Cover Image

HARDBACK
$54.95



View/Hide Left Panel

2
Impact of Late Ordovician Glaciation-Deglaciation on Marine Life

W. B. N. BERRY, M. S. QUINBY-HUNT, and P. WILDE

University of California, Berkeley

ABSTRACT

Sea-level fell at least 50 m during Late Ordovician continental glaciation which centered on the South Pole. Oxygen isotope analyses indicate that ocean surface waters cooled during glaciation. As sea-level fell and surface waters cooled, mass mortalities occurred among most marine benthic faunas, primarily brachiopods and trilobites. Carbon isotope analyses reveal a significant biomass loss at the time of the mass mortalities. The brachiopod-dominated Hirnantia fauna spread widely during glacial maximum. That fauna essentially became extinct during deglaciation. Cold, oxygen-rich deep ocean waters generated at the South Pole during glaciation drove a strong deep ocean circulation and ventilated the deep oceans. Potentially, waters bearing metal ions and other substances toxic to organisms were advected upward into ocean mixed layer during glacial maximum. Graptolite mass mortality apparently was a consequence. Mass mortalities took place among pre-Hirnantia brachiopod and trilobite faunas at the same time as the graptolite mass mortality. Reradiation among graptolites and benthic marine faunas followed after sea-level rise, and deep ocean circulation slowed as deglaciation proceeded. Initially, reradiation rates were slow as unstable environments persisted during the early phases of deglaciation. New colony organization developed among graptolites, but significant originations did not take place until habitats preferred by graptolites stabilized. Conodont mass mortality occurred at the onset of deglaciation. Originations of new taxa were slow initially, but the pace increased as shelf seas expanded and new environments became stable. Similarly, the pace of marine benthic faunal reradiation was slow at first but increased after shelf sea environments stabilized.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 34
Effects of Past Global Change on Life 2 Impact of Late Ordovician Glaciation-Deglaciation on Marine Life W. B. N. BERRY, M. S. QUINBY-HUNT, and P. WILDE University of California, Berkeley ABSTRACT Sea-level fell at least 50 m during Late Ordovician continental glaciation which centered on the South Pole. Oxygen isotope analyses indicate that ocean surface waters cooled during glaciation. As sea-level fell and surface waters cooled, mass mortalities occurred among most marine benthic faunas, primarily brachiopods and trilobites. Carbon isotope analyses reveal a significant biomass loss at the time of the mass mortalities. The brachiopod-dominated Hirnantia fauna spread widely during glacial maximum. That fauna essentially became extinct during deglaciation. Cold, oxygen-rich deep ocean waters generated at the South Pole during glaciation drove a strong deep ocean circulation and ventilated the deep oceans. Potentially, waters bearing metal ions and other substances toxic to organisms were advected upward into ocean mixed layer during glacial maximum. Graptolite mass mortality apparently was a consequence. Mass mortalities took place among pre-Hirnantia brachiopod and trilobite faunas at the same time as the graptolite mass mortality. Reradiation among graptolites and benthic marine faunas followed after sea-level rise, and deep ocean circulation slowed as deglaciation proceeded. Initially, reradiation rates were slow as unstable environments persisted during the early phases of deglaciation. New colony organization developed among graptolites, but significant originations did not take place until habitats preferred by graptolites stabilized. Conodont mass mortality occurred at the onset of deglaciation. Originations of new taxa were slow initially, but the pace increased as shelf seas expanded and new environments became stable. Similarly, the pace of marine benthic faunal reradiation was slow at first but increased after shelf sea environments stabilized.

OCR for page 34
Effects of Past Global Change on Life INTRODUCTION Although many glaciations have occurred during the history of the Earth, the glaciation near the end of the Ordovician stands out because the environmental changes that took place then were accompanied by near extinction of a great number of organisms (Berry and Boucot, 1973; Sheehan, 1973). A massive continental ice cap centered near the South Pole covered a large part of a continent that may be called Gondwanaland for about 2 million years (m.y.) about 435 to 437 m.y. ago (Ma). During that time of maximum glaciation, sea-level fell at least 50 m and perhaps as much as 100 m. The rock record suggests that sea-level fall may have been slower than its rise following glacial melting. These glacioeustatic sea-level changes created significant changes in marine environments. Extinctions among most marine organisms living at the time appear to have been related to the environmental changes. The features typical of continental ice sheets, including glacial pavements, striated pebbles, esker-like ridges, and glacial ice-carried dropstones have been described by Beuf et al. (1971) and Rognon et al. (1972) in Saharan Africa and Saudi Arabia (see summaries in Brenchley, 1988; Vaslet, 1990). Evidence of glaciation appears to have extended as far north from the South Pole as about 40° of south latitude (Brenchley and Newall, 1984; Brenchley, 1988). Tilloids of glaciomarine origin have been recorded from many localities in Spain, Portugal, and France (Brenchley, 1988). The presence of large dropstones that deform finely spaced sediment laminae in deposits in Spain and France indicates that large icebergs floated many miles from the shores of Gondwanaland. The spread of continental ice and the icebergs that calved off from it are analogous to the development of ice and icebergs during Pleistocene glacial maximum (Brenchley and Newall, 1984). Brenchley et al. (1991) described Late Ordovician glaciomarine diamictites in stratal sequences in Portugal and in the Prague Basin, Czechoslovakia. They concluded that "the sequence in the Prague Basin suggests that cold climates with floating marine ice developed early in the Hirnantian, before the main glacioeustatic regression, whereas in Portugal, deposition from marine ice was somewhat later and postdated the regression." Areas that had been sites of shallow marine deposition became lands into which rivers cut channels and across which sands spread in sites of nearby terrigenous materials or karst topography formed on lime rock deposits in tropical areas (Brenchley and Newall, 1984). The rock record suggests that the features formed during glacial maximum and lowstand of sea were covered rapidly as sea-level rose during glacial melting. Postglacial environmental changes seemingly were rapid (Brenchley, 1988). DATA SUMMARIES Cocks and Rickards (1988) assembled an extensive body of data concerning environmental and faunal changes in the Ordovician-Silurian boundary interval. Those summaries, arranged by region and faunal group, provide precise information on environmental changes, near extinctions, and radiations of organisms in the postglacial interval. Brenchley and Newall (1984) summarized the evidence for sea-level changes and related marine environmental changes during and after glaciation. Detailed summaries of changes in specific groups of organisms during the interval of Late Ordovician faunal change may be found in Rickards et al. (1977), Briggs et al. (1988), Fortey (1989), and Sheehan and Coorough (1990). Summaries in Barnes and Williams (1991) enhance the faunal and stratigraphic data for the Late Ordovician glacial interval. These several summaries and the data on which they are based provide the information essential to this overview of faunal changes in the Late Ordovician glacial-postglacial sequence of environmental changes. THE TIME FRAME Six primary divisions, termed Series, have been recognized within the Ordovician Period. Each of the Series is typified by a unique fauna. The youngest Series, the Ashgill, has been divided into four stages based on associations of brachiopods and some trilobites. The youngest Ashgill division or stage, the Hirnantian, is succeeded by the earliest Silurian stage, the Rhuddanian (see Table 2.1). The stages are characterized by fossil faunas that were benthic in life and lived in shelf sea environments. Remains of planktic organisms are rare, with certain exceptions, among the shelly faunas. Graptolites—the remains of colonial, marine plankton—occur in abundance in sequences of thinly laminated black shales and mudstones in which shelly fossils do not occur, except in debris materials derived from environments in which shelly fossils live. Accordingly, graptolite zones have been recognized as divisions of the Ordovician in black shale sequences. Graptolite zone divisions of the Ashgill are shown in Table 2.1. Time synchronous correlation of the graptolite zones with the shelly fossil stages is imprecise because the occurrence of the two types of organisms in the same deposit is so rare. Two sets of graptolite zones are shown for the early part of the Ashgill because graptolites of that time interval were distributed in two faunal provinces. The stages typified by brachiopod-trilobite associations and the graptolite zones provide a temporal context for analysis of patterns in mass mortality and reradiation. Glacial maximum occurred during the Hirnantian Stage, although glaciation probably commenced early in the Ashgill.

