Drawing a Line in the Sand: Identifying and Characterizing Boundaries in the Geological Record

Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

Drawing a line in the sand: identifying and characterizing boundaries in the geological record

ALWYNNE B. BEAUDOIN 1 & MARTIN J. HEAD 2 1 Quaternary Environments, Provincial Museum of Alberta, 12845-102nd Avenue, Edmonton, Alberta, T5N OM6, Canada (e-mail: [email protected]) 2 Department of Geography, University of Cambridge, Downing Place, Cambridge, CB2 3EN, UK (e-mail: [email protected])

Abstract: The identification and characterization of boundaries is a fundamental activity in geoscience. Spatial and temporal boundaries are rarely sharp but are more usually zones of transition, which may have variable characteristics. The examination of palynological and micropalaeontological data is often crucial for the delineation of geological boundaries, especially for the definition of Global Boundary Stratotype Sections and Points (GSSPs). The sixteen papers in this volume highlight many productive methodological approaches to boundary identification. This essay reviews the theoretical background to boundary identification in geology, and provides the contextual perspective for the subsequent papers.

Much of geoscience has to do with classification, guidelines such as the North American Strati- of organisms, of rocks, or of processes. Inevi- graphic Code (North American Commission on tably, classification devolves into an exercise in Stratigraphic Nomenclature 1983), or the Strati- drawing boundaries: putting a limit on what is graphic Procedure (Rawson et al. 2002). The included and what is not. In the geological demarcation of boundaries is generally a con- record, this exercise takes place in three dimen- sequence of the identification of units, rather sions, with space and time interacting to produce than the other way round. Historically, the a plethora of boundaries and boundary defini- identification of boundaries has often arisen tions. Although conventionally, on stratigraphic from the practical exigencies of geological charts, for example, temporal boundaries are mapping: the need to produce visual representa- often shown as sharp lines, in reality they are tions of geology, primarily based on sections or rarely so. Similarly, spatial boundaries on maps surface exposures. Boundaries, therefore, were are seldom as abrupt as they are drawn. Process often synonymous with the limits of mappable boundaries are yet more subtle and usually less units, that were generally field-identified on the clear. Indeed, even to speak of 'a boundary' is basis of lithology. This approach gave rise to the contentious, since boundaries come in many designation of 'body stratotypes', a methodol- forms and types. Boundaries are more often ogy that proved limiting for the subdivision of intervals or spaces of transition, leaving plenty geological time because of gaps in the rock of room for argument over their placement. record (Harland et al. 1990). "erhaps for this reason, some of the more Modern geochronology emphasizes bound- truculent and long-lasting debates in geology aries, especially lower stage boundaries, rather have been over the positioning and identification than units, as fundamental for subdivision and of boundaries. Drawing a line in the sand may correlation (Remane 2003). Under the auspices seem a simple exercise but turns out to be of the International Commission on Stratigra- fraught with complexities! phy (ICS), various subcommissions have been established to examine the placement of parti- ]oundary definitions cularly critical temporal boundaries in the geological record (see list in Gradstein & Ogg In geoscience, the criteria for identifying chron- 2003). Considerable effort has been devoted to ostratigraphic, biostratigraphic and lithostrati- the establishment of Global Boundary Strato- graphic units are outlined in rules set up by type Sections and Points (GSSPs), with more international agreement, such as the Interna- than 40 defined so far (Gradstein et al. undated). tional Stratigraphic Guide (Salvador 1994), or For each boundary, a type section and a specific

From: BEAUDOIN, A.B. & HEAD, M.J. (eds) 2004. The Palynology and Micropalaeontology of Boundaries. _Geological Society, London, Special Publications, 230, 1-10. 0305-8719/04/$15 @ The Geological Society of l,nndc)n "~OOt Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

2 A. B. BEAUDOIN & M. J. HEAD

point within that section are chosen by interna- which this volume is primarily concerned. tional agreement as a means to define the However, it is worth noting that some geological boundary formally. The type section is therefore boundaries may be a consequence of investiga- a particular section at an identified location (ICS tive boundaries, where research or analyses are 2004). Although parastratotype sections may be constrained by jurisdictional or political limits, established at considerable convenience to local such as those that confine national geological and regional stratigraphers, ultimately all surveys. Establishing boundaries may be more securely identified boundaries must be correla- than an esoteric exercise; boundary definitions table to the GSSP. In practice, almost all GSSPs may have important management implications, are defined from shallow-marine sediments. as in the establishment of ecological reserves, or Macrofossil biostratigraphy is used for many legal consequences, as in the definition of ~SSP definitions, particularly in the Mesozoic continental margins (e.g. Hedberg 1979). (see list in ICS 2004). However, changes in Spatial boundaries differ in type and degree. micropalaeontological indicators - especially Boundaries may be abrupt or gradual, solid or conodonts, calcareous nannofossils, and forami- permeable, permanent or ephemeral, constant or nifera - often underpin these definitions. Micro- fluctuating, stationary or moving, narrow or palaeontology is becoming increasingly broad, relatively straight or highly convoluted. important, particularly for GSSPs in the Cen- Where a boundary is defined by a physical ozoic and Palaeozoic, due to the more contin- landscape expression or environmental disconti- uous nature of microfossil recovery. Since it acts nuity, its character may affect the way in which as an exemplar, the selection of a type section is organisms react to it. Moreover, conditions that often a difficult matter, and candidate sections form a boundary for one type of organism may are minutely scrutinised (see Sikora et al. and have no impact on another. Physical boundaries Mei et al.). The establishment of a GSSP rarely can also act as a filter, only allowing certain ends discussion or extinguishes debate; GSSPs organisms to pass through. For these reasons, may be challenged, re-examined, and defended the identification and characterization of a (Zhang & Barnes a). boundary through biotic indicators may depend on what organism is being examined as a proxy ]oundary types and the sensitivity of that organism to change. Boundaries may also regulate flows of materials Surprisingly, despite all the attention paid to or energy (Cadenesso et al. 2003b). Where a boundaries, there is no cohesive theoretical boundary is defined by a perturbation, the treatment of them in the geological literature, magnitude, extent, and duration of that pertur- although stratigraphic principles have been bation may influence biotic response. Some widely discussed (e.g. Hedberg 1976; Salvador biota may show considerable complacency or 1994; Remane 2003; Walsh 2004). This situation resiliency until critical threshold values are differs from ecology, where there is an emerging crossed. body of literature dealing with boundary theory (see Gosz 1991, and Cadenasso et al. 2003a, 2003b). Because ecology deals with living organisms and their environment, this theoretical Analogues from modern ecological and approach is also applicable to palynology and environmental boundaries micropalaeontology. Strayer et al. (2003) distinguish two major types of boundaries: investiga- The identification of boundaries on the modern tive and tangible boundaries. Investigative landscape provides many examples of these boundaries are those that are imposed by different types. Here, we can examine bound- practical or administrative considerations and aries in the simplest spatial cases, with the often have no physical expression in the land- complication of geological time removed. How- scape. Investigative boundaries can also occur ever, the identification of boundaries across the when language barriers preclude discussion or landscape, and the characterization of transi- access to information. Nikitenko & Miekey, for tions, provides analogies for the 'space for time' example, review an enormous quantity of substitution which is the foundation of the literature from Russia, which has not hitherto 'present is the key to the past' approach to been accessible to people who do not read geoscience. Some boundaries are obvious, the ~,ussian. Spatially speaking, tangible boundaries land-sea transition, for example, and have long are associated with some biotic change, environ- been a focus of research. State-change bound- mental discontinuity, or landscape expression. It aries, such as the water-atmosphere or ice-water is the exploration of tangible boundaries with interfaces, are also clearly marked. Other Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

