Explorations in Impact Cratering

A tour of crater types by the Kaguya lunar orbiter

2010-01-06T16:25:00.145-05:00

Up until this point, we have been discussing what are commonly referred to as "simple" craters; that is, impacts that produce a final crater which forms a smooth, paraboloid bowl with a depth which is roughly 1/5 its diameter. The small craters produced by the Deep Impact and LCROSS missions both fall into this category. However, when a hyper-velocity impactor is massive enough to produce a very large impact crater, something rather nifty occurs during the modification stage of the impact cratering process: for a short amount of time, the crater collapses as if it were a fluid, with little to no internal friction present. This unique feature of impact cratering was discovered in the late 1970's [1], and although it is still not well understood (with several competing mechanisms for why and how it happens), this "fluidization" process is generally accepted as the means by which large, complex craters are formed. Unlike simple craters, complex craters develop wall terraces rather than smooth crater walls, surrounding either a flat floor (for small complex craters), or a central peak structure (for medium-sized complex craters), or internal peak-ring structures (for the largest craters and impact basins).

In order to get a feel for what this process looks like, take a look at the below slow-motion video of a water droplet (the "impactor") falling into a container of water (the "target"), and notice the various forms that the water's surface assumes over time, under the force of gravity: first forming a "crater" cavity; which collapses outward to form the first ripple but also inward to form a tall central peak; which subsequently collapses downward and outward to form a peak-ring (the second ripple); some of which collapses again to for a second central peak; which subsequently collapses downward and outward to form a second peak-ring (the third ripple); and so on, in ever diminishing magnitude.

Slow-motion video of a water drop falling into a container of water. Note the various shapes assumed by the water's surface over time, under the influence of gravity.

Unlike the water puddle above, it is very important to realize that while fluidized crater collapse (in a rocky target) is going on, the target material is not, repeat NOT molten: individual rock fragments remain solid. However, during this short time period, frictional forces between solid rock fragments are reduced to near-zero, allowing fluid-like motion (under the force of gravity) of the entire fractured aggregate of rock forming the transient crater rim, sides, and bowl. One potential mechanism for how this happens is called "acoustic fluidization," in which the seismic waves (think "earthquake") produced by the impact and crater excavation flow persist for a time during transient crater collapse (the modification stage), keeping rock fragments separated enough to permit fluid-like flow via the constant vibrational motion between them [2]. Whatever the "fluidizing" mechanism, this frictionless motion only lasts for a finite amount of time, after which the rock flow "freezes" again (figuratively) as normal frictional forces are restored and motion is brought to a grinding halt. The larger the impact (and resulting crater), the longer fluidized motion can continue, thus resulting in different final crater shapes (morphologies) for different final crater sizes.

It is also important to understand that all of the motion that occurs during crater collapse is due to the force of gravity. The impactor's energy has been expended in the formation of the vapor plume, solid-ejecta plume, and transient crater cavity. There is no "rebound" of compressed basement rock to speak of (although this was once a popular hypothesis to explain central-peaks). Impacts do not induce any sort of volcanic reaction or activity at the impact site, unless they just happen to hit a region which is already volcanically active. Gravity does all of the work in forming a complex crater during the short time that fluidized flow is permitted. As such, the higher the gravity field, the faster motion occurs, and the further along in the collapse process a crater of a particular size will get before fluidized motion is brought to a halt. Thus, craters with central peaks first appear at a smaller crater size on Earth (2-4 km diameter) than they do on the Moon (15-20 km diameter), and likewise for the transition from central peaks to peak-ring structures. There are even some bodies in the solar system which are so small, and have such miniscule gravity fields, that complex craters can never form on them, only simple craters: a subject which we will explore in further detail in an later blog entry.

An excellent method for exploring how impact crater size affects the shape or morphology of the final crater was made available a couple of years ago by the Japanese Kaguya (SELENE) lunar orbiter mission, which in addition to a suite of scientific instruments, also carried on-board a High-Definition Television (HDTV) camera mounted so as to have a "pilot's eye" view of the lunar landscape below. Cruising in a low-altitude, polar orbit, Kaguya's HDTV was able to take some stunning footage of the lunar landscape, including several Earth-rise and Earth-set sequences. Also included in the video collection below are computer-generated "tours" of specific bits of the lunar surface, captured in stereo and mapped by Kaguya's high-resolution Terrain Camera. So let's begin!

Simple Craters

We start this tour with the simple craters that we are by now quite familiar with. On the lunar surface, simple craters occur in sizes from < 1 meter to 15-20 km in diameter, where the transition to complex cratering begins. The below video shows some of the smallest impact craters resolved by Kaguya, located at the Apollow 17 landing site (mapped by the Terrain Camera):

Small, simple craters, up to a few hundred meters in diameter, at the Apollo 17 landing site.

Still using data collected via the Terrain Camera, but moving up in scale, next is an impressive collection of simple craters in Mare Mosoviense. Note that the very flat-floored craters in this video were subsequently filled with lava some time after their formation, with unfilled younger craters overlaid on top of them.

The lunar far-side's Mare Moscoviense, showing simple craters 0.5 to 15 km in diameter, as mapped by the high-definition Terrain Camera. Some (flat-floored) small craters have been filled with lava.

Simple to Complex Crater Boundary

At about 15-20 km diameter (on the Earth's Moon), craters begin to show the characteristics of complex craters, beginning with terraced crater walls (rather than smooth), and the emergence of a central peak at slightly larger sizes (around 25-30 km). The video below shows a model of the formation of a small, central-peak crater, using acoustic fluidization as the friction-reducing mechanism (courtesy of Gareth Collins [3]).

An acooustic fluidization model showing crater collapse to the formation of a central peak, produced by Garen Collins, Imperial College, London.

Sitting very close to the Simple-Complex crater boundary is Giordano Bruno, which possesses terraced walls and a flat crater floor. This crater has the rather unique distinction in that its formation may have actually been witnessed by five monks from Canterbury on June 18, 1178, who reported seeing "two horns of light" on the shaded part of the Moon that evening, consistent with a sun illuminated solid-ejecta plume. The crater itself is located on the lunar far-side, not visible from Earth. The narration writer for this video must be a Pink Floyd fan, because he refers to the "Dark Side of the Moon," rather than correctly saying "far-side," as seen from Earth. The Moon's polar axis is tilted by only 5.1 degrees (to ecliptic normal), as compared to the Earth's 23.5 degree tilt, and as such, all portions of the Moon see both "day" and "night" during its 27.3 day rotation (and orbital) period. With the exception of a few permanently shadowed crater floors near the Moon's poles, there is no so-called "dark side," just a near-side and a far-side from our perspective on Earth.

A flight over the very fresh 22 km diameter crater, Giordano Bruno, just above the simple-to-complex crater boundary in size and possessing a slightly flat floor, slightly terraced crater walls, and a bright, rayed ejecta blanket.

Central Peak Craters

Now we tour a visually impressive series of lunar central-peak craters, slowly moving our way up the size scale as we go. This series begins with the extremely bright, 40 km diameter crater, Aristarchus. The even, terraced walls of Aristarchus give the crater a stadium-like appearance.

