1 Introduction

Sandy beaches are known to provide various ecosystem services including increasing coastal protection, sand storage, water filtration, and shaping energy fluxes between biotic and abiotic components [1]. Additionally, they play an important role in the recreation and entertainment of millions of people worldwide. Unfortunately, coastal regions are also one of the most vulnerable regions that are negatively impacted by anthropogenic activities [2,3,4,5]. Beaches are resilient to natural changes as long as they are unimpeded, but the anthropogenic pressure on the coastal regions restricts the ability of beaches to adjust to the changes in the shoreline [6]. The exponential human population growth in coastal regions leading to rapid urbanisation on the terrestrial side, along with stressors from the seaside due to climate change, is resulting in a ‘coastal squeeze’ and negatively impacting marine biodiversity [7]. In addition, the exponential and unregulated growth in coastal tourism and infrastructure leads to the concretisation of coastal areas and further leads to an increase in coastal erosion. All these consequences indicate the need to rethink development strategies and sustainable planning of coastal infrastructure. Apart from coastal erosion, unregulated beach tourism is also responsible for increasing rates of marine littering and pollution [8].

At the habitat level, beach tourism has a direct impact on species composition in a sandy beach ecosystem. This impact is mainly caused by the trampling effect and beach litter (objects/ substances that have been discarded by people). Sources of beach litter are known to be marine (washed ashore), domestic, and industrial waste dumped into the sea or rivers or on the beach site. Most of the beach litter is considered to be plastic waste accounting for millions of tons worldwide and most of it is believed to be recreational litter [9]. Litter containing harmful chemicals and toxins can also pollute the water impacting the microscopic plankton and other benthic fauna. The beach litter impacts marine wildlife in various ways such as entanglement, ingestion, rafting/ allies for introduced species [9], and can also indirectly impact the species composition by attracting scavengers like feral dogs to the beaches [10]. On the other hand, trampling induces a direct impact on the macrobenthic communities and several studies have reported a decline in the abundance of macrobenthic species due to the trampling effect [1, 11,12,13].

The level of impact, however, is species-specific due to varying levels of adaptability of the species to the disturbance [14]. Species like Donax sordidus and Eurydice longicornis are more resistant and known to be affected only at a high intensity of trampling [1, 14]. In addition to the trampling effect, unregulated beach tourism also poses a threat to marine life due to the high visitation rate, waste management, and litter disposal [9]. Beach tourism causing high levels of human-induced disturbances not only affects the benthic communities, but also alters the abundance, diversity, and behaviour of predators like shorebirds, wild mammals, etc. Human activities are also known to affect coastal avian predators either positively or negatively based on the behavioural traits of the birds and the type of interaction with humans [15].

Our study aimed at understanding the impact of unregulated beach tourism on the crab assemblages with two main objectives, (i) estimating direct impact on indicator species caused by trampling and (ii) estimating indirect impact caused by marine litter and predation pressure on the target species. In this context, we achieved the two objectives by evaluating and comparing the three beaches with different levels of tourism in the Indian coastal state of Karnataka. We hypothesized a negative correlation between the abundance of the macro-benthic fauna and the level of tourism across the gradient of pressure on the beach. We assumed the null hypothesis to be that there is no significant correlation between the abundance of the macro-benthic fauna and the level of tourism at the beach in which case, the sample variables would be normally distributed over the three sampling locations. Also, the current research helps achieve the sustainable development goals 14 and 15 which deal with life below water and life on land.

2 Materials and methods

2.1 Sampling location

Three locations—Malpe beach, Padubidri beach and Mattu beach were selected along the coastline of the Udupi district in the Indian coastal state of Karnataka (Fig. 1). Padubidri is approximately 20 km from Mattu on the same coastline resembling the morphology of Malpe and Mattu beaches. All three locations have similar beach profiles and zones consisting of shoreline, swash zone, beach face, berm crest, berm, and dune systems [16]. Malpe beach (13°21′39.9"N, 74°41′51.0"E) is one of the high tourism beaches of Karnataka hosting several adventure water-sports. Easy accessibility, water sports, availability of beach infrastructure, and eateries attract a high number of tourists to Malpe beach. Apart from tourists, lifeguards, water-sports guides, local vendors, and staff of restaurants contribute to the regular human presence on the beach. Beach litter is controlled to some extent by rag pickers, but the beach authorities do not provide any incentives or impose any disincentives regarding litter management.