IDENI'IFYING AND CHARACTERIZING BOUNDARIES 3

boundaries, such as temperature limits, are more these limits, in the krummholz zone, trees often subtle and less visible. These boundaries are occur in isolated 'tree islands' and may be often formed when the gradient or rate-of- deformed, flagged, or stunted. Where elevational change of a parameter becomes steep (Cade- controls are sharp, as on steep slopes, the alpine nesso et al. 2003a). Process boundaries have treeline zone can be quite narrow, in contrast to been a research focus for decades, especially the northern boreal treeline zone, which is often with respect to fluid dynamics in fluvial (Thornes broad. Here, an expanse of 'forest-tundra' is 1979) and aeolian (e.g. Walker & Nickling 2002) identified, which in northern Canada may be up systems. The recognition that unique suites of to about 200kin wide (Timoney et al. 1992). processes occur at boundaries also permeates Thus, the boundary between forested and non- meteorology and climatology (e.g. Oke 1993, forested areas represents the interplay of topo- 1997) and soil science (e.g. Belnap et al. 2003). graphical, latitudinal, biotic, temporal and Despite their definition with respect to the ecological factors. present landscape, spatial boundaries are usually Such definitional minutiae may seem dynamic and do often also incorporate a time picayune, but they have significant implications dimension, albeit short term. Boundaries that for interpretation of palynological data. Because are formed by physical discontinuities, such as altitudinal treeline shows a strong correlation breaks in slope, tend to have greater longevity with growing season temperature, many studies and be fixed to a specific landscape position. have attempted to characterize its modern Dther types of boundaries, including process position through palynological signatures, and and biotic boundaries, however, are often more then use these signatures to interpret past pollen mobile and short lived. They may oscillate or records as a 'proxy' temperature signal (e.g. fluctuate around a mean position on several Beaudoin 1986; Pellatt et al. 2000). There are scales from diurnal, such as tidal limits, to several problems to this approach, not least subseasonal, such as wetland margins (Shay & being the assumption that the present boundary Shay 1986; Kantrud et al. 1989), to seasonal, is in equilibrium with, or controlled by, current such as the active layer-permafrost boundary conditions and thus is a surrogate measure for (French 1993). Other ecological boundaries may contemporary conditions. But because tree move at decadal or century scales and can often response to a perturbation or a temperature be directional. These may include, for example, change may be lagged, this assumption may be the migration front for plant species, such as the suspect. The response may also vary depending post-glacial expansion of lodgepole pine in on the floristic composition of the treeline, western North America (MacDonald & Cwynar because not all taxa will necessarily react in the 1985). Boundaries which are most likely to have same way. Computational issues may also over- correlates in the geological record are ones simplify the situation. Usually, some form of where the mean position is relatively fixed for regression between modern pollen signatures a long time, or where the directional shift is and growing season temperatures is performed accompanied by distinctive biotic changes. and the results used to interpret signatures from Of all modern ecological boundaries, perhaps past records. This approach forces a one-to-one none is as well studied as the treeline, both the correspondence (albeit with some probability alpine (altitudinal) treeline and the northern attached) with the values from the past record, boreal (latitudinal) treeline (e.g. Frenzel et al. and does not allow for the fact that the character 1996; Arno & Hammerly 1984). This boundary of the treeline may have varied over time, for exemplifies many of the issues of identification example, through floristic change. The inferred and characterization that pervade ecological and record may therefore appear more sharply geological boundary definition. Wardle (1974, p. focused and delineated than it probably was. 396) emphasized that 'timberline is the sharpest Modern analogies are always laden with such temperature-dependent boundary in nature', ambiguities, but nevertheless remain the best and thus its location is climate-related. But, approach for ecological or environmental because of the physiological plasticity and boundary characterization. response of trees to environmental stress, the treeline is not, in fact, a sharp line, but rather a Geological boundaries zone representing a complex transition between forested and treeless areas. Strictly, the alpine In the geological past, boundary identification treeline (or tree limit) is the upper elevation limit becomes even more complex with the addition of of krummholz forms of nominally arboreal the time dimension. The same dichotomies species, and the timberline is the limit of apply, although we can also add swift or slow, standin~ tree ~rowth (Wardle 1974). Between abrupt or ~radual, continuous or gapped to the Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

