<<

COMMENTARY

Ecological disruption precedes mass COMMENTARY Steven M. Hollanda,1

Mass are dramatic features of the record in which extinction risk is substantially elevated above background levels. Although extinction risk varies mark- edly over geologic time, as well as geographically, it was particularly elevated and global in extent during the so-called Big Five events: the Late , Late , end-, end-, and end-Creta- ceous (1). These events were originally recognized by variations in extinction rate in marine families, and their importance remains in analyses at the level and that account for variable preservation over geo- logic time (2, 3). Increasing attention has concentrated on understanding the ecological effects of mass extinc- tion and other lesser but still significant extinction epi- sodes (4–10). In PNAS, Sheets et al. (11) document the ecological changes in marine planktonic communities not only during, but preceding the Late Ordovician (447–444 Ma) mass extinction. Examining the ecological changes during a mass extinction would seem to be straightforward: go to a stratigraphic column spanning the mass extinction and describe the changing ecological composition of succes- sive sedimentary layers through the extinction episode. This would be a direct history of ecological changes related to the extinction if those layers recorded the same habitat through time, such as the distance from shore or water depth for marine benthic organisms, or the over- lying water masses for marine plankton. Unfortunately, this simple scenario is rarely the case, as numerous studies of Fig. 1. Effects of rising and falling sea level on the spread of mesopelagic water sedimentation over the past 40 y have shown (12). Pro- across the and on the neodymium isotopic ratio of seawater, used by Sheets et al. (11) as a proxy for water depth. Graptolites, extinct colonial cesses of sediment accumulation create two challenges macroplanktonic hemichordates with chitinous skeletons, lived in distinct for studying ancient mass extinctions (13). communities in depth-stratified epipelagic and mesopelagic marine waters. First,thewaterdepthatanygivenlocationvaries continuously over geological time, from the combined effects of changes in sea level, sediment accumulation, only ecological effects associated with the mass extinc- and the subsidence or uplift of the ’s surface (12). tion, but also those caused by the changing habitats that Because the distribution of benthic marine invertebrates can be sampled at that location. Ideally, one would track is primarily controlled by factors correlated with water a particular habitat through the extinction interval, but depth (13), these variations in water depth will cause limited rock exposure rarely makes that possible (14, changes in the composition of benthic marine commu- 15). Studying the effects of extinction on planktonic or- nities at any given location, even if the distribution and ganisms faces similar problems, because the water abundance of remains static within any particular masses in which these species live also move laterally habitat. Stratigraphic columns will therefore record not through time (Fig. 1).

aDepartment of Geology, University of Georgia, Athens, GA 30602-2501 Author contributions: S.M.H. wrote the paper. The author declares no conflict of interest. See companion article on page 8380. 1Email: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1608630113 PNAS | July 26, 2016 | vol. 113 | no. 30 | 8349–8351 Downloaded by guest on September 26, 2021 Second, the rates at which habitats move laterally over time Sheets et al. (11) find that the graptolite communities experienced vary markedly, as do sediment accumulation rates. As a result, the a series of changes not only during the extinction, but leading up to it. habitat preserved in the stratigraphic record at any given location Particularly in the mesopelagic zone, the authors document a can change abruptly and solely as a result of how sediment progressive shift toward communities with an increasingly uneven accumulates (12). Therefore, fossil communities can appear to abundance distribution: that is, dominated by a few species. In ad- change abruptly as a result of changes in the rates of subsidence, dition, Sheets et al. show that these communities were progressively sedimentation, and sea-level change (16). These stratigraphically invaded by a different suite of graptolites known as the Neograptina, controlled community changes can be difficult to distinguish from which were previously limited to cooler-water high-latitude settings. those caused by a mass extinction (13), particularly so if the mass This study (11) has several important implications, not only for extinction was caused by the same processes that control sedi- understanding how mass extinctions are recorded in the stratigraphic ment accumulation, in what is known as the “common cause hy- pothesis” (17, 18). Sheets et al. present direct evidence that the In PNAS, Sheets et al. (11) account for these effects in their Late Ordovician mass extinction was a study of planktonic graptolite (an extinct hemichordate) commu- nities leading up to and through the Late Ordovician mass extinc- prolonged event, one preceded by substantial tion. Their study focuses on two well-studied sections in the ecological degradation before a tipping point western United States, which was situated just south of the equa- was reached. tor during the Late Ordovician. Previous studies have shown that this extinction was associated with a sea-level fall of 70–100 m record, but more broadly for understanding the ecological dy- caused by geologically short-lived continental glaciation, a drop namics of mass extinction and the prelude to them. in sea-surface temperatures of ∼6 °C, increased delivery of oxy- Because sea level has changed continuously through geo- gen to the deep ocean, and reduced supply of nutrient-rich and logic time (19), the fossil record of most mass extinctions is denitrified water. Two aspects of the Sheets et al. (11) study are complicated by patterns of stratigraphic accumulation (13). In particularly novel. particular, the depositional response to sea-level change gen- First, Sheets et al. (11) developed an innovative method for erates abrupt local changes in fossil assemblages that reflect determining the water masses inhabited by each graptolite spe- only the lateral movement of habitats and prolonged pauses cies. Graptolites are inferred to have lived in two depth-stratified in sedimentation. When a mass extinction is caused by climate oceanic water masses: an epipelagic zone of well-lit and well- change that results in sea-level change, as in the Late Ordovi- mixed surface water and an underlying mesopelagic zone, with cian, a literal reading of local stratigraphic records can suggest the two water masses separated by a sharp increase in water not only that the mass extinction was more abrupt than it actu- density (Fig. 1). Upon death, graptolites settled to the seafloor, ally was, it can mislead as to the timing of the event (20). Nu- forming relatively shallow-water fossil assemblages containing merical simulations of sedimentation suggest that most ancient epipelagic species and progressively deeper-water assemblages mass extinctions were prolonged, lasting several hundred thou- bearing an increasing proportion of mesopelagic species in addi- sand , much longer than a literal reading of the strati- tion to epipelagic species. Based on Late Ordovician graptolite graphic record would suggest (13). Sheets et al. (11) present assemblages from around the world, Sheets et al. constructed a direct evidence that the Late Ordovician mass extinction was maximum-likelihood model of finding each species at a shallow- a prolonged event, one preceded by substantial ecological degra- water site and a deep-water site, then used a Markov chain-Monte dation before a tipping point was reached. Recognizing that at least Carlo search to find the posterior probability that each species some mass extinctions are protracted events will significantly guide occupied the epipelagic or mesopelagic zone. future considerations of the causes of mass extinction, the pro- Second, Sheets et al. (11) accounted for changes in water cesses by which species go extinct during them, and the selectivity depth over time at each site using neodymium isotopes. Times of mass extinction (21). of rising global sea level allow the spread of more radiogenic The recognition of ecological degradation leading up to mass 143Nd-bearing water from the ocean onto continental shelves, extinction has important implications for modern ecological change. raising the ratio of 143Nd to 144Nd. During low stands of global The decline of , resulting in lower evenness, coupled with sea level, more continental area is exposed to erosion, increas- widespread biotic invasions, is cause for concern even if the magnitude ing the supply of Nd with a smaller radiogenic component from of modern extinction has not reached the intensity of the Big Five mass land, thereby lowering the 143Nd/144Nd ratio. Using 143Nd/144 extinctions. The recognition that theLateOrdovicianextinctionwas Nd as a proxy for water depth, a reasonable approximation in the distal preceded by substantial ecological change also raises the intriguing settings of this study, allowed Sheets et al. to disentangle the changes question of how much ecological change must be accrued before an in graptolite communities associated with the from extinction tipping point is reached. Given the extensive record of those that merely reflect lateral shifts in the mesopelagic zone regional and global ecological change in the fossil record, this is an as sea level rose and fell (Fig. 1). exciting avenue for future work.