OCR for page 34
Effects of Past Global Change on Life TABLE 2.1 Time Units for the Late Ordovician (Ashgill) and Earliest Silurian Period Age Stage Zone   Silurian Llandivery Rhuddanian acuminatus   Ordovician Ashgill   persculptus       Hirnantian           extraordinarius       Rawtheyan           pacificus bohemicus         uniformicus       complexus           mirus     Cautleyan             typicus       complanatus       Pusgillian     Deglaciation started in the graptolite Glyptograptus persculptus zone. At least part of that zone is coeval with the latter part of the Hirnantian. The Ordovician-Silurian boundary is at the base of the Parakidograptus acuminatus zone. The glaciation was thus entirely within the latter part of the Ashgill. THE PALEOGEOGRAPHIC FRAMEWORK The paleogeographic maps in Figures 2.1 and 2.2 were supplied by C. R. Scotese. The Ashgill and Early Silurian (Llandovery) paleogeography (i.e., the positions of the lands, shelf seas and open oceans of the time) has been developed from remnant magnetism data, glacially derived sedimentary materials and features, and the locales of carbonates and evaporites. The carbonates most likely formed in tropical shallow marine environments. Ocean water masses and ocean currents are proposed based on paleogeography and consideration of modern oceanic circulation patterns. Three primary features stand out (see Figure 2.1 and 2.2): (1) no record of land or shallow marine shelf environments in the Northern Hemisphere north of the tropics; (2) a significant number of plates bearing shallow marine environments aligned essentially east-west within the tropics; and (3) a large land mass, Gondwanaland, that covered the South Pole, with a portion of it extending northward into the equatorial region. The positions of the plates bearing lands and shallow shelf seas, as well as the zoogeographic relationships of the faunas present in the rocks formed in these environments, allow the major features of ocean circulation to be proposed. Late Ordovician-Early Silurian ocean circulation (see Figures 2.1 and 2.2) is based on the assumption that similar insolation patterns existed in the Ashgill as at present. The presence of a large part of Gondwanaland in southern temperate and tropical latitudes suggests that ocean surface circulation near it was influenced by seasonal monsoons. The positions of pressure systems over land and nearby seas would have shifted seasonally, leading to seasonal reversals in surface ocean circulation, as is seen today in the Indian Ocean. Accordingly, surface ocean circulation patterns are proposed for winter and summer seasons (Figures 2.1 and 2.2). Ocean surface circulation north of the Northern Hemisphere tropics would have been zonal. Surface circulation in the tropics would have been influenced greatly by the several plates bearing shallow marine environments. The relatively long north-south orientation of the Laurentian (North America), Baltoscania, and Gondwanaland plates would have resulted in major oceanic circulation that was essentially within a single ocean. The ocean bordered by Gondwanaland and Baltoscania (see Figures 2.1 and 2.2) would have had a unique surface circulation that included a relatively cold polar western boundary current that flowed northward on the west side of Gondwanaland. Upwelling would have been strong and continuous there as a result of Coriolis deflection. The modern analogue of that current is the Humboldt Current off Peru-Chile. A similar but weaker western boundary current probably flowed along the west side of Baltoscania. Pre-Hirnantian Ashgill and Hirnantian brachiopod zoogeography provides clues to potential changes in surface circulation in the ocean between Baltoscania and Gondwanaland. Sheehan and Coorough (1990) recognized four unique zoogeographic faunas, prior to the Hirnantian, along the Baltoscanian-Gondwanaland coasts. A single fauna,