A flight over the bright, 40 km diameter central-peak crater, Aristarchus, with a bright crater rim and ejecta blanket.

Next comes 60 km diameter Bullialdus. During these fly-overs, keep an eye out for interesting smaller, simple craters, and see if you can spot the boundary between simple and complex as slightly larger craters begin to display terraced walls and flat floors. Note also that many very old craters have been filled with mare lava, such that only the top of the rims remain.

A flight over the 60 km diameter central-peak crater, Bullialdus.

The 84 km diameter crater Tycho is perhaps the best known crater on the Moon, with a very extensive, bright ray system easily visible from Earth when the Moon is near full. The below "tour" of Tycho is one of my favorite Kaguya videos, but beware: the people who wrote the narrative obviously hold to old views of wall terrace formation (they attribute them to melting and resolidification) and central peak formation (they attribute it to post-impact volcanism): hypotheses which have been discounted for some time now.

A detailed examination of 84 km diameter central-peak crater, Tycho, as mapped by the high-definition Terrain Camera.

Never seen from Earth, the 96 km diameter, far-side crater Icarus has an impressively high central peak, which is unusual in that it extends above the crater's rim -- most lunar central peaks rise to roughly half of the floor to rim height.

A flight over the 96 km diameter central-eak crater, Icarus.

Familiar to many backyard observers is the 100 km diameter central-peak crater, Copernicus, which, like Tycho, has a brightly visible system of rays.

A flight over the 100 km diameter central-peak crater, Copernicus.

Moving to even larger sizes, the next video features Earth-rise over the 110 km diameter crater, Plaskett.

A slow, spectacular flight over the 110 km diameter central-peak crater, Plaskett, with Earth-rise shown in the background.

Last in this series is the 127 km diameter Langrenus crater, which comes into view from beneath, at the beginning of this video.

A flight over the 127 km diameter central-peak crater, Langrenus.

Central Peak to Peak-Ring Crater Boundary

At around 140-150 km diameter (on the Earth's Moon), craters begin to transition from central peak craters to single peak-ring craters. The video below shows a model of the formation of a small, peak-ring crater, using acoustic fluidization as the friction reduction mechanism (courtesy of Gareth Collins [3]).

An acoustic fluidization model showing the formation of a peak-ring crater, produced by Gareth Collins, Imperial College, London.

Right on the line between having a central peak or a peak-ring is the 140 km diameter crater, Pythagoras. Pythagoras has both a central peak and the beginnings of a peak-ring, just outside of the central peak.

A flight over the 140 km diameter crater, Pythagoras. In addition to a central peak, the crater shows the beginnings of a peak-ring just outside (and cencentric to) the central peak.

Peak-Ring Craters

On the Moon, the transition between central peak craters and peak-ring craters tends to be a bit irregular, resulting in many oddly shaped, small peak-ring structures. At larger sizes, the peak-rings themselves tend to be rather low in height (altitude above the crater floor), making them easy to cover over by later lava flows. These two features make small, pristine, peak-ring craters a bit rare on the Moon. However, the 143 km diameter peak-ring crater, Antoniadi, is an exception, with a nicely formed, visible peak-ring (along with a tiny, residual central-peak).

A flight over the 143 km diameter peak-ring crater, Antoniadi.

The next two craters, the 149 km diameter crater, Drygalski and the 167 km diameter crater, Hausen, both fall into the "irregular" central peak / peak-ring category.

A flight first over the 149 km diameter crater, Drygalski, then over the 167 km diameter crater, Hausen. Both craters have oddly shaped, lop-sided, peak-ring structures.

Nearby to the 84 km diameter central-peak crater, Tycho (where the narrator again incorrectly cites volcanism as the formation process for its central peak), is the much larger 245 km diameter crater, Clavius, the site of the fictional, manned Moon-base in Arthur C. Clark's 2001; A Space Odyssey. While an impressive crater, Clavius's peak-ring structure has been almost entirely buried by lava and eroded by subsequent impacts.

A flight first over the 84 km diameter central-peak crater, Tycho, then over the 245 km diameter peak-ring crater, Clavius. The peak-ring is almost entirely buried in lava and subsequently pock-marked by smaller craters.

Peak-Ring Basins

Now we get into the really big craters, often called basins or impact structures, which generally have an inner-ring structure along with an outer, ruggedly terranced and broken crater rim structure. The below model depicts the formation of one of these very large peak-ring craters, again using acoustic fluidization as the mechanism for near-frictionless crater collapse (courtesy of Gareth Collins [3]).

An acoustic fluidization model showing the formation of a peak-ring basin, produced by Garen Collins, Imperial College, London.

Below, the 320 km diameter Schrödinger basin presents an impressive and well defined single peak-ring structure.

A flight over the 320 km diameter Schrödinger basin, which possesses two visible rings: one interior peak-ring and a heavily fractured and terraced rim, or "outer ring."

Continuing to move up in scale, the 480 km diameter Apollo basin presents a battered, lava filled structure which is much more difficult to define, but still consisting of a single peak-ring and outer rim, as did the smaller Schrodinger basin.

A flight over the 480 km diameter Apollo basin, which has a cratered, single peak-ring structure (plus outer crater-rim).

Multiring Basins

Our final catagory of impact structures are the enigmatic multiring basins, which not only posses inner peak-rings as in the previous examples, but also additional, concentric rings which formed outside of the final crater rim. These outer rings on the Moon often occur as a series of rugged, steep scarps which face towards the basin's interior. The mechanism of formation for these outer ring structures is not yet understood, but is thought to be due to a separate "fluidzation" process from the one discussed above for the formation of inner peak-rings.

At 590 km in diameter, the Hertzsprung basin is our first look at a crater with an outer rim and two concentric peak-rings, giving it an overall three-ring structure.

A flight over the 590 km diameter Hertzsprung basin, which has a broad, double peak-ring structure (plus the terraced rim). Note the fresh simple crater with a very bright ejecta blanket that appears just right of center around the 45-50 second point.

And finally, at 930 km diameter, the Orientale basin, gives us a look at an impact structure with an outer rim (the Montes Cordiller lunar mountain range) and three concentric peak-rings, giving it an overall four-ring structure.

A flight over the 930 km diameter Orientale basin, which has a prominent, interior three-ring structure (plus the mountainous outer rim).

The lunar surface possesses several impact basins which are still larger in size, most of which were later filled by mare lavas to form the dark "seas" and "oceans" visible on the near-side of the Moon. All that remains visible of their impact structures are curved mountain ranges around the Mare edges: the remains of the basin rims. The largest and oldest basin on the Moon is the South-Pole Aitken Basin, which is roughly 2500 km in diameter. Visibly, it is difficult to distinguish via the "Leibnitz mountains" which roughly mark the old basin's rim, but detailed topographic and surface composition studies are able to reveal the basin for what it is: an ancient lunar impact structure.

And thus ends our tour of the various crater morphologies (types) and the sizes on the lunar surface which create them. Please remain seated, with your seatbelt fastened, until the spacecraft has come to a complete stop, and be sure to congratulate the Japan Aerospace Exploration Agency (JAXA) for a successful and visually spectacular mission!