Fig. 1
figure 1

Map showing the three sampling locations Malpe, Mattu, and Padubidri in Udupi district of India’s coastal state, Karnataka

Full size image

Mattu beach (13°27′36.1"N, 74°72′49.3"E) is used regularly by local fishers to carry out nearshore fishing and dock their motorized boats. The influx of tourists on the beach is lowest compared to Malpe and Padubidri due to low accessibility and the absence of water-sports and adventure activities. The beach also lacks beach maintenance staff. The area beyond the dune, towards the coastal road, is used as a garbage dump yard by local people, leading to high levels of litter on the beach.

Padubidri Endpoint Beach (13°06′57.1"N, 74°45′56.3"E) is one of the 10 beaches certified as Blue Flag Beaches in India (Blue flag global, n.d.). The beach is a plastic-free zone and tourist activity is regulated by the state tourism department. The beach is equipped with non-concrete and eco-friendly infrastructure made of bamboo and wood. During the study duration, the influx of tourists was moderate-low compared to Malpe because of low accessibility, restricted timings, and paid entry into the beach site. Unlike in Malpe beach, the tourists are concentrated in a single designated area in the swash zone as the beach regulations limit the tourist activities to certain areas of the beach. Though fishing is carried out in this region, it is prohibited within the boundaries of the beach further reducing the human presence inside the designated area.

2.2 Sampling method

2.2.1 Sampling time

We conducted sampling twice a day in three selected locations Malpe, Mattu and Padubidri for 5 days each in April 2021. Owing to the hot climatic conditions, April month is annually declared as summer vacation period for educational institutions by several state/Central governments across most of the country. Hence leading to high tourist influx on the beaches especially during early mornings and late evenings. The level of tourism was assessed based on a review of literature, and informal interviews with stakeholders. Post-hoc the areas were assigned into low, medium, and high disturbed areas, based on the level of tourism present at these locations. While selecting locations for study, we ensured that all three locations had similar physical and morphological characteristics, with the only difference being the level of tourism. For sampling purposes, the beach area was divided into three zones, A, B, and C from the lowest low tide mark (Fig. 2). The zones were delineated based on regular observations of the moisture content of sand with visual references. To minimise the temporal variations in environmental, physical, and morphological characteristics of the beaches, we chose a short sampling duration. The timings of sampling per day were selected in such a way that there was a notable difference in the tourist influx on the beaches so that we can capture the impact of tourism on crab assemblages. As these tropical beaches are hot, especially during the summer season, beaches are most preferred by tourists during the early mornings and evenings and are less frequented during noon. In this case, April is known to be the warmest month of the year in Mangalore with mean temperature 28.9° (climate data.org.). We based our first sampling of the day in the noon from 12:00 pm to 3:00 pm, and the second sampling was in the evening from 4:00 pm to 7:00 pm. The approximate duration of each sampling was 2 h ± 1 h at each zone.

Fig. 2
figure 2

Layout of sampling zones at each study location, where A, B, and C represent zones along the beach width and 1, 2, 3, 4, and 5 represent zones along the beach length. Each sampling plot name (5 × 2 sq. m) indicates the area that falls under respective zones along beach length and width

Full size image

2.2.2 Sampling zones

The ‘A’ zone was assumed in a swash zone consisting of continuous wave action. This zone is usually saturated and exhibits a thixotropic effect as it is constantly exposed to the waves [17]. The ‘B’ zone was in the moist region, which was exposed to the waves during the previous high tide. This region possesses moist and compact sand but is not saturated with water. The ‘C’ zone lies on and beyond the berm where the sand is completely dry. This zone comprises slightly moist sand compared to the dune sand (Fig. 2).

2.2.3 Plot-wise sampling

Within each zone, five plots along the length and 3 plots along the width of the beach were drawn accounting for 15 plots on a spatial scale (Fig. 2). As the sampling was carried out twice a day, the total number of samples was 30/day. The number of samples obtained from all the beaches for 15 days (five days each) accounted to 450. The dimensions of each plot were 5 × 2 m with a spacing of ~ 50 m between each plot in the same zone. Owing to the smaller width of beaches in Mattu and Padubidri, a low distance was maintained between each zone. Moreover, the dynamic nature of the wave action along the shoreline was responsible for the varying distance between the zones. We did not collect live samples of macrobenthos to avoid disturbance [18] and used a non-invasive method of sampling.

