Linking Spatial and Temporal Scales in Paleoecology and Ecology
Andrew S. Cohen, Department of Geosciences, University of Arizona, Tucson, Arizona
James Brown, Department of Biology, University of New Mexico, Albuquerque, New Mexico
Dale Springer, Department of Geography and Earth Science, University of Bloomburg, Bloomburg, Pennsylvania
Peter Holterhoff, Houston, Texas
The GSA Penrose Conference , "Linking Spatial and Temporal Scales in Paleoecology and Ecology," was held in Solomons, Maryland, May 1418, 1998. This conference was cosponsored by the Paleontological Society and the Ecological Society of America. It brought together 76 paleoecologists and ecologists from eight countries to consider how ecological interpretation and synthesis are affected by the spatial and temporal scale at which data are collected and models are constructed. This conference provided an exciting venue for members of the earth and biological sciences communities to look for common scientific ground, and to identify interdisciplinary research directions that will cross the traditional boundaries between the fields of paleontology and ecology. Owing to the diverse background of the participants, the conveners organized the meeting around a series of activities intended to maximize cross-disciplinary interaction. Field trips, dedicated poster sessions, and panel discussions were emphasized throughout the meeting, whereas formal talks were used primarily to introduce general topics and provide fodder for conversation.
The first day of the conference was an all-day field trip examining the Virginia Coastal Reserve-Long Term Ecological Research Site (LTER), located along the barrier islands, coastal marshes and beaches of eastern Virginia, on the Delmarva Peninsula. This trip was organized by Bruce Hayden, and run by John Porter, Ray Dueser, Aaron Mills, Linda Blum, Bob Christian, and Michael Fenster. The premeeting field trip served to introduce the themes of the conference, through an understanding of the evolution of the coastal ecosystem of Virginia at varying temporal and spatial scales, and through discussion of how physical processes (tectonism, eustasy, and sediment supply) interact with ecosystem development at varying scales. The formation of an Eocene impact crater, centered on what is now southern Chesapeake Bay, set the stage for subsequent ecological events, whose effects are still being felt today. A combination of long-term relative subsidence in the southern Delmarva Peninsula associated with crater-induced structures, along with both low sediment supply and eustatic sea-level rise, is creating a rapid modern relative sea-level rise of ~3mm/yr along this coast. A primary function of the Virginia Coastal LTER is to monitor the effect of this sea-level rise, as it induces forest die-backs, transgressive salt marsh migration, and the formation of new storm flooding surfaces in what were previously forested coastal regions. Field trip participants also saw how human-induced change (changing agricultural methods and conversion of economic bases) is playing out in the context of this environmental change. In the afternoon, the field trip moved south, across the Chesapeake Bay Bridge to Cape Henry. Here, less than 50 km south of the LTER sites, a progradational Holocene coastline is developing, under the combined influence of considerably greater sediment input and lower subsidence rates. Formation of a succession of shoreward-migrating beach ridges has sequentially caused the development of bald cypress swamps in the swales between ridges. Changes in the scale and rate of landscape evolution during beach migration has caused vegetational zones to also migrate at varying rates.
Day two of the conference marked the beginning of the formal sessions. Convener Andy Cohen introduced the background and themes of the conference. Ecological processes occur over a vast range of spatial and temporal scales, and it is increasingly evident that our perception of how these processes and changes play out is dependent on the scales at which we observe them, as well as our scientific and cultural differences. "Deep Time" (i.e., pre-Quaternary) paleoecologists, Quaternary paleoecologists, and neoecologists collect observations and generate theory at variable scales and with variable information bases. Can these observations be synthesized by common methodology and theory, or are there fundamental discontinuities crossing between scales? The morning's speakers confronted these issues through introductions to the various approaches to ecological understanding employed today. Mike Rosenzweig gave an introduction to large-scale ecological patterns and our current understanding of what these patterns signify in terms of dynamics at the regional and global level. Advances in statistical methods have greatly improved our ability to assess patterns of diversity in biotas, particularly with regard to species that are relatively rare. Convener Jim Brown presented a series of neoecological data sets gleaned from a variety of temporal and spatial scales, in an attempt to clarify for the paleoecologists in the audience the complexities in understanding pattern and its mechanistic interpretation. Richard Bambach gave participants an overview of the fossil record. He framed this discussion around a dichotomy of those processes and patterns that are familiar to both paleoecologists and neoecologists, and those that are only evident from the vantage point of long time intervals that paleoecologists can observe. Karl Flessa attacked the thorny question of temporal resolution in the fossil record. Paleoecology is only recently moving away from a long interlude of hand wringing over the problems taphonomy and time averaging of fossil faunas pose for paleoecology, to a realization that the fossil record can be resolved much better than traditionally thought and that time averaging isn't all bad news. Roy Plotnick discussed the roles of models and modeling in paleoecology, considering their potential for providing linkages between scales. He also showed the importance of a solid grounding in biological understanding in which to embed the development of theory, and the misinterpretations that may await modelers who don't have this grounding.
