Habitat Matters: Behavior and Activity of Round Goby (Neogobius melanostomus) at Different Substrates

Abstract

This study focuses on the behavior of the Round Goby in relation to various types of hard substrates, which may be crucial in the context of the emergence of hydrotechnical structures in their habitat. The aim of this study was to determine the behavior and habitat preferences of the Round Goby in relation to specific types of hard substrates. Three types of substrates were used in the experiments: bare, colonized by benthic organisms, and rocky. Laboratory observations were conducted and divided into three variants, each with two cycles—daytime and nighttime. Each variant involved different combinations of two substrates. It was found that during the day, fish spent more time on vegetated substrates than on bare or rocky ones. At night, the opposite trend was observed. The lowest activity was recorded when the fish had a choice between bare and rocky substrates. The largest fish in the experiments changed their habitat most frequently, indicating that body size influences the activity level. However, no significant differences were found in either the time spent on specific substrates or the number of substrate changes. In addition, it was observed that the total length of Round Gobies did not affect their activity level.

1. Introduction

The southern region of the Baltic Sea is characterized by sandy or sandy–muddy sediments [1]. Hard substrates occur naturally in this part of the Baltic in relatively small areas. This type of seabed, along with the organisms that inhabit it, is more heterogeneous compared to hard, smooth substrate before it is colonized or bare stones. As such, it provides shelter and breeding grounds for many fish species, including the Round Goby [2] (Figure 1). The Round Goby (Neogobius melanostomus) is an alien species, first recorded in the Baltic Sea in 1990 [3]. The Round Goby is a benthic fish that prefers habitats with hard substrate, particularly sites where mussels are present. These environments not only provide shelter but also serve as feeding grounds, contributing to the successful establishment and proliferation of the species in the Baltic Sea. The Round Goby has demonstrated remarkable adaptability, colonizing various types of substrates, including artificial structures such as piers and coastal reinforcements, which have become more prevalent as a result of the intensification of human activities [4]. Depending on the region of occurrence, N. melanostomus may reach different sizes, but its total length does not exceed 25 cm [5]. It is characterized by a distinctive black spot on the first dorsal fin and fused pelvic fins, which form a suction disc that allows it to attach to rocky substrates and other hard surfaces [6]. Its native range includes the Black Sea and the Caspian Sea, but its spread, including through maritime transport, has led to its establishment in several regions of Europe and North America. Its spread may result in competition with native fish and their elimination [7,8]. It is a eurytopic species, tolerating a wide range of salinities, from freshwater bodies to brackish and marine environments [9].

It is likely to have been introduced into the Baltic Sea via ballast water from ships passing through canals connecting the Black Sea and the Caspian Sea with the Baltic Sea [3]. The species exhibits various behaviors depending on the type of activity. In addition to swimming, horizontal movements include so-called “hopping” and “darting”. In search for shelter, the Round Goby can burrow into the substrate or bury itself completely in it [10].

Despite the growing body of research on the distribution and ecological impact of the Round Goby in both native and invasive ranges [6,11], there is still a lack of studies presenting quantitative, detailed data on the behavior and activity patterns of this species. Understanding behavioral traits, such as movement, feeding times, habitat use, and interactions with other species, is essential for assessing the ecological role of the species and for developing effective management strategies [12,13]. Several studies have examined movement patterns, particularly related to seasonal migrations, habitat shifts, and invasion fronts [13,14,15]. Habitat preferences of N. melanostomus are well studied in relation to substrate, water flow, and temperature. In the Great Lakes, the Round Goby was observed to prefer rocky, structured habitats [16]. Interactions with other species are probably the best studied aspect, especially with regard to competition and predation [6,17,18]. There are some studies that have investigated the effects of substrate complexity on the behavior, habitat selection, shelter use, and reproductive strategies of the Round Goby. Substrate complexity is an important factor shaping the ecological success of individuals of this species, especially in invasive ranges. The species selects microhabitats with more crevices and shelter availability, especially in predator-rich environments [19]. Specimens living on complex substrates displayed more aggressive and territorial behavior [20]. This type of habitat also provides better protection for eggs from predators and currents [10].

Increasing human impact on the marine environment has led to an increase in the proportion of hard substrates in the Baltic Sea, i.e., hydrotechnical structures such as ports, wind farms, piers, and coastal reinforcements. Does the Round Goby prefer only hard substrate in advanced stages of succession? Does the structure of substrate matter? Most studies of the habitat preferences and behavior of N. melanostomus are based on underwater observations and fish catches in specific regions. The stage of benthic succession on hard substrates in the Baltic Sea seems to affect the quality of these substrates for animals.

