The Scientific Mission and Background

Introduction

This research cruise focuses on the southern end of the Reykjanes Ridge, located ~1000 km south of Iceland in the north Atlantic.  The Reykjanes Ridge is part of the Mid-Atlantic Ridge spreading center, part of the global mid-ocean ridge system that extends for over 70,000 km around the entire earth.  The previous cruises were also along spreading centers, but in those cases they were backarc spreading centers that were all situated on the upper plate of the Tonga or Mariana subduction zones (see old posts for a brief introduction to subduction zones and backarc spreading centers). 

Mid-ocean ridge basics

Mid-ocean ridges are in most ways much simpler than backarc spreading centers, because you remove the complications introduced by the plate subducting into the mantle, which has major effects on the characteristics of the spreading center.  At mid-ocean ridges, the lithosphere (which includes the crust and the upper-most part of the mantle) is pulling apart, causing it to thin and stretch, and crack.  In the upper few km of the crust where the rock is cool and brittle, this extension is accommodated by faulting.  In the deeper parts of the crust and lithosphere, extension is accommodated by ductile stretching (like silly putty or toothpaste).  This thinning of the lithosphere causes the underlying mantle to rise (a process called upwelling) and fill the space.  As it rises, the pressure is reduced on the mantle, causing it to partially melt, intrude into the plate boundary and create new crust.  On the earth’s surface, most people think of melting as a simple process of adding heat until the material reaches its melting temperature, but in reality the process of melting depends on three factors: temperature, pressure, and composition.  If you brought a piece of mantle up to the surface, it would already be at a temperature well above its melting point, but the extreme pressure in the interior of the earth prevents the hot mantle from melting.  Most people think of the mantle as a liquid because it does flow over long timescales, but in fact, due to the high pressures it is actually a solid.  So, at a mid-ocean ridge, as the mantle rises and pressure decreases, the components of the mantle with the lowest melting temperature melt first in a process called partial melting (i.e. the entire mantle does not melt).  These components (which are the chemical components of basalt) separate from the remaining mantle, rise upward intrude into the crust, and accumulate in a magma chamber a few km under the ridge axis.  Usually, less than 10% of the mantle volume actually melts, and the remainder becomes “depleted” in the elements that melt easily.  Once enough melt has accumulated in the magma chamber and driving pressure is adequate, these materials break through the upper crust in dikes, and once they reach the seafloor spread out as lava flows.  So while geologists often talk of spreading rates as being a few cm per year, in reality this spreading mostly happens in discrete short events, with long intervening periods of little to no activity.  Spreading rates are just an average over long time periods of these discrete, shorter events.  The duration of these events and the time between them is determined by the spreading rate, with higher spreading rates favoring shorter events with much shorter time gaps between them.  Spreading rate is also the primary controlling factor on the morphology of the spreading center, with broad, smooth, shallow volcanic peaks associated with fast spreading rates, and deep faulted valleys associated with slow spreading rates. Backarc spreading centers in subduction zones are more complicated because you have this same process (called “depressurization melting”) interacting with the water and other materials introduced into the mantle by the subducting plate.

The Reykjanes Ridge and the Iceland hot spot

While the majority of the variations along the global mid-ocean ridge system can be explained by variations in spreading rate, the Reykjanes Ridge has an extra complicating factor.  Iceland, like Hawaii and numerous other island chains around the world, is underlain by a hot spot, also known as a mantle plume.  Hot spots are plumes of anomalously hot upwelling mantle that are thought to rise all the way from the core-mantle boundary (although the source of hot spots is still very much debated).  In Iceland, you have two sources of melt interacting, depressurization melting from the spreading center, and additional melting due to the extra heat from the hot spot.  This is why Iceland is a huge island sitting well above sea level, while the rest of the Mid-Atlantic Ridge is 1000+ m under water.  The Reykjanes Ridge extends through the western portion of Iceland and because of the influence of the hot spot, it becomes shallower and more volcanically active toward the north as it approaches the hot spot. 

           
If you look at the majority of the Mid-Atlantic Ridge (just look at google maps), it has a characteristic “stair-step” appearance.  This stepped appearance is caused by spreading center segments (oriented roughly N-S), with intervening fracture zones (oriented roughly E-W) that connect the ends of the segments.  You can see that the overall trend of the ridge is not N-S; in the north Atlantic the overall ridge is oriented in a NE-SW direction, but the spreading direction is ~E-W.  Spreading centers tend to form perpendicular to the spreading direction, so instead of having a linear spreading center oriented in a NE-SW direction (oblique to the E-W spreading), the overall oblique orientation is accommodated by the stair-step pattern.  Just think of it as using stairs to go up/down a hill rather than a straight ramp.