4 A. B. BEAUDOIN & M. J. HEAD mix. We can also distinguish between abiotic a discontinuity, it may also act as a filter, adding (lithostratigraphic) and biotic (biostratigraphic) further complications to the recognition of boundaries. Biostratigraphic boundaries are indicator taxa. The incorporation of biotic analogous to modern ecological boundaries remains in sediments may be complicated by because they are defined by changes in biota. taphonomic factors, such as downslope trans- These could, of course, coincide with lithostrati- port (Zhang 8,: Barnes a), and by post-deposi- graphic changes, but this need not be the case. tional factors affecting preservation (Van These boundaries can be both spatially and Eetvelde & Dupuis). temporally transgressive, and may be of local or Sampling factors also influence the potential regional significance. Critical boundaries, such for boundary identification. First and last as those that define stages, must be of more than appearance of taxa in a stratigraphic record local extent (Remane 2003). This can make their may be influenced by how much of the record identification in geological records problematic, has been examined, and are fundamentally not least because geological studies often have a statements of probability, associated with con- very limited spatial view. Often, boundary fidence limits (Holland 2003). Over millennia, inferences are made from samples obtained geological processes, primarily erosion, reduce from one or a few cores or boreholes. So the the amount of any exposed stratigraphic unit. geological or subsurface view of boundaries is Hence, only some of the original sediment, and often highly patchy and discontinuous and therefore part of the original variability, is represents only a limited subset of the available available for examination. Thus, the probability environmental variance. Where multiple views of finding remains of any organism is influenced are available (e.g. Zhang & Barnes a, Nikitenko by how much of the sediment unit is left. Peters & Miekey), it may be important to tease out the & Foote (2002), for example, suggest that the ecological variability, such as shallow v. deep evaluation of extinction rates may be influenced water or warm water v. cool water, so that the by these sampling factors. Using data from more substantive regionally or globally signifi- marine sedimentary formations, they conclude cant variation can be clearly identified. that derived extinction rates may be spurious, an artefact of the amount of the stratigraphic record available for examination. It may be Techniques for boundary identification more difficult therefore to identify boundaries Boundaries in the geological record are identi- within intervals for which the sedimentary fied using many techniques, but the examination record is slender or poorly preserved. of palynological and micropalaeontological data Numerical or statistical methods are often is often crucial. Perhaps the easiest boundaries useful in distinguishing meaningful and consis- to identify are those where there is an abrupt tent biotic changes. These methods are most lithostratigraphic change or where there is a applicable where large data-sets are available, clear temporal break in the record - as at an and they can help to identify patterns in erosional unconformity. The identification of otherwise overwhelming amounts of data. chronostratigraphic boundaries may be more Numerical methods can be used to answer two difficult, and in some cases relies on biotic important biostratigraphic questions. First, events. Thus, chronostratigraphic and biostrati- given changing assemblages through time or up graphic boundaries may coincide. Significant a stratigraphic section, where are the significant biostratigraphic changes are often marked by breaks or most important changes? Such judge- the appearance or disappearance of important ments can be made qualitatively, by visual indicator taxa or, more often, assemblage inspection for instance, but numerical methods changes. have the advantage of being reproducible, given The reliance on indicator taxa for boundary the same data-set, and following defined rules. definition raises questions of consistency, fidelity Numerical zonation methods have been widely and reproducibility, especially when these taxa employed in Quaternary palynology to identify form only a small part of an assemblage. For assemblages (e.g. Birks & Gordon 1985), espe- example, in a multi-core pollen study of a well- cially when stratigraphic constraints are defined Holocene temporal level from Lake included (Grimm 1987). Pollen zone boundaries D'Hara (Beaudoin & Reasoner 1992) showed may be ecologically or chronostratigraphically that minor taxa (present as less than 1% of the important (e.g. Fernfindez-Marr6n et al.). Sec- assemblage) were unreliable indicators, because ond, given several assemblages and some knowl- their occurrence was highly variable and they edge of contemporaneous environmental did not show high fidelity between contempora- conditions, can we make inferences about neous samples. Where a boundary is defined by assemblage-environment linkages or controls? Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

