microfossils from a cave in Hungary Emilia Jarochowska & Wilma Wessels (Utrecht University)
Voles are rodents that are predominantly herbivorous and live in burrows. Vole populations can grow very large in a short time due to a high reproduction rate. Each species has a distinctive dentition (with prismatic hypsodont teeth) and is associated with a specific type of food and environment. Voles are widely used in the terrestrial fossil record to interpret paleoecology and paleoclimate, because their ecology makes them good environmental indicators:
- High abundance - it is possible to make quantitative studies and evaluate statistically the precision of ecological reconstructions
- Small size - it is possible to find a large fossil assemblage in a relatively small volume of rock or sediment (e.g. when outcrop area is limited or when we have to ship the samples to the lab - try this with a triceratops!)
- High preservation potential - as you already know, teeth are covered with a highly mineralized layer of enamel, which contains the mineral apatite. This makes teeth very resistant to chemical dissolution and mechanical fragmentation.
- High rates of morphological evolution - each species has different teeth
- Clear link between the morphology and diet (and thus vegetation available). By studying teeth alone we can tell what food was available in the animal’s habitat.
This practical consists of two parts:
- The mammal succession in Pleistocene and Holocene deposits in the Yankovich cave (Hungary) is used to reconstruct the climate in the area. The succession comes from 11 successive layers in a 3 meters high section. The total material collected consists of some 20 000 fossils of large and small mammals.
The Yankovich Cave is situated about 40 kilometers NW of Budapest (Hungary) near Bajót. Its altitude is about 330 m.
Hungary has a relatively dry continental climate, with cold winters and warm summers. Average temperatures range from -1°C in January to 21°C in July. Rainfall is heaviest in early summer, and the average amount decreases from 787 millimeters along the western frontier to 508 millimeters in the east.
The first study using quantitative data from a succession of fossil rodent assemblages for paleoecological and paleoclimatic interpretations was by Kretzoi (1957).
- The reconstructed biotopes are compared with biomes calculated from a global climate model.
After finishing this practical, you should be able to:
- Understand how the nearest living relative principle is used to reconstruct past environments
- Use the present-day distributions and habitats of rodents for the interpretation of a late Pleistocene to Holocene faunal succession
- Carry out a quantitative analysis of changes in the composition of ecological groups of fossil fauna
- Use a dataset from a published article to extract climate information you need
- Document your analyses in a document (notebook or Markdown file) that and allows any scientist or fellow student to reproduce them without you being around.
Table2.csv
contains the relative frequencies of voles in the Yankovich
cave. To make the findings comparable, first lower molars were counted,
so e.g. if one individual left multiple teeth preserved, we wouldn’t
count it twice. You will use this data to construct an area chart. On
the horizontal axis should be the layer and on the vertical axis should
be the relative proportion of each species in the assemblage. After you
made this plot, make a second version of it, but this time with the
absolute age on the horizontal axis. You might encounter problems
resulting from the fact that some layers are so close to each other that
they have been dated as the same age. Based on your geological knowledge
so far, can you think of why this could happen? Having two assemblages
with the same age will make the plot hard to read. You can modify the
age slightly so that the two samples are plotting close together, but
not one on top of the other.
Alternative version: A table with absolute counts
(data/Table2_abs.csv
) has been also generated. It can be used to ask
students to generate relative counts themselves.
You will discover that some beds have the same age. Why could this be? What are geological and methodological explanations for this? Propose a solution on how to plot it so the plot remains legible.
Use the nearest living relative principle to reconstruct the environment of the fossils from the cave. Make a new version of Table 2, but instead of showing the proportions of taxa, extract the biotope from Table 3 and apply it in the new version of Table 2. You are aiming for an area chart that shows the relative proportions of biotopes. So you may need to re-calculate the proportions. You can modify the code from 3.1, but make sure the labels are adjusted accordingly.
In Table 4, also attached as a CSV file, the presence of other vertebrates other than voles is shown. Make the same set of area charts as in 3.1 and 3.2, but for these taxa. You will again have to modify the code and use the information from Table 3 to plot changes in organisms living in different biotopes. Is the biotope area chart the same as in 3.2?