OCR for page 34
Effects of Past Global Change on Life FIGURE 2.1 Paleogeographic maps for Ashgill time showing positions of lands, shelf seas, and open oceans. Ocean surface circulation is suggested. The size of Gondwanaland suggests that monsoonal conditions near it generated seasonal differences in ocean surface circulation. As sea level fell during the latter part of the Ashgill, large areas of the shelves were exposed and ocean surface currents may have been enhanced along shelf margins of the time. Base map provided by C. R. Scotese. the Hirnantian fauna, replaced three of the four provincial faunas during the Hirnantian Stage. This zoogeographic change suggests that the current flowing from the tropics south along Baltoscania strengthened to such an extent that it had an influence on the Gondwanaland shelf. The Gulf Stream may be a somewhat analogous modern current. The southerly flowing current would have flowed from the tropics into high latitudes, carrying warm surface water into a cold air regime. When the Gulf Stream originated to carry warm water northward in the past 1.5 to 2 m.y., the infusion of warm water led to enhanced evaporation in a cold air regime and so was a factor in the development of glacial ice. Potentially, the ocean surface current circulation implied by the wide geographic spread

OCR for page 34
Effects of Past Global Change on Life FIGURE 2.2 Paleogeographic maps for the Llandovery showing positions of lands, shallow shelf seas, and open oceans. Ocean surface water circulation currents are indicated. The size of Gondwanaland suggests that monsoonal conditions were generated near it. Seasonal monsoons would lead to seasonal differences in surface circulation. The shelf seas were sites of marine transgression that commenced in the latest Ordovician and continued into the early part of the Llandovery. Return of stable shelf sea environments during the Llandovery was followed by significant reradiation among marine communities, both benthic and planktic. Sheehan (1982) described development of Llandovery brachiopod communities in shelf sea environments that had become stable following deglaciation and transgression. Base map provided by C. R. Scotese. of the Hirnantia fauna functioned as the Gulf Stream did during the onset of North Polar glaciation in the Pleistocene. Glaciation during the Hirnantian Stage resulted in lower sea-level. Surface water current impinged on only the outer margins of plates, and shelf seas were significantly narrower than prior to glaciation. Surface currents may have swept along shelf margins more strongly that when waters were widespread across the shelves. Currents may have slowed when sea-level rose as a result of deglaciation.

OCR for page 34
Effects of Past Global Change on Life More zoogeographic provinces among benthic faunas would have resulted as a consequence of slowed surface circulation. Sheehan and Coorough (1990) noted that the Early Silurian brachiopod faunas were characterized by more provinces than the Hirnantian Stage brachiopod faunas. Deep ocean circulation probably increased significantly during glacial maximum. Deep ocean circulation would have been driven by both evaporation and cooling of large volumes of ocean water situated around the margins of a glaciated Gondwanaland. Cold, dense water formed in this manner sinks and flows along the ocean floor. Such water bears more oxygen than warmer surface water. Thus, during glacial maximum, the deep oceans would have received a significant and prolonged increase in oxygen. As deep water flowed north, the positions of the plates (aligned essentially east-west in the tropics and south of the tropics) would have created baffles against which deep ocean water would have advected upward. A consequence of that advection would have been a general toward-the-surface movement of the thermocline and of the zone of oxygen-depleted waters or oxygen minimum zone waters. Upward motion of oxygen-depleted water would have forced waters that contained metal ions and other substances toxic to many organisms into the mixed layer. Breaking of internal waves, as well as upwelling of waters bearing unconditioned nutrients (nutrients in proportions or in oxidation states such that organisms cannot take them in) or toxic trace metals, would have resulted in episodic incursion of water toxic to nearly all phytoplankton and zooplankton into those waters inhabited by most plankton and nekton, the mixed layer. If glacial development waxed and waned, as suggested by Vaslet (1990) and Brenchley et al. (1991), then these incursions may have had a greater or lesser effect on organisms, depending on the influence of bottom-water generation, which was related to glacial developments. Such episodic stronger and weaker incursions of toxic waters into plankton and nekton habitats could have had a greater impact on the long-term survival or extinction of many organisms than a single incursion. Wilde and Berry (1984) and Wilde et al. (1990) described how regional- to global-scale vertical advection of deep ocean waters into near-surface mixed layer water can create an environmental change crisis for many marine organisms. Such vertical advection into the mixed layer can result in the following (Wilde et al., 1990): (1) direct toxicity of mixed layer water; (2) modification or reduction of nutrients and food resources through inhibition of photosynthesis; (3) chronic debilitation through continued contacts with toxic waters; and (4) increased predation by more adapted organisms. Such environmental crises for most organisms could also result in new ecologic opportunities for organisms that had been ecologically suppressed under prior environmental conditions. Whereas deep circulation was vigorous during glaciation, it slowed markedly with the onset of deglaciation. A characteristic of Pleistocene glacial to interglacial change is rapid development of deglaciation (W. Broecker, Lamont-Doherty Earth Observatory, oral communication, 1990). Relatively rapid deglaciation results in rapid change in deep ocean circulation. Marked vertical advection of the glacial maximum was followed by ocean conditions in which the zone of oxygen-depleted water expanded and descended somewhat in the ocean. Upward vertical advection from that zone diminished as a result. Mixed zone water expanded downward and rapidly became more hospitable to life. Sea-level rose as a consequence of deglaciation. As sea-level rose and oxygen-depleted waters expanded, these waters spread anoxia across the outer parts of shelves and platforms. Vertical advection of toxic waters during glaciation would have created inhospitable environments not only for nektic and planktic organisms, but also for many benthic organisms. Reduction or absence of vertical advection of toxic waters during deglaciation would have reopened many environments in the mixed layers to resettlement by organisms. GEOCHEMICAL EVIDENCE OF DEEP OCEAN VENTILATION Although the ocean surface and deep circulation suggested for the Late Ordovician glacial and subsequent nonglacial interval is essentially speculative and derived from proposed paleogeographic reconstructions, some direct geochemical evidence has been developed to support the proposed model. Quinby-Hunt et al. (1989) summarized the results of approximately 300 neutron activation analyses of dark shales. Many of the samples came from the Late Ordovician-Early Silurian succession at Dob's Linn (Wilde et al., 1986; Quinby-Hunt et al., 1989). Dark shales in which the calcium concentration is less than 0.4% were selected for close scrutiny. The low calcium concentration in such samples allows the assumption that the Fe and Mn contained in them are in oxides and sulfides that reflect reducing conditions. The Fe and Mn in such low-calcium rocks are not bound in carbonates. Accordingly, Fe and Mn concentrations may be used as indicators of the intensity of reducing conditions. Under oxic conditions (environments in which oxygen is relatively plentiful), Fe and Mn concentrations are relatively high. As oxygen availability diminishes to a condition in which it is no longer present, Mn is reduced before Fe and becomes more soluble. As a consequence, Mn concentration diminishes because it may form oxides and sulfides. Manganese diminishes in the early stages of onset of reducing conditions. As reducing conditions be

