Use of Microbiological and Chemical Data to Evaluate the Effects of Tourism on Water Quality in Karstic Cenotes in Yucatan, Mexico

Cenotes are spectacular karst formations in Yucatan, Mexico, often used for recreation. However, their impact on water quality has yet to be explored in detail. Therefore, during Easter, water samples were collected from four cenotes to identify variations in water quality associated with the presence of tourists. PCO of water quality, before (PH) and during Holy Week (HW) in 2019, explained 49.02% of the total variation. The indicators contributing to the first principal coordinate’s variation were Sr, K, sulfate, and chloride (0.89). Whereas, alkalinity, temperature, conductivity, nitrate, and ORP contributed to the second PC. PERMANOVA indicated a significant interaction between “cenote” and “condition” factors, and post hoc paired comparisons indicated significant differences between PH and HW conditions. Significant correlations varied among the four cenotes as the result of hydrogeological differences. Whereas, numbers of visitors were correlated with at least one fecal-matter indicator, demonstrating anthropogenic influence on the cenotes’ water quality.


Introduction
The Yucatan peninsula is characterized by its karstic nature, a region without rivers or surficial streams and slight topographic elevation (Schmitter-Soto et al. 2002;Herrera-Silveira et al. 2004;Arcega-Cabrera et al. 2014). Thus, groundwater is the only freshwater source in the state (Aranda-Cirerol et al. 2011). This karst aquifer hosts large amounts of groundwater resources that contain precious and highly vulnerable environments such as sinkholes, mangroves, and seagrass (Bauer-Gottwein et al. 2011, Camacho-Cruz et al. 2020, Cejudo et al. 2020. It is one of the world's unique regions concerning the character and manifestation of karst processes (Lebedeva et al. 2017), since it is dominated by limestones and dolomites with abundant inundated cave systems and geological formations produced by rock dissolution (Scholz et al. 1995).
In the 1980s, Yucatan experienced the takeoff of tourism development (Lebedeva at al 2017). Tourism is one of the activities that many communities in the Maya area have embraced because it provides jobs and income. As part of this, government programs promote the use of cenotes as tourist attractions. As of 2020, 113 cenotes had been certified for tourism-recreational use (SDS 2021, SEFOTUR 2021. Therefore, cenotes, located both in and outside the RC, have become an essential part of the activities offered by communities that depend on eco-tourism in Yucatan. Unfortunately, this growing interest has resulted in some cenotes experiencing an exponential increase in visitors, which may negatively impact water (Enseñat-Soberanis et al. 2020). Thus, water quality for recreational use is of interest since Mexico lacks regulations for recommended values of human exposure to contaminated water, which can generate an environmental and public health problem. Thus, measurements should be taken to promote the sustainable use of cenotes (Hoogesteijn Reul et al. 2015).
Tourism associated with cenotes brings large numbers of visitors, requiring the installation of significant infrastructure, like doors, pathways, lights, roads supporting vehicular traffic, and restrooms, all of which can be sources of contamination. Lights that are used to highlight certain features in caves and help guide people, could cause algal growth which is unsightly and can dissolve the underlying rock. Moreover, visitors may directly impact the cenotes' ecosystems in several ways. The most immediate and visible is litter, both above and under the water. They can also negatively impact caves through respiration, which increases levels of CO 2 and potentially leads to the dissolution of speleothems (Onac and van Beynen 2021). As well, some cosmetics contain heavy metals (HMs), which lead to contamination and represent a risk for aquatic life and human health (Ayenimo et al. 2010;Arshad et al. 2020). Inorganic groundwater contaminants, like nitrate and heavy metals, have been detected in groundwater in the Yucatan Peninsula; and, they have been attributed mainly, but not solely, to waste from agricultural activities and animal husbandry (Arcega-Cabrera et al. 2021). Karstic aquifers are known to be particularly vulnerable to anthropogenic contaminants such as nitrate (Visser et al. 2021). Therefore, considerable efforts have been made to promote the reduction of nitrate inputs, as well as enhance attenuation processes in groundwater in some parts of the world (Yang et al. 2020). However, increases in groundwater nutrient loads are generally attributed to domestic wastewater discharges from nearby urban areas. Direct contributions from human excrements have been ignored and remain unknown, although some studies on water quality in cenotes located around tourist zones in the Yucatan Peninsula have reported nutrients (Camacho-Cruz et al. 2019) with ammonium (average values of 0.01 mg L -1 ), nitrate (7.90 mg L -1 ), nitrite (0.01 mg L -1 ), and orthophosphate (0.01 mg L -1 ). Average reported results in the physicochemical parameters of conductivity and total dissolved solids (TDS) are 1410 µS cm -1 and 865 mg L -1 respectively (Hernández-Mendoza et al. 2022). Additionally, Leal-Bautista et al. (2011) reported total coliform from 4 to 298 CFU 100 mL -1 . Whereas, Borbolla-Vazquez et al. (2020) reported total coliforms from 3 to 240 CFU 100 mL -1 and fecal coliforms above 2400 CFU 100 mL -1 .
Water-based tourism, including tourism at cenotes, has increased the presence of contaminants in groundwater systems (Nava-Galindo 2015; Arshad et al. 2020;Mattos 2020, Casas-Beltrán et al. 2021). Besides, ill-constructed sanitary services may release their contents to the cenotes (Hoogesteijn Reul et al. 2015, Arcega-Cabrera et al. 2021. However, the solution is not to forbid people from swimming in the cenotes but to apply measures that control and minimize this contamination (Enseñat-Soberanis 2020). Since there is no sanitation infrastructure adapted to the karst system in the tourist-recreational cenotes in Yucatan, the preservation of water quality depends on regulation strategies, monitoring programs, and site-specific management.
Therefore, in this study, we examine water quality parameters, including pH, dissolved oxygen (OD), electric conductivity (EC), alkalinity, nitrate (NO 3 -), nitrite (NO 2 -), ammonium (NH 4 + ), orthophosphate (PO 4 = ), chlorides, sulfates, copper (Cu), cadmium (Cd), chromium (Cr), arsenic (As), total coliforms, fecal coliforms, and Enterococcus. These parameters were measured in water samples collected from four eco-tourist cenotes. Sampling was conducted two weeks before the Easter vacation and over a day-long period in each sampled cenote during Easter vacation in 2019. Easter was chosen because it is the period with the highest influx of visitors. The results from these analyses were used to evaluate the response of water quality variables between the pre-tourist season and the high season (Easter vacation) to document probable water quality perturbations related to this activity.