Alternative version: A table with absolute counts
(data/Table4_abs.csv
) has been also generated. It can be used to ask
students to generate relative counts themselves.
Tables 3 and 4 contain taxa found in each layer in the cave. Make two bar plots, showing the species richness and genus richness across time. For that purpose, you need to generate a variable showing the number of species and genera, respectively, found in each layer. This will be the height of the bar (vertical axis).
Bonus task: this point requires making the same operations for each layer. Can you turn these operations into a function and apply it to all layers? This would make your code faster and less prone to mistakes.
Rank abundance curves are used in ecology to show how balanced an ecosystem is: is it dominated by one taxon or are there similar proportions of each taxa? In this step, you have to sort the species based on how abundant it is in the fossil assemblage. It has to be calculated for each layer in the cave. The most abundant taxon has the lowest rank (1) and the least abundant - the lowest. Once you sorted the species based on abundance, you can make a plot where the rank is on the horizontal axis and the abundance on the vertical axis. Generate the plot one after another, in stratigraphic order, i.e. the oldest at the bottom. How did the ecosystem change over time? Did it go from less to more balanced or vice versa?
Bonus task: this point requires making the same operations for each layer. Can you turn these operations into a function and apply it to all layers? This would make your code faster and less prone to mistakes.
Are fossil mammals good indicators of past climate and vegetation? We can find out by comparing with past vegetation. But how? The most important sources of information are pollen records and climate models. The latter often include information from the former. Here we show you one of several possible examples of obtaining data on past vegetation. This example relies on Open Science: an approach that requires authors of all studies to make their data and code available to anyone who would like to reproduce their results. Beyer et al. (2020) generated a model reconstructing biomes on the Earth for the last 120 000 years. Biomes in palaeoclimatology have very standardized definitions and their occurrence can be predicted for palaeoclimatic data (Kaplan et al. 2003). The data generated by the model can be downloaded from a public repository (warning: big file!) and the authors provided the code to extract the information from it in R and Python.
Your tasks are to: 1. Download the data and the code for the language
you are using for this assignment. 2. Install the packages required to
open the data” netCDF
using the command specific for the language or
environment that you are using. 3. Change the geographical coordinates
to the area where the Yankovich cave is located. 4. Extract the annual
temperature in June using the code provided by the authors of the model,
but changing the are to the cave and the time interval to that
represented in the cave. Make a plot of the annual temperature across
the years represented in the layers sampled in the cave. Do you see the
temperature pattern of the climate model reflected in the biotope data
of the cave? 5. Adapt the script provided by the authors of the model to
modify the plot of biomes in Europe. You can chane the range of the
plotted area to focus on a smaller geographical area, e.g. only on
Europe. Make this plot for at least two relevant points in time, e.g. at
the beginning and at the end of the interval sampled in the cave.
Compare with the area plots and discuss your results.
Beyer, R.M., Krapp, M. & Manica, A. (2020) High-resolution terrestrial climate, bioclimate and vegetation for the last 120,000 years. Sci Data 7, 236 https://doi.org/10.1038/s41597-020-0552-1
Kaplan, J. O., et al. (2003) Climate change and Arctic ecosystems: 2. Modeling, paleodata-model comparisons, and future projections, J. Geophys. Res., 108, 8171, doi:10.1029/2002JD002559, D19.
Kretzoi M. (1957) Folia Archaeologica, 9, pp. 16 - 21
Hand in an document as recommended by the teacher (e.g. Jupyter notebook, Pluto notebook, R Markdown etc.). Write a report text as you would for any analysis, but include the code corresponding to the steps you’ve taken in the analysis and the output.
Address the following topics: What is the difference between a biotope and a biome and how does this affect the comparison between the two methods? What are the limitations of both ways of reconstructing palaeoclimate? Are there any limitations of the nearest living relative principle? What in the nature of the geological record, and specifically in caves, could bias this analysis?
Use formatting to split your document into chapters, pay attention to the spelling, especially to the use of taxon names. You don’t have to include the files provided with the assignment unless you have modified them.