The objective of this study was to determine the behavior of the Round Goby and its habitat preferences for specific types of hard substrate. The hypothesis was that hard substrate in the early stage of succession is a more attractive habitat for the Round Goby than hard substrates with no fouling organisms, i.e., smooth, bare substrate or pebbles.

2. Materials and Methods

Fish were collected using a beach seine in the coastal zone of the Gulf of Gdańsk (southern Baltic Sea) in August and September 2024. A total of 96 specimens of N. melanostomus were collected. The specimens were subjected to acclimatization for two weeks to adapt to laboratory conditions. During this time, they were fed natural food (invertebrates) collected from the environment. Throughout the acclimatization period and between experimental procedures, the fish were kept in tanks with filtered (mechanically and biologically) and aerated artificial seawater (salinity of 7 PSU). The tanks also contained shelters. A series of laboratory experiments was conducted on individuals of the Round Goby with a total length of 4.8 cm to 10.4 cm. The fish were subadults that had not yet developed all adult characteristics. The reason for this choice was to provide the specimens with sufficient space in the experimental chambers and to avoid inducing behaviors that could be related to sex. We used three types of substrates—bare smooth substrate, bare pebbles, and substrate with fouling organisms in the early stage of succession—which were tested in three different combinations (Figure 2):

• Bare smooth substrate and bare pebbles (bare substrate vs. rocky substrate);
• Bare smooth substrate and substrate with fouling organisms in the early stage of succession (bare substrate vs. colonized substrate);
• Substrate with fouling organisms in the early stage of succession and bare pebbles (colonized substrate vs. rocky substrate).

Figure 2. Bare (uncolonized) substrate (A), rocky substrate (B), colonized substrate (C), and experimental setup with two different substrates in eight chambers (D).

Figure 2. Bare (uncolonized) substrate (A), rocky substrate (B), colonized substrate (C), and experimental setup with two different substrates in eight chambers (D).

The number of benthic species and their percentage cover were determined on the colonized substrate. The experiment was conducted under laboratory conditions. The temperature was maintained at 17 °C, and the light cycle was set to 16 h of light and 8 h of darkness (16L:8D). The temperature and light cycle are characteristic of late summer in the southern inshore waters of the Baltic Sea and were used because fish activity during this period is the highest. This allows us to capture changes in fish behavior, which should be more noticeable than in autumn and winter. Fish tanks with bottom dimensions of 79 cm × 58 cm were used, and each tank was divided into eight separate chambers of equal size. Two different substrates and one Round Goby were placed in each chamber (bottom: 15.5 cm × 31 cm). The behavior of the fish was recorded using two Samsung SNB-6004P (Samsung Electronics Poland, Warsaw) (having a maximum image resolution of 1920 × 1080 pixels and the ability to transmit video at up to 60 fps) cameras mounted on a frame 1.25 m above the tanks. An IR LAB LIR-IC88 infrared emitter (IR LAB Poland, Warsaw) with a wavelength of 850 nm and a range of up to 180 m was also mounted above the tanks to track fish movements at night. Each experiment was recorded for 24 h, including footage during two day cycles (2 × 150 min) and two night cycles (2 × 150 min). During the video analysis, the location of each individual was recorded for a period of 150 min, with data collected every minute. The data were collected by an observer, who noted the location of the fish based on a video recording. A total of 32 repetitions of the experiment (16 during the day and 16 at night) were performed for each substrate combination. After each repetition, the fish were replaced. Between recordings, the fish were kept in separate tanks with shelters, where they were fed before and after each trial. Statistical analyses were conducted in MS Excel 365 (Version 2505 Build 16.0.18827.20102 64-bit) and Dell Statistica 13.0. The proportion of time spent by fish on different substrates was presented for all trials in one experiment (16 chambers × 150 min; n = 2400 min). The Shapiro–Wilk test was used to check the normality of the data distribution (normal distribution at p > 0.05). Non-parametric tests (Wilcoxon test and Mann–Whitney U test, at p < 0.05) were used to compare samples for significant differences. The Kaiser–Meyer–Olkin (KMO) test and Bartlett’s test of sphericity were used to determine whether the variables are suitable for factor analysis. Exploratory factor analysis was used to identify the relationships between variables (time spent on a specific substrate, light conditions, number of substrate changes).

