Introduction

In recent years, the environmental pollution directly threatening the ecosystem quality has become an important subject for researchers. Many ecosystems in the world have degraded due to the effects of environmental pollution. Because of the effects of pollution, some of the species in those ecosystems either went extinct or are in danger of extinction. Damage to or extinction of any species in an ecosystem affects other organisms in the life cycle. Caves, which are among the non-renewable natural resources, are one of the important ecosystems affected by pollutants (Du Preez et al., 2016). Geological formations in our country provide suitable conditions for cave ecosystems, and there are approximately 40,000 caves in our country (Nazik et al., 1996). Due to their nature, their low temperature, high humidity, nutrient scarcity, light absence, and season-independent stable ecological conditions make those caves unique ecosystems (Gèze, 1965; Mejía-Ortíz et al., 2021).

Acting as an ecological filter between surface and underground, caves are categorized by making use of continuity of darkness, scarcity of nutrient sources, and shortness of seasonal fluctuations (Culver & Pipan, 2009). Rather than considering these ecosystems as empty spaces below the ground, it is very important to know that they are habitats, where many organisms live (Rachmawati et al., 2021). These ecosystems, which are seen as the final border of the world by some people, constitute a part of the underground ecosystem. Moreover, they might host single and endemic species (Culver & Pipan, 2009).

Since heavy metals, which are one of the pollutants that are among the main problems of the current era, can accumulate through the food chain without degradation, it is very important to determine their concentrations, follow them, and reveal the accumulation rates for the health of ecosystems (Ferrante et al., 2018). Although some metals (Cu, Fe, Zn, Co, etc.) are necessary for organisms, they would have a toxic effect on organisms when they exceed a specific threshold level (Mansour et al., 2016). Caves became one of the ecosystems experiencing the pressure of heavy metal pollution in recent years. For this reason, many researchers investigated the current metal accumulation at the levels of water, sediment, and organic tissue (Cuculić et al., 2011; Ferrante et al., 2018; Printarakul & Meeinkuirt, 2022).

Amphipoda ordo is represented by approximately 6000 species worldwide. It is divided into 4 sub-ordos as Caprellidea, Hyperiidea, Ingolfiellidea, and Gammaridea. More than 80% of Amphipod species are included in the Gammaridea sub-ordo. The dominant body structure observed in Amphipoda ordo is the one compressed by the sides (Gammaridea). Besides that, they are characterized using the differentiation that they have three pairs of uropods and the first two pairs play role in nutrition. These species live in almost all habitats. The ones living in rapidly flowing waters live in the sand and under the small stones on the ground (Gledhill et al., 1993). Those living on the weeds in seas and freshwaters are generally herbivores, whereas those living in mud and sand feed on bacteria living on surface particles. Others are detritivores or scavengers feeding on plant-animal carcasses (Aydemir, 2007).

Analyzing metal contamination in sediment is crucial for predicting the status of contaminated aquatic environments. Different sediment quality standards have been developed for ecological risk assessments (Ergen, 2020; U.S.E.P.A., 1996). Sediment Quality Guides (SQGs), which were preferred in this study, have also been preferred by many researchers due to their short duration and low cost (Kükrer et al., 2015; Tunca et al., 2018; Yong et al., 2020; Fikirdeşici-Ergen et al., 2021). SQGs are a set of mathematical formulations developed and used to assess the risks posed by heavy metals (Bat & Şahin, 2018; Birch, 2018). It provides important outputs in revealing the potential of heavy metals to cause harm (Kwok et al., 2014). Determining and interpreting the amount of heavy metals in sediment is very important for the health of ecosystem and SQGs have been developed in order to meet exactly this need (Yong et al., 2020). This method incorporating a series of mathematical formulations is used in analyzing the risks posed by heavy metals and foreseeing the potential risks (Birch, 2018).

