Like a Rock in a River

Stones and large rocks often scatter across river beds, some protruding above the surface, and some below. If you pay close attention you will see that the water reacts in different ways as it flows past each of these obstacles. The shape, size and depth of the rocks along with the speed of the river determines the patterns that we see downstream.

The vortex forming on the wake of the Big Island on November 8 2012 (source: EOSDIS NASA)

The vortex forming on the wake of the Big Island on November 8 2012 (source: EOSDIS NASA)

A big rock, sticking out above the surface will block the flow completely. The water builds up on the upstream of the rock, splits and flows around each side. Maybe you have noticed, the water coursing around sides flows faster than normal. This is what we call a jet. The water meets again on the downstream side, often causing some kind of turbulence. If you look closely you will often see whirlpools forming for short periods of time.

A rock that is well below the surface still affects the river flow. The water is still forced over these rocks and we can see the effect of this as ripples, which sometimes extend far downstream from where the rock is. These ripples are called gravity waves.

Modern satellite imagery and data offer us the incredible opportunity to see that some islands in the world act much like rocks in a river. Their size and shape determine the patterns that we can sometimes in the atmosphere downstream.

There are regions in the world where we find the atmospheric equivalents of rivers on land; places where we can imagine an equivalent of a water surface and a flow. And sometimes in the middle of this flow we find islands, which act just like rocks in a river. In the tropics our river surface is known as the Trade Wind Inversion, which marks the border between air masses of differing densities.

The tropical atmospheric river

The dry subsiding air of the subtropical cyclone – the descending arm of the Hadley Cell – meets the moist air over the Pacific Ocean and the trade wind inversion forms. Below the inversion the trade winds blow predominantly from the north-east towards the equator.

The islands of Hawai’i sit in the middle of this flow.  With the cumulus clouds that form below the inversion, we can sometimes see the effect of the islands in the clouds patterns as the air is forced over or around.

The height of the trade wind inversion fluctuates, but is usually 2-3000 m over the islands. The mountains on Maui and the Big Island usually protrude above this level and into the clear blue skies above. The Big Island is still volcanically active and Hawaiians often have problems with fog caused by the volcanic ash, which they call vog. Occasionally it can help reveal beautiful patterns and show us exactly how the Big Island is interacting with the tropical atmospheric river. The image is retrieved from one of the MODIS satellites on November 8 2012 where we can see that the Big Island is acting like a big rock in a river. To the east the cloud patterns indicate where the air starts splitting. No vog gets into the jets of fast flowing air around the northern and southern tips of the island. The huge swirl we see to the west occurs downstream of the island where the speed of the jet – and the shear created by it – causes the air to spin back on itself. Without the vog, the evidence for these patterns would be very hard, if not impossible to see.

To the west of some of the other smaller islands like Molokai and Oahu, we can see indications of the wave-like patterns in the cloud formations. However, since the vog is absent in this area, we cannot see exactly how the air is circulating.

From observations Smolarkiewicz and others constructed the first conceptual model of this wake with large eddies on the downstream side of the Big Island, strikingly similar to the MODIS image we see here. Xie and others showed that the effects of Hawaiian islands can be detected up to 3000 kilometers across the Pacific. They saw that the wake caused by the islands helps drive ocean currents over 8000 kilometers west from the volcanic archipelago.

The wake of Madeira shown in the cloud layer on

The wake of Madeira shown in the cloud layer on March 25 2004 (left) and April 28 2002 (right) (source: EOSDIS NASA)

There are many other tropical islands that sit in this tropical atmospheric river. In the Atlantic, Madeira and the Canary Islands sit in the flow below the trade wind inversion. Their smaller size and the fact that the inversion is lower here (usually 500-1000 m), means that sometimes we see vivid whirlpools or vortices downstream. The image on the left shows a straighter wake (from March 25 2004) with eddies forming towards the end. This could mean that the wake is at an early stage of development. Schar and Smith  simulated the nonlinear evolution of these wake formations starting with a pattern surprisingly similar to what we see downstream of Madeira on March 25 2004 and ending with a pattern resembling April 28 2002.  

Further to the south, the Cape Verde islands are in a similar situation, often sitting in the trade wind flow with a relatively low inversion. Again, if the cloud patterns are optimal, we can clearly observe the effects of the islands and the vortices that form (see the images MODIS images from April 27 2004 and May 16 2007). Sometimes the huge sand storms from the Sahara can blow far out across the Atlantic. Much like the vog on Hawaii, the sand is probably a tedious health risk for the local inhabitants, but in satellite imagery it can help us see the patterns caused by the islands in the flow (see the image from January 1 2007). We also see gravity waves forming in the suspended Saharan sand, showing a pattern much like the wake of a sailing ship.

The wakes of the Cape Verde islands. The islands interacting with a suspended Saharan sand (left) on the DATE. The other images are from DATE (middle) and DATE (right)

The wakes of the Cape Verde islands. The islands interacting with a suspended Saharan sand (left) on January 1 2007. The other images are from April 27 2004 (middle) and May 16 2007 (right) (source: EOSDIS NASA)

The fact that I am at the University of Hawaii on a research stay, gave me the idea to write about this phenomena in the Tropics. These wakes and vortices are also very prevalent at high latitudes and Polar regions. Here we also find inversions caused by different mechanisms than the trade wind inversion. The second part of this snack will look at how this inversion develops and what happens when the flow interacts with considerably colder islands. All the images here are taken from the MODIS rapid response webpages. I would thoroughly recommend looking through their galleries or finding the daily images over the region where you live.

 

Share on Facebook0Tweet about this on TwitterShare on LinkedIn0Share on Tumblr0Share on Google+0Pin on Pinterest0Share on Reddit0

Mathew Stiller-Reeve

I am a postdoc researcher at Uni Research Climate and the Bjerknes Centre for Climate Research. I research the monsoon in Bangladesh and I am the founder of ClimateSnack; an community that hopes to give all young and early career climate scientists an opportunity to practice and improve their scientific communication skills.

Latest posts by Mathew Stiller-Reeve (see all)

SciSnack Disclaimer: We write in SciSnack to improve our skills in the art of scientific communication. We therefore welcome comments concerning the clarity, focus, language, structure and flow of our articles. We only accept constructive feedback. All comments are manually approved and anything slightly nasty will not be accepted.