Ecology of the past

Britte reconfiguring the chronology for Lake Pata

The second and third weeks of the Environments Through Time course at the University of Amsterdam has focused on obtaining practical experience of developing chronologies, analyzing multi-variate data-sets, and conducting time series analysis. The focus of the course has been on Quaternary environmental change, however, the skills learnt can be applied to almost any time-scale so long as you have time control points you want to tie together, and multiple things you can track changing through that time.

Over the two week period the students worked on a previously published paper that they had selected that contains: (i) chronological information (at least 3 control points), and (ii) multiple variables that change through the time series (at least 9 variables). In week two they deconstructed the chronologies and generated their own revised versions. For example students have (re-)calibrated radiocarbon dates, made different decisions on dates to include/exclude, and used different approaches to constructing the age vs. depth model, e.g. contrasting linear point-to-point vs. Bayesian methodologies. In week three they have taken the data-set(s) associated with their paper and re-evaluated it in light of the revised chronologies using cluster analysis, ordination techniques, and wavelets.

The joy of wavletes

Through this exercise students have gained experience of how to critically assess scientific literature and gained an appreciation of where re-analysis of data-sets can (and cannot) make a difference. Personally I have be delighted with the high level of engagement and enthusiasm for the material and have been excited to have a chance to delve into literature that I would not otherwise be aware of.

For more information on this course see: Environments Through Time – week 1