2.2.4 Sampling human presence and predation pressure

For sampling the human presence and predation pressure, the point count method was used by setting a radius of 25 m from the selected plot. The number of humans and predators within the radius were counted for 5 min. As they are living entities tending to move from time to time, to avoid recounting the individuals, a 25 m radius was considered as opposed to a 5 × 2 sq. m area in the quadrat method used to sample other parameters. Hence, the radius set for the point count was larger than the plot size. Observations on the number of humans and predators encountered were noted from each plot along with the time of sampling, and zone of sampling. The number of humans per location was grouped into five categories (i) Very low (0–20), (ii) low (21–40), (iii) moderate (41–60), (iv) high (61–80), and (v) very high (81–100). The predators encountered during the sampling were classified into two major types (a) Human commensal species (Crows, Pigeons, feral dogs, and pet dogs) and (b) non-commensal species (Raptors, and waders).

2.2.5 Sampling litter composition

We used the quadrat method of 5 × 2 sq. m for quantifying litter. We noted the observations of litter present on top layers of the sand on each plot and categorized the litter into 4 categories i) non-degradable (plastic waste, Styrofoam, etc.) ii) moderately degradable (discarded clothes, mixed material cloth bags, etc.), iii) biodegradable (food waste, paper waste, etc.) and iv) toxic waste (medical waste, e-waste, etc.). The litter composition includes the types of items belonging to each component found within the sampling plot and does not depict the actual number.

2.2.6 Sampling sand crabs and ghost crabs

We used the quadrat method to record the abundance of sand crabs (Emerita sp.) and ghost crabs (Ocypode sp.) where quadrat dimensions were 5 × 2 sq. m. Sand crabs are usually found in aggregations with many individuals usually associated per aggregation. Using visual observations, the number of aggregations per plot was recorded and categorized into small (200–500 individuals), medium (500–1000 individuals), and large aggregations (1000–1500 individuals). The upper limit of the range was used to calculate the abundance of sand crabs. Video clips were taken whenever possible to add accuracy to the count.

A non-invasive method of sampling was employed to calculate the abundance of ghost crabs, involving the burrow count along with the visual observations of the crabs present outside the burrows. Ghost crab burrow count and number of live ghost crabs helped in analysing the abundance as well as the impact of disturbance on the behaviour. Ghost crabs are diurnal and show a consistent pattern of behaviour [19] but show altered behaviour in the presence of anthropogenic disturbance [20]. All the observations of ghost crab count were noted for each plot separately.

3 Data analysis

The data was organised into columns in nested format i.e., date, time of sampling, location/ beach name, plot name, human count (categorized into five groups i.e., 1–20, 21–40, 41–60, 61–80, 81–100 people), predator pressure (categorized into natural and human commensal species), litter composition (categorized into (i) non-degradable (ii) moderately degradable, (iii) biodegradable and (iv) toxic waste), sand crab count, ghost crab, and burrows.

To analyse the normality of the data, a one-sample K-S test (Kolmogorov–Smirnov test) was carried out using IBM SPSS Statistics software (version 26). In this test, the test statistics is the maximal absolute value (D), and the probability value (p-value) less than 0.05 (p < 0.05) indicates that the distribution of the data is not normal, concluding that the null- hypothesis can be rejected. The abundance of each component was calculated individually for each plot per beach. Using Microsoft Excel software, the mean and standard deviation of each variable to compare the variations among the study locations, time of sampling and zone of sampling and plotted them on bar graphs with the error bars of 95% Confidence Interval (CI) using the IBM SPSS Statistics version 26 software.

To assess the factors affecting the abundance of the target species, spearman’s rank-order (rho) correlation was performed with two-tailed significance in IBM SPSS Statistics version 26 software. It resulted in the correlations for each pair of the selected variables showing the degree to which the variables are correlated and the direction of correlation.

4 Results

In the results, we present our study findings indicating the following parameters: the normality of the data, abundance of each variable, the dependency, and the correlation between variables.

4.1 Normality test

One sample KS test of all the variables i.e., total human count, natural predators, human commensal predators, total predation pressure, total litter, total sand crab count, ghost crab count, and total number of ghost crab burrows did not show normal distribution (Table 1) suggesting rejection of null hypothesis.