References

[1] H.J. Melosh (1977). Crater modification by gravity - A mechanical analysis of slumping, Impact and explosion cratering: Planetary and terrestrial implications; Proceedings of the Symposium on Planetary Cratering Mechanics, Flagstaff, Arizona, September 13-17, 1976. New York, Pergamon Press Inc., pp. 1245-1260.
[2] H.J. Melosh, B.A. Ivanov (1999). Impact Crater Collapse, Annual Review of Earth and Planetary Sciences, Vol. 27, pp. 385-415.
[3] G.S. Collins, H.J. Melosh, J.V. Morgan, M.R. Warner (2002). Hydrocode Simulations of Chicxulub Crater Collapse and Peak-Ring Formation, Icarus, Vol. 157, pp. 24-33.

Shootin' the lunar icebox: the LCROSS impact

2009-12-17T14:28:00.076-05:00

Water Ice on the Moon

The notion that impact craters located in polar regions of the Moon could potentially harbor hidden reserves of water ice within their permanently shadowed depressions has been around since the year that man first entered space [1]. For many years, however, we lacked the means to test this hypothesis. The Apollo landings were, by necessity, concentrated in relatively flat, Earth-facing, low-latitude regions, and thus found no evidence of water in their collected samples. The regions of the Moon visited by humans were bone-dry, and many assumed that this must also be true for the entire lunar surface.

The prevailing dry-Moon view, however, began to change with the 1994 Clementine mission. This military lunar orbiter performed radar mapping of the lunar surface, and it was discovered that radar signals returned from permanently shadowed regions of lunar craters near the poles seemed to indicate the presence of water ice, exactly where Watson et al. [1] predicted it should be. These results, however, were rather controversial and inconclusive. A few years later in 1998, the Lunar Prospector spacecraft specifically searched for evidence of water on the Moon by mapping the abundance of hydrogen-rich minerals on the Moon's surface using a neutron spectrometer. Lunar Prospector found that the permanently shadowed regions of polar craters indeed showed high concentrations of hydrogen: perhaps due to the presence of water ice. At the end of Lunar Prospector's mission, the probe was purposefully impacted near the Moon's south pole in hopes that enough water vapor might be liberated for Earth-based instruments to detect it. None was, leaving some to continue to question where the hydrogen in the Moon's polar regions originated: either from water ice, or from hydrated minerals.

Most recently, the Lunar Crater Observation and Sensing Satellite (LCROSS) mission purposefully sent a Shepherding Spacecraft (S-S/C), powered by a Centaur rocket engine, on a collision course with one of the permanently shadowed craters near the Moon's south pole (see Fig. 1 below). Following separation of the S-S/C from the Centaur rocket, the S-S/C was slowed sufficiently to permit it to make detailed observations of the Centaur's collision with the interior of the crater Cabeus (from a safe distance of about 600 km), using a variety of ultraviolet, visible-light, and infrared instruments. With a nominal impact mass of 2305 kg and impact speed of 2.5 km/sec, the Centaur rocket delivered a total of 7.2 gigajoules of impact energy (equivalent to 1.72 tons of TNT) to the lunar surface. After sending data from the Centaur rocket's impact back to Earth, the Shepherding Spacecraft then followed the Centaur rocket into the lunar surface itself, a few minutes later. The purpose of these collisions was to utilize the impact cratering process to expose the material hidden in the shadows of these polar craters through both vapor plume expulsion and the excavation of solid-particle ejecta during crater formation. The LCROSS impact took place on 9 October 2009 at 11:31 UTC with near perfect impact targeting and instrument pointing for both Centaur rocket and Shepherding Spacecraft, an engineering feat that we tend to take for granted these days, but which is still impressive, nonetheless.

Figure 1: A view of the Moon's south polar region as seen by the LCROSS Shepherding Spacecraft (S-S/C) during its approach, taken from a range of 770 km. The Centaur rocket impact point is within the shadowed region of the crater Cabeus, slightly up and to the left of where the arrow is pointing. (Source: NASA LCROSS mission site)

Impact day: 9 October 2009

In addition to observations conducted by the Shepherding Spacecraft, the LCROSS mission team requested that Earth-based professional and amateur astronomers participate in the observation campaign in order to maximize the potential science return from the mission. The visibility of the vapor plume and solid ejecta plume to the backyard observer (with a large scope: > 10"-12" diameter), was questionable, but intrepid enthusiasts were welcome to try it, with the following caveat:

"To re-iterate, impact event visibility variables include surface topography (soft dust or hard rock) both at surface and 1 to 2 meters down at the location of impact, impact angle (we are controlling fairly well), velocity and mass (both pretty much fixed). Key elements for the backyard observer are how high will the visible ejecta plume go, and how bright will it be. What we can see will be some intersection of these two elements in the visible spectrum. Large aperture earth and orbiting instruments will be able to observe/resolve to very low altitudes where there will be much greater plume/dust mass/brightness (we can't resolve well to that low level with most backyard scopes but if sunlight is low enough, may be able to see a bright spot), and very dim high altitude spectra of the ejecta (we won't be able to see the low brightness parts of this very diffuse vapor cloud)."

On the morning of the planned impact, which occurred at 7:31 EDT / 4:31 PDT, many others of us who were following the mission (and not actively monitoring telescopes) watched the live feed from the Shepherding Spacecraft, very conveniently web broadcast on NASA TV, and witnessed the following:

Note that near the time of Centaur impact, the NASA TV feed switched to one of the Mid-Infrared instrument's false color views (colored by temperature) during "impact flash" mode, and then back again to the visual camera for "curtain" mode. These mode terms refer to the two parts of the cratering excavation stage that we've discussed previously; that is, the eruption of the brightly incandescent vapor plume (impact "flash"), and the emergence and expansion of the solid-particle ejecta plume (or ejecta "curtain"). Following this broadcast, Emily Lakdawalla wrote in the Planetary Society Blog:

"Everything seemed to go according to plan, except for one thing: as far as I can tell, nobody anywhere saw any visual confirmation of the impact. That doesn't necessarily mean bad news; the real data will come from thermal instruments and spectrometers and such, and the reduction of that data may take days to weeks. It was a bit disappointing not to see anything though."

The most encouraging point in the above broadcast occurred near it's end, when the science team reported that they had confirmation of a "thermal signature of the crater" in the Mid-Infrared camera. At a press conference later on that day, the below image of vapor plume eruption ("impact flash") was released:

Figure 2: Initial press release image showing the signature of the incandescent vapor plume produced by the LCROSS Centaur impact. Source: Emily Lakdawalla (Planetary Society)

Vapor Plume Observations

About one month after the LCROSS impact, NASA held a press conference to announce the preliminary results from the LCROSS science team. These results centered on images and spectrometer data from the vapor plume produced by the Centaur impact, and showed that over 100 kg (about 24 gallons) of water had been liberated and detected in at least two instruments on-board the Shepherding Spacecraft: validating the hypothesis that permanently shadowed craters at the lunar poles act as "cold traps" and posses stores of water there -- a very exciting and satisfying result.

