-28 September 2016 [Click to animate]

The area of disturbed weather that has been moving across the tropical Atlantic for the past week, shown above in six-hour steps using visible (0.63 µm) imagery during the day and shortwave infrared (3.9 µm) imagery at night, has developed into a strong tropical storm, named Matthew, as it moves through the Windward Islands. Matthew is embedded with the rich moisture source of the Intertropical Convergence Zone, as shown in the animation below of MIRS Total Precipitable Water, taken from this source. Matthew is also heading towards a region rich in moisture — it appears that dry air should not influence Matthew’s evolution in the near term.

Morphed MIRS Total Precipitable Water, 25-28 September 2016 [Click to animate]

Morphed MIRS Total Precipitable Water, 25-28 September 2016 [Click to animate]

Matthew’s predicted course into the Caribbean is over very warm water with high heat content (below, imagery from the CIMSS Tropical Weather site). Wind shear over the storm is low, with larger values to the west; the region of high shear in advance of Matthew has been preceding the storm for the past 72 hours and it is forecast to continue its retreat from Matthew. The forecast from the National Hurricane Center is for slow strengthening.

Observed Sea Surface Temperatures and Oceanic Heat Content [Click to enlarge]

Observed Sea Surface Temperatures and Oceanic Heat Content [Click to enlarge]

Wind Shear, 1500 UTC on 28 September [Click to enlarge]

Wind Shear, 1500 UTC on 28 September [Click to enlarge]

Late in the afternoon on the 28th, GOES-13 detected cloud tops cooler than -80º C associated with some of the overshooting tops (the purple enhancement). Overshooting tops can be detected automatically and shown at this website.

GOES-13 Infrared (10.7 µm) imagery, 1445-2045 UTC on 28 September 2016 [Click to animate]

GOES-13 Infrared (10.7 µm) imagery, 1445-2045 UTC on 28 September 2016 [Click to animate]


============== Added, 29 September 2016 =================

Suomi NPP overflew the eastern Caribbean early on 29 September, and the Day Night Band imagery, below, (from RealEarth) shows Matthew west of the Windward Islands. A lone lightning strike is visible in the convective clouds. There appears in this very low light image to be a low-level circulation exposed to the west of the deep clouds, suggesting a sheared storm.

Suomi NPP Day Night Band Visible (0.70 µm) imagery, ~0540 UTC on 29 September 2016 [Click to enlarge]

Suomi NPP Day Night Band Visible (0.70 µm) imagery, ~0540 UTC on 29 September 2016 [Click to enlarge]

A GOES-13 Visible Image animation from just after sunrise shows a the circulation center exposed to the west of the deep convection.

GOES-13 Visible (0.63 µm) imagery, 1045-1245 UTC on 29 September 2016 [Click to enlarge]

GOES-13 Visible (0.63 µm) imagery, 1045-1245 UTC on 29 September 2016 [Click to enlarge]

NOAA-19 overflew Matthew at about 1900 UTC on 29 September 2016, shortly after the storm was upgraded to a Hurricane, and visible imagery from that pass (data courtesy of the AOML Direct Broadcast antenna) shows far less evidence of a sheared storm. The Central Dense Overcast is above the surface circulation.

NOAA-19 Visible (0.63 µm) imagery, 1931 UTC on 29 September 2016 [Click to enlarge]

NOAA-19 Visible (0.63 µm) imagery, 1931 UTC on 29 September 2016 [Click to enlarge]