Table 1 Showing the parameters considered for sampling and the One sample KS stats
Full size table

4.2 Abundance

4.2.1 Human count

The number of humans varied significantly with the sampling location (Malpe, Mattu, and Padubidri) and the time of sampling (noon and evening). Malpe showed a high number of tourists (mean = 36.5 ± 17.9) compared to Mattu (mean = 6.5 ± 0.7) and Padubidri (mean = 9.6 ± 0.9). The human count was found to be higher during evenings (mean = 22.6 ± 1.3) compared to noon (mean = 12.4 ± 1.1). With reference to the zone of sampling, the human count was almost the same in all three zones (mean≈17). The tourists were frequently found to be clustered on the beach (Fig. 3).

Fig. 3
figure 3

Infographic showing the response variables—human disturbance, litter, and predation pressure and their impact (colour coded) on the abundance of indicator species—sand crabs and ghost crabs based on location, zone, and time of sampling

Full size image

4.2.2 Litter composition

Litter quantification was carried out for each type (non-degradable, moderately degradable, bio-degradable, and toxic litter) individually as per the same metrics. The composition of the litter varied at 3 sampling locations. Mean total litter was found to be higher in Malpe (mean = 5.52 ± 0.21), moderate in Mattu (mean = 3.42 ± 0.24) and lowest in Padubidri (0.49 ± 0.66). Non-degradable litter was highest in Malpe (mean = 3.0 ± 0.1). Moderately degradable and toxic litter was lowest in Padubidri (mean = 0.0, 0.01 respectively). The litter composition did not vary much with respect to the time of sampling (Total litter mean ≈ 3.0). However, with respect to the Zone of sampling, the C- zone had the highest total litter mostly consisting of non-degradable litter (Fig. 3).

4.2.3 Predation pressure

Total number of predators found in the area i.e., predator count was used as a measure to understand predation pressure. The mean predator count (per 25 sq. m radius/5 min) of each type (Raptors, Crows, Waders, Pigeons, Feral Dogs, Pet Dogs) is individually calculated for sampling locations, zone of sampling and time of sampling. Total predation pressure was found to be highest in Malpe (mean = 1.9 ± 0.2), moderate in Mattu (1.2 ± 0.3) and lowest in Padubidri (0.4 ± 0.1). Total predation pressure was found to be slightly higher during evenings (mean = 1.2 ± 0.2) compared to noon (mean = 1.1 ± 0.1). Concerning the zone of sampling, total predation pressure was higher in C- zone (mean = 1.4 ± 0.2) and nearly equal in A and B (mean ≈ 1.0). Among the two types, human commensal species contributed most to the total predation in all three scenarios (location, time and zone of samples) (Fig. 3).

4.2.4 Sand crab abundance

The mean of total sand crab abundance (per sq. m) was calculated separately for sampling locations, time of sampling and zone of sampling. The sand crab count was highest in Malpe (mean = 556.7 ± 99.8), moderate in Mattu (mean = 530.0 ± 88.5) and lowest in Padubidri (mean = 16.6 ± 11.0). Total sand crab count was higher in the evenings (mean = 504.4 ± 75.5) compared to noon (mean = 231.1 ± 51.3). As sand crabs tend to inhabit swash zone, they were found only in the A- zone (mean = 1103.3 ± 117.1) (Fig. 3).

4.2.5 Ghost crab abundance

To estimate ghost crab abundance (per sq. m), both live crab count and burrow count were calculated separately. Total ghost crab count was found to be negligible in Malpe (0.0 ± 0.1) compared to Mattu and Padubidri (mean ≈ 1.3). Total ghost crab count was higher in the evening (mean = 1.0 ± 2.1) compared to noon (mean = 0.8 ± 1.6). Ghost crabs were abundant in B- zone (mean = 2.0 ± 2.7) compared to A- zone (mean = 0.6 ± 1.1) and C- zone (mean = 0.2 ± 0.6).

The number of ghost crab burrows were higher in Mattu and Padubidri (mean ≈ 10) compared to Malpe (mean = 1.8 ± 3.7). Burrow count did not vary much with respect to time of sampling (mean ≈ 7.0) but with respect to zone of sampling, more burrows were found in B and C- zones (mean = 11.2 ± 8.4 and 10.3 ± 8.8 respectively) compared to A- zone (mean = 0.2 ± 0.6) (Fig. 3).