A time-sequence of images showing the emergence of the impact-produced vapor plume are shown below, as seen by one of the Mid-Infrared imagers. In viewing these, it might be helpful to review this image sequence of the erupting, brightly incandescent vapor plume produced by Deep Impact (which quickly expands and moves away from the comet's surface) and imagine what that vapor plume would have looked like viewed from directly overhead (the view that the LCROSS Shepherding Spacecraft had of the Centaur rocket impact), rather than at the 60 degree tilt that the Deep Impact flyby spacecraft enjoyed.

Figure 3: A false-color image sequence (spectral colors indicate temperatures) of the initial, hot eruption of the vapor plume produced by the Centaur impact, as seen from the Shepherding Spacecraft. Note that the orientation of this image set is upside-down and inverted left-right, compared to the other images presented in this blog entry. (Source: NASA LCROSS mission site)

The vapor plume's expansion was also captured in the ultraviolet/visible-light camera, particularly as the rapidly cooling and condensing water vapor (along with entrained dust particles) rose up out of the shadow of the crater rim and into sunlight. At 15 seconds after the impact, the vapor plume had already expanded to a diameter of roughly 6-8 km:

Figure 4: The expanding vapor plume (along with entrained, fine dust particles) produced the Centaur rocket impact was captured in the ultraviolet/visible-light camera as it emerged into the sunlight above Cabeus crater about 15 seconds after impact. Note the scale-bar in the insert on the lower-left. (Source: NASA LCROSS mission site)

Thus, although vapor plume eruption (including initial "flash") and expansion were not readily visible in the real-time video feeds on the day of impact, with a bit of data enhancement and reduction the results have proven to be quite impressive and robust. And the science team has only gotten started -- other molecules, beyond water, are also present in the vapor plume's spectral signature which look rather intriguing, but which will require a good bit more work to constrain, or "nail down." Stay tuned for further announcements and/or initial papers from the LCROSS science team.

Solid-Particle Ejecta Plume

Noticeably missing from the images released so far is any indication that the solid-particle ejecta plume produced by the LCROSS Centaur impact was observed. This is in marked contrast to the extraordinary solid-particle ejecta plume produced by Deep Impact, as seen in Fig. 7 and Fig. 10 of the previous post. This is not surprising, however, for two basic reasons. First, the cometary surface struck by Deep Impact is composed of an extremely fine (1-10 micron diameter) silicate dust embedded in a matrix of water ice and other substances, while the lunar regolith which makes up the floor of craters such as Cabeus is much coarser: rather like a comparison between a fine talcum power and gravely beach sand. The finer the material, the brighter and more visible it will be when launched upwards to form a cloud of ballistic particulate. Second, the LCROSS Centaur impact took place in the shadowed region of a crater, and unlike a vapor plume, the solid ejecta plume is not incandescent at any stage -- it depends upon sunlight to be visible (although it should appear warmer than background at infrared wavelengths). As such, only those portions of the solid-particle ejecta plume which rose above the shadow of the crater's rim would have been visible, and were expected to be rather tenuous to observe at best. The below movie shows the LCROSS science team's expectation for this feature:

This animation is based upon a variety of predictive models [2] which indicated that about 1x10^3 - 1x10^4 kg of ejecta material should have been launched above the level of the rim shadow and into sunlight. Taking advantage of the impact ejecta plume modeling development that I did for the Deep Impact mission [3], I've run some simulations of my own which may show why the LCROSS impact solid-particle ejecta plume was not visible. To begin, you'll recall from Fig. 5 in the previous post that passage of the impact shock-front through the target material sets up a crater excavation flow-field, in which material close to the impact site is ejected at high velocity and material located farther from the impact site is ejected at slower velocities, until a point is reached that nothing is ejected at all (only uplifted slightly to form the crater rim). For my LCROSS Centaur impact model, I can plot ejection velocity as a function of distance from the impact site to obtain:

Figure 5: A log-log plot of particle ejection velocity as a function of distance from the impact site, shown for a Centaur rocket impact into unconsolidated lunar regolith at 2.5 km/sec (dashed lines indicate target material uncertainties). The minimum distance of 1.4 m corresponds to 1/2 of the mean impactor radius (2.8 m), and represents an approximate lower limit to the location from which solid ejecta could have been launched.

The unusual thing about this ejecta velocity distribution, in comparison to other impacts of similar size, is the extremely large size of the projectile in relation to the size of the crater produced. The Centaur rocket used here was a cylinder roughly 13 m long and 3 m in diameter, composed essentially of empty propellant tanks, having a nominal mass of about 2305 kg and thus a very low mean density of about 25 kg/m^3. With a mean impactor diameter of 5.6 m, this means that target material which would have been ejected at velocities approaching 1000 m/sec (given a smaller, denser impactor), was instead caught up in the contact and compression zone during the coupling stage of the cratering process. Consequently, maximum ejection velocities for the solid ejecta produced by this Centaur rocket impact approach only 100 m/sec. Another way of looking at this is to plot the maximum altitudes achieved by this ejecta as a function of ejected mass:

Figure 6: A log-log plot of cummulative ejected mass, shown as a function of maximum altitude achieved above the lunar surface, for the solid-particle ejecta produced from a Centaur rocket impact into unconsolidated lunar regolith at 2.5 km/sec (dashed lines indicate target material uncertainties). Note that of the 2x10^5 - 7x10^5 kg ejected, less than 1x10^4 kg achieves more than 100 m in altitude, with a maximum altitude of 500-800 m reached by the fastest ejecta.

Thus, according to my own modeling, the solid (that is, non- vapor-plume) ejecta produced by this impact should not have reached the altitudes needed to rise above the crater rim's shadow, thus making it invisible to the Shepherding Spacecraft's visible-light camera. Below is a simulation of what this ejecta plume (or ejecta curtain) would have looked had a stationary camera been perched nearby (150 m away) and with sufficient light available. Note that the vapor plume and crater interior are not included -- this simulation shows only the solid-particle ejecta plume. In this view, the impactor enters from the far-right, at an impact angle of 65 degrees above the horizon:

Following it's flight and landing, the solid-particle ejecta plume produced by the LCROSS Centaur impact formed an ejecta blanket around the final crater, which was imaged in the Infrared (while the freshly landed ejecta was still warm compared to its background):

Figure 7: An infrared view of the normally unlit floor of lunar crater Cabeus, showing portions of the ejecta blanket (bright material) surrounding the (dark) ~20 m diameter crater produced by the LCROSS Centaur rocket impact. (Source: NASA LCROSS mission site)

Compare the ejecta blanket in this image with the ejecta blanket produced in the simulation above.