Among the multivariate statistical techniques (MSTs) allowing the interpretation of very complex water and sediment quality datasets, cluster analysis (CA), principal component analysis (PCA), and factor analysis (FA) are widely used in analyzing the water and sediment quality. These methods are also used in determining the water and sediment pollution sources, revealing the temporal and spatial changes, and interpreting them (Aydın-Uncumusaoğlu, 2018; Mutlu, 2019; Aydın-Uncumusaoglu & Mutlu, 2021).

The present study is the first biological study carried out on this cave system, which has a high tourism potential, in Ayancık district; new results and a new species were introduced to the literature from the İnaltı Cave and the water and sediment quality of the case was reported by using the statistical methods.

Methods and materials

Study area

İnaltı Cave was registered as a Natural Park on 08.03.2020, and it has a surface area of 22.8 ha; it is located in İnaltı Village in Ayancık district of Sinop province (Northern Türkiye) (Fig. 1). Having the coordinates of 41°43′50.60″N, 34°34′7.63″E, the cave has an altitude of 1070 m. The case is not rich in speleothems, and the main point of interest is its height, which exceeds beyond 20 m at some locations, and the wide and high shelf area in the entrance. The average length of the cave is 658 m. Of 358 m part of the cave, only 125 m part was found to be suitable for tourism. The case courses in S-form in the east–west direction. The entrance opens to a smooth gallery that has a length of 125 m, ceiling height of 15 m, and width of 10–12 m. This gallery is the driest and widest part of the cave. In the cave, there are stalactites, stalagmites, pillars, cover and wall dripstones, and dripstone pools. On the other hand, there also are cauliflower- and popcorn-shaped dripstones in the near-ground parts of the cave and right above the waterline. High-grade limestone, which is dissoluble, plays an important role in the formation of caves and pits in this region (S.K.T.M., 2022).

Fig. 1
figure 1

İnaltı Cave location (draw with QGIS, Las Palmas version 2.18.10)

Full size image

Sample collection and analysis methodology

Since the cave entrance and sampling process require specific experience and technical equipment, the field study was carried out in cooperation with MAD-Association for Cave Research. Land study was carried out on 8 September 2019, before the rainy season began and the in-cave water level increased.

Water temperature (T, °C), pH, dissolved oxygen (DO mg/L), and electrical conductivity (EC) values were measured using a Hanna HI 9829 field-type multiparameter device.

Within the scope of this study, water and sediment samples were taken from 10 points (Fig. 2). Samplings were conducted using plastic materials, and only the surface water and sediment were used. Samples were kept in polyethylene containers until delivered to the laboratory.

Fig. 2
figure 2

Map of İnaltı Cave and sampling locations (Nazik et al., 1996)

Full size image

Digestion-analysis of sediment samples and analysis of water samples was carried out by Bureau Veritas Mineral Laboratories Canada (ACME LAB.) by using AQ270 and SO200 methods, respectively. Sediment results were assessed using the sediment quality guide (SQG) (Table 1).

Table 1 Meaning scale of sediment quality indices
Full size table

Anthropogenic effect analysis methods

Data provided by Turekian and Wedepohl were used as reference values (Turekian & Wedepohl, 1961). The results were interpreted using the scale in the study carried out by Fikirdeşici-Ergen et al. (2021) (Table 1).

Contamination factor (\({C}_{f}^{i}\)) (Hakanson, 1980)

$${C}_{f}^{i}={C}^{i}/{C}_{n}^{i}$$
(1)

\({C}^{i}\) = metal amount

\({C}_{n}^{i}\) = reference value

Contamination degree (Cd) (Hakanson, 1980)

$${C}_{d}=\sum_{i=1}^{n}{C}_{f}^{i}$$
(2)

\({C}_{f}^{i}\) = contamination factor

Modified contamination degree (mCd) (Abrahim & Parker, 2008)

$${mC}_{d}=\frac{\sum_{i=1}^{n}{C}_{f}^{i}}{n}$$
(3)

\({C}_{f}^{i}\) = contamination factor n = number of metals examined

Pollution load index (PLI) (Tomlinson et al., 1980)

$$PLI={({C}_{f1} \times {C}_{f2}\times {C}_{f3}\dots .\times\; {C}_{fn})}^{1/n}$$
(4)

