Wednesday, April 22, 2015

On Rock Classification

Two interesting articles:

1) In the Journal of Sedimentary Research (behind paywall) Kitty Milliken proposes a tripartite classification of fine grained sedimentary rocks, those with grain assemblages with greater than 50% of particles by weight or volume less than 62.5 µm (4 Phi). There are a number of names for these types of rock; mudstone, claystone, pelite, argillite to name a few. This classification categorizes the rocks according to the composition, thereby indicating the source of the grains. Composition in turn controls to a large measure bulk rock properties upon burial and interaction with fluids, thus enabling general predictions about their economic and engineering qualities.


A tripartite compositional classification is proposed for sediments and sedimentary rocks that have grain assemblages with greater than 50 percent of a weight or volume of particles smaller than 62.5 µm (4 Phi). Tarl (terrigenous–argillaceous) contains a grain assemblage dominated by more than 75 percent of particles of extrabasinal derivation, including grains derived from continental weathering and also volcanogenic debris. Carl (calcareous–argillaceous) contains less than 75 percent of particles of extrabasinal derivation debris and among its intrabasinal grains contains a preponderance of biogenic carbonate particles including carbonate aggregates. Sarl (siliceous–argillaceous) contains less than 75 percent of particles of extrabasinal derivation and contains a preponderance of biogenic siliceous particles over carbonate grains.

These three classes of fine-grained particulate sediments and rocks effectively separate materials that have distinct depositional settings and systematic contrasts in organic-matter content and minor grain types. In the subsurface the grain assemblages that define these classes follow contrasting and predictable diagenetic pathways that have significant implications for the evolution of bulk rock properties, and thus, assigning a fine-grained rock to one of these classes is an important first step for predicting its economic and engineering qualities. For purposes of description these three class names can be joined to modifier terms denoting rock texture, more precise compositional divisions, specific grain types of notable importance, and diagenetic features. 

2) In Earth Magazine, a delightful article (open access) titled Geologic Column: The Rumpelstiltskin Factor by Ward Chesworth, professor emeritus at the University of Guelph, Canada.

Chesworth muses on the importance of naming objects and whether it is better to be a "lumper" or a "splitter" i.e. whether it is better to organize variation in to as few groups as possible or whether it is better to draw finer and finer distinctions and place a smaller range of variation into its own distinct cubicle.

An excerpt:

Excessive splitting can lead to problems, though. If an overly meticulous taxonomist kept on splitting hairs ad absurdum, we would wind up with a classification resembling an advanced case of logorrhea, the kind of thing guaranteed to drive working geologists to the brink. C.B. Hunt staged his own rebellion against this tendency when he considered the plethora of names invented for minor igneous intrusions. He expressed his displeasure by sarcastically concocting one more, cactolith, which he described as “a quasi-​horizontal chronolith composed of anastomosing ductoliths whose distal ends curl like a harpolith, thin like a sphenolith, or bulge discordantly like an atmolith or ethmolith.” He insinuated it into his 1953 U.S. Geological Survey professional paper on the Henry Mountains of Utah, and from there it crept under the radar into the first edition of AGI’s very own “Glossary of Geology.” Unfortunately, some humorless jobsworth banned it from all subsequent editions.

Sprinkled with more anecdotes, this is a fun read.

Sunday, April 19, 2015

Scientists At Work- The Latest On Dog Domestication

Over at Science Magazine David Grimm has written a very informative article on the status of research on dog domestication. Its about the techniques being brought to bear on the question of the place and timing of dog origins and also about the scientists involved in the research, their pet theories and the conflicts within the field.

An excerpt:

Hulme-Beaman has spent the past 6 months traveling the world in search of ancient dog bones like this one. He's found plenty in this Ohio State University archaeology laboratory. Amid boxes stacked high with Native American artifacts, rows of plastic containers filled with primate teeth, and a hodgepodge of microscopes, calipers, and research papers, a few shoe and cigar boxes hold the jigsaw pieces of a dozen canines: skulls, femurs, mandibles, and vertebrae.

It's all a bit of a jumble, which seems appropriate for a field that's a bit of a mess itself. Dogs were the very first thing humans domesticated—before any plant, before any other animal. Yet despite decades of study, researchers are still fighting over where and when wolves became humans' loyal companions. “It's very competitive and contentious,” says Jean-Denis Vigne, a zooarchaeologist at the National Museum of Natural History in Paris, who notes that dogs could shed light on human prehistory and the very nature of domestication. “It's an animal so deeply and strongly connected to our history that everyone wants to know.”

And soon everyone just might. In an unprecedented truce brokered by two scientists from outside the dog wars, the various factions have started working together. With the help of Hulme-Beaman and others, they're sharing samples, analyzing thousands of bones, and trying to set aside years of bad blood and bruised egos. If the effort succeeds, the former competitors will uncover the history of man's oldest friend—and solve one of the greatest mysteries of domestication.

The main points of contention:

a) did dogs originate in China around 15,000 years ago? this is based on the greater genetic diversity of East Asian breeds suggesting earlier origins than elsewhere, a notion that is opposed by some who say that diversity could reflect later migration.

b) did dogs originate in Europe/ West Asia? between 19,000 and 32,000 years ..  this is based on DNA comparisons of ancient and modern dogs and wolves from Europa and the America's which suggest that modern dogs are most similar to a now extinct population of wolves from Europe. This finding is criticized for its lack of samples of ancient dogs from China (there aren't any found as of yet) and for possible confusion in identifying ancients wolf  from ancient dog remains.

c) do unusual skeletons having wolf/dog mixed features as old as 30,000 from Belgium and Siberia represent ancestors of living dogs or extinct populations indicating failed attempts at domestication? ...  do these samples represent dogs at all or are they just strange looking wolves... a question that is now  being probed  used computer assisted morphological analysis.

there is plenty of think about..  read the full article here.