OCR for page 34
Effects of Past Global Change on Life come somewhat more intense, Fe3+ is reduced to Fe2+, possibly as a result of resolution of oxyhydroxides (Libes, 1992, p. 196). Much of the reduced Fe will form iron sulfide. These developments in Mn and Fe, which follow from a change from oxic to mildly reducing conditions in depositional environments, suggest the existence of three basic environmental situations that reflect the change from oxic to reducing conditions: (1) relatively high concentrations of Fe and Mn in oxic environments; (2) relatively low concentration of Mn and high concentration of Fe in mildly reducing conditions; and (3) relatively low concentrations of both Mn and Fe in slightly more highly reducing conditions. Based on Fe and Mn concentrations reflective of the change from oxic to reducing environments, the low-calcium dark shale sample analyses were studied to ascertain if their Mn and Fe concentrations reflected depositional environment. The Mn and Fe concentrations in about 200 low-calcium dark shale samples revealed three clusters (see Quinby-Hunt et al., 1988, 1990). As indicated in Table 2.2, the three clusters appear to reflect the degree of oxidation in the above three environmental situations. Also indicated in Table 2.2, a fourth cluster is present in the samples analyzed. That cluster of samples is characterized by relatively low concentrations of manganese and iron but high concentrations of vanadium. Samples from modern oxic and anoxic depositional environments were analyzed by using neutron activation to ascertain if Mn and Fe concentrations in them were closely similar to those characteristic of any of the clusters recognized among the ancient shales. Sediment samples from the Santa Barbara and Santa Monica Basins, housed at the University of Southern California, were analyzed. Oxygen content of waters approximately a meter above these sediment had been measured (D. Gorsline, oral communication, 1988.). The Mn and Fe concentrations of samples of basin sediment beneath waters in which the oxygen content varied from 0.1 ml/l to undetectable were comparable to cluster 2. Sediments beneath waters that had an oxygen content of 0.5 ml/l or greater had relatively high concentrations of Mn and Fe (similar to cluster 1). TABLE 2.2 Chemical Clustering of ~200 Low-Calcium Dark Shales Cluster Mn (ppm) Fe (ppm) 1 1300 56,000 2 310 52,000 3 176 19,000 4 Similar to cluster 3 but V ranges from 350 to 1500 ppm   SOURCE: Quinby-Hunt et al., 1988, 1990. Petroleum source-rock samples from the Miocene Monterey Formation recovered from a producing oil well in California also were examined with neutron activation. These source-rock samples had low Mn and Fe and high V concentrations, the unique geochemical signature of cluster 4. Kastner (1983) pointed out that the highly organic-rich, petroleum source-rock shales of the Monterey Formation formed under highly reducing, methanogenic conditions. The analyses of modern sediments from the Santa Barbara and Santa Monica Basins and those of the organic-rich shales from the Monterey Formation suggest certain geochemical aspects of the depositional environments in which dark shales in clusters 1, 2, and 4 accumulated. Berner's (1981) discussion of a geochemical classification of sediments indicates possible geochemical conditions in the depositional environment of cluster 3. Berner (1981) described four primary geochemical categories of environmental conditions under which sediment accumulates: (1) oxic; (2) post-oxic, nonsulfidic; (3) sulfidic or sulfate reducing; and (4) methanogenic. Categories 2, 3, and 4 are indicative of an increasingly greater degree of reducing conditions in the depositional environment (Berner, 1981). The geochemical data from analyses of ancient dark shales and modern sediment are consistent with Berner's (1981) geochemical categories for clusters 1 and 2. If cluster 4 is indicative of Berner's methanogenic zone, then cluster 3 is likely, at least in part, to be a product of sediment accumulation in Berner's sulfidic category. In view of the lack of direct comparison with samples from modern environments, some cluster 3 samples could have been derived from sediment that accumulated in Berner's post-oxic, nonsulfidic interval. Bacterial sulfate reduction generates hydrogen sulfide that may react with iron to form pyrite (Berner, 1981). Pyrite occurs in many dark, graptolite-bearing shales that have Mn and Fe concentrations characteristic of cluster 3 (Quinby-Hunt et al., 1989). Shale samples were taken from closely spaced stratigraphic intervals in the Late Ordovician-Early Silurian shales that comprise the Ordovician-Silurian boundary stratotype at Dob's Linn, southern Scotland. These samples were analyzed by neutron activation (Wilde et al., 1986). The Fe, Mn, and V concentrations (Wilde et al., 1986) in the Ordovician-Silurian boundary interval shales at Dob's Linn are consistent with the assignment of each sample to one of the four clusters reflective of oxic and the sequence of increasingly more reducing anoxic depositional environments (Figure 2.3). The Mn and Fe concentrations in gray shales that bear rare graptolites or are unfossiliferous are similar to the Mn and Fe concentrations in Santa Barbara and Santa Monica Basin sediments that accumulated

