Mysterious Gravity Wave Over the Eastern Pacific Ocean

November 14th, 2011 |

We received the following email from Ken Waters of the National Weather Service forecast office in Phoenix, Arizona:

I noticed something interesting in this morning’s visible imagery off the Baja California coast.

Here’s a link: https://docs.google.com/open?id=0B2ktDMIN5qWfODI2OTYzYjgtZjJkMS00MTU5LTk2ODctYzdhNzY5M2Y2MWIx

I’m looking at the apparent wave pattern that’s going “upstream” towards the northeast whereas the low level flow is mostly towards the south. Is that a gravity wave? If so, what causes it? The vis can only go so far back so I looked at the IR and couldn’t find anything obvious.

Great question Ken — it certainly appears to be some sort of internal gravity wave, but what caused it and why was it propagating against the ambient flow shall remain a bit of a mystery until we can dig into this case a bit further. One more event for the “What the heck is this?” blog category.

GOES-15 6.5 µm water vapor channel images + GOES-15 0.63 µm visible channel images

GOES-15 6.5 µm water vapor channel images + GOES-15 0.63 µm visible channel images

McIDAS images of GOES-15 6.5 µm water vapor channel data (prior to daylight) and then GOES-15 0.63 µm visible channel data after sunrise (above) on 14 November 2011 tell us one thing: this gravity wave was apparently fairly deep in the vertical, since it exibited a signal on both the water vapor channel imagery (which generally senses radiation from the middle troposphere: San Diego 12:00 UTC rawinsonde water vapor chanel weighting function profile) as well as on the lower-tropospheric cloud features seen on the visible channel imagery.

Note that there was a second packet of shorter-wavelength gravity waves that could be seen in the far southwestern portion of the GOES-15 visible image satellite scene toward the end of the animation. This second packet of gravity waves was very evident on a 500-meter resolution Aqua MODIS Red/Green/Blue (RGB) true color image at 21:21 UTC (below).

Aqua MODIS true color image

Aqua MODIS true color image

Gravity waves are usually ducted within a well-defined temperature inversion. A look at the 12:00 UTC rawinsonde profile from San Diego, California (below) did indicate the presence of a few inversions that might have been capable of ducting such a gravity wave — but the inversions existed at multiple levels.

San Deigo, California 12:00 UTC rawinsonde profile

San Deigo, California 12:00 UTC rawinsonde profile

An AWIPS image of 18:00 UTC MODIS 0.65 µm visible channel data with overlays of 1-hour interval MADIS satellite winds (below) did not reveal any atmospheric motion vectors with a southwesterly component – but these would likely have been rejected by the winds quality control algorithms, since such a motion would have differed too greatly from the model first guess wind fields at 850 hPa, 500 hPa, 300 hPa, and 250 hPa.

MODIS 0.65 µm visible channel image + MADIS 1-hour interval satellite winds

MODIS 0.65 µm visible channel image + MADIS 1-hour interval satellite winds

Regarding the effect of the gravity wave seen on the lower-tropospheric clouds bands, a MODIS 11.0 µm IR image detected cloud top IR brightness temperatures around +4ºC, which on a RUC model sounding at that location apparently corresponded to a cloud top height around 12,550 feet (below) — however, this value seemed to be a bit high judging from the appearance of the cloud band features on the GOES and MODIS visible and IR imagery.

MODIS 11.0 µm IR image + RUC model sounding

MODIS 11.0 µm IR image + RUC model sounding

On the other hand, POES AVHRR Cloud Type and Cloud Top Height products indicated that these low-level cloud bands were water droplet clouds, with cloud top heights of around 1 km (below) — much more typical for marine boundary layer cloud features over this region.

POES AVHRR Cloud Type product

POES AVHRR Cloud Type product

POES AVHRR Cloud Top Height product

POES AVHRR Cloud Top Height product

What the heck is this?

April 25th, 2011 |
GOES-13 6.5 µm "water vapor channel" images (click image to play animation)

GOES-13 6.5 µm "water vapor channel" images (click image to play animation)

We received the following in an email message from Paul Fuentes at the National Weather Service forecast office at Key West, Florida:

“Just though you might be interested in a peculiar little feature we noticed at the WFO Key West on AWIPS from GOES-13 IR imagery on 4/23/11 starting at about 22:301Z over Andros Island (24.21N, 77.7W) and persisted to into the afternoon on 4/24/11. The feature looked almost like a smoke ring (several miles across) that was drifting off to the Northeast that emanated off of Andros Island and was also apparent in VIS/WV.”

McIDAS images of 4-km resolution GOES-13 6.5 µm “water vapor channel” data (above) showed the feature in question, as it first appeared along the east coast of Andros Island (the large island in the lower left corner of the images) at 22:45 UTC on 24 April 2011 — and then propagated northeastward, grew in size, exhibited progressively colder brightness temperatures, and at times took on a ring-like shape.

A closer look using 1-km resolution GOES-13 0.63 µm visible channel images (below) again revealed the ring-like structure that was evident at various times during the day on 25 April 2011.

GOES-13 0.63 µm visible channel images (click image to play animation)

GOES-13 0.63 µm visible channel images (click image to play animation)

The feature (as seen on 10.8 µm POES AVHRR IR imagery, below) seemed to follow the CRAS model 300 hPa wind flow — and southwesterly winds were not found until the 375 hPa pressure level on the 12 UTC rawinsonde report from Nassau in the Bahamas.

POES AVHRR 10.8 µm IR images + CRAS model 300 hPa wind flow

POES AVHRR 10.8 µm IR images + CRAS model 300 hPa wind flow

A comparison of the MODIS 0.65 µm visible channel, the 11.0 µm IR window channel, the 6.7 µm water vapor channel, and the 1.3 µm cirrus detection channel images at 15:57 UTC (below) seem to support the idea that this was a high ice cloud feature.