\({C}_{f1}\) = contamination factor n = number of metals examined

Enrichment factor (EF) (Hasan et al., 2013)

$$EF=\frac{{C}_{n}/{C}_{\mathrm{ref}}}{{B}_{n}/{B}_{\mathrm{ref}}}$$
(5)

\({C}_{n}\) = metal amount

\({C}_{\mathrm{ref}}\) = metal amount in reference medium

\({B}_{n}\) = amount of reference element in sample

\({B}_{\mathrm{ref}}\) = amount of reference element in reference medium

Al was preferred as reference element.

Geo-accumulation index (Igeo) (Müller, 1969)

$${I}_{\mathrm{geo}}={\mathrm{log}}_{2}\frac{{C}_{n}}{{1.5 \times B}_{n}}$$
(6)

\({C}_{n}\) = metal amount

\({B}_{n}\) = amount of the metal in reference medium

= natural oscillation coefficient

Mean effect range-median quotient (m-ERM-Q) (Long et al., 2000) and Mean Probable effect level quotient (m-PEL-Q) (Carr et al., 1996)

$$m-ERM-Q=\frac{\sum_{i=1}^{n}{C}_{i}/{ERM}_{i}}{n}$$
(7)
$$m-PEL-Q=\frac{\sum_{i=1}^{n}{C}_{i}/{PEL}_{i}}{n}$$
(8)

\({C}_{i}\) = metal amount

ERM = effect range median of the metal,

PEL = mean probable effect level of the metal, n = number of metals examined.

Benthic macrofauna sample collection

In stations with water, it was aimed to discomfort the macroinvertebrates by muddying the water and mixing the sediment on the bottom by using a manual scoop, and samples were collected using the scoop. Moreover, in order to access more taxa in the cave, the samples taken from the mud on the rocks and stone in the water were poured into a tray and then transferred to 250–500 ml plastic jars by using a small shovel. They were kept by adding 96% ethyl alcohol into the jars.

Then, the samples taken to the laboratory were filtered using a three-sieve system with 0.5 mm, 1 mm, and 2 mm mesh sizes and rinsed with fresh water. The organisms on the sieves were collected and examined using a stereomicroscope. Then, the organisms were put into glass tubes containing 70% alcohol. The samples obtained from Gammaridea sub-ordo were analyzed in the laboratory by using NIKON SMZ 445 stereomicroscope, and species classification was performed using Ruffo’s (1953) classification atlas.

Statistical analysis methodology

Surface water, sediment, and biological analysis results were analyzed using IBM SPSS 25 statistical package program. Using one-way ANOVA analysis of parameters, both descriptive statistical analysis and significance tests (0.01 and 0.05) were performed in order to determine differences between stations. Pearson’s correlation index (PCI) was used in analyzing the relationships between sediment data and species found in the sediment (Alberto et al., 2001).

Multivariate hierarchical cluster analysis (HCA), a combination of techniques for classifying large data sets into similarities or clusters based on differences, was used to spatially represent the water and sediment quality of İnalti Cave (Cvejanov & Škrbić, 2017). HCA also utilized Ward’s method as a similarity measure (Barakat et al., 2016).