3. Results

3.1. Bare Substrate vs. Rocky Substrate

Analysis of the total time spent by the fish on different substrates showed that during the day, individuals spent 52% of their time on bare substrate. At night, the fish spent 43% of their time on bare substrate (Figure 3). The total length of the specimens used in this experiment ranged from 5.6 cm to 9.0 cm during the day cycle and from 4.8 cm to 9.8 cm during the night cycle (Figure 4A,B). In 32 trials (day and night cycles), there were nine cases where fish did not change their position (Figure 4A,B). Among individuals that changed substrates most frequently, the ratio of time spent on each substrate was close to 1:1. Data on time spent on uncolonized and rocky substrate at night followed a normal distribution (p > 0.05). The remaining data were characterized by a non-normal distribution (p < 0.05). Sampling adequacy was assessed using the KMO test, which yielded a value of 0.626, indicating meritorious sampling adequacy. Bartlett’s test of sphericity was significant (p < 0.001). Neither the number of substrate changes nor light conditions were correlated with the time spent on any of the substrates. It was found that the type of substrate and the level of fish activity explained 76.3% of the variability in the data. Fish activity was correlated with the type of substrate and was higher on rocky substrate (Figure 4C).

3.2. Bare Substrate vs. Colonized Substrate

Thirteen taxa of zoobenthic and phytobenthic organisms were found on substrates that had undergone natural colonization. Bryozoans dominated in the communities that colonized the substrates. Analysis of the total time spent by the fish on substrates showed that during the day, Round Gobies spent more than 64% of their time on the vegetated substrate. The opposite was observed at night, with fish spending only 43% of the observation time on the vegetated substrate (Figure 5). The total length of the specimens used in this experiment ranged from 5.5 cm to 10.4 cm during the day cycle and from 5.4 cm to 9.5 cm during the night cycle (Figure 6A,B). Analyzing the time spent on each substrate during the day (Figure 6A) and night (Figure 6B), as well as the number of substrate changes for individual fish, we found that most individuals spent a similar amount of time on both substrates. Of the 32 trials, only two fish did not change their position even once, and their total lengths were 6.8 cm and 7.9 cm (Figure 6A,B). The time spent by fish on both substrates during the day and night, as well as the number of substrate changes during the day, followed a normal distribution (p > 0.05). In contrast, the data describing the number of substrate changes at night had a non-normal distribution (p < 0.05). The KMO value was 0.663, and Bartlett’s test was significant (p < 0.001). A positive correlation was found between light conditions and time spent on the colonized substrate. The number of substrate changes was not correlated with the time spent on any of the substrates. It was determined that the type of substrate and the level of fish activity explained 81.8% of the variability in the data. There was no clear correlation between fish activity and the type of substrate; however, fish were more active at night (Figure 6C).

3.3. Colonized Substrate vs. Rocky Substrate

It was observed that, during the day, the fish spent 65% of their time on vegetated substrates and 35% on rocky substrates. At night, the observed individuals spent most of their time on the rocky substrate (59%; Figure 7). The total length of the specimens used in this experiment ranged from 5.2 cm to 8.8 cm during the day cycle and from 5.9 cm to 9.7 cm during the night cycle (Figure 8A,B). Comparing the amount of time spent on each substrate and the number of changes in each chamber, it was found that fish that changed their location frequently spent a similar amount of time on both substrates (Figure 8A,B). Of the 32 trials (day and night), there was only one case when a fish did not change substrates—an individual in chamber 1 (Lt = 7.7 cm) during the second day cycle (Figure 8A,B). The time spent on substrates and the number of substrate changes during the night followed a normal distribution (p > 0.05). The time spent on substrates and the number of changes during the day were characterized by a non-normal distribution (p < 0.05). The KMO measure of sampling adequacy was 0.738, and Bartlett’s test of sphericity was significant (p < 0.001). Positive correlations were found between light conditions and time spent on the colonized substrate, and between the number of substrate changes and time on the rocky substrate. It was determined that the type of substrate and the level of fish activity explained 83.1% of the variability in the data. The activity of the fish was more varied on the rocky substrate (Figure 8C).

Differences in the time spent on two substrates (Wilcoxon test) during the day and night and in the number of substrate changes (Mann–Whitney U test) were not statistically significant (p > 0.05).