IDEN IIFY INtJ AND CHARACI ERIZIN(5 BOUNDARIES 3

Such environmental inferences can then be Palaeozoic and the diversification of phyto- extended to assemblages where the contempora- plankton in the Late Triassic. This open: neous conditions are not well known. Ordina- intriguing possibilities of linkages and feedback tion or classification methods have been used to mechanisms between evolutionary events, cli- explore these patterns in palynological or mate changes, and extinctions - perhaps point- micropalaeontological data, using techniques ing the way to a more integrated theoretical such as cluster analysis or principal components perspective. analysis (see Kovach 1989). These techniques Doran et al. also adopt a 'big picture' can often highlight environmental shifts, such as approach and also examine evolutionary trends, sea-level or water temperature changes, in in this case in post-extinction planktonic for- complex data. Several studies in this volume, aminifera, concentrating on two boundaries, the including Zhang & Barnes a, Elewa & Morsi, Cenomanian-Turonian (C-T) and the Cretac- and Nikitenko & Miekey, use numerical methods eous-Tertiary (K-T). Their concern is with the for this purpose. rate and pattern of faunal recovery after extinction. If extinction is a boundary, it can be regarded as the quintessential filter, only Using palynology and micropalaeontology to allowing certain life-forms to pass. What hap- pens to the taxa that make it through extinction? identify boundaries in the geological record Doran et al.'s analysis shows rapid evolutionary From the widest perspective, there are certain change in foraminifera following these major biological events in the geological past that extinction events. They suggest that this pattern stand out as being pivotal: the occurrence of the relates to specific characteristics, which they first fossils, the appearance of the first terrestrial term 'passport' characteristics, that allowed fauna, the rise of flowering plants. However, no some taxa to pass through or survive the events have arguably generated more debate extinction events. These taxa may not necessa- than mass extinctions and their causes. By their rily be best adapted to the post-extinction very nature, extinctions, reflecting significant environment, but they form the foundation for changes in biota, are often defining events for subsequent populations. boundaries. MaeLeod takes this 'big picture' After much debate, the GSSP for the Cam- approach by examining the pattern and periodi- brian-Ordovician boundary was located at city of extinction events shown by marine Green Point, western Newfoundland, Canada, invertebrate genera through the sweep of Pha- and defined on the basis of conodont biostrati- nerozoic time. He investigates the association graphy (Cooper et al. 2001). This GSSP was between these and the five most widely invoked contentious because of differing opinions about explanatory factors - bolide impacts, continen- the conodont record, in particular the degree of tal flood-basalt eruptions, eustatic changes, and transport and mixing. To clarify this issue, marine anoxia events. His analysis shows that Zhang & Barnes a analyse conodont commu- two factors - marine regression and volcanic nities associated with different environmental eruptions - explain most of the observed events. settings, from shallow platform to distal slope, MaeLeod observes that terrestrial mechanisms across this boundary in western Newfoundland. provide sufficient control to account for most They use multivariate statistical techniques to extinction events through the Phanerozoic. The identify consistent patterns and gradational popular-culture image of extinction, exemplified relationships, and show that conodont commu- by startled dinosaurs staring skywards as a nities were partitioning the environment accord- flaming fireball approaches, apparently needs ing to slope position and water depth. Therefore, some revision! conodont community change may reflect sea- MacLeod's analysis of extinction intensity level changes affecting water depth, in particular through time also leads to some provocative sea-level rise in the Early Ordovician. Over- suggestions. Rather than being a steadily declin- printed on this are community changes resulting ing trend through the Phanerozoic, MaeLeod from rapid evolution and diversification. Both shows that the extinction intensity became factors are important to an evaluation of the marked around the end of the Devonian, with conodont record across this boundary. a notable reduction in extreme-intensity events The end of the Ordovician is marked by the around the end of the Triassic. MaeLeod links second-most severe mass extinction in the these to changes in the global carbon cycle as a Phanerozoic, resulting in the estimated loss of consequence of evolutionary events. He offers some 85% of all marine species (Jablonski 1991; two possibilities for the causal mechanism: the Sheehan 2001). This was brought about by a diversification of land plants in the Late continental ~laciation in North Africa that Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

6 A. B. BEAUDO1N & M. J. HEAD resulted in cooler oceanic temperatures and a borehole in the southern North Sea, McLean more aerated bottom waters: changes that had et al. have been able to assess the potential of profound implications for marine organisms miospores for recognizing the boundary else- (Hallam & Wignall 1999; Sheehan 2001). Zhang where. Results show that although the Vander- & Barnes b have examined the depletion of beckei Marine Band and can be recognized conodont taxa at the end of the Ordovician, and palynologically, a clear floristic break does not particularly the nature and timing of their post- occur at its base. Indeed, a gradual turnover of extinction recovery. Their analysis of deposits of taxa occurs across the boundary. McLean eta[. Early Llandovery (earliest Silurian) age from the conclude that marine flooding events in the essentially complete succession on Anticosti Upper Carboniferous coal measures may not Island, Quebec, has revealed an unexpectedly exclusively provide the impetus for evolutionary complex series of speciation, extinction, immi- change in the flora, which apparently continued gration and emigration events that can be in part to evolve throughout transgressive-regressive correlated to eustatic and other ocean-climate cycles. changes. The Anticosti Basin may indeed have In the Lower Jurassic, the Early Toarcian was been an important centre of evolutionary radia- characterized by a mass-extinction event, tion for conodont animals during the earliest marked by rising sea-levels with widespread Silurian. anoxia, and consequent deposition of black Taxonomic matters often lie at the heart of shales (Hallam & Wignall 1999; Harries & Little boundary definitions, as in the study by Mei et 1999). Nikitenko & Mickey review studies on a al. They present the case for establishment of a broad regional scale from Russia and Alaska, ,~SSP for the base of the Changhsingian Stage concentrating especially on foraminifera and (Upper Permian) at Meishan, China, based on ostracodes through the Pliensbachian-Toarcian the analysis of conodonts. They suggest that the interval. Numerical analysis allows them to boundary should be defined at the first appear- identify biogeographical units of ostracodes ance of Clarkina wangi. This species was and foraminifera within the Arctic and Boreal- established following the authors' examination Atlantic realms. Nikitenko & Mickey distinguish of conodonts from the Meishan section, using consistent patterns of zonation within ostracode characteristics of the sample-population rather and foraminifera that allow correlation both than individuals to distinguish taxa: focusing on across this region and with the microfossil carinal morphology. Taxa are defined when the sequence from Western Europe. In the Early population as a whole exhibits a predominance Toarcian in the Arctic, ostracode genera and of the particular morphology. In this instance, families completely changed, and more than the transition from one carinal configuration to 80% of the foraminiferal species were replaced. another occurs over a narrow interval, allowing Interestingly, the authors note that the reduction the boundary to be relatively well constrained. in species and generic diversity began in the Late Their sample-population analysis is an interest- Pliensbachian in the Arctic - somewhat earlier ing way of approaching a question that is than in Western Europe. relevant to many micropalaeontological indica- The exact placement of the Jurassic-Cretac- tors - how much morphological variation can be eous boundary remains contentious, largely allowed before a new taxon is recognized? owing to difficulties in correlating between The Carboniferous System is internationally Tethyan and Boreal realms. These realms occur defined (by GSSPs) at its top and base, but the at a time of marked provincialism in the marine inter-regional correlation of stages continues to biota, brought about by low global sea-levels. present challenges. The Duckmantian is a The lowermost stage of the Cretaceous is the 2uropean stage within the lower Upper Carbo- Berriasian, whose stratotype is at Berrias, niferous. It overlies the Langsettian Stage, and France. Correlating this stratotype beyond the its base is defined by the base of the Vander- Tethys has been problematic, owing to the beckei Marine Band and in a boundary strato- localized nature of the ammonite fauna. Hunt type in Derbyshire, England, UK. The position has used dinoflagellate cyst stratigraphy at of this marine band has been correlated with classic sites in Dorset, southern England, UK, important changes in macro- and microfloral as a means of correlating these sites with the assemblages across northwestern Europe. type Berriasian. Dorset was within the Tethyan McLean et al. have analysed the Duckmantian Realm, but close to the northern Boreal Realm. stratotype in detail, using miospores to char- This has allowed Hunt to use miospores to acterize the Vanderbeckei Marine Band and the correlate the Dorset sections with the Terschel- =angsettian-Duckmantian boundary in particu- ling Basin in the Netherlands, which is within lar. By then analvsin~ this same marine band in the Boreal Realm. This approach achieves a Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