Today, multivariate statistical techniques (MSTs) such as cluster analysis (CA), principal component analysis (PCA), factor analysis (FA), and discriminant analysis (DA) are widely used to evaluate and interpret very complex and large data sets (Aydın-Uncumusaoğlu & Mutlu, 2022; Shrestha & Kazama, 2007; Singh et al., 2004).These methods also facilitate the identification and interpretation of natural or anthropogenic factors controlling temporal and spatial variations in water and sediment pollution sources and quality. Furthermore, these methods can provide rapid solutions to pollution problems and have been recognized as invaluable tools for reliable and rapid assessment and management of these natural resources to help improve water and sediment quality (Aydın-Uncumusaoglu & Mutlu, 2022). Principal component analysis (PCA/FA) and Pearson’s correlation analysis were used in analyzing the relationships between Niphargus sp., which was the biological parameter, and other parameters in the sediment in which it lives and to reveal the pollution sources. Prior to principal component analysis (PCA/FA), Kaiser–Meyer–Olkin (KMO) and Bartlett’s fit tests were performed. Since this value was 0.652 and Bartlett’s test (P = 0) for this cave, PCA test was found significant. In order to avoid misclassification due to large differences in datasets, analyzes were performed using standardized experimental image data with Z-scale transformation. Principal components were validated with eigenvalues (eigenvalues > 1) summarizing the total variance in the data set, that is, by the number of major components passing through 1 before a clear break of the scree plot (Jahin et al., 2020). The obtained PCA factor loadings were classified as “strong,” “moderate,” or “weak” corresponding to absolute loading values above 0.75, in the range of 0.75–0.50 or in the range of 0.50–0.30, respectively, as indicated in the study by Liu et al., 2003.

Results and discussion

Water and sediment quality assessment

Both physicochemical and biological parameters were studied in water and sediment samples collected from 10 stations in the cave. Statistically significant differences were found between the stations in terms of annual mean values of all parameters (P < 0.05).

Mean values of water quality parameters were found to be as follows, and the mean values were found respectively: pH: 7.80 ± 0.1, DO: 5.35 ± 0.22, EC: 685 ± 0.26 ms/cm, water temperature (WT): 7.43 ± 0.1 °C. The levels of heavy metals detected in the water were as follows: Al = 16.2 ug/L, As = 0.52 μg/L, Cd = 0.04 μg/L, Cr = 9.66 μg/L, Cu = 1.27 μg/L, Fe = 9.55 μg/L, Mn = 0.067 μg/L, Ni = 0.19 μg/L, Pb = 4.22 μg/L, and Zn = 0.49 μg/l (Table 2). The cave is in the high-quality water class according to the World Health Organization (WHO) and our country’s surface water quality management regulation criteria, and no parameter was found to be beyond the limit values (SWQR, 2016; WHO, 2011). No statistically significant difference was found between 10 stations in terms of the water quality (ANOVA, P > 0.05).

Table 2 Concentrations of elements detected (µg/L)
Full size table

Mean concentrations of heavy metals found in the sediment were as follows: Al = 3.96 μg/L, As = 10.9 μg/L, Cd = 0.92 μg/L, Cr = 49.8 μg/L, Cu = 57.66 μg/L, Fe = 1.79 μg/L, Mn = 289.4 μg/L, Ni = 40.45 μg/L, Pb = 13.1 μg/L, and Zn = 123.4 μg/L (Table 2). Considering the sediment, Cd results were found to be above the crust, whereas Ni results were found to be close (S.W.Q.R., 2016; W.H.O., 2011). Cadmium is a metal, which has toxic effects on organisms and ecosystem even when at low doses. It has a high capacity to accumulate within an organism’s system thanks to its water-solubility and load, ionic radius, and chemical behavior that are identical to calcium (Fikirdeşici et al., 2012; Thornton, 1986). Uncontrolled use of Cd-containing fertilizers is among the most important anthropogenic sources. Moreover, combustion emissions, mining, and atmospheric accumulation play an important role in Cd concentration. Its natural release into nature occurs through the abrasion of cadmium from the rock structure (Kubier et al., 2019). Nickel is an element, which exists in the ecosystem from both natural and anthropogenic sources (Cempel & Nikel, 2006). Phosphatic fertilizers, mud combustion, mining activities, and combustion of diesel fuel, coal, and fuel oil are among the anthropogenic factors playing role in the release of Ni into nature. Moreover, plant cover also triggers the nickel release (Von, 1997; Kabata-Pendias & Pendias, 1992).