OCR for page 34
Effects of Past Global Change on Life FIGURE 2.3 Diagram illustrating the results of the neutron activation analyses of shales in the Ordovician-Silurian boundary section at Dob's Linn, southern Scotland. The Mn and Fe concentrations of cluster 1 indicate relatively oxic depositional environments. Concentrations of Mn and Fe suggest that cluster 2 is mildly anoxic. Mn and Fe concentrations of clusters 3 and 4 are indicative of relatively highly anoxic depositional environments. Shales with Mn and Fe concentrations suggestive of oxic conditions formed during glaciation, an interval during which deep ocean circulation was strong and ocean water ventilation great. The data indicate that samples bearing extraordinarius zone graptolite accumulated under anoxic conditions, perhaps during an interglacial. Anoxic environments developed in the area during persculptus zone time, probably reflecting the onset of deglaciation and global warming. Oxic conditions appeared in the depositional environment during sedgwicki zone time, possibly reflecting shallowing of the basin and/or influx of currents with oxic water. under waters with some oxygen. Dob's Linn section shales bearing numerous graptolites representative of numbers of different taxa commonly have Mn, Fe, and V concentrations characteristic of clusters 3 or 4 (Quinby-Hunt et al., 1989). Shales having Mn and Fe concentrations characteristic of cluster 2 contain fewer graptolites than those with Mn and Fe concentrations typical of clusters 3 and 4. Shales with Mn and Fe concentrations indicative of oxic depositional environments (cluster 1) contain only specimens of climacograptids of C. miserabilis and C. normalis groups (normalograptids) in the Dob's Linn Ordovician- Silurian boundary interval. These graptolites were survivors of the Late Ordovician mass mortality among graptolites (Berry et al., 1990). The shales with Mn and Fe concentrations indicative of oxic depositional environments, containing only normalograptids, were deposited during the Late Ordovician glaciation in the Southern Hemisphere. That is, they were deposited during the Hirnantian Stage of the post-D. anceps into the G. persculptus zone interval. Deep ocean water circulation would have been at its maximum during glacial maximum, and the oxygen content of that water would have been as its greatest. Orth (see Wang et al., 1990) analyzed closely spaced black shale samples from the Ordovician-Silurian boundary interval at two localities in south China using neutron activation. The Mn, Fe, and V concentrations from the south China samples reflect a pattern of change from anoxic to oxic and a return to anoxic conditions in the deposi-

OCR for page 34
Effects of Past Global Change on Life tional environment that is similar to the change seen in coeval rocks at Dob's Linn. The Mn and Fe concentrations in south China indicate that oxic waters were present in the area during the glacial maximum. The graptolite faunas obtained from several Ordovician-Silurian boundary sequences in south China suggest that environmental conditions favored by many graptolites persisted longer in south China than in Dob's Linn in the Late Ordovician (Mu, 1988; Berry et al., 1990). Furthermore, those waters preferred as habitats by many graptolites returned more rapidly to areas in south China than they did in southern Scotland (Berry et al., 1990). The geochemical evidence from two sites that were distant from one another in the Late Ordovician appears to be consistent with significant deep ocean circulation and ventilation during glaciation. The geochemical data are also consistent with the hypothesis that only certain graptolites lived in near-surface and relatively oxic waters, and that many others lived at some depth close to oxygen-poor but nitrogen oxide rich waters (see Berry et al., 1987). The geochemical data indicate that graptolites disappeared from areas in which they had been plentiful during glacial maximum when the oxygen-poor, nitrogen oxide-rich waters were diminished and forced upward into locations nearer the surface than in nonglacial times. Lateral spread of oxygen-poor and possibly toxin-bearing waters across shelf-marginal sites during glacioeustatic sea-level fall may have been a factor in mass mortalities among certain benthic shell dwellers as well as graptolites living relatively low in the oceanic mixed layer. PATTERNS IN EXTINCTION AND RERADIATION Graptolites The graptolite extinction and reradiation pattern is linked closely to the presence and absence of richly graptoliferous black shales. Such shales both at Dob's Linn and in south China bear Fe and Mn concentrations of cluster 3 or 4 (Quinby-Hunt et al., 1989). The most abundant and richly diverse graptolite faunas occur in sediments deposited under anoxic ocean water. Analogous conditions are found today in the eastern tropical Pacific (Berry et al., 1987). There, oxygen-poor, nitrogen oxide-rich waters are inhabited by zooplankton that may be modern ecologic analogues of many ancient graptolites (Berry et al., 1987, 1990). Sulfate-reducing bacteria have been recorded from sediment accumulating under the oxygen-poor, nitrogen oxide-rich waters (Gallardo, 1963). Stratigraphic study of Ordovician-Silurian boundary sections indicates that such conditions were diminished markedly during glacial maximum (see summaries in Cocks and Rickards, 1988). Koren (1991) pointed out that the Late Ordovician graptolite mass mortality occurred at the end of the P. pacificus zone, a time when black shales disappeared from nearly all Ordovician-Silurian boundary sequences. Commenting on the Late Ordovician graptolite mass mortality, Melchin and Mitchell (1988) pointed out that the "Late Ordovician graptolites experienced nearly total extinction," and that "post-extinction morphological radiation stemmed from only a few species, most of which had the same pattern in colony development." Graptolite faunas during glacial maximum included mostly normalograptids (i.e., small climacograptids of the C. normalis and C. miserabilis types, glyptograptids of the G. persculptus group, and Climacograptus extraordinarius) as well as Diplograptus bohemicus and a few other rare diplograptid (biserial) species. All taxa except the climacograptids of the C. normalis and C. miserabilis type and glyptograptids of the G. persculptus group became extinct prior to the appearance of those glyptograptids characteristic of the G. persculptus zone. Rickards et al. (1977) and Rickards (1988) drew attention to graptolite development at the onset of reradiation. Reradiation commences when black shales become more widespread during G. persculptus zone time. Species diversity is low, but a number of glyptograptids and climacograptids appear. The most significant innovation among graptolites in G. persculptus zone time is the appearance of the uniserial scadent or monograptid colony (Atavograptus ceryx and similar species). The base of the superjacent Parakidograptus acuminatus zone is characterized by the appearance of a number of biserial scandent graptolites (new genera) with a slender, elongate proximal region (initial part of the colony). The sicula in these taxa is a relatively long, slender cone, as are the first few thecae or zooidal cups. Commonly, the origin of the first zooidal cup is high enough on the conical sicula so that much of the sicula is exposed. Taxa with slender initial regions spread widely as sea-level rose and platform areas flooded. Most genera and species are new. Their appearances are relatively slow, taking most of the time duration of three graptolite zones for a fauna that is relatively species rich to redevelop. The morphological innovation that is most striking among the reradiation faunas is that of the monograptid colony organization. At least three graptolite zones elapsed; however, that new style of colony development led to many new taxa. Shelly Faunas Brachiopods were numerous and widespread during the latter part of the Ordovician. Mass mortalities occurred among them in two phases in the Late Ordovician. The first phase took place at or near the Rawtheyan-Hirnantian Stage boundary. At that time, sea-level was falling as