Final Crater Observations

Unlike the Deep Impact mission, the LCROSS Shepherding Spacecraft did manage to take images of the final crater produced by the Centaur rocket impact, as seen in the infrared on the unlit floor of Cabeus crater:

Figure 8: A mid-infrared view of the still-warm final crater produced by the LCROSS Centaur rocket impact. (Source: NASA LCROSS mission site)

Figure 9: A closer, near-infrared view of the still-warm final crater produced by the LCROSS Centaur rocket impact. (Source: NASA LCROSS mission site)

The size of the final crater produced gives us important clues about the type of material that the impactor struck. Both images above indicate a final crater diameter of roughly 16-24 meters, which is consistent with an impact into loosely consolidated, low material strength, lunar regolith. In the lunar gravity, this produced a transient crater which formed in about 1.5-2.0 seconds, and then collapsed to form a final crater in about 5-10 seconds. If you're interested in seeing the effects of using different target materials, I'd recommend two websites to you: The first is a relatively simple crater-size calculator, developed by my Ph. D advisor, H. Jay Melosh, and the second is a somewhat more complex (but nicely detailed) "Crater Sizes from Explosions or Impact" calculator developed by Keith Holsapple, another highly-respected researcher in the field of impact cratering. What you will find is that only the weakest target materials will produce a crater approaching 20 meters in final diameter for the LCROSS Centaur rocket impact -- hard rock targets produce craters which are roughly one half that size (or smaller). Key parameters that you will need are:

Impact speed: 2.5 km/sec
Impact angle: 60-70 degrees above the horizon
Impactor mass: 2305 kg
Actual impactor dimensions: 12.7 m long x 3.0 m diameter cylinder
Mean impactor diameter: 5.6 m (mean radius = 2.8 m)
Mean impactor density: 25.0 kg/m^3
Target surface gravity: 1.62 m/sec^2
Target regolith density: 1700-2500 kg/m^3 (variable)
Target surface strength: 50-500 kPa (variable)

I should mention one caveat when playing with these crater-size calculators: the analytical formulae that they use (called impact-crater scaling relationships) are based upon experimental data and mathematical point-source solutions, which means that the formulae assume that the final impact crater is much larger than the impactor which produced it. Obviously, this is not true for the Centaur impact, although the formulae are still quite useful for producing rough estimates. I suspect that the large, 13 m by 3 m cylindrical impactor (with it's very low mean density), most likely produced an asymmetrical impact crater and ejecta blanket, which, in turn, probably accounts for some of the asymmetries present in Figures 7-9 above. It would be most interesting to return for a closer look one of these days.

Conclusion

In pictorial summary, take another look at one of the basic "concept" diagrams for what the LCROSS mission was attempting to do:

Figure 10: A basic concept diagram of the way in which the LCROSS misison would take advantages of the the impact cratering process to liberate and detect water ice on the floor of Cabeus crater. The initial, quickly erupting vapor plume is labeled "Plume" here, while the slower, solid-particle ejecta plume is labeled "Blanket." (Source: NASA LCROSS mission site)

The LCROSS mission team succesfully accomplished and observed (collected useful data on) pretty much everything depicted above. The only portion of this concept diagram which was not actually observed was the formation and expansion of the solid-particle ejecta plume (labeled "Blanket" in the left-most frame), although it was observed once it had fallen out, as a warm ejecta blanket around the final crater. The liberation and detection of over 100 kg of water (as either H2O or OH molecules) in the vapor plume was particularly well done.

I'd like to offer a very heartfelt congratulations to the entire LCROSS team for a very well engineered and executed mission, as well as my thanks for the presentation of some spectacular images and impressive early science results in such a short time frame. You folks have done a marvelous job!

References

[1] K. Watson, B.C. Murray, and H. Brown (1961). The Behavior of Volatiles on the Lunar Surface, Journal of Geophysical Research, 66(9), 3033–3045.
[2] D.G. Korycansky, C.S. Plesko, M. Jutzi, E. Asphaug, A. Colaprete (2009). Predictions for the LCROSS mission, Meteoritics & Planetary Science, 44(4), 603-620.
[3] J.E. Richardson, H.J. Melosh, C.M. Lisse, and B. Carcich (2007). A ballistics analysis of the Deep Impact ejecta plume: determining comet Tempel 1's gravity, mass, and density, Icarus, 190, 357-390.

Deep Impact and the impact cratering process

2009-12-04T12:25:00.031-05:00

In the previous post, we touched on the fundamental similarity between a hyper-velocity impact crater and an explosion crater. But how exactly is a crater formed when some super-sonic chunk of asteroid or comet strikes a much larger solar system body? In this post, we will "dig a bit deeper" and discuss the basic stages of the impact crater formation process itself, and as a way of keeping it interesting, I will also describe how these stages relate to the observations made by the Deep Impact flyby-spacecraft of the cratering event produced when NASA purposefully colliding a 366 kg impactor-spacecraft with the 6 km diameter nucleus of Comet 9P/Tempel 1, at a closing speed of 10.2 km/sec, on 4 July 2005. Although the terminology used by scientists varies somewhat, in general, an impact cratering event can be loosely divided into three basic stages: a coupling stage, an excavation stage, and a modification stage. In reality, these stages do not have clear-cut boundaries, but instead gradually move from one to the next with areas of overlap between them. Nonetheless, dividing the event into these stages gives us a convenient set of "handles" on the process.

Coupling Stage

The first, coupling stage (also called the "contact and compression" stage) of an impact event begins the instant that the impactor touches the target surface. During the coupling stage, the kinetic energy and momentum of the impactor are transmitted, or coupled, into the target material as the impactor pushes into and rapidly accelerates the target material, while at the same time the target material rapidly deforms and decelerates the impactor. These rapid velocity changes produce two shock-waves, which begin at the point of contact between impactor and target, and then rapidly propagate both forward into the target material and backward into the impactor. It is the forward propagating, hemispherically-shaped, compressive shock-wave in the target material that transmits the majority of the energy initially contained in the impactor to the target. This compressive shock-wave is followed almost immediately by a rarefaction-wave, which is the reflection of the compressive shock-wave off the upper, free surface of the target and impactor. This coupling-phase energy transfer from impactor to target takes place within a small region of the target, roughly a few times the volume of the impactor, and in a very short amount of time, on the order of d/v, where d is the diameter of the impactor and v is its velocity. This process is depicted in the simple diagram below:

Figure 1: A simplified diagram of the Coupling Stage of the cratering process, during which the kinetic energy and momentum of the impactor is transmitted to the target material. Note that the remains of the impactor end up lining the wall of the initial crater cavity. Source: E.H. Christianson, Exploring the Planets, 2nd Ed., Pearson Education Inc., 1995.

In the case of the Deep Impact mission, this purposeful impactor-comet collision resulted in the deposition of ~1.9 x 10^10 Joules of energy (equivalent to 4.6 tons of TNT) in a target volume about 1-10 cubic meters in size, and in only 1-10 msec (milliseconds), depending upon impactor penetration depth (5-50 meters). As such, this first stage of the cratering process was not directly observable by the flyby-spacecraft, which at that time had an image spatial resolution of only 35 meters per pixel and a time resolution of 59 msec. In the low-resolution image frames below, the coupling stage begins at time t = 0, between the top two images, after which the impact site became visible as a slightly brighter pixel-cluster for a couple of frames prior to vapor plume eruption, which is discussed next. Exactly why there was this relatively long, 150-180 msec "pause" between first contact and vapor plume eruption remains an open research question.

Figure 2: High-speed Medium Resolution Imager (MRI) frames showing the initial stages of the Deep Impact collision. The coupling stage of this cratering event lasted about 1-10 msec, and took place between the top two frames in this sequence. Times are given in reference to the time of impact, t = 0.