Poster presentations that followed the morning talks emphasized the general theme of data acquisition in paleoecology and ecology. Posters provided an opportunity for participants to better understand the nature of data sets from fields in which they may have had little prior experience, as well as seeing how such data are manipulated statistically by other fields. Posters emphasized a wide variety of ecological subjects, including patterns of diversity or community structure data at varying scales, taphonomic effects of scaling interpretations, and body-size dynamics at varying scales and their relationship to evolution and climate. Two panel discussions were convened following each of the poster sessions of the day. Panel discussions allowed a different group of participants than those who had just spoken or presented posters to comment on what had been presented and to field questions from the general audience. This activity, which continued throughout the meeting, proved a highly successful and stimulating way to generate excitement about what had been covered up to that point. There was a buzz in the air following the first day's panel sessions, as participants recognized the great potential for cross-disciplinary research efforts that might arise by linking the perspectives of paleoecologists and ecologists around common research questions. For example, participants pondered how analogous time and space really are in ecology, whether time averaging is really a problem or, in fact an opportunity, what other types of issues besides diversity we should be considering in terms of temporal and spatial scaling, and at what conceptual levels paleoecology and ecology might interact. That evening, Jim Reichman from the National Center for Ecological Analysis and Synthesis (NCEAS) gave an after-dinner talk on how his center works, and its eagerness to entertain innovative proposals for sabbatical visits and collaborative workshops spanning the boundary between paleoecology and ecology.
On the third day, the conference moved to a detailed discussion of community structure and stability. Participants got down to considering major and contentious issues concerning the dynamics of diversity change and community structure over varying time and space scales. Linda Ivany and Carl Brett presented their case for coordinated stasis in the fossil record. This idea postulates that large assemblages of marine benthic organisms evolve at about the same time, co-occur over millions of years, and then simultaneously undergo extinction. Ivany reviewed evidence from the Devonian of New York and the Eocene of the U.S. Gulf Coast that suggests that benthic assemblages persisted over 18 Ma time intervals, even in the face of sea-level fluctuations. Mark Patzkowsky argued against the coordinated stasis model based on his work on similar Paleozoic ecosystems. In the Ordovician of the eastern United States he finds that species turnover largely corresponds to pulses of environmental change, and that significant extinction and origination events are not always simultaneous. Background turnover within stratigraphic units seems to be much higher than what has been reported by Brett, Ivany, and others. These divergent views lead to the question of how common patterns of individualistic vs. coordinated change really are, and to what extent our perception of these patterns are regulated by the temporal resolution of individual data sets and the episodicity of environmental change. Russ Graham and Steve Jackson both made a case for individualistic responses in explaining community composition in the Quaternary, in particular the occurrence of "no modern analog" biotas. Graham argued for using the Quaternary as a bridge between the deep paleoecological record and modern ecology. He noted that in the Quaternary, temporal resolution is now good enough that although he can't say that two individuals of co-occurring "no-analog" species saw each other "eyeball to eyeball," they can be linked within 50 years in time. Both Graham and Jackson emphasized that patterns of individualistic change should not be confused with a notion of random association. Jackson argued for a greater understanding of potential yet unrealized niche space in organisms that may well explain these "anomalous" communities. Fred Grassle gave the conferees a fascinating glimpse into the species diversity and community composition in the deep sea. Although our understanding of this environment is limited by the difficulties involved in obtaining broad spatial and temporal coverage, it is evident that deep-sea organisms are highly diverse and their distributions very patchy. Even small-scale environmental heterogeneities, such as the deposition of a log, can create habitat patches with distinctive characteristics that persist for many years.
Afternoon posters and panel sessions and the evening lecture continued exploring the themes of structure and stability, with greater attention to details of changes across scale boundaries. Questions tackled included: (1) Do "rules" exist in ecology, either within scales or transcending scale boundaries? (2) How analogous are time and space in ecology? (3) How does scaling affect our understanding of other issues such as body size structure in communities? (4) To what extent do terminological differences impede understanding between ecologists and paleoecologists? This point was illustrated by Joan Roughgarden in a discussion of the meaning of "stability." Jack Sepkoski discussed patterns of global marine biotic organization, building on his earlier compilation analyses of Phanerozoic diversity patterns. He presented the problems of decomposing Phanerozoic marine diversity as a "chicken or egg" problem. Are the dominant controls at the local or regional level building up to creating a global pattern, or are there global controls on diversity that filter down to the local level? Both panel discussions and informal conversations on this day and the next day generated a lively discussion of specific research questions that might be usefully pursued by small working groups as followups to the conference, and the general excitement level surrounding this discussion led many to view the Penrose conference as but the first step in what could become an ongoing dialog among ecologists and paleoecologists with similar research question interests. There was considerable discussion of how working groups might continue the dialog through a series of NCEAS-type workshops.