Materials and Methods
The Ring of Cenotes (RC) is in the Yucatan Peninsula's (YP) northwest region, between parallels 88.5-90.5° W and meridians 20.5-21.5° N (Pérez-Ceballos et al. 2021). A subhumid tropical climate characterizes this region, with temperatures ranging from 25 to 35 °C and annual cumulative rainfall from 555 to 1500 mm y − 1 (Gonneea et al. 2014). Within this region, the RC extends from the southwest in Celestun to Dzilam in the northeast, forming a large arc across the middle of the state of Yucatan (Arcega-Cabrera et al. 2014). It is characterized by a complex hydrological system with high permeability and porosity, which makes it vulnerable to contamination (Derrien et al. 2015). Two tourist cenotes along the RC and two to the southeast of the RC were selected for this study ( with a depth from 2 to 50 m; and Yokdzonot, an open-type of free fall cenote with a water lens that is 40 m long by 40 m wide and a maximum depth of 45 m (SDS 2021). All these cenotes are used for ecotourism and have ladders and platforms to access the water.
Samples were collected twice. The first collection, a single sampling operation to establish a reference baseline for conditions during the pre-tourists' season (PH), was conducted two weeks before Easter vacation 2019, in the absence of visitors. The second collection was carried out during Holy Week (HW), as follows: X'Batun and Kankirixche were sampled on April 19th, and Yokdozonot and X'Canche on April 20th. This second sampling campaign was designed to obtain an 8-hour time series (with grab samples every two hours starting at 8 am and finishing at 6 pm, resulting in a total of 6 samples from each cenote, while counting the number of visitors) to assess the potential correlation between water quality variables and touristic activity. The samples from the 2-hour sampling intervals were analyzed separately to obtain an average value for the parameters.
Physicochemical parameters were analyzed in situ using a YSI Professional Plus multiparametric probe (temperature, Fig. 1 Map showing the location of the study area, the four cenotes sampled, the Ring of Cenotes area, and the limits of the State of Yucatan