4.3 Chi-square test

From the Chi-square test results for all the parameters observed, it was evident that, all the variables, total human count, abundance of natural predators, abundance of human commensal predators, total litter count, total sand crab count, ghost crab count, and ghost crab burrow count are dependent on the location of sampling with p- value less than 0.05 (Table 2). However, only total human count, abundance of natural predators, and ghost crab burrow count are dependent on the time of sampling with p-value less than 0.05 (Table 3). The variables, abundance of natural predators, total litter count, total sand crab count, ghost crab count, and ghost crab burrow count, are dependent on the zone of sampling with p- value less than 0.05 (Table 4).

Table 2 Chi-square test for dependency on the location of sampling results
Full size table
Table 3 Chi-square test for dependency on the time of sampling results
Full size table
Table 4 Chi-square test results for dependency on the zone of sampling
Full size table

4.4 Correlation

The bivariate spearman’s correlation of all the variables showed a number of significant correlations. Human count showed significant positive correlation with human commensal predators (r = 0.3, p < 0.05) leading to significant correlation between total predation pressure and human count (r = 0.23, p < 0.05). Mean total litter was found to be significantly correlated with human count (r = 0.3, p < 0.05). Total sand crab count did not show any significant correlation with human count but showed a significant positive correlation with natural predators indicating that natural predators were abundant in the presence of sand crabs (r = 0.13, p = 0.01). On the other hand, ghost crabs (r = -0.3, p < 0.05) and burrows (r = -0.3, p < 0.05) showed a negative correlation with a human count indicating a negative impact of human presence on the abundance of ghost crabs. In addition, the ghost crab abundance was also found to be significantly and negatively correlated with human commensal predators (r = -0.3, p < 0.05). Ghost crab abundance showed strong negative correlation with total litter (r = -0.2, p < 0.05) (Table S1-S6).

5 Discussion

Based on the findings, Malpe can be categorised as a highly disturbed beach, Padubidri and Mattu as moderately and least disturbed beaches respectively. Malpe beach has several recreational activities and amenities which makes it the most preferred beach among the three study locations. This high tourist influx results in high disturbance levels on the beach. Interestingly, while the number of tourists visiting Mattu and Padubidri was almost equal, it was observed that the human count in Mattu was largely contributed by artisanal fishermen who drag their fishing crafts into the sea on a small portion of the beach, rather than trampling the beach like tourists in Malpe and Padubidri do. In addition to trampling, we found out that littering was also a direct consequence of human disturbances in crab assemblages.

5.1 Impact of litter

Other studies have reported that over 663 marine species have been affected by marine litter out of which 15% are included in the International Union for Conservation of Nature (IUCN) Red List. This impact is mostly through entanglement and ingestion [9, 21]. Although we observed that litter quantity was high on tourist beaches, we were unable to derive the source. This could be either caused by the tourists littering on the beaches or the marine litter washed up onto the shores. Most litter in our study locations was contributed by non-degradable waste majorly consisting of plastic debris, as indicated by other studies conducted on the coast [22].

Being a well-managed beach where the litter was cleared from time to time, the litter quantity observed in Padubidri was low compared to the other two beaches. However, the landward side of Mattu Beach is used as a garbage dump yard from which litter is carried into the beach, despite it being a non-tourist beach. In all three study locations, this beach litter was found to indirectly influence the predator pressure on the crab assemblages. While, it has such an impact on the beach ecosystem, studies also suggest that this impact is not unidirectional, i.e., high tourism results in high beach litter but the continuous and uncontrolled addition of litter on the beaches also tends to impact the tourism sector [23].

5.2 Impact of predation pressure

Human-induced disturbances can potentially alter the habitat selection of birds and other natural predators on sandy beaches. The alteration can be both a positive and/ or negative influence based on the type of interaction with the humans [15]. In this study, we observed that human- commensal species such as crows, feral dogs, pet dogs, few raptors (Black kites) and pigeons dominated the beaches whereas, non-commensal species such as few waders (whimbrel, oriental pratincole), Raptors (White-bellied sea- eagle), etc. were not abundant where the disturbance levels were high. The threat is high especially from the feral dogs as they feed on local wildlife and also compete with natural predators/ scavengers for the resources [24]. This can alter the ecosystem's food web and energy flow on a sandy beach. In addition to increasing the predation pressure on the crab assemblages, human commensal predators also alter the habitat selection of the natural predators leading to community restructuring [15].