IDEN lIP ~ INIJ AND CHARAC l ERILIN(5 BOUNDARIES / novel correlation between the Berriasian, includ- although the scale and rate of biotic extinctions ing its base (and hence the Jurassic-Cretaceous is debated. boundary), and the ammonite biostratigraphy of An intriguing aspect of the K-T transition is the Boreal Realm. the contrast between North American (and Hart's historical account of the recognition of Pacific) and European pollen records. North a mid-Cenomanian non-sequence in the chalk American records reveal a pronounced extinc- succession of southern England and northern tion event near the boundary (Nichols 1996), France is a salutary reminder of the practical followed at some sites by a brief increase in fern application of boundary problems. The non- abundance (the 'fern spike') that is thought to sequence was originally identified during geolo- relate to short-term environmental disturbance gical studies, especially micropalaeontological (Tschudy et al. 1984; Fleming & Nichols 1990; work on the foraminifera, associated with the Sweet 2001). In contrast, no appreciable changes development of the Channel Tunnel. Work have been recorded in pollen records from the throughout southern England and northern Old World, including Europe. To test this France during subsequent decades showed that apparent lack of change, Fernfindez-Marr6n et this non-sequence represents a hiatus, in some al. have analysed spore and pollen data from places an unconformity, of regional extent. two sections that span the K-T transition in Above it, the foraminifera include a greater Spain. Because these sections represent differing abundance of planktonic forms, indicating more palaeoenvironmental settings, any effects caused open-marine conditions. This study raises some by local factors and taphonomy should be fascinating issues with respect to boundary detected. While no noticeable extinctions could definitions. Should this non-sequence or deposi- be linked to the boundary, a statistical analysis tional break be considered a sequence bound- reveals significant differences in assemblage ary? Can a boundary be formed by an absence? composition. These changes, which include an The definition of a GSSP relies on its identifica- increase in trilete fern spores across the K-T tion within a conformable sequence (Walsh transition and a reduction in the Danian 2004), but Hart's study shows that other types samples, offer a new means to identify the of boundaries may also be useful in regional K-T boundary in terrestrial deposits of the correlations. region. Macrofossil biostratigraphy provides one line Dinoflagellate cysts are relatively unaffected of evidence for Sikora et al. as they tackle an by short-term environmental disruption, owing aspect of the Jurassic-Cretaceous boundary- to their ability to remain dormant for several the establishment of a GSSP. To be useful as a years. This presumably explains their continuous standard, a stratotype must be continuous record across the K-T boundary, which makes across the interval, be widely correlatable, and them well suited for studying environmental be accessible to the scientific community. Sikora changes through this interval (Brinkhuis et al. et al. examine and compare the macrofossil and 1998). Most sites presently studied represent microfossil biostratigraphy of two sections shelfal facies (e.g. Brinkhuis & Leereveid 1988; proposed as potential GSSPs: at Wagon Mound, Brinkhuis & Schioler 1996; Brinkhuis et al. New Mexico, USA, and Salzgitter-Salder, cen- 1998). The study by Gedl a examines dinofla- tral Germany. Sikora et al. are able to show that gellate cysts at a deep-water site in the Czech different indicator types provide different tem- Republic, thereby offering a new perspective on poral signatures, and that, more significantly, this critical boundary. Gedl a concludes that a the micropalaeontological indicators show a warm, stable, marine climate prevailed across different temporal pattern to macro fossil the boundary. Although no major changes in remains. Their study suggests that neither of assemblages were found, minor changes might the proposed stratotypes are likely to be good relate to gradual sea-level fall or increasing candidates for a GSSP. This example shows the nutrient availability. A peak abundance of value of using multiple indicators for boundary heterotrophic dinoflagellate cysts near the characterization. boundary appears to indicate upwelling in this The Cretaceous-Tertiary (K-T) boundary is part of the Tethys. perhaps the most studied of all geological The Palaeocene-Eocene transition is marked boundaries, and probably, because of its well- by significant changes in many Earth systems, publicized association with dinosaur extinction, including global climate, with a marked carbon the most well-known outside geoscience (see isotope anomaly, and ocean circulation, accom- Alvarez et al. 1980; Hildebrand 1993). It panied by palaeogeographical changes (Norris & represents a profound disruption to terrestrial R6hl 1999; Zachos et al. 2001). Both marine and and marine ecosystems on a ~lobal scale, terrestrial organisms show considerable evolu- Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