Friday, April 10, 2015

Do Marine Animal Lineages Evolve Toward Larger Body Size Over Time

aka Cope's Rule-

I like these big questions about the history of life and I am fascinated and very impressed when palaeontologists take up such questions. It is incredibly laborious and time consuming work, to go through archival data on fossils and often generate new data from museum specimens and older compilations describing fossil taxa.

A recent study in Science Magazine:

Cope’s rule in the evolution of marine animals - Noel A. Heim, Matthew L. Knope1, Ellen K. Schaal, Steve C. Wang, Jonathan L. Payne

Cope’s rule proposes that animal lineages evolve toward larger body size over time. To test this hypothesis across all marine animals, we compiled a data set of body sizes for 17,208 genera of marine animals spanning the past 542 million years. Mean biovolume across genera has increased by a factor of 150 since the Cambrian, whereas minimum biovolume has decreased by less than a factor of 10, and maximum biovolume has increased by more than a factor of 100,000. Neutral drift from a small initial value cannot explain this pattern. Instead, most of the size increase reflects differential diversification across classes, indicating that the pattern does not reflect a simple scaling-up of widespread and persistent selection for larger size within populations.

What that means is that the size increase is not due to a uniform increase across all animal groups. Rather, groups that were larger very early in animal evolution have diversified disproportionally more than smaller sized groups. Why should that happen? The authors suggest that there may be advantages to being larger, such as, ability to move faster, to capture larger prey and to burrow deeper for protection and exploiting additional food resources.

That would seem to make larger animals more resilient to background extinction and make larger sized lineages longer lived. But why would that make larger animals more speciose? i.e. why would larger sized animals species split into more new species than smaller sized animal species? ..because that is what is the claim, that throughout the history of animal evolution larger sized species gave rise to more new species than smaller sized ones (differential diversification). In fact, one could make arguments favoring higher rates of speciation in smaller sized organisms, such as, their ability to disperse over greater geographic area resulting in greater chances of populations getting reproductively isolated resulting in new species, their ability to survive better during environmental crises (survivor fauna after mass extinctions tend to be smaller bodied, mass extinctions seems to kill of larger bodied species disproportionately). If mass extinctions differentially kill off larger bodied species, then is the observed trend really a series of trends, each reset at the aftermath of the crises, resulting in small pioneer /survivor fauna evolving towards larger size. There could be a physical limit to how small one could become and the only direction for size to vary (either through drift or natural selection) would be towards a larger size. I am just speculating without even reading the paper, the authors do mention that drift from a small initial size does not explain their findings, but it would interesting to know what role mass extinctions might be playing in disruption or amplifying trends.

So although a trend is apparent, the answers are not all clear cut. It would also be interesting to group the trends according to life habits, i.e. planktonic versus benthic, sessile versus mobile  and see if any of these life styles particularly favors evolution towards larger size.

Eurekalert has a summary of the study

Monday, April 6, 2015

Hyperspectral Remote Sensing- Open Access Papers In Current Science

Current Science has a special section (open access) on Hyperspectral  Remote Sensing with papers on applications in geology, soil mapping, glacial dynamics, forestry and planetary sciences. Hyperspectral Remote Sensing involves measuring the energy from visible and infrared spectrum at very narrow intervals or channels. For example, the Hyperion sensor collects spectral information in 220 spectral bands from between the 0.4 to 2.5 µm (micrometer) bandwidth with a 30-meter ground resolution. This means scientists can use this data to identify surface objects with subtly different spectral properties. I've written about the applications of hyperspectral remote sensing in geology in an earlier post on the use of this technique in Afghanistan.

Here is a short excerpt from that post on the utility of hyperspectral imaging for mineral identification:

For the map of Afghanistan, the USGS used a sensor known as HyMap imaging spectrometer loaded aboard an aircraft. It collected spectral data covering 128 bands of 15-20 nm  (nanometer) bandwidth in the 0.4 to 2.5 µm (micrometer) range i.e the visible and the near infrared spectrum with a 5 meter ground resolution. Minerals have distinctive absorption signatures,  meaning that when sunlight strikes the surface of the earth the O-H or C-O3 or Si-O2 or Fe-OH bonds in the mineral absorb energy at a distinct wavelength, each covering a very narrow portion of the spectrum . Because conventional multispectral sensing collects energy averaged over a  broad interval it cannot discriminate between individual minerals. Hyperspectral sensing is fine grained enough (15-20 nm bandwidth) to be able to resolve the distinctive absoption signatures of several minerals.

here is the list of papers in Current Science-

Advances in spaceborne hyperspectral imaging systems
Hyperspectral image processing and analysis
Algorithms to improve spectral discrimination from Indian hyperspectral sensors data
Hyperspectral remote sensing of agriculture
Hyperspectral remote sensing: opportunities, status and challenges for rapid soil assessment in India
Monitoring of forest cover in India: imaging spectroscopy perspective
Hyperspectral remote sensing and geological applications
Snow and glacier investigations using hyperspectral data in the Himalaya
Simulating the effects of inelastic scattering on upwelling radiance in coastal and inland waters: implications for hyperspectral remote sensing
Hyperspectral remote sensing of planetary surfaces: an insight into composition of inner planets and small bodies in the solar system