Results and Discussion
Changes in water quality parameters, monitored before (PH) and during holy week (HW), show clear variation (Table 1).
PCO applied to water quality parameters explained 49.02% of the total variation in the first two principal coordinates (Fig. 2). Also, PERMANOVA indicated a significant interaction between the factors "cenote" and "condition" (Pseudo-F = 3.77, p = 0.0001, 9921 permutations), indicating that the water quality parameters varied differently in the cenotes. Furthermore, post hoc paired comparisons indicated significant differences between the conditions before (PH) and during vacation (HW) for Kankirixche (p = 0.022), X'batun (p = 0.009), X'Canche (p = 0.032) and Yokdzonot (p = 0.01) although these significant differences could be the result of touristic use, there is still an uncertainty related with the environmental hydrogeochemistry and location of each cenote that should be kept in mind.
The indicators that contributed most to the PCO variation in the first principal coordinate (horizontal) were Sr (0.97), K (0.91), sulfates (0.94), and chlorides (0.89). The indicators with the highest contribution to the second coordinate (vertical) were alkalinity (-0.86), temperature (-0.78), conductivity (-0.69), nitrate (-0.73) and ORP (0.69). The ordination showed three groups. The first group includes samples from the Yokdzonot cenote, which showed the highest reducing condition (ORP), as well as high ammonia, and nitrites. These conditions may be present in this cenote because of its depth, as well as the significant amounts of degrading organic matter at the bottom could promote reducing conditions in the water column (Schmitter-Soto et al. 2002).
The second group is formed by the X'batún and Kankirixche cenotes. It is characterized by higher concentrations of silicate, Ca, Sr, Mg, chlorides, and sulfates. This could be a result of location, since in that region sulfates, sodium, and chlorides increase towards the coast (Pérez-Ceballos 2021). Also, these cenotes were the ones with the highest concentration of bacteria, especially fecal coliforms. This could be expected, because these cenotes were the ones that had the highest number of tourists ( Table 2).
Following the classification by Schmitter-Soto et al. (2002), nutrient concentration in lotic cenotes is expected to be lower than in lentic ones, because of the difference in turnover rate. Group three includes the X'Canche cenote. It showed the highest concentration of NO 3 − (7.52 mg L − 1 ), yet all the measured concentrations were below the Mexican guideline values (9.9 mg L − 1 of NO 3 − -NOM-127-SSA1-1994). Historically, nitrate concentrations in the groundwater of the state of Yucatan have ranged between 15.51 and 70.61 mg L − 1 for Merida (Fabro et al. 2015) and 0.6 and 17.7 mg L − 1 for the ring of cenotes area (Pérez-Ceballos et al. 2021); these high concentrations have been dissolved oxygen, pH, TDS, ORP, and conductivity) previously calibrated according to the manufacturer's guidelines. The grab samples of water were collected in the central zone of the water lens' occupation zone (at approximately 0.5 m below the surface) for nutrients and metals. One liter of water for metals analyses was placed in previously washed high-density polypropylene bottles with HNO 3 . For nutrient analysis, 1 L of water was collected in amber glass bottles, which had been acid cleaned and rinsed with deionized water prior to use. Nitrate (NO 3 − ), nitrite (NO 2 − ), ammonium (NH 4 + ), and orthophosphate (PO 4 = ) were analyzed by microplate with the methods described in Ringuet et al. (2011). To validate these data, Simple Nutrients in Seawater-whole volume from Sigma-Aldrich (QC3179-500ML, Lot LRAA9336) was used as certified reference material. Water for fecal and total coliforms was sampled in 250 mL sterile polyethylene bags and then analyzed by the membrane filtration method, according to the Mexican Standard NMX-AA-102-SCFI-2006, following the regulations from the Mexican legislation for water destined for human use and consumption (NOM-127-SSA1-1994), which admits reported values both in CFU 100 mL − 1 and MPN 100 mL − 1 . Hardness, sulfates, and chlorides quantification were sampled in 1 L polyethylene bottle, then analyzed and interpreted according to Mexican standards NMX-AA-072-SCFI-2001, NMX-AA-074-SCFI-2014, and NMX-AA-073-SCFI-2001, respectively. For metals analysis 2 mL of HNO 3 were added to the sample to prevent precipitation or adsorption. Then, the concentration of metals was analyzed using an atomic absorption spectrophotometer (Perkin Elmer Analyst 800) coupled with the reading of a deionized water blank (Arcega-Cabrera and Fargher 2016).
Spearman Rho was calculated to evaluate correlations, using the data from the 2-hour sampling interval, between water quality descriptive variables and the cumulative number of visitors. Data analyses were performed using Origin-Pro 2019b v9.6.5.169 (OriginLab Corporation).
For multivariate analysis, permutational MANOVA (PERMANOVA) and Principal Coordinate Analysis (PCO) were performed using Primer v 7.0 + PERMANOVA. Data were transformed using the function log10(x + 1) and normalized. The resemblance was calculated using the Euclidean distance of samples (Legendre and Legendre 1998); PERMANOVA was conducted using the permutation of the residuals under a reduced model with 9999 permutations to generate the pseudo-F (Anderson et al., 2001). Permutational multiple pair-wise tests with Monte Carlo test were used to compare the centroids of the combination of two factors, the site (cenotes, four levels) and the condition (PH, HW, two levels).

