Richard Binzel from MIT is the lead author of a new Nature paper titled Earth encounters as the origin of fresh surfaces on near-Earth asteroids (abstract only). They address a long-standing problem in meteoritics: why does the color of meteorites found on Earth so rarely match that of asteroids in the Main Belt? Binzel and his colleagues demonstrate that it is exactly the proximity to Earth that causes changes in the color of the meteorite parent bodies – a tidy solution to a puzzle, and it bears on a few of the most important processes on planetary bodies and the tools we use to interrogate them.
First, some background. The surface composition of a planetary body determines the color of the visible, infrared, and ultraviolet light it reflects. We observe this light, with ground telescopes or spacecraft instruments, measure and quantify it using reflectance spectrometry, and try to deconvolve the spectra to pull out compositional information*. Of course, our solar system is a dynamic place, and the color of a surface changes with time from exposure to solar and cosmic radiation and meteorite impacts (both large and microscopic). This is known as space weathering**, or maturation of the surface. Mature planetary surfaces look darker to the naked eye, and this is referred to as reddening because of the shift in dominance to the larger wavelength light in the infrared.
Consider our Moon. There are two crustal domains visible to the naked eye: the highlands are light due to the abundance of plagioclase feldspar and the mare are largely dark basalt. These are primary compositional differences. (When you examine broader spectra including nonvisible light, much more compositional variation is evident.) However, within a given lunar crustal domain, the craters and the rays that extend from them will be lighter in color. This is because the impacts excavate and expose the less mature, and thus lighter and less red, regolith. Understanding how mineralogical and chemical composition controls reflectance and how to distinguish that from the superimposed effects of maturation is a major undertaking in planetary science.
The Earth’s Moon. The craters and the rays that radiate from them are lighter than their host regions because they excavate relatively fresh material.
Over 80% of the meteorites we’ve collected on Earth are ordinary chondrites (which are more-or-less primitive samples of the early solar system). There are some relatively unweathered asteroids that have similar spectral characteristics to ordinary chondrite meteorites, and they are known as Q-types. However, Q-type asteroids are so far undetected in the Main Belt and compose only a small minority of the near-Earth asteroids. This is the “ordinary chondrite problem” that Binzel et alia address. It’s important to distinguish that “ordinary chondrite” is a petrological classification based on mineralogy, texture, and bulk chemistry – properties we can only measure on meteorites in hand, while Q-type is a classification based on the spectral properties which we can measure on asteroids and meteorites. In fact, the NEAR Shoemaker mission landed on 433 Eros, a non-Q-type near-Earth asteroid, in 1998 and found the bulk chemistry to be similar or identical to ordinary chondrite meteorites. This raised the possibility that it is only the unweathered nature of meteorites that is underrepresented in the solar system, and not the primary composition.
The relative reflectance spectra of ordinary chondrite meteorites, Q-type asteroids (which are relatively unweathered) and increasingly weathered Sq- and S-type asteroids. The more weathered asteroids have spectra more reflective in the infrared (above 0.7 µm in wavelength). The dips in the spectra at 1 µm and 2 µm are due to the absorption of Fe-bearing minerals olivine and pyroxene, respectively. These two absorption bands become more subdued with increasing weathering/maturity. From Binzel et al. (2010), Figure 1.
Binzel and his colleagues tested the hypothesis that near-Earth encounters cause the partial or complete resurfacing of those asteroids which become or shed meteorites. They examined 95 asteroids whose paths cross that of Earth or Mars, and limited their sample group to asteroids of diameter 0.2 to 4 kilometers. They had spectral data for each asteroid, and 20 of the 95 were Q-type asteroids. They also had orbital data, with which they calculated the Mean Orbit Intersection Distances (MOIDs) – a measure of the potential closeness between each asteroid and Earth. Asteroidal orbits can be perturbed by the gravity of any of the eight planets, so the authors used planetary orbits as input and modeled MOID for each asteroid for the last 500,000 years, with an output for every 50 years of model time. It’s known from lunar and asteroidal studies that the processes of maturation take less than one million years – a rapid pace, geologically speaking, but a longer period than the modeled time span. They then examined the closest calculated MOID for each of the 95 asteroids.
They found that 75 of the 95 objects could have come within one Earth-Moon distance in the last half million years. All 20 of the Q-type asteroids fell in this group and, statistically, there is only a 0.9% chance of this occurring randomly. Recall that the MOID calculates possible closest distances so that a low MOID doesn’t mean that the object actually came within that distance; thus, it isn’t necessarily problematic that 55 of the modeled orbits had low MOIDS but weren’t Q-type asteroids. So the correlation of fresh surfaced asteroids with proximity to Earth is robust at 99.1%.
No one has yet done the detailed calculations, but the proposed mechanism is tidally induced seismicity on the asteroid. This seismicity is envisioned to shake and stir the surface regolith, exposing fresh material. Density measurements taken by NEAR Shoemaker and from the ground suggest that some asteroids may be as much as 50% empty space; this data has led to the rubble pile model for asteroids that suggests they are mechanically weak agglomerations held together by gravity. Such loose material would presumably be particularly susceptible to shaking and resettling. Furthermore, events as relatively small as landslides in asteroid craters have been observed to expose fresh regolith. Although there is insufficient data for 0.2-4 km Main Belt asteroids to compare directly to this near-Earth dataset, the lack of any detected Q-type Main Belt asteroid strongly suggests that proximity to a large body, i.e., a planet, is necessary to unredden (my word, not theirs) an asteroid.
The S-type asteroid 433 Eros as photographed by the NEAR Shoemaker probe (NASA-JPL).
*Iron has disproportionate effects on the spectra of geologic materials. Beyond space weathering effects**, the Mg/Fe of the mafic minerals also exhibit control over the subtleties of absorption bands. Lunar plagioclase contains, on average, more FeO than terrestrial plagioclase, though still generally less than 1.5 wt.%. This allows lunar plagioclase abundance to be more easily measured by spectral techniques than it otherwise would.
**The surfaces of planets with atmospheres are shielded from most meteorite impacts, and the surfaces of those with magnetospheres are protected from most effects of the solar wind. Among the primary effects of space weathering are (1) mixing and churning of regolith at a range of scales by meteorite impact; (2) implantation of solar wind ions, largely H+, into the particle surfaces; and (3) very localized melting, i.e., glass formation, on the scale of microns to millimeters by micrometeorite impact. The localized melting of (3) in the presence of implanted H+ from (2) results in glass with reduced iron spherules. This glass coats other particles with micron-thick rims. The metallic iron spherules the rims contain are often nanometers in size, and they contribute to the reddening of the mature regolith.
Binzel RP, Morbidelli A, Merouane S, Demeo FE, Birlan M, Vernazza P, Thomas CA, Rivkin AS, Bus SJ, & Tokunaga AT (2010). Earth encounters as the origin of fresh surfaces on near-Earth asteroids. Nature, 463 (7279), 331-4 PMID: 20090748
0 comments:
Post a Comment