OCR for page 34
Effects of Past Global Change on Life glaciers expanded. Ocean surface water circulation and chemistry changed as well. Sheehan and Coorough (1990, p. 184) point out that ''. . . of the 211 genera recorded in the early-middle Ashgill, 92 (44 percent) did not survive into the Hirnantian. A substantial portion of these genera probably died during the early-middle Ashgill rather than at the base of the Hirnantian." Of 124 brachiopod genera known from the Hirnantian Stage, only 21 originated during that stage (Sheehan and Coorough, 1990, p. 185). Approximately one-half the Hirnantian brachiopod genera become extinct by the end of the Stage; thus, the Hirnantian extinctions comprise the second phase of Late Ordovician brachiopod mass mortality. Taken together, approximately two-thirds of the early-middle Ashgill brachiopods became extinct by the end of the second phase of mass mortality. Apparently, the shelf sea brachiopod fauna typified by members of the genus Hirnantia spread widely during glacial maximum conditions. Many members of the fauna may have been adapted to somewhat cooler ocean surface water temperatures than were present either before or after glacial maximum. Oxygen isotope determinations made from brachiopod shells in Sweden, which lay within the tropics in the Late Ordovician, indicate that surface water temperatures were cooler in the Hirnantian than earlier (Marshall and Middleton, 1990). Marshall and Middleton (1990) also note that the carbon isotope values from the same shells suggest "enhanced deposition of organic carbon, a process which would have decreased" the partial pressure of CO2 in both the ocean and the atmosphere, contributing to rapid global cooling. Sheehan and Coorough (1990) note that of 130 genera recorded from Early Silurian strata, about 40% are new and 60% range into the Silurian from the Ordovician. Cocks (1988) noted that the earliest Silurian brachiopod faunas have fewer species and fewer individuals in similar samples than Hirnantian Stage faunas, based on a precise review of Ordovician-Silurian boundary interval brachiopods. Sheehan (1982) pointed out that the Early Silurian brachiopod communities changed in species composition relatively rapidly. The time interval of the most rapid change in community organization was during sea-level rise and surface ocean water warming. After the transgression of seas across formerly emergent platforms had stabilized and surface waters warmed, brachiopod community organization stabilized as well. Late Ordovician trilobite mass mortality took place during and at the Rawtheyan-Hirnantian Stage boundary (Briggs et al., 1988). Hirnantian trilobites lived on the outer parts of marine shelves and were highly provincial (Lesperance, 1988). Shelf sea-dwelling trilobites generally had high rates of generic extinction as waters cooled with the onset of glaciation (Fortey, 1989). Survivors are those trilobites that either lived among reefs or "reef-like calcareous habitats, or were commoning inshore clastics around Ordovician Gondwana" (Fortey, 1989, p. 106). Lesperance (1988, p. 359) noted that Early Silurian trilobites appear to be holdovers from or survivors of the mass mortality at the end of the Rawtheyan with a few new taxa. Early Silurian originations were rare, with the result that Early Silurian trilobite faunas include primarily relatively long-ranging taxa with broad environmental tolerances. Trilobites adapted to particular habitats disappeared in the Late Ordovician mass mortality, and no taxa adapted to discrete environments appeared until the latter part of the Early Silurian. Barnes and Bergstrom (1988) summarized conodont faunas in the Ordovician-Silurian boundary interval, noting striking differences between Ordovician and Silurian faunas. The position of most intense conodont faunal turnover seems to be within the Glytograptus persculptus zone, at a position about coeval with the beginnings of graptolite reradiation. The stratigraphic position coincides with the onset of rising sea-level coincident with the commencement of deglaciation. Late Ordovician conodont mass mortality "was not a sudden catastrophic event although only a few species survived into the Silurian; rather, during the Ashgill there was a gradual disappearance involving many characteristic and long-established stocks and the new taxa that appeared were considerable fewer than those that became extinct" (Barnes and Bergstrom, 1988, p. 334). Clearly, origination rates of conodont species were reduced markedly during the Late Ordovician and earliest Silurian. The extinction rate seemingly increased during the Hirnantian Stage and the early part of the G. persculptus zone (Barnes and Bergstrom, 1988, Figures 4 and 5). Demonstrably, conodonts survived the environmental changes coincident with glaciation that had so profound an impact on extinction and origination rates among other organisms. Interestingly, the environmental changes related to sea-level rises and global warming had the greatest influence on conodont mass mortality and the origination of new taxa. Late Ordovician chitinozoans are primarily taxa with long stratigraphic ranges. Most chitinozoans in Ordovician strata disappear near the end of the Rawtheyan (Grahn, 1988). Chitinozoans are rare in the Hirnantian Stage-G. persculptus zone interval. New chitinozoan taxa appear in the Early Silurian, with reradiation commencing near the base of the Silurian (Grahn, 1988). ENVIRONMENTAL-ORGANISMAL CHANGES: A SUMMARY Late Ordovician glaciation involved several significant environmental changes, including the following: a lowering and then a rising sea-level; marked deep and midocean