3 A. B. BEAUDOIN & M. J. HEAD tionary turnover through this interval (Berggren in Patagonia is associated with this event, ot al. 1998; Hallam & Wignall 1999). Despite the although its precise dating has remained abundant evidence for change, the establishment questionable. Guerstein et al. have used paly- of a GSSP for the Palaeocene-Eocene is nology to assign a Late Oligocene and early zurrently subject to vigorous debate (see Aubry Early Miocene age to the marine Centinela et al. 1999; Walsh 2004). Two studies in this Formation deposited near the margin of this volume shed further light on this interval. transgression. A maximum flooding surface is Van Eetvelde & Dupius use diatoms to indicated by high ratios of dinoflagellate cysts examine the Upper Palaeocene to Lower Eocene in the earliest Miocene. interval in two localities from northern France. In conclusion, the sixteen papers summarized -Iere, examination of the record is complicated here highlight many methodological and defini- by preservation issues because the diatoms are tional issues with boundaries in geology. heavily pyritized, requiring specialized extrac- Although not all problems are resolved, the tion techniques. Nevertheless, three distinct papers point the way to productive investigative diatom assemblages can be recognized, and and analytical approaches that may prove these offer the prospect of correlation between worthwhile in other situations. the sequences of the Dieppe-Hampshire and North Sea basins. Elewa & Morsi also examine the Palaeocene- =~ocene interval, in this case based on ostracodes References recovered from sediment sequences in northeast ALVAREZ, L. W., ALVAREZ, W., ASARO, F. & Egypt. Numerical analysis permits the identifi- MICHEL, H. V. 1980. Extraterrestrial cause for cation of ecozones, distinguished by environ- the Cretaceous-Tertiary extinction. Science, 208, mental parameters, including water depth, 1095-1108. temperature, turbulence and dissolved oxygen ARNO, S. F. & HAMMERLY, R. P. 1984. Timberline: Mountain and Arctic Forest Frontiers. The Moun- content. Elewa & Morsi conclude that changes in taineers, Seattle, Washington, 304 pp. ostracode assemblages were mainly the result of AUBRY, M. P., BERGGREN, W. A., VAN COUVERING, changing local environmental conditions, rather J. A. & STEININGER, F. 1999. Problems in than speciation or extinction. However, they do chronostratigraphy: stages, series, unit and detect faunal changes probably associated with boundary stratotypes, global stratotype section the Palaeocene-Eocene thermal maximum (Ken- and point and tarnished golden spikes. Earth- nett & Stott 1991), which elsewhere is associated Science Reviews, 46(1-2), 99-148. with extinctions. BEAUDOIN, A. B. 1986. Using Picea/Pinus ratios from The Eocene-Oligocene transition is generally the Wilcox Pass core, Jasper National Park, Alberta, to investigate Holocene timberline fluc- characterized by falling temperatures and a drop tuations. Gdographie physique et Quaternaire, 40, in global sea-level. However, the boundary itself 145 152. is contentious, owing to disagreements in defini- BEAUDOIN, A. B. & REASONER, M. A. 1992. tion (Berggren et al. 1995, pp. 197-198; Brin- Evaluation of differential pollen deposition and khuis & Visscher 1995). It is marked either by pollen focussing at three Holocene intervals in the highest occurrence of the foraminiferal genus Lake O'Hara, Yoho National Park, British Hantkenina or that of the dinoflagellate cyst Columbia, Canada: intra-lake variability in pollen Areosphaeridium diktyoplokum. The latter percentages, concentration and influx. Review of datum, which is stratigraphically higher than Palaeobotany and Palynology, 75, 103 131. BELNAP, J., HAWKES, C. V. & FIRESTONE, M. K. the former, is used by Gedl b in his study of 2003. Boundaries in miniature: two examples dinoflagellate cysts from the Carpathian Moun- from soil. BioScience, 55, 739-749. tains of Poland. Gedl's study identifies the BERGGREN, W. A., KENT, D. V., SWISHER III, C. C. position of the Eocene-Oligocene boundary in & AUSRY, M.-P. 1995. A revised Cenozoic the Leluch6w section, and infers a drop in geochronology and chronostratigraphy. In: relative sea-level that might correlate with Early BERGGREN, W. A., KENT, D. V. & HARDENBOL, Oligocene eustatic lowering. Sea-surface tem- J. (eds) Geochronology, Time Scales and Global peratures are found to drop prior to the Eocene- Stratigraphic Correlation. SEPM Special Publica- Oligocene boundary at this site. tions 54, Tulsa, Oklahoma, 129-212. BERGGREN, W. A., LUCAS, S. & AUBRY, M.-P. 1998. The Oligocene-Miocene boundary represents Late Paleocene Early Eocene climatic and biotic one of the most important eustatic rises in the evolution: an overview. In: AUBRY, M.-P., Cenozoic, with high sea-levels continuing LUCAS, S. & BERGGREN, W. A. (ed.) Late throughout the early Early Miocene (Haq et Paleocene-Early Eocene Climatic and Biotic Events al. 1987, 1988; Hardenbol et al. 1998). In in the Marine and Terrestrial Records. Columbia South America, a major marine transgression University Press, New York, 1-17. Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