Sediment quality assessment

It was determined that the study area was mid-level polluted according to contamination degree (8.63), while very low level of pollution was found according to modified contamination degree (0.78). Examining the contamination factors of metals separately influencing this contamination degree, Pb, Ni, Mn, As, and Cr were found to have low contamination factors, whereas Cu and Zn were found to have mid-level results and Cd to have a high contamination factor. Using the pollution load index, no pollution was found in the study area (PLI < 1). Considering from the aspect of the enrichment factor, it was determined that Cd (62.02; > 50) had an excessively high enrichment level. Cu and Zn metals were determined to have a very high level and enrichment and Pb, Ni, As, and Cr to have a high-level enrichment. Geo-accumulation index values of all the metals, except for Cd, were found to be lower than 0. In other words, no contamination could be found in relation to other metals, but geo-accumulation value of Cd was found to be 1.03. Given the geo-accumulation index results, Cd contamination in the sediment of the study area was found to be at medium level. Given the mean effect range-median quotient (m-ERM-Q) results, the toxicity probability of the sediment was 21%. Given the mean probable effect level quotient (m-PEL-Q) results, it was determined that the sediment was affected at medium level (Table 3).

Table 3 Assessment results by sediment quality guide (SQG)
Full size table

Examining the statuses of elements which have potential to exhibit intense toxic effects in the study area, it was determined that the elements having the highest level of risk were Cd and Ni. The same result was also achieved in different studies (Remeikaitė-Nikienė et al., 2018; Tunca et al., 2018). Du Preez et al. (2016), in a study carried out in South Africa, found the concentration of Cd element to be very high. Cd in the carbonate rocks dissolves because of the rain; some of them accumulate in sediment and some can directly reach water masses (Gabarrón et al., 2018). Thus, Cd can reach organic tissues by accumulating in sediment.

Biological assessment

To investigate the organic life in deep mud in the cave, the samples taken from water and mud by using a small hand scoop were rinsed with sieves in the laboratory to eliminate the mud, and then, they were examined using a microscope. As a result, 40 juvenile individuals and 24 mature individuals from only the Niphargidae family could be detected (Fig. 3). The order Amphipoda currently contains 10,226 described species most of which are free-living and benthic (Horton et al., 2019). The accumulation of heavy metals in sediments usually results in high concentrations of these metals in the sediment. As a consequence, these metals are considered a toxic risk in benthic systems.

Fig. 3
figure 3

Photo of detected Niphargus sp. (Original)

Full size image

Hierarchical cluster analysis (HCA) was used in order to determine temporal and spatial differences/similarities and define the groups. HCA analysis was conducted using seasonal and station averages of the parameters. During the analysis, the Ward method and Euclidian distance showing the relationship between distances and correlation the best were used. Given the HCA results based on the mean values of stations, the similarities and differences of stations were clustered as seen in Fig. 5.

Accordingly, the waters in the cave can be examined under three groups (Fig. 4). Cluster A consists of stations 3, 6, 5, 2, and 8; Cluster B consists of stations 4, 7, 1, and 10; and cluster C consists of station 9. There is no clear spatial distinction. Considering the similarities or differences for sediment samples, among the two clusters, cluster A consists of the first 6 stations. Cluster B includes the other stations, and the figure shows that the two groups clearly distinguish spatially in the sediment (Fig. 4).

Fig. 4
figure 4

Water and sediment variable clusters by dendrogram stations

Full size image

Since the living organism was reached in the sediment, principal component analysis and Pearson’s correlation analysis were conducted in order to reveal the relationships between the organism and its ecological conditions (Mutlu & Aydın-Uncumusaoğlu, 2022; Aydın-Uncumusaoglu & Mutlu, 2022).

At the end of the PCA test, it was determined that 5 principal components could represent the sediment data of the cave (Table 4). Those principal components explain 90.22% of the total variance in the dataset, and the eigenvalues were found to be > 1 (Fig. 5).