1 Raup DM, Sepkoski JJ, Jr (1982) Mass extinctions in the marine fossil record. Science 215(4539):1501–1503. 2 Foote M (2003) Origination and extinction through the : A new approach. J Geol 111(2):125–148. 3 Alroy J (2014) Accurate and precise estimates of origination and extinction rates. 40(3):374–397. 4 Plotnick RE, McKinney ML (1993) Ecosystem organization and extinction dynamics. Palaios 8(2):202–212. 5 Roy K (1996) The roles of mass extinction and biotic interaction in large-scale replacements: A reexamination using the fossil record of stromboidean gastropods. Paleobiology 22(3):436–452. 6 Droser ML, Bottjer DJ, Sheehan PM (1997) Evaluating the ecological architecture of major events in the Phanerozoic history of marine invertebrate . Geology 25(2):167–170.

8350 | www.pnas.org/cgi/doi/10.1073/pnas.1608630113 Holland Downloaded by guest on September 26, 2021 7 Droser ML, Bottjer DJ, Sheehan PM, McGhee GR (2000) Decoupling of taxonomic and ecologic severity of Phanerozoic marine mass extinctions. Geology 28(8): 675–678. 8 McGhee GR, Sheehan PM, Bottjer DJ, Droser ML (2004) Ecological ranking of Phanerozoic crises: Ecological and taxonomic severities are decoupled. Palaeogeogr Palaeoclimatol Palaeoecol 211:289–297. 9 McGhee GR, Clapham ME, Sheehan PM, Bottjer DJ, Droser ML (2013) A new ecological-severity ranking of major Phanerozoic biodiversity crises. Palaeogeogr Palaeoclimatol Palaeoecol 370:260–270. 10 Christie M, Holland SM, Bush AM (2013) Contrasting the ecological and taxonomic consequences of extinction. Paleobiology 39(4):538–559. 11 Sheets HD, et al. (2016) Graptolite community responses to global and the Late Ordovician mass extinction. Proc Natl Acad Sci USA 113:8380–8385. 12 Catuneanu O (2006) Principles of Sequence Stratigraphy (Elsevier, New York). 13 Holland SM, Patzkowsky ME (2015) The stratigraphy of mass extinction. Palaeontology 58(5):903–924. 14 Patzkowsky ME, Holland SM (2012) Stratigraphic Paleobiology (Univ of Chicago Press, Chicago). 15 Smith AB, Gale AS, Monks NEA (2001) Sea-level change and rock-record bias in the : A problem for extinction and biodiversity studies. Paleobiology 27(2):241–253. 16 Holland SM (2000) The quality of the fossil record—A sequence stratigraphic perspective. Deep Time: Paleobiology’s Perspective, eds Erwin DH, Wing SL (Paleontological Society, Lawrence, KS), pp 148–168. 17 Peters SE, Foote M (2002) Determinants of extinction in the fossil record. 416(6879):420–424. 18 Finnegan S, Heim NA, Peters SE, Fischer WW (2012) Climate change and the selective signature of the Late Ordovician mass extinction. Proc Natl Acad Sci USA 109(18):6829–6834. 19 Miller KG, et al. (2005) The Phanerozoic record of global sea-level change. Science 310(5752):1293–1298. 20 Ghienne JF, et al. (2014) A Cenozoic-style scenario for the end-Ordovician glaciation. Nat Commun 5:4485. 21 Jablonski D (2005) Mass extinctions and . Paleobiology 31(2):192–210.

Holland PNAS | July 26, 2016 | vol. 113 | no. 30 | 8351 Downloaded by guest on September 26, 2021