OCR for page 34
Effects of Past Global Change on Life ventilation; advection upward of waters potentially toxic to many organisms into the lower part of the ocean mixed layer; the cooling and then warming of ocean surface water temperatures; and changes in nutrient availability in many nearshore marine environments. The most prominent organismal (Table 2.3) mass mortality in the Late Ordovician took place at or close to the Rawtheyan-Hirnantian Stage boundary. Many graptolites, brachiopods, trilobites, corals, and chitinozoans became extinct or were markedly reduced in numbers and taxonomic diversity at that time. Isotopic studies of brachiopod shells from Sweden (Middleton et al., 1988; Marshall and TABLE 2.3 Significant Physical Environmental Changes, Organismal Mass Extinctions, and Clade Turnovers in the Latest Ordovician Stage Zone Event Rhuddanian acuminatus — Hirnantian persculptus Conodont turnover     Graptolite reradiation     Sea-level rise     Onset of glacial melting     Brachiopod turnover   extraordinarius Glacial maximum Rawtheyan pacificus Trilobite mass extinction     Brachiopod major extinction   complexus   NOTE: Prominent depletions in 13C (Wang Xiaofeng and Chai Zhifang, 1989; Marshall and Middleton, 1990) have been noted in samples from near the Rawtheyan-Hirnantian boundary and close to the base of the persculptus zone. Significant numbers of brachiopod extinctions may have taken place throughout the late Rawtheyan to early Rhuddanian, although the majority seemingly occurred at the levels indicated. Prominent trilobite mass mortalities appear to have taken place near the end of the Hirnantian Stage in the tropics and at the Rawtheyan-Hirnantian boundary outside the tropics. Conodont mass mortality or faunal turnover occurred at about the same time the graptolites commenced reradiation during the time of the persculptus zone. The prominent graptolite mass mortality took place close to the end of the Rawtheyan, as did that of the chitinozoa. Mass mortality among corals occurred in the tropics near the Rawtheyan-Hirnantian boundary as sea-levels dropped significantly. The patterns in mass mortality and reradiation differ from organism to organism, depending on their mode of life and tolerance to change in the physical environment. As Wilde and Berry (1984) proposed, significant faunal changes took place near both the beginning and the end of glaciation. Organisms responded to major changes in ocean circulation and thermohaline density stratification at those times. Middleton, 1990) using 12C and 13C indicate that a significant sequestering of 12C in sediment took place at about the Rawtheyan-Hirnantian Stage boundary. Hirnantian brachiopod shells are enriched in 13C. Wang Xiaofeng and Chai Zhifang (1989) described 13C enrichment in the Hirnantian in their study of dark shales in the Ordovician-Silurian boundary interval in south China. Faunal and geochemical studies appear to be consistent in indicating a marked biomass change near the Rawtheyan-Hirnantian boundary. Slowed rates of origination characterized most organismal stocks during the Hirnantian Stage and Glytograptus persculptus zone. Extinction rates slowed in post-Rawtheyan-Hirnantian Stage boundary interval time, but slow origination rates during the Hirnantian into earliest Silurian resulted in marked faunal changes between the Late Ordovician and Early Silurian. Recovery and reradiation were slow until sea-level rose significantly such that many shelf sea habitats not only reopened but became stable during the early part of the Silurian. Much of the Hirnantian and subsequent Early Silurian Rhuddanian was typified by environmental instabilities resulting from glaciation followed by relatively rapid global warming. Each major organismal stock responded to these environmental changes somewhat specifically, depending on its tolerances for the environmental changes. REFERENCES Barnes, C. R., and S. M. Bergstrom (1988). Conodont biostratigraphy of the uppermost Ordovician and lowermost Silurian, in A Global Analysis of the Ordovician-Silurian Boundary, L. R. M. Cocks and R. B. Rickards, eds., British Museum (Natural History) Bulletin 43 (Geology Series), pp. 325-343. Barnes, C. R., and S. H. Williams, eds. (1991). Advances in Ordovician Geology, Geological Survey of Canada Paper 909, 336 pp. Berner, R. A. (1981). A new geochemical classification of sedimentary environments, Journal of Sedimentary Petrology 51, 359-365. Berry, W. B. N., and A. J. Boucot (1973). Glacio-eustatic control of Late Ordovician-Early Silurian platform sedimentation and faunal change , Geological Society of America Bulletin 84, 275-284. Berry, W. B. N., P. Wilde, and M. S. Quinby-Hunt (1987). The oceanic non-sulfidic oxygen minimum zone: A habitat for graptolites? Geological Society of Denmark Bulletin 35, 103-114. Berry, W. B. N., P. Wilde, and M. S. Quinby-Hunt (1990). Late Ordovician mass mortality and subsequent Early Silurian reradiation, in Extinction Events in Earth History, E. G. Kauffman and O.H. Walliser, eds., Springer-Verlag, Berlin, pp. 115-123. Beuf, S., B. Biju-Duval, O. de Chapperal, R. Rognon, O. Gariel, and A. Bennacef (1971). Les Gres du Paleozoique inferieur au Sahara—Sedimentation et discontinuities, evolution structurale d'un Craton, Institute Francais Petrole—Science et Technique du Petrol 18, 464 pp.