            
If you look at the Reykjanes Ridge south of Iceland, it does not have this stair-step appearance.  Instead it appears to be a linear oblique spreading center with no fracture zones.  If you zoom in far enough, you can actually see that the individual spreading segments are still oriented ~N-S, but instead of long (~100’s of km) segments separated by fracture zones, there are short (a few 10’s of km) segments that overlap with no fracture zones between them.  We are looking at the southern end of this ridge, where it transitions from an oblique spreading center back to the normal stair-step pattern toward the south.  The question we are trying to answer is what has caused the previous stair-step pattern to be erased and replaced with this completely different seafloor fabric.  The currently accepted (though not well-supported) theory is that it is entirely a hot spot effect.  This theory claims that hot mantle from the hot spot has migrated south along the ridge, softening the lithosphere and erasing the fracture zones as it migrates southward.  The competing theory (which is favored by my advisor and the other professor at UH that he wrote the proposal with) is that it is largely due to the Reykjanes Ridge propagating southward at a slightly different angle than the previous spreading center, erasing the stair-step pattern and replacing it with this new seafloor fabric.  The reason that a propagating spreading center forms is mostly thought to be due to a change in overall plate motion, in this case the new spreading direction has been rotated slightly clockwise.  Instead of rotating the original spreading center in a clockwise direction, the earth responds to this change in plate motion by forming a new spreading center at an oblique angle (rotated slightly clockwise) compared to the previous one, which then propagates and replaces the old spreading center.  If you look at a map, you can see that the Reykjanes Ridge is indeed rotated slightly clockwise relative to the rest of the Mid-Atlantic Ridge toward the south.  Mapping the southern end of the Reykjanes Ridge across the zone where the seafloor fabric changes should help us to clearly distinguish between these two models.

The planned survey area showing the ship tracks in black lines, with waypoints as white circles, overlain on a very coarse (~2 km resolution) bathymetry map. Two prominent fracture zones, the Pendragon, which has been mostly erased by the Reykjanes Ridge propagation, and the Bight fracture zone, which is just beginning to be erased, are labeled.  The red lines outline the exclusive economic zones of Iceland and Greenland. Iceland would like that red line extended to the bottom of our survey area.

The discussion of exactly what we’re looking for requires a lot of background to understand, some of which I’m not entirely familiar with, but I’ll attempt to explain some of the main things that will help us distinguish between the two models.  One of the major features created by a propagating spreading center is called a pseudofault.  As the rift propagates, it creates a zone of new crust that has a v-shape, because the older part of the rift continues spreading and opening as the tip continues propagating, somewhat like a zipper.  Along the boundaries of this v-shaped zone is a sharp edge that looks somewhat like a fault (hence the name pseudofault), separating the old seafloor from that which is created along the propagating spreading center.  On our way south from Iceland to the survey area, we attempted to map this boundary, and we’ll go along the same area on the way up, so hopefully we will be able to see this boundary in detail and determine whether it is indeed a pseudofault.  Another even more definitive structure that we will look for is a zone of transferred lithosphere.  Imagine a simple N-S spreading center, spreading in an E-W direction, dividing the North American and Eurasian plates.  Then imagine a new spreading center forming 100 km E of the old one, starting in the N and propagating in a slightly SW direction.  Now, that 100 km slice of lithosphere that was once on the Eurasian (E) side of the original spreading center is on the North American (W) side of the new spreading center, meaning that a piece of Eurasian lithosphere has been “transferred” over to the North American plate. This piece of transferred lithosphere rotates as the spreading center propagates, so if we see a zone of seafloor with faults and volcanic ridges that are at an oblique orientation to the surrounding seafloor, this will be pretty much slam dunk evidence that the propagating rift model is correct.  With the scale of mapping that we have right now, it simply isn’t possible to see this, so we need more detailed maps to see if there indeed is a zone of transferred and rotated lithosphere.

