GEOL 331 Principles of Paleontology

Fall 2016

Biostratigraphy II: Limits and prospects

Reasons for caution

Biostratigraphy opened the door to global correlation of strata, but is, nevertheless subject to biases and filters that make it more reliable on a local scale than a global one.

Biases of the Rock Record

• There are no perfect index fossils.
  • Even good ones are subject to some substrate/facies constraints. Thus, all fossils are, to some degree, facies fossils. An ammonoid, for example might have a global marine distribution, but will still never turn up in terrestrial deposits.

  • Species' geographic ranges change over time. Thus, many critters' records Many contain some biogeographic signal, such that local FADs and LADs may record immigration and extirpation. In effect, critters, as well as rock units, can be diachronous.
  
  Henodus chelyops from Eberhard Karls Unversität Tübingen
• The rock record is inconsistent and no rock unit preserves ancient environments on a global scale. Thus:
  • changes in depositional rate
  • depositional hiatuses
  • local facies changes
  impose their own non-biological signal. Consider:
  • Abrupt "special-creation-like" appearances. E.G. the placodont Henodus (right). (Compare to a normal placodont.)

  Lesson - Evolution (without which biostratigraphy would be impossible) can be difficult to interpret.

Mass Extinctions: Abrupt simultaneous extinctions of numerous taxa, if real, are powerful biostratigaphic markers and are interesting ecological phenomena in their own right. Alas, identifying them isn't straightforward. It requires continuous deposition and a reasonably dense fossil record.

• Unconformities create the impression of an abrupt extinction event when in truth, a gradual turnover is occurring. Unconformities are more common than true mass extinctions. Consequently, they are the most likely cause of abrupt simultaneous disappearances.
  

  

• A sparse sample introduces statistical uncertainty into an otherwise good depositional record. Consider the Signor-Lipps effect in which a simultaneous mass extinction is made to appear gradual by random sampling from a poor record.
  

• Confidence intervals: Until now we have concentrated on FADs and LADs, but actually every horizon in which a given taxon occurs is a datum that can be used to contrain its confidence interval statistically. A very sparse record yields wide 95% confidence intervals above and below observed FADs and LADs. A dense record yields narrow 95% confidence intervals. In a case of a single occurrance, the confidence interval is infinite.
  
  The coelacanth Latimeria from Smithsonian USNM
• Lazarus Taxa: Taxa that temporarily "disappear" and then reappear in fossil record. This might be because of environmental changes, or local extirpation and re-immigration. E.g. North American horses, coelacanths (right).
  
  
  Miocene tiger shark fossil in Holocene sediments from Polar Trec
• Zombie effect: Post-extinction reworking of specimens into younger sediments (right). As in Fasstt et al., 2000, these can be significantly misleading.
  

Bias of the Organisms

Flabellum impensum, a modern scleractinian coral from Smithsonian USNM
• Elvis Taxa: Taxa that converge on extinct forms, giving false impression of Lazarus taxa. (A particularly common problem with planktonic forams, whose morphology is strongly biomechanically constrained. Also reef forming organisms, consider Cambrian archeocyathid sponges, late Paleozoic rugose corals and Cretaceous rudistid clams.)

Good News and Bad:

The bad news: Constraints on biostratigraphy

• In a biological sense, FADs are arbitrary. Assuming that evolution is gradual, (no consensus on that!) how do we identify the first occurrance of a new species? Indeed, how do we distinguish an evolving lineage from a branching phylogeny (right)?
  
  
  Hipparion gracile from Wikipedia
• Likewise, how do we identify true global FADs from immigrations?
  • Migrating species are highly time transgressive. Also, the immigration of a taxon into a region is more nearly instantaneous than a taxon's evolution in situ.
  • Even so, we are constrained by the chronological resolution of the rock record. For example, the Miocene Hipparion event - the immigration into western Eurasia of three-toed horses ended up being two widely separate events.
  