Excavation Stage

In the immediate vicinity of an impact site, shock pressures will greatly exceed the yield strength of the target material, and at the same time, the amount of internal energy deposited in the target will greatly exceed that necessary to vaporize and/or melt this material (called shock heating). Additionally for Deep Impact, crushing of the highly porous cometary surface may also have transformed a large amount of kinetic energy into internal (heat) energy. The result of this rapid crushing and heating is the creation and expulsion of an extremely hot, rapidly expanding bubble of vaporized target material and entrained melt droplets from the impact site, called a "vapor plume," which moves quickly away at speeds near to that of the original impactor. This marks the beginning of the excavation stage of the cratering process, and is shown in simplified form below:

Figure 3: A simplified diagram of the first phase of the excavation stage of crater formation: showing the eruption of the vapor plume from the immediate vicinity of the impact site, where shock pressures are the highest. Source: H.J. Melosh, Impact Cratering, A Geologic Process, Oxford University Press, 1989.

The amount of material vaporized and melted by a given impact is highly dependent upon the nature of the impactor, impact velocity, and target material, but in the case of the Deep Impact event, this amount is estimated to be on the order of several impactor masses, encompassing material within a few meters in radius of the impact point (the impactor itself was about 1 m in diameter). The vapor plume produced by Deep Impact was initially brightly incandescent due its extremely high temperature, and has been determined by H. Jay Melosh to be primarily composed of water vapor and silicate melt droplets (pointing to a water-ice and silicate dust surface composition):

Figure 4: High-speed Medium Resolution Imager (MRI) frames showing the emergence of the vapor plume produced by Deep Impact, which was initially so brightly incandescent that it saturated parts of the CCD camera and caused "bleed-over" into adjacent pixels. This bubble of cooling and expanding material rapidly moved away from the impact site at roughly 5-7 km/sec.

The movie below shows the MRI image sequence of the Deep Impact vapor plume emergence, played at about 1/2 actual speed. Note the bright, incandescent flash as the vapor plume initially erupts, followed by the rapid departure of this bubble of vapor and melt droplets. At the same time, the solid-particle ejecta plume begins to emerge, casting a shadow on the comet's surface, whose formation will be discussed next.

Outside of the vaporization and melt zone, the rapid compression and rarefaction produced by the passage of the expanding (and weakening) shock front does two things: first, it severely fractures and damages the target material as it passes, and second, it injects a large amount of residual kinetic energy into this material. This results in the establishment of a hydrodynamic (fluid-like) flow-field of solid ejecta fragments which then excavates and forms the crater. This process is depicted pictorially in Fig. 5 below, in which the excavation flow moves material upward and radially outward to (a) open up a paraboloid crater cavity in the target and (b) form a rapidly expanding, hollow, conical plume (or curtain) of expelled ejecta fragments.

Figure 5: A simplified diagram depicting the crater excavation flow-fleid established by the rapid passage of the impact shock-front through the targat material. The fastest fragments, closest to the impact site, are ejected first but will land on the ground last, far from the impact site (if they land at all). The slowest fragments are ejected last, but will land first and closest to the crater rim. Source: H.J. Melosh, Impact Cratering, A Geologic Process, Oxford University Press, 1989.

As the shock front in the target expands away from the impact site (advancing far ahead of the excavation flow-field that it sets up), it weakens rapidly, such that the amount of damage done to the target material, and the amount of kinetic energy deposited in this material, rapidly falls off with increasing distance from the impact site. Therefore, although the crater initially grows quite quickly and the early ejecta are launched at high velocities, the crater growth rate falls off rapidly, accompanied by rapidly slowing ejecta velocities with increasing distance from the impact site. The excavation flow finally comes to a halt, and the transient, pre- modification-stage crater formed when the upward and outward, hydrodynamic excavation flow is overcome by either the force of gravity, the post-shock strength of the target material, target material viscosity (usually negligible in impacts involving solid, rocky materials), or a combination of these three factors.

Figure 6: A simplified diagram showing the solid-particle phase of the excavation stage: transient crater growth and ejecta plume formation. Note that the crater is formed by both the ejection of target material, launching it radially upward and outward, and the displacement of target material, pushing it radially downward and away from the impact site (see Fig. 5). This process forms a paraboloid "transient" crater, which initially has a depth of about 1/3 of its diameter. Source: H.J. Melosh, Impact Cratering, A Geologic Process, Oxford University Press, 1989.

In the case of the Deep Impact cratering event, the excavation stage is estimated to have lasted up to about 300 seconds (5 minutes) in duration, and was therefore observed in its entirety during the 800 seconds (13.3 minutes) of approach phase imaging. Unfortunately, the copious amount of fine particulate produced by the cratering event obscured the spacecraft's view of the impact area itself, such that the crater formation process was not directly observed. However, there is still much information about this process that was gleaned from these observations, made as the flyby-spacecraft peered directly into the interior of the funnel- or cone-shaped ejecta plume. Jessica Sunshine and the team in charge of Deep Impact's Infrared Spectrometer have determined that the bulk of the solid ejecta produced by the impact is composed of a mixture of water ice and extremely fine silicate dust particles.

Figure 7: A view of Deep Impact's second phase of crater excavation -- the formation and expansion of the solid-particle ejecta plume -- as seen from almost directly overhead as the flyby-spacecraft approached the comet. The extremely fine, 1-10 micron diameter ejecta particles where so copious as to obscure our view of the impact crater itself, which is contained within the dark, oval portion of the "keyhole" shaped shadow in the right-hand image.

Modifcation Stage

The halt of crater excavation marks the beginning of the final stage in the cratering process, called the modification stage. This stage comprises two processes which occur simultaneously. First, the plume of solid ejecta fragments expelled during the excavation stage will gradually fall out under the influence of the target body's gravity onto its surface, and will thus form a blanket of material extending outward from the rim of the transient crater:

Figure 8: A simplified diagram showing the expansion and fall-out of the solid ejecta plume, in radial cross-section. Although each fragment is following its own ballistic path under the influence of the target's gravity field, together the fragments form a hollow, expanding, cone-shaped cloud whose bottom maintains contact with the target surface (at the fall-out "front"). Times (T) are in terms of the crater's formation time. Source: H.J. Melosh, Impact Cratering, A Geologic Process, Oxford University Press, 1989.

Second, the transient crater itself, which is gravitationally unstable due to its steep sides, will collapse, with crater wall materials sliding downward and inward toward the center of the bowl and causing the crater to become wider and shallower as it attains its final shape: for small, simple craters, the final crater diameter is larger than the transient crater diameter by a factor of about 1.1-1.3, with a final depth to diameter ratio of about 1 to 5. Recall that the nuclear explosion craters that we examined in this blog post have depth to diameter ratios of about 1 to 4, and are thus more steeply sided. The modification stage is complete when the crater has attained its final, stable form, and all of the impact ejecta have either been redeposited on the surface of the target body or have escaped from the target body's gravity-well.

Figure 9: A simplified diagram showing the modification stage of crater formation. Note the "lens" of previously excavated material that collects in the bottom of the final crater. Source: H.J. Melosh, Impact Cratering, A Geologic Process, Oxford University Press, 1989.