5.3 Nature of impact on crustaceans

Sand crabs are filter feeders often found burying themselves in the sand with feathery antennae protruding out to feed on planktonic contents brought by the waves in the swash zone [25]. They are known to be highly sensitive to pollution which makes them a good indicator of anthropogenic disturbances in the intertidal zones of sandy beaches [26]. However, our study findings indicate that sand crab abundance was unaffected by human count. The methodological drawback of sampling the sand crabs based on visual inspection limits the possibility of drawing strong conclusions on the correlation between their abundance and human-induced disturbances. To overcome these limitations, future studies could collect their samples along with their behavioural data, which could provide deeper insights into their role as bioindicators for human-induced disturbances.

Ghost crabs respond to disturbances easily; hence they are considered bioindicators [20]. This response has been earlier recorded in research that ghost crabs declined in number significantly with trampling [20, 27]. Building evidence to this, in our study, we found least number of ghost grabs and their burrows on the highly disturbed Malpe beach. Furthermore, they are diurnal and have consistent activity patterns throughout the day. They construct new burrows from early dawn till afternoon. During the afternoon, they plug the burrows with damp sand and wait till sunset to emerge out of burrows and scavenge for food during the night [19, 28]. Thus, studying ghost crab and burrow count separately helped us to understand the effect of trampling on their abundance as well as their behaviour.

Interestingly in our study, we also found that ghost crabs were not only affected by trampling but also affected by predators. As high tourism leads to higher rates of littering, this attracts many human commensal species on the beach, which impose additional predation pressure on the macro-benthic species on a sandy beach. This indicates that human presence affects ghost crabs directly and indirectly.

Several studies across various landscapes have identified that the presence of human commensal species can be detrimental to the local ecosystems in several ways including competition for resources, predation pressure and overall changes in the community structure of the region [24]. In our study, we observed that natural predators like waders were abundant where the sand crab population was high. This predator–prey interaction when interrupted by human commensal species like feral dogs can have a potential negative impact to the local ecosystem. Similarly, we observed scavenging species like crows and feral dogs near the trash cans while studies have reported that ghost crabs tend to seek refuge near trash cans when human-induced disturbance is high making them prone to high predation pressure [19].

6 Conclusion

There are several studies from various countries focusing on the disturbance caused by human visitation on the beaches primarily focusing on trampling effect caused by off-road vehicular traffic, dredging, and coastal defence structures. However, the beach morphology, dynamics and coastal tourism differ significantly across the landscapes. Our study location in India, for instance, have low off-road vehicular traffic and more human presence on the beaches compared to other countries. Thus, the study provides more regional insights into the impacts of unregulated beach tourism on sandy beach ecosystem of Indian west coast.

Among the study location, Malpe is a highly disturbed tourist beach with high amounts of litter attracting more predators. Mattu is a low tourist and unmanaged beach but contains moderately higher litter. Padubidri is also a low tourist well-managed beach with low levels of litter. On all these beaches, we observed that human presence negatively impacts crab assemblages both directly (trampling) and indirectly (attracting human commensal predators through beach litter). We observed that ghost crabs tend to seek refuge near trash cans where scavengers like crows and feral dogs are often found and such high disturbances make them prone to high predation risk. This impacts the behaviour and abundance of macrobenthos and leads to changes in the community structure. Among other target species, sand crab abundance is not much influenced by human presence. A long-term study of sand crabs, however, must consider both biotic and abiotic factors to understand these trends better. Species-level identification of the fauna was not possible since no biological samples were collected from the sampling location. Due to which we could not quantify the impact of uncontrolled beach tourism on each species, trophic interactions, and community assemblages of our study locations. To gain a deeper understanding of this, we recommend more studies incorporate biological sample collection, species-level identification, and behaviour of target species.

From the insights obtained so far from this study, we conclude that disturbances caused by tourist visitation negatively impact the crustaceans both structurally and functionally leading to further changes in a sandy beach ecosystem. Hence, we recommend that there is an urgent need to implement effective crowd and waste management plans at high tourist locations to avoid the impacts of tourism on macrobenthic communities.