  
  
  Eofallotaspis, a Cambrian index taxon
• The most useful index taxa, although varying with geologic time, have certain characteristics that give them global, but not absolutely universal applicibility. Absolutely no taxon lives in every possible Earth environment. We rely most heavily on:
  • Cenozoic: planktonic microorganisms, especially foraminiferans
  • Mesozoic: Ammonoids predominate
  • Late Paleozoic: Ammonoids and conodonts
  • Ordovician - Devonian: Conodonts and graptolites
  • Cambrian - Ordovician: Trilobites

The good news:

• Comparisons with abiotic criteria such as magnetostratigraphy or stable isotope ratios suggest that planktonic organisms in the marine realm, at least, are reasonably reliable.

• Species may be time transgressive but assemblages are typically not. In fact, the co-occurance of different taxa is strongly controlled by global climatic factors and geographic factors. E.G. discordant faunas of the Pleistocene. For example, in the modern world, the ranges of:
  • Blanding's turtle
  • the wood turtle
  • the musk turtle
  Barely overlap or don't overlap at all. Phillips, 2006, however, showed that during the last Pleistocene glacial, their ranges overlapped in northern Alabama and Mississippi. Thus, changing temperature was only one environmental factor explaining changing species ranges with the last deglaciation, and the Pleistocene glacial maximum and the Holocene present regional faunas that are distinct from one another, even when the species that make them up are the same.

Biostratigraphic Conventions:

Mindful of these considerations, we see why biostratigraphers employ a variety of zone definitions despite their invocation of conjecture and assumptions: In many circumstances, the ability to bring more data to bear on a problem is simply more important than the avoidance of the fuzziness that follows from inference and conjecture. The biostratigrapher seeks the optimal tradeoff for the specific situation.

Lower boundary of Cambrian Global Standard Stratotype and Point from America Pink

Biostratigraphic nomenclature and golden spikes:

Biostratigraphy is the principal determinant of such important things as period boundaries. Boundaries between periods are arbitrarily decided, but usually involve biostratigraphic markers. Some conventions:

• Decisions are made by a committee of the International Union of Geological Sciences (IUGS)
• Typically, they are based on the first appearance of a diagnostic taxon but not the last - i.e. they are "topless" boundaries.
• The boundary is identified with reference to an agreed upon location somewhere in the world called the Global Standard Stratotype and Point (GSSP.) This is physically marked by a spike driven into the rock. (Referred to as golden spike, but not really gold. Sorry, treasure-hunters.)

E.G. the Silurian-Devonian boundary (right) was decided to be the base of the Monograptus uniformis (a graptolite) zone. Over forty such golden spikes have been established.

How does this help us with the stratigraphy of environments in which fossil preservation is spotty and intermittent?

Biochronology:

When biostratigraphic data is combined with numeric age information we can use biozones as the basis for biochrons - time units! (as opposed to biozones which are rock units)).

Rancho La Brea during the Rancholabrean NALMA from The Page Museum

One famous application: Land Vertebrate Ages. Originally developed during the early and mid-20th century as North American Land Mammal Ages (NALMAs), these were based on the recognition of the regular co-occurrance of distinct mammalian assemblages. NALMAs answered the need for some kind of biochronological tool for North American Cenozoic terrestrial rock units, but were limited by the fact that Cenozoic mammal fossils tended to occur in isolated pockets, often in places with little precise stratigraphic context. (Cave deposits, for example.) As a result, they could not be used as the basis for proper rigorous biozones based on the rocks in which they were found. Instead, they are tied to the time intervals in which the assemblages are inferred to have lived. For regions like the continental deposits of North America, however, NALMAs are much better than nothing, and more frequently mentioned than proper biozones. The NALMAs of the Pleistocene Epoch, for example are good examples and often cited. E.G.:

• Rancholabrean: Named after the Rancho La Brea locality in Los Angeles, CA, characterized by the presence of Bison in association with characteristic Pleistocene forms like Mammuthus. 0.024 - 0.011 ma.