OCR for page 34
Effects of Past Global Change on Life Brenchley, P. J. (1988). Environmental changes close to the Ordovician-Silurian boundary, in A Global Analysis of the Ordovician-Silurian Boundary, L. R. M. Cocks and R. B. Rickards, eds., British Museum (Natural History) Bulletin 43 (Geology Series), pp. 377-385. Brenchley, P. J., and G. Newall (1984). Late Ordovician environmental changes and their effect on faunas, in Aspects of the Ordovician System, D. L. Bruton, ed., Palaeontological Contributions from the University of Oslo No. 295, pp. 65-79. Brenchley, P. J., M. Romano, T. P. Young, and P. Storch (1991). Hirantian glacio-marine diamictites—Evidence for the spread of glaciation and its effects on upper Ordovician faunas, in Advances in Ordovician Geology, C. R. Barnes and S. H. Williams, eds., Geological Survey of Canada Paper 90-9, 325-336. Briggs, D. E. G., R. A. Fortey, and E. N. K. Clarkson (1988). Extinction and the fossil record of the arthropods, in Extinction and Survival in the Fossil Record, G. P. Larwood, ed., Clarendon Press, Oxford, pp. 171-209. Cocks, L. R. M. (1988). Brachiopods across the Ordovician-Silurian boundary, in A Global Analysis of the Ordovician-Silurian Boundary, L. R. M. Cocks and R. B. Rickards, eds., British Museum (Natural History) Bulletin 43 (Geology Series), pp. 311-315. Cocks, L. R. M., and R. B. Rickards, eds. (1988). A Global Analysis of the Ordovician-Silurian Boundary, British Museum (Natural History) Bulletin 43 (Geology Series), 394 pp. Fortey, R. A. (1989). There are extinctions and extinctions: Examples from the lower Palaeozoic, Philosophical Transactions of the Royal Society of London B325, 327-355. Gallardo, A. (1963). Notas sobre la densidad de la fauna bentonica en el sublittoral del norte de Chile, Gayana Zoologica 8, 3-15. Grahn, Y. (1988). Chitinozoan stratigraphy in the Ashgill and Llandovery, in A Global Analysis of the Ordovician-Silurian Boundary, L. R. M. Cocks and R. B. Rickards, eds., British Museum (Natural History) Bulletin 43 (Geology Series), pp. 317-323. Kastner, M. (1983). Origin of dolomite and its spatial and chronological distribution—A new insight, American Association of Petroleum Geologists Bulletin 67, 2156. Koren, T. N. (1991). Evolutionary crisis of the Ashgill graptolites, in Advances in Ordovician Geology, C. R. Barnes and S. H. Williams, eds., Geological Survey of Canada Paper 90-9, 157-164. Lesperance, P. J. (1988). Trilobites, in A Global Analysis of the Ordovician-Silurian Boundary, L. R. M. Cocks and R. B. Rickards, eds., British Museum (Natural History) Bulletin 43 (Geology Series), pp. 359-376. Libes, S. M. (1992). An Introduction to Marine Biogeochemistry, John Wiley & Sons, Inc. , New York, 733 pp. Marshall, J. D., and P. D. Middleton (1990). Changes in marine isotopic composition in the Late Ordovician glaciation, Journal of the Geological Society London 147, 1-4. Melchin, M. J., and C. E. Mitchell (1988). Late Ordovician mass extinction among the Graptoloidea, in Abstracts of the Fifth International Symposium on the Ordovician System, H. S. Williams and C. R. Barnes, eds., St. John's, Newfoundland, p. 58. Middleton, P., J. D. Marshall, and P. J. Brenchley (1988). Isotopic evidence for oceanographic changes associated with the Late Ordovician glaciation, in Abstracts of the Fifth International Symposium on the Ordovician System, H. S. Williams and C. R. Barnes, eds., St. John's, Newfoundland, p. 59. Quinby-Hunt, M. S., P. Wilde, W. B. N. Berry, and C. J. Orth (1988). The redox-related facies of black shales, Geological Society of America Abstracts with Programs 20, A193. Quinby-Hunt, M. S., P. Wilde, C. J. Orth, and W. B. N. Berry (1989). Elemental geochemistry of black shales—Statistical comparison of low-calcic shales with other shales, in Metalliferous Black Shales and Related Ore Deposits-Program and Abstracts, R. I. Grauch and J. S. Leventhal, eds., U.S. Geological Survey Circular 1037, pp. 8-15. Quinby-Hunt, M. S., W. B. N. Berry, and P. Wilde (1990). Chemo-facies in low-calcic black shales, Abstracts of Papers, 13th International Sedimentological Congress, 444. Rickards, R. B. (1988). Graptolite faunas at the base of the Silurian, in A Global Analysis of the Ordovician-Silurian Boundary, L. R. M. Cocks and R. B. Rickards, eds., British Museum (Natural History) Bulletin 43 (Geology Series), pp. 345-349. Rickards, R. B., J. E. Hutt, and W. B. N. Berry (1977). Evolution of the Silurian and Devonian Graptoloids, British Museum (Natural History) Bulletin 28 (Geology Series), 120 pp. Rognon, P., B. Biju-Duval, and O. de Charpal (1972). Modeles glaciaires dans l'Ordovicien superior saharien: Phases d'erosion et glacio-tectonique sur la bordure nord des Eglab, Revue Geographie Physical Geologie Dynamique 14, 507-527. Sheehan, P. M. (1973). The relation of Late Ordovician glaciation to the Ordovician-Silurian changeover in North American brachiopod faunas, Lethaia 6, 147-154. Sheehan, P. M. (1982). Brachiopod macroevolution at the Ordovician-Silurian boundary , in Third North American Paleontological Convention Proceedings 2, B. Mamet and M. Copeland, eds., pp. 477-481. Sheehan, P. M., and P. J. Coorough (1990). Brachiopod zoogeography across an Ordovician-Silurian extinction event, in Palaeozoic Palaeogeography and Biogeography, W. S. McKerrow and C. R. Scotese, eds., The Geological Society London Memoir 12, pp. 181-187. Vaslet, D. (1990). Upper Ordovician glacial deposits in Saudi Arabia, Episodes 13, 147-161. Wang Kun, B. D. E. Chatterton, C. J. Orth, M. Attrep, Jr., and Jijin Li (1990). Geochemical analyses through the Ordovician-Silurian mass extinction boundary, Anhui Province, South China, Geological Society of America Abstracts with Program 11, 87. Wang Xiaofeng, and Chai Zhifang (1989). Terminal Ordovician mass extinction and discovery of iridium anomaly—An example from the Ordovician-Silurian boundary section, eastern Yangtze Gorges area, China, Progress of Geosciences of China 1985-1988, Vol. III, Geological Publishing House, Beijing, pp. 11-16. Wilde, P., and W. B. N. Berry (1984). Destabilization of the ocean density structure and its significance to marine "extinc

OCR for page 34
Effects of Past Global Change on Life tion" events, Palaeogeography, Palaeoclimatology, Palaeoecology 48, 143-162. Wilde, P., W. B. N. Berry, M. S. Quinby-Hunt, C. J. Orth, L. R. Quintana, and J. S. Gilmore (1986). Iridium abundances across the Ordovician-Silurian stratotype, Science 233, 339-341. Wilde, P., M. S. Quinby-Hunt, and W. B. N. Berry (1990). Vertical advection from oxic or anoxic water from the main pycnocline as a cause of rapid extinction or rapid radiation, in Extinction Events in Earth History, E. G. Kauffman and O. H. Walliser, eds., Springer-Verlag, Berlin, pp. 85-98.