With regard to the Deep Impact event, it was expected that very little, if any, of the transient crater collapse would be observed, for two reasons: first, this stage would only be captured in the last portion of the 800 seconds of approach phase imaging, after the excavation stage was complete; and second, the very low gravity field of comet Tempel 1 (estimated at 0.25-50 mm/sec^2) would cause such crater gravitational collapse to proceed quite slowly, and therefore would not be visible in the limited time available. However, it was expected that much of the expansion and fallout of the ejecta plume would be visible, during both the approach and look-back imaging phases, provided that the viewing geometry was favorable and the ejecta particle distribution was fine enough to produce an easily visible ejecta plume, as proved to be the case. Observations of the ejecta plume produced by the Deep Impact mission lasted from the moment of first emergence (~340 msec after impact) all the way to the final look-back images taken 75 minutes following the impact.

Figure 10: A High Resolution Imager (HRI) frame of the solid-particle ejecta plume produced by Deep Impact, shown at 48 minutes after the impact in profile-view, as the flyby-spacecraft speed away from the comet. At this point, the plume has a base diameter of about 3 km as it slowly expands and falls-out in the comet's extremely low gravity field. The particles in the lower (brighter) portions of the plume were ejected at less than the comet's escape velocity, and will eventually land back on the surface, while the particles in the upper (darker) portions of the plume will eventually escape and be swept away by solar radiation pressure.

The below movie shows the complete image sequence for the Deep Impact HRI instrument from 18 minutes before impact to 64 minutes after (impact occurs at time t=0), showing the full formation and much of the evolution of the solid-particle ejecta plume. The first 800 seconds (13.3 minutes) after impact show "approach-phase" images, which was followed by 31 minutes of spacecraft "safe-mode" (no images taken) during closest approach to the comet, and finally a "look-back" phase of images taken from 44 to 75 minutes after the impact: so be ready for the sudden leap from t=788 to t=2659 seconds in the image sequence. The imager's field-of-view is constantly changing, as is spacecraft's pointing direction (as other instruments were pointed at the impact site); however, this particular animation keeps the impact site at the center of the frame, and the frame-size is kept at the instrument's maximum field of view -- resulting in a smoother animation sequence.

Summary

So in brief, the impact cratering process looks like this:

Coupling Stage
- impactor contact with target
- compressive shock-wave expansion in impactor & target
- reflected, rarefaction-wave expansion in impactor & target
Excavation Stage
- hydrodynamic flow-field establishment
- vapor plume formation & expulsion
- crater growth & solid ejecta plume formation
Modification Stage
- solid ejecta plume expansion & fall-out
- crater gravitational collaspse to final form

Not quite as easy as just digging a hole in the ground with a shovel, eh? I'll admit that impact cratering is a multi-staged and complicated event, but getting a good feel for this process will help tremondously in following our future discussions here, and showing how they fit into the overall, big picture. So, bear with me for now, and if you need to, come on back to this post for reference.

I feel the need, the need for speed

2009-11-29T12:30:00.010-05:00

One common feature of nearly all impact craters is that they are essentially circular in shape. But why is that? Every kid who has ever plunked stones into soft mud realizes that in order to get a round "crater," you have to throw your stone down from directly overhead -- if you hit the target at an angle, you get an oblong, or elliptical "crater." This same fact was observed with artillery back when cannons were still in use: most landing balls produced oblong gouges or holes in the ground -- not circular ones. And yet, when we look at the Lunar surface through a telescope, we sees circular, or nearly circular craters, despite the fact that very few impacts on the moon came in from directly overhead:

Figure 1: A telescopic view of the Lunar Highlands. These craters appear slightly elongated due to a tilt in the surface relative to the cameraa. Image courtesy of Lunar Photo of the Day (LPOD).

The answer lies in the velocity, or speed at which such impact occur. During the debate surrounding the origin of the Barringer meteor crater which took place in the early part of the 20th century, geologist Herman LeRoy Fairchild and astrophysicist Ray F. Moulton were two of the first scientists to recognize that objects traveling at "cosmic velocities" possess specific kinetic energies (energy per unit mass) which exceed those of conventional chemical explosives (which possess potential chemical energy). As such, impact craters are roughly equivalent to explosion craters; that is, such impacts result in the very rapid release of a large amount of energy within a very small space just below the target's surface. Due to this nearly instantaneous, "point-source" energy release, impact craters are circular regardless of impact entry angle (unless the angle is extremely low). To gain an idea as to the levels of specific energy (in Joules per kilogram) involved, and the velocities (in kilometers per second) which give an equivalent level of kinetic energy, take a look at the following list:

Energy Source	Specific Energy	Equiv. Velocity
Chemical: TNT	4.610 MJ/kg	3.036 km/sec
Chemical: Dynamite	7.500 MJ/kg	3.873 km/sec
Kinetic: Earth escape velocity	62.72 MJ/kg	11.2 km/sec
Kinetic: typical Lunar impact speed	200.0 MJ/kg	20.0 km/sec
Kinetic: mean meteor speed (Earth)	450.0 MJ/kg	30.0 km/sec
Kinetic: fastest meteor shower (Earth)	2.592 GJ/kg	72.0 km/sec
Fission: Natural Uranium (0.7% U-235)	443.0 GJ/kg	--
Fission: Bomb Uranium (100% U-235)	88.25 TJ/kg	--
Fusion: Bomb/Sun Hydrogen	645.0 TJ/kg	--

As you can see, inter-planetary impactors in the vicinity of the Earth (and also throughout our solar system) possess roughly 10-100 times the specific energy of conventional explosives when their high impact velocities are taken into account. These are called hyper-velocity impacts, which occur when the impactor is traveling at super-sonic speeds as compared to the speed of sound within the target material (a rocky target has a sound-wave speed of about 2-4 km/sec). While specific energy gives us an idea as to the efficiency of a particular crater-producing energy source, what determines the final outcome is the total amount of energy released, or the energy yield, which is determined by multiplying the specific energy by the mass of the bomb, in the case of explosives, or the mass of the incoming projectile, in the case of an impact.

By way of example, let's look at a couple of man-made craters, beginning with the Schooner crater within Area 20 of the Nevada Test Site. Schooner crater was formed by the detonation on 8 December 1968 of a nuclear fission bomb having a yield of 172 TJ, equivalent to 41 kt (kilotons) of TNT. The bomb was buried 110 m (350 ft) deep in a shaft drilled into rock obstensibly for the purpose of producing an explosion crater for study as part of Operation Plowshare. The result was a crater 260 meters (850 ft) in diameter and 63 meters (210 ft) deep, which was later visited by the Apollo 14 & 16 astronauts as part of their training program, including "lunar rover" testing.

Figure 2: The Schooner nuclear explosion crater, Area 20, NV

Upping the ante a notch, the largest man-made crater is the Sedan crater within Area 10 of the Nevada Test Site. Sedan crater was formed by the detonation on 6 July 1962 of a 30% fission / 70% fusion nuclear bomb having a yield of 435 TJ, equivalent to 104 kt of TNT. The bomb was lowered into a shaft drilled into the desert alluvium 194 m (640 ft) deep obstensibly for the purpose of investigating the peaceful use of nuclear weapons for ground excavation, as part of Operation Plowshare. The resulting crater is 100 m (330 ft) deep with a diameter of about 390 m (1,300 ft).