Over a century of development, competing criteria were used in definitions of ages. Today, biostratigraphers must:

• Formally resolve contradictions that arise as new information becomes available. E.G: The Chadronian was originally defined by the last appearance of titanotheres and the top of the Chadron formation. Alas, titanotheres are now known from above the Chadronian.
• Address the circular reasoning that can result from not being able to tie the ages to well defined stratotypes. (One 1998 study linked a Early Triassic vertebrate biochron with the presence of a primitive ichthyosaur thought to characterize the biochron, however the rocks containing the ichthyosaur had only been deemed Early Triassic because of the presence of a primitive ichthyosaur.)

Nevertheless, as a quick-and-dirty, NALMAs/Land Vertebrate Ages are sufficiently useful that they have been expanded globally and pushed confidently into the Late Cretaceous as Land Vertebrate Ages. But note: The use of Land Vertebrate Ages is a tradition among North American paleontologists and stratigraphers originally born of necessity but now, perhaps, just a tradition. Prothero, 1995 reviews the application of new data from improved radiometric dating and magnetostratigraphy, and concludes that the time is ripe for incorporation of NALMAs into proper biozone-based biostratigraphy.

All rest on the assumption that biostratigraphic units are good proxies for time. As indicated above, this seems to be a good first-order approximation.

Final thoughts

• Whatever their limitations, biozones are very useful stratigraphic and chronological markers.

• Unlike radiometric dating methods, biozones don't lose precision or resolution with increasing age. In this way, they resemble magnetostratigraphic or chemostratigraphic zones.

• Can be used in conjunction with other dating techniques. See magnetobiochronology.

• Can provide quantitative information about rates of deposition using the methods of quantitative biostratigraphy.

Quantitative Biostratigraphy

Besides hopefully constraining their age and sequence, does biostratigraphy add to our kowledge of the deposition of sediments? Actually, yes.

Graphic correlation: method for stratigraphic correlation based on statistical correlation of first and last appearances, but not biozone terminology. Facilitates comparison of locality sections containing local FADs and LADs of the same taxa. Used to:

• Identify errors and outliers.Consider the following data:

  Taxon	Section X FAD	Section X LAD	Section Y FAD	Section Y LAD
  A	0	3	0	2
  B	0	6	0	4
  C	1.5	12	1	6
  D	4.5	6	3	8
  E	4.5	10.5	3	7
  F	7.5	10.5	5	7
  G	9.75	13.5	6.5	9
  H	10.5	15	7	10
  I	10.5	15	8	10
  J	12	15	8	10

  
  

• Characterize differences in depositional rate: Consider the following data:

  Taxon	Section X FAD	Section X LAD	Section Y FAD	Section Y LAD
  A	0	1	0	2
  B	1	3	2	6
  C	3	4	6	8
  D	5	6	10	12
  E	5	8	10	13
  F	7	9	12.5	13.5
  G	10	13	14	15.5
  H	11	12	14.5	15
  I	12	14	15	16
  J	14	18	16	18

  
  

• Identify depositional hiatuses. Consider the following data:

  Taxon	Section X FAD	Section X LAD	Section Y FAD	Section Y LAD
  A	0	2	0	2
  B	1	4	1	4
  C	2	5	2	7
  D	3	5	3	8
  E	4	5	4	9
  F	5	6	7	10
  G	5	9	8	12
  H	9	11
  I	10	14	12	14
  J	13	16	13	18

  
  

Composite standards: The examples above correlate teilzones from pairs of localities. On a larger scale, the data used to achieve this can be combined into substantial composite standard databases that:

• Represent the summation of large volumes of regional data. (Thus, they are neither teilzones nor taxon range zones.)
• can also incorporate numerical age data from reliable marker beds.

Comparison of information from a particular locality with the composite standard facilitates robust identification of range extensions, hiatuses, and outliers.

Moral: No method of biostratigraphy lacks significant biases and limitations, however biostratigraphy compensates by enabling stratigraphers to bring an overwhelming quantity of data to bear, making it cumulatively very powerful.