Instrumentation and Data Collection

Compared to my previous cruises, this is one is pretty basic as far as instrumentation.  We are not doing any seismic work, there are no AUV’s or towed cameras, and we likely will not be taking any rock samples unless the multibeam sonar fails for some reason.  The main instrument is the multibeam sonar system, which all research ships are equipped with, and is mounted to the hull.  I’ve explained this before in past posts, but I’ll give a brief summary here as well.  Sonar systems both emit and receive sound pulses that bounce off of the seafloor.  It is called a multibeam system because the sound waves are produced from multiple transducers that emit sound in a fan-shaped pattern giving us a wide swath of data on the seafloor rather than a single track.  There are two types of data that can be recovered from the sonar system: bathymetry and backscatter.  Bathymetry is simply a measure of the time it takes for the sound to travel to the seafloor and back to the receiver array.  Knowing the speed of sound in water and how it changes with depth, the travel times can be converted into a distance measurement, and you can get an image of the topography of the seafloor.  Along with the travel time, the receivers also measure the intensity of the reflected sound waves, which is the backscatter.  If the sound reflects off of a hard surface (a lava flow or a sunken ship for example) the intensity of the return is very high, but if the sound reflects off of a soft or rough surface (sediment for example), much of the energy is scattered and the intensity is weaker.  Backscatter images essentially look like a black and white image of the seafloor and allow us to see structures such as faults and lava flows, and distinguish them from sedimented regions.

A much higher resolution (~100 m) bathymetry map of the northern portion of the Reykjanes Ridge to give you an idea of how much better resolution you can get when mapping with a ship compared to just satellites.  Now you can see (hopefully) that although the overall orientation of the ridge is NE-SW, in detail the actual segments of the spreading center (along the central red portion) are oriented in a more N-S direction, perpendicular to the spreading direction.  This is from two combined surveys, one with dense tracks right along the axis, and the other with further-spaced tracks that go much further off the axis.  The rows of yellow dots are aligned along "flow-lines" that are parallel to the spreading direction.

Along with the multibeam sonar data, which is the primary mapping tool, we will be collecting both gravity and magnetic data as well.  Gravity data is collected with a gravimeter, which is mounted in the main computer lab as close as possible to the center of the ship (to reduce the accelerations due to ship motion).  There are different designs for various gravimeters and I won’t get into the technical details, but basically they measure slight variations in gravity due to local variations in mass.  For instance, if we are passing over a large volcanic seamount, the local increase in mass causes a slight increase in gravity.  Gravity data is useful for looking at variations in crustal thickness and can help illuminate structures in the lithosphere that may be obscured by sediment and therefore are not visible in the sonar data.

Satellite gravity map showing the entire Reykjanes Ridge. It looks superficially similar to the bathymetry, but you can see more structure in the off-axis areas because the gravity can "see through" the low density sediments that obscure the underlying structure in the bathymetry data.  The fracture zones are clearly visible toward the southern end as the blue "stripes" going roughly horizontally, and you can clearly see that they progressively disappear as you go north.

Magnetic data is measured with a magnetometer that basically looks like a missile, towed a few 100 m behind the ship to reduce the magnetic effects of the metal hull.  Magnetic minerals (particularly magnetite) in the seafloor rocks produce a magnetic field, which varies in intensity mostly due to the quantity of these minerals in the rock.  But the more important piece of data is the orientation of the magnetic field.  While the lava is still molten, the magnetic minerals align with the orientation of the earth’s magnetic field, and when the lava cools this orientation is permanently frozen into the rock.  The earth’s magnetic field has switched polarity thousands of times over its history, and the rocks formed at various points in earth history still record the orientation of the field at the time they formed.  This allows geologists to correlate these reversals recorded in rocks all over the earth. Along spreading centers, these reversals create stripes of positively or negatively magnetized rocks as they form at the spreading center and migrate further off-axis.  By dating lava flows near these boundaries and measuring how wide the stripe of crust is between them, we can determine both current and past spreading rates.  For instance, the last reversal has been dated at 780,000 years ago, so if we know that this reversal is 7.8 km away from the axis (to make the math easy), we can estimate the spreading rate at 10 cm per year.

An example of magnetic data from the northern part of the Reykjanes Ridge.  There are two forms of the same data shown here.  The black wiggly lines are the actual data collected along the straight black lines that are the ship tracks, which also represent the "0" line if these were plotted as a graph. Where the wiggles are above the line, the seafloor is positively magnetized and where the wiggles fall below the line, it is negative.  The colored background is a grid produced from interpolating between the survey lines, with warm colors representing positive and cool colors representing negative magnetization.  You can see the first reversal as the first transition from the positive (red) magnetization right along the axis, to the dark blue just off the axis, and at least 7-8 more reversals are recorded as you go further away from the axis.