MODIS visible, IR window, water vapor, and cirrus detection channel images

MODIS visible, IR window, water vapor, and cirrus detection channel images

So what exactly was this interesting satellite feature? Until an explanation is found, this blog posting shall remain in the “What the heck is this?” Category…

“Plume of unknown etiology” moving over Alaska

March 17th, 2011 |
GOES-11 0.65 µm visible channel images (click image to play animation)

GOES-11 0.65 µm visible channel images (click image to play animation)

McIDAS images of GOES-11 0.65 µm visible channel data (above; click image to play animation) showed an interesting “dark plume” feature that was moving in an arc from far northeastern Russia, across the East Siberian Sea and Chukchi Sea, and finally over far northwestern Alaska on 16 March – 17 March 2011.

When viewed from a more western angle using MTSAT-2 0.73 µm visible channel images (below; click image to play animation), the plume feature (which can be seen moving over far northwestern Alaska in the upper right portion of the images) also exhibited a darker appearance, similar to that seen on the GOES-11 visible imagery. This darker appearance was due to backward scattering of light from the particles within the plume.

 

MTSAT-2 0.73 µm visible channel images (click image to play animation)

MTSAT-2 0.73 µm visible channel images (click image to play animation)

AWIPS images of POES AVHRR 0.86 µm visible channel data (below) provided more of a “direct view from above”,  and revealed that the main body of the plume was basically transparent (allowing details of the sea ice to be seen through the plume).  However, the plume edges appeared to have some vertical structure, being thick enough to cast shadows onto the sea ice below.

POES AVHRR 0.86 µm visible channel images

POES AVHRR 0.86 µm visible channel images

It is interesting to note that this plume feature did not exhibit any notable signature on POES AVHRR 12.0 µm IR images (below).

POES AVHRR 12.0 µm IR images

POES AVHRR 12.0 µm IR images

A series of MODIS true color Red/Green/Blue (RGB) images (below; courtesy of the GINA, University of Alaska) again showed the transparent nature of the main body of the plume feature, except for the thicker edges which  were casting shadows.

MODIS true color Red/Green/Blue (RGB) images (courtesy of University of Alaska, GINA)

MODIS true color Red/Green/Blue (RGB) images (courtesy of University of Alaska, GINA)

Could this feature have been an aged volcanic plume that was being transported aloft over the Arctic? AWIPS images of the MODIS Volcanic Ash Mass Loading product (below) did display a few isolated very small patches exhibiting 1-10 tons per square kilometer of loading at 04:44 UTC on 17 March, but there was no temporal continuity when examining the Ash Mass Loading product before or after this particular time.

MODIS Volcanic Ash Mass Loading product

MODIS Volcanic Ash Mass Loading product

Volcanic Ash Height product

Volcanic Ash Height product

The corresponding MODIS Volcanic Ash Height product (above) indicated that these features were located at an altitude of 3-4 km, while the MODIS Ash Mass Effective Particle Radius product (below) showed values in the 3-5 µm range.

Volcanic Ash Particle Effective Radius product

Volcanic Ash Particle Effective Radius product

However, rather than an aged volcanic ash plume, a more plausible explanation of the feature seen on satellite imagery is the long-range transport of smoke and pollution from industrial sources in northeastern China. A calculation of 96-hour backward trajectories using the NOAA ARL HYSPLIT model (below) indicated that air parcels arriving at 3 points along the plume at an altitude of 6-km had originated within the boundary layer over northeastern China on 13 March. MODIS images showing the thick haze over that region can be found on the US Air Quality “Smog Blog”.

NOAA ARL HYSPLIT back trajectories arriving at  the 4km, 6km, and 8km altitudes

NOAA ARL HYSPLIT back trajectories arriving at the 4km, 6km, and 8km altitudes

Small-scale vortex over the tropical Atlantic Ocean

September 25th, 2010 |
GOES-13 0.63 µm visible image (courtesy of Tony Cristaldi, NWS Melbourne FL)

GOES-13 0.63 µm visible image (courtesy of Tony Cristaldi, NWS Melbourne FL)

Tony Cristaldi (National Weather Service Forecast Office in Melbourne, Florida) sent the following email (with the attached GOES-13 visible image shown above):

There is a very small cyclonic swirl to the southwest of Lisa near 12N 35W, that has been firing off central convection since Friday afternoon. Now, one wouldn’t expect a feature of this magnitude to be carried operationally as a TC, since these features are often times transient, being very vulnerable to even modest amounts of low level deformation and mid of upper level wind shear. Nevertheless, since it has persisted since early this morning, I thought I’d pose a couple questions: 1) Looking at visible imagery, does it appear that convection is persisting due to a warm core TC process (CISK) on a very small scale, or a more traditional low level mesoscale convective process? 2) Would you call this feature a “micro-midget?”

At his request, we created a 2-day animation covering the period 24-25 September 2010, using GOES-13 0.63 µm visible images during the daytime and GOES-13 3.9 µm shortwave IR images at night (below). The low-level “swirl” appeared to be propagating southeastward, and it was indeed firing off some impressive convective bursts (with a few cloud top IR brightness temperature values as cold as -70º C).

GOES-13 0.63 µm visible images (daytime) and 3.9 µm shortwave IR images (night-time)

GOES-13 0.63 µm visible images (daytime) and 3.9 µm shortwave IR images (night-time)

Not quite sure exactly what to call it…but thanks to Tony for bringing this interesting feature to our attention!