A field trip on the fourth day to the well-known Miocene Calvert Cliffs, led by Patricia Kelley and Susan Kidwell, provided an opportunity for the participants to consider, on the outcrop, many of the scaling, data collection, and taphonomic issues confronting paleo-ecologists that had been discussed over the last few days. Kidwell and Kelley gave detailed talks on the depositional setting of the marginal marine and shelf outcrops visited, the taphonomic context of the fossils (which varies in significant ways related to sea-level fluctuations along the Atlantic coastal margin), and the paleoecological problems currently or previously under study. The Miocene deposits of this region have been the focus of intensive investigations on the interface between ecology and evolution, most notably concerning processes of predator-prey interactions over time, and the tempo and mode of evolutionary change. Ecologists attending the meeting with little prior paleo-ecological experience gained a great deal of insight into the practical considerations of sampling.
An evening lecture on the fourth day by Mark Westoby looked at ecomorphic classification schemes for plants and plant communities. Westoby put forward a new classification scheme which characterizes various aspects of total leaf size, plant height, and seed size. The expression of all of these features involves fundamental ecological compromises across environmental gradients of moisture and nutrient availability. Some, though probably not all of these variables, could in principle be measured for fossil plants, thereby extending the comparative value of Westoby's scheme from strictly between floras or regions to cross time periods.
On the final day of the conference the theme of relationships between scales was more thoroughly explored. Joan Roughgarden looked at the evidence for linkage across scales in the recruitment and growth of intertidal barnacle communities. She demonstrated that local or patch scales of benthic intertidal adult barnacle and starfish interactions are strongly mediated by the vastly different scale over which planktonic barnacle larvae are dispersed and subsequently recruited to the benthos as adults. The latter results from oceanic current circulation and migration of Ekman transport systems with respect to the coastline. David Jablonski looked at the evolution of onshore to offshore trends in the origination of major benthic invertebrate biotas. He asked what upward or downward effects may influence this pattern. After showing that these patterns are not artifactual, resulting from taphonomic biases, he proceded to demonstrate that the dynamics of invasion at the higher, ordinal level differ in substantive ways from what is observed at lower hierarchical levels. Bill DiMichele discussed some of the scaling problems involved in understanding the evolution of terrestrial floras. Ecomorphic exploration of plant form and habitat preference on the broadest levels are strongly linked to phylogeny in plants. Although turnover may occur at lower taxonomic scales during the evolution of a flora, the fundamental character of a plant community with respect to more inclusive clades remains remarkably stable over long geologic time intervals. Community assembly "rules" may change through the history of a clade, and incumbency limits the degree to which new floras can invade an ecosystem. Arnie Miller reviewed the nature of the global Ordovician radiation of marine invertebrates, moving from global down to regional scales. Miller showed that an apparent global diversification event, when dissected into its component parts, is actually a composite of regions with highly variable rates of diversification, possibly driven by profound differences in tectonic setting of the various continents. This examination of processes at progressively smaller scales suggests a need to evaluate, in similar detail, the tempo of other presumptively global events such as mass extinctions. Bruce Patterson gave the final talk of the meeting, looking at altitudinal diversity gradients across the Peruvian Amazon. Although data from bird surveys have come to be a sort of benchmark against which biogeographic models are compared, Patterson showed that many other patterns of diversity occur, particularly among small mammals. Spatial ranges and gradients in extant organisms display many properties analogous to the biostratigraphic ranges familiar to paleoecologists. Using his nested subset methodology, Patterson argued that historical factors and differences in dispersal ability, as well as species-area relationships, play a role in structuring mammal distribution patterns.
During the final panel discussion, a consensus arose among the panelists and audience as to the need for continued dialog among paleoecologists and ecologists. Scientists from both disciplines were urged to seek out colleagues with complementary skills to solve scaling problems of mutual interest while simultaneously drawing on the strengths of their respective knowledge. Following on the favored metaphor of the meeting, one conferee suggested that "we had to become polygamous" in developing collaborations among multiple fields. During the conference, participants had made tentative but important steps in learning each others' scientific "languages," critical for sustained dialog. It was widely agreed that the need for such a dialog has never been greater. The combined issues of global change and biodiversity loss on Earth both dramatize the need for understanding processes of diversification, invasion, and extinction at all scales. Our old modus operandi, of making simplistic interpretations of data gleaned from each other's fields, should be replaced by truly collaborative efforts to understand these most serious of environmental problems. This Penrose Conference provided an exciting opportunity for all participants to begin what we hope will become a sustained and fruitful conversation between paleoecology and ecology.
Penrose Conference Participants
J. Bret Bennington
G. -Lynn Brewster-Wingard
| Susan Kidwell
Christopher G. Maples
Brian A. Maurer
| Roy Plotnick