IDENIII ~ ~(IN~J AND CHARACIERILIN(5 BOUNDARIES 9

BIRKS, H. J. B. & GORDON, A. D. 1985. Numerical GRADSTEIN, F. M., FINNEY, S. C. & OGG, J. G. Methods in Quaternary Pollen Analysis. Academic undated. ICS on Stage. 15 pp, PDF manuscript, Press, New York, 317 pp. World Wide Web Address: www.stratigraphy. BRINKHUIS, H. & LEEREVELD, H. 1988. Dinoflagel- org/stage.pdf, last accessed 13 March 2004. late cysts from the Cretaceous/Tertiary boundary GRADSTEIN, F. M. & OGG, J. G. 2003. International sequence of E1 Kef, northwest Tunisia. Review of Union of Geological Sciences International Com- Palaeobotany and Palynology, 56, 5-19. mission on Stratigraphy (ICS) Consolidated BRINKHUIS, H. & SCHIOLER, P. 1996. Palynology of Annual Report for 2003. 130 pp, PDF manu- the Geulhemmerberg Cretaceous/Tertiary bound- script, World Wide Web Address: www. ary section (Limburg, SE Netherlands). Geologie stratigraphy.orE, last accessed 7 February en M~nbouw, 75, 193-213. 2004. BRINKHUIS, H. & VISSCHER, H. 1995. The upper GRIMM, E. C. 1987. CONISS: a Fortran 77 program boundary of the Eocene Series: a reappraisal for stratigraphically constrained cluster analysis based on dinoflagellate cyst biostratigraphy and by the method of incremental sum of squares. sequence stratigraphy. In: BERGGREN, W. A., Computers & Geosciences, 13, 13-35. KENT, D. V. & HARDENBOL, J. (eds) Geochronol- HALLAM, A. & WIGNALL, P. B. 1999. Mass extinc- ogy, Time Scales and Global Stratigraphic Corre- tions and sea-level changes. Earth-Science lation. SEPM Special Publications, 54, Tulsa, Reviews, 48, 217-250. Oklahoma, 295-304. HAQ, B. U., HARDENBOL, J. & VAIL, P. R. 1987. BRINKHUIS, H., BUJAK, J. P., SMIT, J., VERSTEEGH, Chronology of fluctuating sea levels since the G. J. M. & VISSCHER, H. 1998. Dinoflagellate- Triassic. Science, 235, 1156-1166. based sea surface temperature reconstructions HAQ, B. U., HARDENBOL, J. & VAIL, P. R. 1988. across the Cretaceous-Tertiary boundary. Palaeo- Mesozoic and Cenozoic chronostratigraphy and geography, Palaeoclimatology, Palaeoecology, cycles of sea-level change. In: WILGUS, C. K., 141, 67-84. HASTINGS, B. S., KENDALL, C. G.St. C., POSA- ~ADENESSO, M. L., PICKETT, S. T. A., WEATHERS, K. MENTIER, H. W., ROSS, C. A. & VAN WAGONER, C., BELL, S. S., BENNING, T. L., CARREIRO, M. J. C. (eds) Sea-Level Changes: an Integrated M. & DAWSON, T. E. 2003a. An interdisciplinary Approach. SEPM Special Publication, 42. Tulsa, and synthetic approach to ecological boundaries. Oklahoma, 71-108, plus one separate chart. BioScience, 53, 717-722. HARDENBOL, J., THIERRY, J., FARLEY, M. B., 2ADENESSO, M. L., PICKETT, S. T. A., WEATHERS, K. JACQUIN, T., DE GRACIANSKY, P.-C. & VAIL, C. & JONES, C. G. 2003b. A framework for a P. R. 1998. Mesozoic and Cenozoic sequence theory of ecological boundaries. BioScience, 53, chronostratigraphic framework of European 750-758. basins. In: DE GRACIANSKY, P. C., HARDENBOL, ~OOPER, R. A., NOWLAN, G. S. & WILLIAMS, S. H. J., JACQU1N, T. & VAIL, P. R. (eds) Mesozoic and 2001. Global Stratotype Section and Point for the Cenozoic' Sequence Stratigraphy of European base of the Ordovician system. Episodes, 24(1), Basins. SEPM Special Publication, 60, Tulsa, 19-28. Oklahoma, 3-29, plus 8 separate charts FLEMING, R. F. & NICHOLS, D. J. 1990. Fern-spore HARLAND, W. B., ARMSTRONG, R. L., COX, A. V., abundance anomaly at the Cretaceous-Tertiary CRAIG, L. E., SMITH, A. G. & SMITH, D. G. 1990. Boundary: a regional bioevent in western North A Geologic Time Scale 1989. Cambridge Uni- America. In: KAUFFMAN, E. G. & WALLISER, O. versity Press, Cambridge, UK, 262 pp. H. (eds) Extinction Events in Earth History. HARRIES, P. J. & LITTLE, C. R. S. 1999. The early Springer Verlag, Berlin and New York, 347-349. Toarcian (Early Jurassic) and the Cenomanian- ZRENCH, H. M. 1993. Cold-climate processes and Turonian (Late Cretaceous) mass extinctions: landforms. In: FRENCH, H. M. & SLAYMAKER, similarities and contrasts. Palaeogeography, O. (eds) Canada's' Cold Environments. McGill- Palaeoclimatology, Palaeoecology, 154, 39 Queen's University Press, Montreal and King- 66. ston, Canada, 143-167. HEDBERG, H. D. (ed.) 1976. International Strati- FRENZEL, B., BIRKS, H. H., ALM, T. & VORREN, K. graphic Guide." a Guide to Stratigraphic Classifica- D. (eds) 1996. Holocene Treeline Oscillations, tion, Terminology, and Procedure by International Dendrochronology and Palaeoclimate. Palfioklima- Subcommission on Stratigraphic Classification of forschung Band 20/Palaeoclimate Research lUGS Commission on Stratigraphy. Wiley, New Volume, 20, Special Issue: ESF Project 'European York, 200 pp. Palaeoclimate and Man' 13. Gustav Fischer HEDBERG, H. D. 1979. Ocean floor boundaries. Verlag, Stuttgart, Germany, x + 303 pp. Science, 204, 135-144. ~osz, J. R. 1991. Fundamental ecological character- HILDEBRAND, A. R. 1993. The Cretaceous/Tertiary istics of landscape boundaries. In: HOLLAND, M. boundary impact (or the dinosaurs didn't have a M., RISSER, P. G. & NAIMAN, R. J. (eds) chance). Journal of the Royal Astronomical Ecotones: the Role of Landscape Boundaries in Society of Canada, 87, 77-118. the Management and Restoration of Changing HOLLAND, S. M. 2003. Confidence limits on fossil Environments. Chapman and Hall, New York, ranges that account for facies changes. Paleobiol- 8-30. o,,v, 29, 468-479. Downloaded from http://sp.lyellcollection.org/ by guest on September 25, 2021

