# Informatic application to characterise and identify small mammal species: Arvicolinae (Cricetidae, Rodentia, Mammalia)

## TL;DR

A new software tool helps classify small rodent species using molar shape analysis, improving accuracy and efficiency in species identification.

## Contribution

A novel MATLAB-based application using geometric morphometrics for accurate and efficient classification of Arvicolinae species.

## Key findings

- The application achieves high accuracy in classifying Arvicolinae species using molar morphology.
- The tool automates linear measurements commonly used for species identification.
- It can reduce the time required for manual classification of specimens with unclear morphologies.

## Abstract

The classification of rodent species can be challenging due to high morphological similarities observed among them. This problem is further increased in palaeontological systematics, where classification is traditionally based on the molar morphology. The subfamily Arvicolinae (Rodentia, Mammalia) is one of these rodent groups, whose classification being important for biostratigraphic and climatic studies of the Quaternary period is challenging. We present an application developed using the MatLab informatic algorithm, designed to classify the Arvicolinae species using Geometric Morphometrics (GMM) analyses of the first lower molar. Moreover, the application includes an option to automatically obtain the linear measurements that are commonly used for the identification of these species. This method shows a high degree of accuracy in the species classification, which is expected to increase as the reference database is further developed. This application can serve as an alternative tool for the classification of specimens with unclear morphologies. It can also be used to reduce the time required to manually obtain the linear indices necessary for their classification.

We have developed a new application in the MatLab environment to address the challenging classification of Arvicolinae species, using images of the first lower molar, achieving high accuracy in the species classification.