Background & Summary

The European cave salamanders (genus Hydromantes; see1 for taxonomic discussion) are a group of eight amphibians species endemic to Italy and to a small part of south-eastern France2. Three species (H. strinatii, H. ambrosii and H. italicus) are distributed along the northern and central Apennine chain (H. strinatii being the only species present in France), whereas five (H. flavus, H. supramontis, H. imperialis, H. genei and H. sarrabusensis) are endemic to Sardinia island, where geomorphology represents the main drive of their allopatry2,3. Only two mainland species, H. ambrosii and H. italicus, come into contact naturally and hybrid populations occur in a small area4. Hydromantes salamanders often have epigean activity during cold and wet seasons, but exploit subterranean habitats (such as caves, mines, small cervices and springs) to avoid unfavourable climatic conditions (when too hot and/or dry)5,6,7. These salamanders are lungless and require a specific combination of relatively low temperature and high moisture to efficiently carry out their cutaneous respiration;2,8 these conditions are often found in subterranean habitats9, thus Hydromantes are able to maintain stable populations and even reproduce there10,11, moving outdoors mostly to reach areas with high prey abundance12,13.

During the last few decades several studies have shed light on some life history traits of Hydromantes, such as the reproductive behaviour, trophic niche and population dynamics14,15,16,17. However, to collect data on these species can be extremely complex. First, the subterranean habitats are not human-friendly; some can be explored only by skilled speleologists and, in any case, the constant low temperature and the air moisture close to saturation represent a challenge for researchers spending prolonged time there18,19. Second, all the Hydromantes are strictly protected by both national and international laws20,21 and thus, studies cannot be performed without the proper authorisations. These salamanders are sensitive to multiple treats such as climate change, habitat degradation and poaching2,22. Furthermore, Hydromantes are sensitive to the deadly chytrid fungus, Batrachochytrium salamandrivorans, thus manipulation must be limited and must adhere to strict protocols to avoid spreading pathogens23. Indeed, researchers are continuously developing and testing different methodologies allowing them to limit their impact on these animals, without reducing the quality of their scientific researches24,25,26. Here we used a non-invasive method to build a large dataset related to Hydromantes, which can be used to investigate some of their unexplored life traits avoiding manipulation. Using photography, we here provide a dataset covering all Hydromantes species occurring in Europe, which also represents one of the few photographic datasets available for the animal kingdom. Our dataset is not made up of simple pictures (e.g.27,28), but we have adopted an ad hoc methodology to obtain standardised photos that can be used for future comparisons and analysed with multiple software29,30,31. There is a growing demand for publication of standardised dataset32,33,34, and those relating to endangered or not yet assessed species have a particular value35,36; however, special care should be taken when publishing data on sensitive species37. This dataset represents a snapshot of multiple Hydromantes populations, providing information on the minimum population size38, the age of individuals and morphometry39,40, and the variability of dorsal colouration41,42,43. Digitisation is a practice adopted to make manuscripts stored in various collections easily accessible to the public44,45,46. More recently, this trend also considers animal species31,47, providing undeniable help for the entire scientific community, as well a window for the public to discover the wonders of nature. Therefore, with our dataset we will provide the first digital collection showing the morphological diversity of all European Hydromantes species.

Methods

Experimental design

We collected photographic information on all European Hydromantes species, also including hybrid populations. For each species and for the hybrid zone between H. italicus and H. ambrosii, we photographed ≥55 individuals, collecting data from at least two different sites (Table 1 and Fig. 1); this allowed us to gather information on a large number of individuals from different areas2,4. We also considered a population introduced into the French Pyrenees, which probably includes individuals of different Hydromantes species, if not hybrids;48 there, we photographed 47 individuals. The surveyed sites included forest, natural caves, mines and artificial springs (see Table 1); in all of them, the presence of Hydromantes was previously assessed4,48,49. Each site was surveyed once between August and October 2018 in order to avoid repeated photos on the same individual. Prior to each survey, equipment and shoes were cleaned and disinfected with bleach to avoid the spread of potential pathogens.

Table 1 Qualitative data of the Hydromantes photographic dataset50.
Fig. 1
figure 1

Map showing the locations where the Hydromantes salamanders were photographed.

Individuals sampling

In a dark area of the cave we built a photographic set to take standardised photos of salamanders. We used a soft white fabric box (hereafter, soft box) with a piece of white and flat plexiglass covering the inner bottom of the box. The soft box was placed suspended on the floor of the cave, with flash units on the left, on the right and under the soft box; such arrangement allows the light to be evenly distributed, limiting the presence of shadows on the subject (Fig. 2). The bottom flash produces a clear white background, a condition enabling a quick and easy way to isolate the salamander from the background in post-production. The two lateral flashes were arranged above the subject and inclined with 45° to fully and evenly illuminate it. The salamanders were collected and placed in fauna boxes until they were photographed. Before being photographed, each salamander was visually inspected and all debris adhering to the skin were removed. The salamanders were then placed in the soft box on the plexiglass and a photo was taken from above, keeping the camera perpendicular to the surface of the plexiglass. Salamanders were photographed with a Pantone colour card (see below) next to them to have a standard size reference and to correctly calibrate the colours and light during post-production. After the photo shoot, the salamanders were released where they were collected.