Figure 3: The Sedan nuclear explosion crater, Area 10, NV. Note the small obversing cage on the rim of the crater, shown on the lower right portion of the lip in this image.

Pretty impressive, huh? Now let's take another look at the Barringer meteor crater, which was formed by the impact of a roughly 50 m diameter iron meteorite striking the ground at about 15 km/sec. That's a kinetic energy yield of 5.8 x 10^16 Joules, equivalent to 14 Mt of TNT, or roughly the equivalent of 1000 Hiroshima-sized nuclear bombs -- placing this event on the same order as the largest thermonuclear (fusion) bomb tests conducted by humanity. The result is an impact crater which is now (after about 50,000 years of erosion) about 1,200 m (4,000 ft) in diameter, some 170 m deep (570 ft), and is surrounded by a rim that rises 45 m (150 ft) above the surrounding plains. Keep in mind that the Barringer crater is tiny, compared to the size of the other known impact structures on the Earth.

Figure 4: The Barringer meteor crater, Flagstaff, AZ. The parking lot and visitor's center are shown on the left. Source: http://www.flickr.com/photos/walterarce/

On the truly mind-boggling scale of things, the Chicxulub impact structure was produced by the impact of a roughly 10 kilometer (6 mi.) diameter stony asteroid about 65 million years ago, forming a basin which is more than 180 kilometers in diameter. The impact is estimated to have released 4 x 10^23 Joules of energy, equivalent to roughly 100,000,000 Mt of TNT. This is 2 million times more powerful than the largest nuclear fusion bomb ever detonated by humanity, the 50 Mt Tsar Bomba, set off by the Soviet Union on October 30, 1961.

In each of the above two man-made cratering events, you may have noticed that the nuclear device was buried in order to produce a large excavated mass of material. If the device is placed directly on the surface, only a shallow crater will result, with most of the energy dissapated in the form of an air-blast. In the classic example, the Trinity nuclear test, conduced on 16 July 1945, denotated a plutonium fission bomb having a yield of 84 TJ, equivalent to 20 kt of TNT. This event produced only a shallow crater, as the picture of "ground zero" below shows:

Figure 5: The center of the shallow depression produced by the Trinity nuclear test, Alamogordo Bombing and Gunnery Range, NM

Thus, in order to produce a cratering event, the energy source must be buried below the surface somehow, such that the energy released is primarily transmitted to the target soil or rock, and not the air or vacuum above.

A fun place to numerically play around with the size of craters that can be produced by either explosions or impacts is Keith Holsapple's "Crater Sizes from Explosions or Impacts" website, which lets you input s wide variety of starting conditions and then quickly computes parameters for the resulting crater, based upon experimentally and analytically derived formulae (which we'll talk about in some later post).

One final note: despite the huge amounts of energy released in the impacts which produce visible craters, and in light of my previous comparisons to man-made nuclear test craters, it should be pointed out that the comet and asteroid stock which make up the bombarding material for impact craters are NON-radioactive, and thus the craters they produce are likewise NON-radioactive (Superman's meteorite notwithstanding). Hollywood may have fun with green-glowing lumps of rock or mutated creatures/humans emerging from impact sites, but such stories are pure fantasy, and simply a handy writer's "vehicle" for creating said mutated creatures/humans. The effects of large impacts can be devestating enough, without resorting to radiation, but that's a different blog entry for a later time.

Craters, craters everywhere

2009-11-26T16:07:00.005-05:00

Impact cratering is perhaps the most common geologic process on the surface of solid bodies in our solar system. And yet, looking around on the surface of our familiar home, Earth, one would be hard pressed to recognize this fact. A few easily recognizable impact sites do exist, such as the 1.2 kilometer diameter Barringer crater (also called Meteor Crater) near Flagstaff, Arizona:

Figure 1: The Barringer meteor crater, near Flagstaff, AZ

but plate tectonic activity (earthquakes), volcanism, oceans and other bodies of water, and the erosional effects of wind, rain, and ice all combine to quickly "cover the evidence" whenever the Earth takes a hit from some wandering bit of asteroid or comet. A classic example of this is the picturesque, 26 kilometer (16 mile) wide circular "valley" containing the German village of Nördlingen:

Figure 2: The Ries impact crater, Nördlingen, Germany

which is, in fact, the Ries impact crater, formed about 14 million years ago during the Miocene epoch. The present floor of the crater is only about 100–150 meters below the eroded remains of the crater rim. As stated in this Wikipedia article: "it was originally assumed that the Ries [crater] was of volcanic origin. In 1960 two American scientists, Eugene Shoemaker and Edward Chao, proved that the depression was caused by meteorite impact. The key evidence was the presence of coesite (shocked quartz), which, in natural unmetamorphosed rocks can only be formed by the shock pressures associated with meteorite impact. The coesite was found in the building stone (suevite) of the Nördlingen town church, constructed from locally derived stone." Across the surface of the Earth, only a couple hundred impact craters / structures have been identified, ranging in size from the (relatively) tiny Barringer meteor crater to some a few hundred kilometers in diameter, including the well-known Chicxulub impact structure, which formed about 65 million years ago: coincident with the worldwide mass-extinction of about 95% of all animal- and plant-life, including the dinosaurs.

On airless bodies without atmospherically-driven weathering processes, active tectonics, or other resurfacing mechanisms, the scars left behind by impacts can last for billions of years, and will, in fact, begin to overlap so heavily that the primary way in which old impact craters are erased is by new impact craters. The Earth's moon has many regions which display this condition, called crater saturation, and examples can be found throughout our solar system, such as in this 2006 Cassini spacecraft image of Saturn's icy moon, Rhea, which is 1528 kilometers in diameter:

Figure 3: The battered surface of Saturn's satellite, Rhea, imaged by the Cassini spacecraft in 2006

Even the smallest objects in the solar system are not immune to the effects of impacts by still smaller particles. It simply becomes a matter of relative size as to whether an object is the impact-or or the impact-ee. The below image depicts the tiny asteroid Dactyl, which is 1-1.5 kilometers in diameter -- about as wide as the Barriage Crater depicted above -- sporting a few impact craters of its own. Dactyl is a satellite of asteroid 243 Ida, which was imaged by the Galileo spacecraft in a 1993 flyby:

Figure 4: The tiny asteroid Dactyl, a satellite of asteroid 243 Ida, imaged by the Galileo spacecraft in 1993

Over the past fifty years or so, our understanding of these most ubiquitous features in the solar system has increased dramatically: ranging over such topics as determining what type or size of impactor produces a particular crater; to the mechanics of how a crater is formed; to what effects, beyond the obvious crater, an impact has on surrounding terrain (and sometimes on a global scale); to how repeated bombardment over time changes the nature of a given target body. It is my desire to share some of these findings with you, and hopefully some findings of my own, over the course of this blogging effort. You are most welcome to join me on this journey!