Palynological and temporal inference in the context of prehistory,with special reference to the evidence from lake and peat deposits

There is an increasing use of palynological data to infer prehistoric agricultural activity. The uncritical acceptance of the technique of pollen analysis by archaeologists and palynologists is questioned. Attention is focussed upon the nature of clearance activity and problems of its inference from pollen spectra; the spatial distribution of palaeoenvironmental sites; and the proximity of such sites to those of prehistoric activity. The resolution apparent in peat and lake sediment pollen profiles together with problems of dating and temporal inference, are examined. In response to a perceived lack of formalization in palynological research, some simple models of clearance activity are presented. The complexity involved in environmental reconstruction is emphasized by reference to the use of supplementary data to augment the orthodox pollen record.     

Spatial Autocorrelation and the Classic Maya Collapse: One Technique, One Conclusion

A pair of articles appearing recently in this journal (Whitley & Clark, Journal of Archaeological Science12, 377-395, 1985; Kvamme, Journal of Archaeological Science17, 197-207, 1990) apply spatial autocorrelation analysis to the distribution of terminal long-count dates from southern Lowland Classic Maya monuments. The authors employ similar techniques yet arrive at contradictory conclusions regarding the presence of geographical patterning in the collapse of the Classic Maya civilization in this region. Kvamme's contention, however, that Whitley & Clark conducted an inappropriate analysis and arrived at an erroneous conclusion is unsubstantiated. Both articles present appropriate analyses and report results which support the presence of spatial patterning in the Lowland Maya dates.     

Bayesian Updating of Seismic Demand Models and Fragility Estimates for Reinforced Concrete Bridges with Two-Column Bents

Probabilistic models have been developed in a previous study by the authors to estimate the seismic deformation demands on structural components of reinforced concrete (RC) bridges with two-column bents. However, such models should be updated to reflect the latest laboratory of field data. Using a Bayesian approach, this article updates a currently available probabilistic model for the deformation demands of columns in bridges with two-column RC bents. The updated model incorporates information from newly available experimental data from shake table tests conducted based on a record of the 1994 Northridge Earthquake for a structural system with three bents with two columns per bent. The updated model is more accurate than the previous one in predicting the deformation demand of bridges with two-column RC bents and reduces the statistical uncertainty due to the addition of new data. As an application, fragility estimates for an example bridge are computed using the updated model both at the component (column) and system (bridge) levels.