10 A. B. BEAUDOIN & M. J. HEAD

"CS (International Commission on Stratigraphy) 2004. RAWSON, P. F., ALLEN, P. M. et al. 2002. Strati- The Global Boundary Stratotype Section and graphical Procedure. Professional Handbook, Point (GSSP). World Wide Web Address: Geological Society of London, 57 pp. www.stratigraphy.org, last accessed 7 February REMANE, J. 2003. Chronostratigraphic correlations: 2004. their importance for definition of geochronologic JABLONSKI, D. 1991. Extinctions: a paleontological units. Palaeogeography, Palaeoclimatology, perspective. Science, 253, 754-757. Palaeoecology, 196, 7-18. T(ANTRUD, H. A., MILLAR, J. B. & VAN DER VALK, SALVADOR, A. (ed.) 1994. International Stratigraphic A. G. 1989. Vegetation of wetlands of the prairie Guide, Second Edition. A Guide to Stratigraphic pothole region. In: VAN DER VALK, A. (ed.) Classification, Terminology, and Procedure. Inter- Northern Prairie Wetlands. Iowa State University national Union of Geological Sciences and the Press, Ames, Iowa, 132-187. Geological Society of America, 214 pp. T(ENNETT, J. P. & STOTT, L. D. 1991. Abrupt deep-sea SHAY, J. M. & SHAY, C. T. 1986. Prairie marshes in warming, palaeoeoceanographic changes and western Canada, with specific reference to the benthic extinctions at the end of the Palaeocene. ecology of 5 emergent macrophytes. Canadian Nature, 353, 225-229. Journal of Botany, 64, 443~454. KOVACH, W. L. 1989. Comparisons of multivariate SHEEHAN, P. M. 2001. The Late Ordovician mass analytical techniques for use in pre-Quaternary extinction. Annual Review of Earth and Planetary plant palaeoecology. Review of Palaeobotany and Sciences, 29, 331-364 Palynology, 60, 255-282. STRAYER, D. L., POWER, M. E., FAGAN, W. F., MACDONALD, G. M. & CWYNAR, L. C. 1985. A fossil PICKETT, S. T. A. & BELNAP, J. 2003. A pollen based reconstruction of the late Quaternary classification of ecological boundaries. history of lodgepole pine (Pinus contorta ssp. BioScience, 53, 723-729. latifolia) in the western interior of Canada. SWEET, A. R. 2001. Plants, a yardstick for measuring Canadian Journal of Forest Research, 15, 1039- the environmental consequences of the Cretac- 1044. eous-Tertiary boundary event. Geoscience NICHOLS, O. 1996. Vegetational history in Western Canada, 28(3), 127-138. Interior North America during the Cretaceous- THORNES, J. 1979. Fluvial processes. In." EMBLETON, Tertiary transition. In. JANSONIUS, J. & MCGRE- C. & THORNES, J. (eds) Process in Geomorphol- GOR, D. C. (eds) Palynology: Principles and ogy. Edward Arnold, London, 213-271. Applications. American Association of Strati- TIMONEY, K. P., LA ROI, G. H., ZOLTAI, S. C. & graphic Palynologists Foundation, Dallas, Texas, ROBINSON, A. L. 1992. The high subarctic forest- 3, 1189-1195. tundra of northwestern Canada: position, width, NORRIS, R. D. & ROHL, U. 1999. Carbon cycling and and vegetation gradients in relation to climate. chronology of climate warming during the Palaeo- Arctic, 45, 1-9. cene/Eocene transition. Nature, 401, 775-778. TSCHUDY, R. H., PILLMORE, C. L., ORTH, C. J., North American Commission on Stratigraphic GILMORE, J. S. & KNIGHT, J. D. 1984. Disrup- Nomenclature 1983. North American Strati- tion of the terrestrial plant ecosystem at the graphic Code. American Association of Petroleum Cretaceous-Tertiary boundary, Western Interior. Geologists Bulletin, 67, 841-875. Science, 225, 1030-1032. OKE, T. R. 1993. Boundao, Layer Climates. 2nd WALKER, I. J. & NICKLING, W. G. 2002. Dynamics of edition. Routledge, London, 436 pp. secondary airflow and sediment transport over OKE, T. R. 1997. Surface climate processes. In." and in the lee of transverse dunes. Progress in BAILEY, W. G., OKE, T. R. & ROUSE, W. R. Physical Geography, 26, 47-75. (eds) The Surface Climates of Canada. McGill- WALSH, S. L. 2004. Solutions in chronostratigraphy: Queen's University Press, Montreal and King- the Paleocene/Eocene boundary debate, and ston, Canada, 21 43. Aubry vs. Hedberg on chronostratigraphic PELLATT, M. G., SMITH, M. J., MATHEWES, R. W., principles. Earth-Science Reviews, 64(1-2), 119- WALKER, I. R. & PALMER, S. L. 2000. Holocene 155. treeline and climate change in the subalpine zone WARDLE, P. 1974. Alpine timberlines. In." IVES, J. D. near Stoyoma Mountain, Cascade Mountains, & BARRY, R. G. (eds) Arctic and Alpine Environ- southwestern British Columbia, Canada. ments. Methuen, London, 371402. Arctic, Antarctic and Alpine Research, 32, 73 83. ZACHOS, J. C., PAGANI, M., SLOAN, L. C., THOMAS, PETERS, S. E. & FOOTE, M. 2002. Determinants of E. & BILLUPS, K. 2001. Trends, rhythms, and extinction in the fossil record. Nature, 416, 420- aberrations in global climate 65 Ma to present. 4"~4 ~ci, ne, ~9"~ 686-693