Table 1
Values for the physical, chemical, and bacteriological water parameters from the sampled cenotes, two weeks before (PH) and during Holy Week ( Therefore, nitrates increase in the X'Canche cenote (present as nitrates as a result of the OD and ORP) during the dry season could be related to human activities, as has been previously reported for other sinkholes (Arcega-Cabrera et al. 2021). Human discharges from inland and littoral activities linked to human activities (Long et al. 2018). Nevertheless, the opposite was documented for X'Batun, Kankiriche and Yokdzonot, since sampling was conducted during the dry season. During this season, in absence of runoff, organic matter inputs are limited, and the type of nitrogenated species (nitrates, nitrites, or ammonia) will depend on the redox All the variables listed in the methods were used for the analysis, but only those with correlations higher than 0.5 are depicted parameters during HW, and also that the number of tourists augments the presence of bacteria.
The presence of Enterococcus in X'Batun and Kankirixche (with the smallest water lens diameters of 20 and 25 m, respectively), which is comparatively more resilient, may indicate leaks from septic tanks or even inputs from other anthropogenic activities in the area (Pérez-Ceballos 2012, Arcega-Cabrera et al. 2014. Variables showing a moderate to strong correlation with the total number of visitors in the sampled cenotes are shown in Table 3. The number of significant correlations to the total number of visitors varies among sites; this could be the result of differences in water lens area, depth, water residence time, cenote shape, etc. Number of visitors is moderate to strongly correlated with at least one of the fecal matter indicators in three cenotes. Results show that a higher number of visitors promoted a higher concentration of coliforms and/or Enterococcus, sometimes surpassing normative values. This outcome may have a negative effect on the cenote and on tourists' health (Almeida et al. 2007).

Conclusions
Monitoring programs in cenotes play a significant role in a sustainable use, promoting good water quality, and safe recreational use. The use of cenotes for water-based tourism purposes is increasing in Yucatan. However, even though their use for recreational activities can be beneficial for an individual's health, it can also be damaging if the water is contaminated and eventually becomes unsafe. Our findings concerning the impact of visitors on water quality provide helpful information for informing decision-making and the creation of stronger regulations for the protection of karstic cenotes in Yucatan. These include; i) specifications for infrastructure nearby and in the cenotes, ii) regular monitoring of water quality, iii) improvement of wastewater treatment systems in nearby urban areas, and, iv) prevention rules for tourists visiting cenotes, among others. Also, in situ management programs according to a specific cenote's characteristics and its tourist use are strongly recommended.

Conflict of Interest
No potential conflict of interest was reported by the author (s).
in Yucatan, generate around 2% of the nitrogen (Aranda- Cirerol et al. 2010). Therefore, tourism should not be underestimated as a nutrient source since sudden increases in nutrient concentrations at a specific site, resulting from crowding people during a comparatively short period of time (10 h), may not be buffered by the ecosystem. This could promote an overload of nutrients in time with consequences such as the sudden growth of opportunistic species.
Regarding fecal contamination, since there is no specific legislation for inland recreational waters in Mexico, we use the values for fecal coliform previously established in the US EPA's 1986 recreational waters criteria (US EPA 2012). According to this fecal coliform standard (200 CFU 100 mL − 1 ), Kankirixche and X'Batun exceeded the limit during Holy Week. This could be a result of the sudden increase in visitors (Table 2), bathing in a smaller water lens area, since these cenotes have the most restricted surface area. In Fig. 2, it is possible to note the increase of all the microbiological  Values in bold represent a positive correlation between the water quality parameter and the number of visitors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.