Table 4 Varimax rotated factor matrix for the whole data set
Full size table
Fig. 5
figure 5

Component plot in the rotated space

Full size image

The first principal component explaining 38.31% of the total variance has a strong positive load with Cu, ORP, DO, pH, EC, As, and Cd TH (r > 0.75) parameters. Niphargus sp. has a negative load with the first two principal components. Amphipods have a relatively high Cu requirement as oxygen-transport occurs by the copper-based, respiratory protein hemocyanin (Veltman et al., 2008). Detoxification of Cu is a characteristic seen in decapod crustaceans, but not observed in the other Crustacea orders (Guven et al., 1999). With a percentage of 14.17%, the second principal component has a strong negative load only with Zn and Ni (r > 0.75). Niphargus sp. had a strong positive load with the third principal component of the cave sediment. The fourth principal component was found to have a strong positive load with Al. The last (5th) principal component was represented with a strong positive load with WT, which is a physical parameter (Fig. 5).

Pearson’s correlation analysis was conducted in order to reveal the relationship of Niphargus sp. with other parameters in the sediment and, given the results achieved, this species has a strong positive relationship only with Cd (Table 5). Among the parameters analyzed in the sediment, very strong positive relationships were found between Cd, Cr, Cu, Al, Fe, and Ni.

Table 5 Pearson correlation matrix among the Niphargus sp. and other variables (Sediment)
Full size table

Biological indicators and monitoring species are used as a tool to observe how the natural environment changes in response to pollution and other environmental degradation. The use of living organisms to monitor pollution is preferred over traditional water chemical analyses as it provides accurate information about the bioavailability of metals in the environment (Guerra-Garcia et al., 2009; Marin, 2017; Turemis et al., 2018). Bioindicators are organisms that show an ecological effect or change merely by their presence or absence. In this context, amphipod species can also be used as a bioindicator of contamination.

Forty percent of Türkiye’s tectonic units consists of rocks that are suitable for cave formation. Located in the Alpen-Himalayan mountain belt, it is thought that there are more than 40,000 caves in Türkiye. However, since the speleological and bio speleological studies were not sufficient, although 2850 caves could be detected in Türkiye, only 1450 of them could be examined to be registered in Türkiye’s Cave Inventory (Nazik, 2008). The studies in biospeleology started only in the early twentieth century, this discipline is still in its infancy in Türkiye (Kunt et al., 2010). Considering the organisms, caves are interesting habitats from an ecological aspect and caves constitute ecosystems on their own. Since they are semi-isolated environments and there are dark zones, some of which get no light at all, the competitive pressure on organisms living there has decreased and the organisms have evolved adapting to the absence of light (Ozansoy & Mengi, 2006).

In a previous study, Plénet (1999) emphasized that Niphargus species (hypogean) were affected by Cd exposure less than epigean species. However, Cd enrichment in sediment suggests that Cd might become toxic to those organisms too in the future. Amphipods can uptake trace elements through both absorption and ingestion (Barka et al., 2010; Ivanciuc et al., 2006). According to a study by Borga et al. (2006), there may be significant differences in the biological characteristics and interactions of amphipod species with elements. Therefore, it can be concluded that different amphipod species in the same region may differ from each other, especially in terms of element concentrations. Thus, it is important to classify and identify amphipods at the species level. This way, the determination of the characteristics of different species and the understanding of the effects of natural environmental element cycling and biological interactions on species become easier.

Conclusion

The results showed that except for Cd and Ni, there was no elemental accumulation that can be dangerous. For Cd and Ni, the comparisons made using SQGs suggest that those two elements should be given specific attention. These two elements constitute 50% of the toxic effect in the sediment in the study area. A biological indicator species can serve as an indicator of an ecological effect simply by its presence or absence in the environment. As a result of the principal components analysis on the sediment data, the fact that Niphargus sp. was found to have a strong positive load with the first two principal components proves that the existence of living organisms in this cave was already at very low dominance. The exposure of amphipods to heavy metals can disrupt their movement, feeding, and reproductive processes, leading to increased mortality rates and negative effects. It is clear that the organisms detected in the cave would be affected especially by Cd exposure. The data revealed that it is very important to investigate metal accumulation in organic tissues in future studies. Moreover, since organisms might convey the stress even if the exposure ended, this subject is suitable for epigenetic studies.