Scaling and the Palaeogeographical Distribution of
Stratigraphic Events
EP. Agterberg and EM. Gradstein
1
Introduction
The RASC computer program for ranking and scaling of biostratigraphic events
(Agterberg 1990; Gradstein et al. 1985) can be used to construct a composite zonation (standard) for the sedimentary record of a basin. This zonation of nonrecurrent stratigraphic events is primarily based on averaged first and last occurrences of fossil taxa in the stratigraphic record but other types of events (logmarkers, seismic events, sequence boundaries) can be incorporated. The basic requirement for an event to be included in the final zonation is that it can be uniquely identified in at least one stratigraphic section or well. The zonation is initially
constructed using the observed superpositional relationships for coexisting pairs
of taxa that are relatively abundant. Events that can be determined with certainty
in the sections are weighted more strongly. Selected rare events that occur in one
or a few sections only within the basin but are of global significance, can be inserted into the final solution by invoking the RASC unique (rare) events option.
Recent examples of application of the RASC method include Gradstein and
Backstrom's (1996) study on Cainozoic biostratigraphy and palaeobathymetry
of the northern North Sea and Haltenbanken using microfossils in 26 wells, and
Gradstein and Agterberg's (in press) paper dealing with the underlying Cretaceous in the same basin. The second paper introduces event variance analysis
which is based on deviations between observed and predicted occurrences. The
purpose of the present paper is to discuss new methods of scaling and correlation developed after 1994, and to show how logistic regression can be used to
construct palaeogeographic patterns of events.
Intervals between successive events in the standard can be graphed in a dendrogram. For example, Fig. 1 shows the Albian-Cenomanian scaled optimum sequence, offshore mid-Norway, as derived from 87 dinoflagellate, foraminiferal
and a log event in 29 wells (cf. Gradstein and Agterberg, in press). Clusters in the
dendrogram represent stratigraphic events that are approximately coeval on the
average. Gaps between clusters can indicate unconformities accompanied by
stratigraphic hiatuses. The biggest breaks in Fig. 1, occurring at events 92 and
52, coincide with latest Albian and mid Cenomanian highstands, respectively,
well-known from NW European sequence stratigraphy.
Stratigraphic Events
EP. Agterberg and EM. Gradstein
1
Introduction
The RASC computer program for ranking and scaling of biostratigraphic events
(Agterberg 1990; Gradstein et al. 1985) can be used to construct a composite zonation (standard) for the sedimentary record of a basin. This zonation of nonrecurrent stratigraphic events is primarily based on averaged first and last occurrences of fossil taxa in the stratigraphic record but other types of events (logmarkers, seismic events, sequence boundaries) can be incorporated. The basic requirement for an event to be included in the final zonation is that it can be uniquely identified in at least one stratigraphic section or well. The zonation is initially
constructed using the observed superpositional relationships for coexisting pairs
of taxa that are relatively abundant. Events that can be determined with certainty
in the sections are weighted more strongly. Selected rare events that occur in one
or a few sections only within the basin but are of global significance, can be inserted into the final solution by invoking the RASC unique (rare) events option.
Recent examples of application of the RASC method include Gradstein and
Backstrom's (1996) study on Cainozoic biostratigraphy and palaeobathymetry
of the northern North Sea and Haltenbanken using microfossils in 26 wells, and
Gradstein and Agterberg's (in press) paper dealing with the underlying Cretaceous in the same basin. The second paper introduces event variance analysis
which is based on deviations between observed and predicted occurrences. The
purpose of the present paper is to discuss new methods of scaling and correlation developed after 1994, and to show how logistic regression can be used to
construct palaeogeographic patterns of events.
Intervals between successive events in the standard can be graphed in a dendrogram. For example, Fig. 1 shows the Albian-Cenomanian scaled optimum sequence, offshore mid-Norway, as derived from 87 dinoflagellate, foraminiferal
and a log event in 29 wells (cf. Gradstein and Agterberg, in press). Clusters in the
dendrogram represent stratigraphic events that are approximately coeval on the
average. Gaps between clusters can indicate unconformities accompanied by
stratigraphic hiatuses. The biggest breaks in Fig. 1, occurring at events 92 and
52, coincide with latest Albian and mid Cenomanian highstands, respectively,
well-known from NW European sequence stratigraphy.