Fig. 2
figure 2

Example of Hydromantes picture from the database. Here an individual of H. flavus photographed using the described methodology; scale bar 10 mm.

Photo calibration

Correctly balancing the white is crucial to obtain standardised and comparable images. To guarantee an accurate and standardised white balance to our images, we shot a reference photo in RAW format (.CR2) of the Pantone colour card X-Rite Colorcheker Passport 2 at the beginning of each photographic session; shooting in RAW creates high quality files containing all the unprocessed data captured by the sensor. Images were then uploaded on a computer and organized in folders, each corresponding to the single photographic session. Using the function “White Balance Tool” of the software Adobe Camera Raw, for each site we created a custom white balance profile using the respective picture of the Pantone colour card as reference. We then applied the profiles to the respective images and converted them into JPEG format, which enables reduction in the size of pictures without compromising their quality.

Data records

The dataset (The European Hydromantes salamanders50) includes 1,052 photographed individuals of all species and hybrids of European Hydromantes. Sites are within species natural range if not differently stated2. In particular, the dataset is composed as follows:

  • H. strinatii (Strinati’s cave salamander): 82 individuals (40 females, 33 males, 9 juveniles) from five sites, two outside its natural range;

  • H. ambrosii (Ambrosi’s cave salamander): 137 individuals (60 females, 47 males, 30 juveniles) from four sites;

  • H. italicus (Italian cave salamander): 141 individuals (54 females, 53 males, 34 juveniles) from four sites;

  • H. flavus (Monte Albo cave salamander): 171 individuals (64 females, 73 males, 34 juveniles) from six sites;

  • H. supramontis (Supramonte cave salamander): 112 individuals (46 females, 25 males, 41 juveniles) from three sites;

  • H. imperialis (Imperial cave salamander): 116 individuals (26 females, 63 males, 27 juveniles) from four sites;

  • H. genei (Gene’s cave salamander): 122 individuals (43 females, 51 males, 28 juveniles) from two sites;

  • H. sarrabusensis (Sette Fratelli cave salamander): 69 individuals (30 females, 27 males, 12 juveniles) from two sites;

  • Hydromantes hybrids: 55 individuals (32 females, 14 males, 9 juveniles) from two sites within the hybrid zone4;

  • Hydromantes (?) allochthonous population: 47 individuals (28 females, 11 males, 8 juveniles) from one site48.

Together with the photographic dataset we include information relating to each individual and their location (Qualitative data of the Hydromantes photographic dataset50; Table 1). Some information related to the salamanders (i.e., age, sex and total length) is critical to increase the quality of the dataset. Adult Hydromantes can be sexed with high confidence only by checking the presence of the mental gland on the chin of sexually mature males;2 this is a part of the body not visible from our images. With this additional information, salamanders can be divided into juveniles, adult males and adult females.

Technical Validation

This dataset shows a unique collection of multiple individuals belonging to all Hydromantes species present in Europe. Collecting data on these salamanders can be challenging, as all species are strictly protected20, and even simple manipulation requires ministerial authorisations. The single survey performed on each site provides data on 1,052 different individuals. The methodology applied here enables production of standardised high quality images with low impact on the species29,30. The overall time required to shoot each salamander was usually <15 seconds, thus limiting the stress caused by handling51. The white calibration before each session avoided potential divergence in light condition and thus, providing standardised pictures. Our methodology allowed to avoid the hurdles due to the use of flash on animals with moist skin, being thus widely applicable. Blind measurements of salamanders were performed to reduce possible bias52. Salamanders were measured entirely or at the furthest visible point. In few cases, the tail was severed or its tip covered; these cases are indicated in Table 150. Possible outliers were identified by plotting the data; the measurement of related individuals was taken twice to check whether the abnormal value was due to measurement errors. Considering that in these species individuals are aged according to their body size, and adults are sexed basing on the presence/absence of the male mental gland2, we added such information for each individual in Table 150. To identify juveniles, we used the size of the smallest male observed as reference. Mainland species, hybrids and H. genei have all comparable size35, therefore we used 68 mm as threshold for these salamanders. The other four Sardinian species are defined as “giant”; for these species the size of the smallest male was 77 mm. All salamanders smaller than the respective reference male were considered juveniles. Hydromantes can live more than 10 years2,53 and thus, considering that our dataset provides a snapshot of individuals from 2018, it can be employed in comparative studies for multiple years ahead. The morphometrics obtained from our dataset can be compared with those of the same species published >20 years ago54 to assess whether any potential change occurred and which may be the cause55,56.

Usage Notes

The pictures can be used in R environment (http://www.R-project.org/) to perform analyses on colouration (e.g.57) and geometric morphometrics (e.g.58), and with the program ImageJ to record multiple salamanders’ morphometrics (e.g.59). Furthermore, considering that, at least, in adult Hydromantes the dorsal pattern does not change throughout time60, this dataset can be used over time as a reference for the recognition of individuals belonging to the same population61,62, allowing to study growth rate, home range and other life traits. This also limits the manipulation of individuals, which represents a potential source of both stress and pathogens23,51,63,64.