michelaleonardi Research, Seminars

Talk for the Computational and Digital Archaeology seminar series, Cambridge

If you are free today at 4 pm (UK time), join us for the first Computational and Digital Archaeology seminar of the term at the Department of Archaeology, Cambridge.

Palaeoecology made easy: two new tools to make the best use of big data in archaeology

Monday, 9 October, 2023 – 16:00 to 17:00 (UK time)

I will be talking about palaeoecology, #paleoclimate, and big data in archaeology. I will also present two new tools that I co-developed to make palaeoecological analyses easier: pastclim and tidysdm.

Abstract: In recent years, we witnessed the publication of both large datasets of archaeological occurrences and palaeoclimatic data series covering hundreds of thousands of years. This opened the door to new, exciting possibilities in the field of palaeoecology. 

To make the best use of such data, I co-developed two new R packages that facilitate their use and allow simpler pipelines for their analyses.

The first package is pastclim. It is designed to easily access and manipulate climatic data and palaeoclimatic/future climate reconstructions. It contains a set of functions to recover the climate for time periods of interest, crop to specific areas, extract data from locations scattered in space and/or time, retrieve time series from individual sites, and manage the ice or land coverage.

The second package is tidysdm. The rationale behind it is to take advantage of the tidymodels framework in R to perform species distribution modelling (also known as habitat suitability/ecological niche modelling). This is the first software specifically designed to work with occurrences scattered in time, a task that with other tools, is either impossible or requires extensive tweaking. The integration with pastclim allows hassle-free access and handling of climatic data from the past, the present, and the future.

I will also present some applications of such tools, as an example of the new opportunities they could offer to scholars studying the past (e.g. archaeologists, archaeozoologists, palaeoecologists, palaeoanthropologists).

Event location: McDonald Institute for Archaeological Research Seminar Room
Join online: https://zoom.us/j/94826858436?pwd=N1JCeU9Id3dUdkJ4ZDVyUjhtQ0FMQT09

Event series: Computational and Digital Archaeology Laboratory Series

Update: a couple of pictures of my talk

michelaleonardi Research

The palaeoecology of European ungulates at INQUA 2023 in Rome

In these days, there is the Congress of the International Union for Quaternary Research (INQUA 2023) in Rome.

Yesterday morning I gave a presentation about “Time matters: testing for ecological niche changes in Late Quaternary European ungulates“. It was based on our work on the ungulates (scientific paper, blog post), but I have also presented a sneak peek of a resource that will be out very soon, so… watch this space!

Here are a few pictures of my presentation, and a happy INQUA mammoth picture taken after the end of it!

michelaleonardi Research

Webinar about pastclim, an R package to work with palaeoclimate

A couple of days ago, my PI, Prof. Andrea Manica (University of Cambridge), presented a webinar. He talked about pastclim, our piece of software to easily access and manage palaeoclimatic data. He gave a general overview about the palaeoclimatic reconstructions included in it, presented the package and gave examples of how palaeoclimate can be used to address questions related to evolution, palaeoecology and migrations.

Here is the video of the event. You will see some of my work presented there (especially my paper on the palaeoecology of European ungulates).

Mapping Ancient Africa: Andrea Manica “pastclim 1.2: an R package for paleoclimatic reconstructions”

The seminar was part of the project “Mapping Ancient Africa: Climate, Vegetation & Humans“, a multi-year project funded by the Palaeoclimate commission (PALCOM) and Humans & Biosphere commission (HABCOM) within the International Union for Quaternary Research (INQUA).

michelaleonardi Preprints, Research

Low ecological differentiation in leopard subspecies

It is now out in bioRxiv the preprint resulting from the part II project of Sidney Leedham:

Sidney Leedham, Johanna L. A. Paijmans, Andrea Manica, Michela Leonardi
Niche conservatism in a generalist felid: low differentiation of the climatic niche among subspecies of the leopard (Panthera pardus)
bioRxiv 2023.01.26.525491

The leopard (Panthera pardus) is a generalist species with a very wide geographic range: it can be found in most of Africa and part of Eurasia. It is subdivided into one African and eight Asian subspecies, which are the result of an ancient expansion from Africa.

We collected published observations of leopards across the entire historical range to see if the Asian subspecies live in the same climate as the African one, or if, in their expansion, they adapted to new climatic conditions.

We visualised the niche occupied by each subspecies in the climate space and compared them to see how much the Asian subspecies diverged from the African one. In most cases, there is great or total overlap, with the exception of the Persian leopard which suggests niche expansion.

This is supported by the fact that, when modelling the range of the species using only African presences, only the most northern part of the distribution is not retrieved.

These results help us better understand how the ecology of the leopard varies across its range, a knowledge that is vital for the effective conservation of its most distinct and vulnerable populations.

Preprint

Sidney Leedham, Johanna L. A. Paijmans, Andrea Manica, Michela Leonardi
Niche conservatism in a generalist felid: low differentiation of the climatic niche among subspecies of the leopard (Panthera pardus)
bioRxiv 2023.01.26.525491. doi: https://doi.org/10.1101/2023.01.26.525491

Abstract

Aim Species distribution modelling can be used to reveal if the ecology of a species varies across its range, to investigate if range expansions entailed niche shifts, and to help assess ecological differentiation: the answers to such questions are vital for effective conservation. The leopard (Panthera pardus spp.) is a generalist species composed of one African and eight Asian subspecies, reflecting dispersal from an ancestral African range. This study uses species distribution models to compare the niches of leopard subspecies, to investigate if they conserved their niches when moving into new territories or adapted to local conditions and shifted niche.

Location Africa and Eurasia

Methods We assembled a database of P. pardus spp. presences. We then associated them with bioclimatic variables to identify which are relevant in predicting the distribution of the leopard. We then constructed a species distribution model and compared the distribution predicted from models based on presences from all subspecies versus the ones built only using African leopards. Finally, we used multivariate analysis to visualise the niche occupied by each subspecies in the climate space, and to compare niche overlaps to assess ecological differentiation.

Results Niche comparisons and model predictions suggest a general lack of niche separation between all subspecies. Most Asian subspecies have overlapping niches and occupy subsets of the niche of the African leopard. Nevertheless, we found the Persian leopard Panthera pardus saxicolor to have the most distinct niche, giving some evidence for niche expansion in more Northern Asian subspecies.

Main conclusions These results suggest little ecological differentiation among leopard subspecies and a lack of adaptation to novel climates after dispersal from Africa. This finding complements recent genetic studies in implying that the taxonomy of Asian leopards may not reflect biological differentiation, an issue that is important to resolve due to its relevance for the conservation of the species.