4. Discussion

The Round Goby was thought to be a primarily nocturnal species [21]. Due to the presence of superficial neuromasts, the species is more sensitive to water movements than other perciforms, which enables it to navigate and feed effectively in the dark [22]. In this study, the individuals spent most of their daytime on colonized substrates, which is consistent with the findings of Dubs and Corkum [23]. Experiments conducted by Dubs and Corkum confirmed that the Round Goby specimens spent less time in shelters at night compared to the daytime. While no shelters were used in this study, the colonized substrate was considered a place where fish might seek refuge due to its three-dimensional structure. Although the differences between the time spent by fish on different substrates (habitats) at night and during the day were not statistically significant, we noticed a clear trend concerning their habitat preferences. It should be noted that the individuals involved in the experiment were not hungry, so they likely did not search for food during substrate selection. From a behavioral perspective, patchy habitats (in the case of our experiment—colonized substrate) can promote site fidelity and territoriality by facilitating the establishment of discrete home ranges that optimize trade-offs between foraging efficiency and predator avoidance. Males in particular rely on structurally suitable nesting substrates to attract females and provide parental care [24]. Analysis of the size of individuals and the number of substrate changes across all variants raised many questions, especially after comparing our results with the available literature. The smallest individuals were expected to be the most active as they had more space to move in the experimental chambers compared to larger individuals. During the experiment with uncolonized and rocky substrates (pebbles), most individuals remained stationary throughout the experiment (Figure 4). However, in both the night and day cycles, these were medium-sized individuals, measuring between 5.3 cm and 9 cm (Figure 4A,B). On the other hand, the individuals that changed their positions most frequently (second night cycle—chamber 1, second day cycle—chamber 3) were also the largest in their respective trials. The same situation occurred in the second variant (uncolonized vs. colonized substrate) during the night cycle, where the largest individual in the trial changed its position most frequently. During the daytime cycle, the most active individual was of medium size compared to other specimens involved in the trial. Only two individuals did not change their position throughout the observation period (Lt = 7.7 cm and Lt = 8.2 cm) (Figure 6A,B). Research by Jude and Deboe [16] showed that individuals of the Round Goby most often inhabit rocky areas (rocky outcrops) covered with vegetation or existing crevices that provide them with shelter and a place to build nests. The distribution of juveniles is rather non-random, as they favor shallower, densely vegetated areas, likely as a refuge from predators and to gain better access to food resources [25]. In our experiment with colonized and rocky substrates, the fewest individuals exhibited no activity—only one individual from the second daytime cycle in chamber 1 did not change its position (Lt = 7.4 cm; Figure 6B). It was also observed that the individual that changed its position the most during the night cycle was also the largest (Lt = 9.7 cm), while the most active individual from the daytime cycles was of a medium length of 6.2 cm (Figure 6A,B). However, there were no statistically significant correlations between the length of the fish and other variables (type of substrate, light conditions, number of substrate changes) at p < 0.05.

Contrary to previous assumptions, this study showed that four out of six individuals that changed substrates the most frequently were also the largest within specific trials. This finding is not consistent with the results obtained by Behrens et al. [26], who demonstrated that fish identified as bolder (exhibiting greater activity) were smaller. Galli et al. [27] reached similar conclusions in their study on the behavior of Round Gobies, noting that body size significantly affects activity levels, with smaller fish being more active. In our study, individuals showing zero activity in each repetition were of medium size. Galli et al. [27] found that smaller gobies exhibited a longer freezing time after exposure to a stressor compared to larger individuals. However, it should be noted that the authors analyzed fish measuring between 10 cm and 18 cm, which were larger than those observed in this study. Taking the above information into account, it was concluded that there is no clear relationship between fish size and the number of substrate changes in experiments with fish 4.8–10.4 cm long. This result is consistent with the conclusions of Marentette et al. [10], who found in their research on the activity of N. melanostomus that body size does not affect mobility. As for the size of the Round Goby, it is important to consider that these fish exhibit complex reproductive strategies, including the presence of alternative male morphs such as sneaker males. These smaller males mimic females in appearance and behavior to stealthily access nests guarded by larger, territorial males and fertilize eggs [20,28]. This tactic reduces their need to compete physically, which may partly explain their different activity patterns. In contrast, larger males often invest in defending territories and nests, a behavior that can involve frequent movements between substrates during mate attraction or nest defense [10]. In addition, at high population densities, larger specimens could cover longer distances and colonize the most optimal territories [29]. These divergent strategies may contribute to the observed variability in activity across different size classes and help clarify why size alone may not be a reliable factor for predicting behavioral trends in this species.

In summary, our findings underscore the complex interaction between environmental factors and individual behavioral traits in shaping the habitat use and activity patterns of the Round Goby. While the results provide valuable insights into the species’ behavioral flexibility and potential ecological strategies, it is important to note that this study was conducted in an artificial environment. Such conditions may not fully replicate the complexity of natural habitats and could affect certain behavioral responses, even if minimally. Nonetheless, the observed ability of the Round Goby to interact with different substrates and to adapt their activity patterns suggests a level of ecological versatility that likely contributes to their success as an invasive species.

5. Conclusions

During the day, the Round Goby spent more time on colonized substrate than on smooth bare or rocky substrate. At night, the opposite trend was observed.

The lowest fish activity was recorded when they had a choice between smooth bare and rocky substrate. This suggests that the Round Goby may treat new, uncolonized hydrotechnical structures similarly to bare rocky substrate.

The variability in the results was primarily determined by the type of substrate, which explained more than 50% of the data variability in each experiment.

No effect of fish size was observed on the activity of the Round Goby with a total length of 4.8 cm to 10.4 cm.