Environmental Preferences and Fish Handling Practice among European Freshwater Anglers with Different Fishing Specialization Proﬁles

: Recreational specialization is characteristic of many activities, including recreational ﬁshing, which is a popular and important form of recreation throughout the world. The pressure on the environmental resources used by anglers is increasing. It is becoming crucial to understand the preferences and behaviors of anglers, which can accumulate and multiply the risk of damage to ﬁsh stocks and aquatic habitats. The main objective of the study was to comprehensively analyze the differences between anglers with diverse specialization proﬁles. By investigating the ﬁshing preferences within the context of the socioeconomic, demographic and engagement factors, three groups of anglers were identiﬁed: anglers who specialize in predatory ﬁsh, anglers who specialize in non-predatory ﬁsh and unspecialized anglers. Specialized anglers, regardless of type, were found to be more supportive of releasing caught ﬁsh (71% on average) and were less likely to keep them (16%) than unspecialized anglers (55% and 27%, respectively). Unspecialized anglers (26.5% of the surveyed population) show less commitment to ethical values. The potential negative environmental impact of the least specialized anglers is an accumulation of the preferences for the use of ground bait, artiﬁcial lures, and live ﬁsh as bait. Anglers specializing in predatory ﬁsh are the most critical in their evaluation of ﬁsh resources and water quality and are least attached to speciﬁc ﬁsheries. This work provides insight into angler experience and it may help to better identify anglers who are disrespectful towards the ﬁshing laws in place. The results of this work may be incorporated into ﬁshery management strategies, including strategies to reduce naïve anglers and deter disrespectful anglers, which are lucrative in the ﬁshing process.


Introduction
Recreational fishing is a very popular way of spending free time. In the industrialized world, in countries with reliable data, the participation of the population in recreational fishing is 10.5% on average [1]. Its attractiveness is determined by the relatively low barriers to entry (both economic and skill) and the possibility of realization in a natural environment. The ancillary benefits, such as an additional source of food or profit (depending on the legislation in each country), are also important. However, according to most definitions (e.g., see, [2][3][4][5][6]) despite different approaches to the importance of particular aspects of angling, such as fun or sport, recreational angling cannot put the consumer-commercial approach in first place.
In every form of recreation, indoor or outdoor, there are various models of leisure activities that require various equipment, may have different environmental effects and are practiced by different groups of people [7]. This is manifested in various specialization behaviors. Many specific forms of recreation have been studied in this regard, e.g., [8][9][10][11][12].
Angling is no exception. Anglers around the world are not the same and researchers have classified them in different ways depending on who they are and how they act.
The first study of recreation specialization, made by Bryan [13], concluded that recreation specialization is a "continuum of behavior from general to particular, reflected by the equipment and skills used and activity setting preferences". There are several different methods for catching fish, including passive and active ones, using different types of equipment and skills (rods, lines, nets, kites, different lures, ground bait, etc.), variously affecting the environment and, of course, with different costs. On the basis of the findings of Bryan [14], it is reasonable to believe that fishing technique and equipment, as well as the time spent on the activity, affect final recreational success, with those who have more specialized equipment and who spend more time fishing likely to be more successful.
In view of the research that has been conducted to date, anglers vary in terms of socioeconomic backgrounds [15][16][17], the avidity expressed with activity and engagement [18,19], the motivation for angling [20,21], and expenditures on angling [22,23]. It has also been shown that anglers demonstrate a range of different attitudes towards the natural environment, with varying potential impacts on the environment [2,16,[24][25][26], including consumptive behaviors towards the fish caught [21,27]. Under individual preferences and specialization behaviors, anglers differ in the choice of the fishing site itself [28,29] as well as the fish species [16,17,30,31]. From the reported findings, it can be concluded that it is particularly important to understand the preferences and behaviors of anglers, which can accumulate and multiply the risk of damage to fish stocks and aquatic habitats. by contrast, the unexplored area of specialization for freshwater anglers involves the phenomena and mechanisms of the fundamental choice determining the catch of species from different ecological formations, i.e., predatory and non-predatory fish.
The main goal of the current research was to comprehensively analyze the differences in the preferences and behaviors between anglers, targeted for predatory and non-predatory fish and for those who do not exhibit specialization preferences. It was assumed that this knowledge would contribute to a better understanding of the intensity of each group's potentially negative environmental impacts and their level of engagement.
It was hypothesized that socioeconomic factors may modify the specialization preferences of anglers or affect the lack of such preferences. This was based on the results of research into the determinants of the motivations of anglers, which may be related to their choices [21]. The second assumption was that, in accordance with recreation specialization theory, specialized anglers positively differ from unspecialized anglers in terms of engagement and scale of the potentially negative environmental impacts.
To verify these hypotheses, the following research questions were formulated: (1) To what extent do socioeconomic and engagement factors determine membership in a particular specialization group? (2) What is the difference in approach to the problem of handling caught fish between specific groups of anglers? (3) Which groups' preferences and behaviors potentially have the most negative environmental impact?

Design of the Questionnaire and Data Collection
The questionnaire, constructed on the Google Forms platform in two languages (English and Polish), was composed of 24 items divided into 3 parts. The first part measures sociodemographic and economic characteristics (7 items), including "gender", "age", "education level", "marital status", "place of residence", "employment status" and "earnings". The second part consisted of 4 items measuring engagement in angling ("experience", "frequency of angling", "money spent on the hobby" and "distance from the favorite fishing site"). The third part included 13 questions measuringfishing site preferences (whether anglers prefer to fish in lakes or rivers and whether they usually fish in one favorite spot), fish handling (releasing of fish according to the "no-kill" principle, i.e., catch-and-release, hereinafter referred to as C&R and the importance of keeping caught fish, i.e., catch-andkeep, hereinafter referred to as C&K), opinions on the environment (fish stock and water quality of anglers' favorite fishing grounds), and preferences in the choice of equipment and fishing methods (common use of ground bait, artificial bait, live fish as a tackle, and preference for fishing from a boat). Those preferences and opinions were measured on a typical 5-point Likert scale, where "1" was equivalent of "strongly disagree", "5" meant "strongly agree", and "3" meant "I have no opinion or it is difficult to determine it" (neutral opinion). The 5-point Likert scale is widely used in preference and sociodemographic research (e.g., see [21,32,33]).
The fishing specialization of anglers was determined by two questions about preferences listed in Part 3 of survey: (1) "I usually fish for predatory fish"; (2) "I usually fish for non-predatory fish". The responses on a 5-point Likert scale were analyzed through a classification matrix (Table 1). On the basis of this, three groups were identified among the respondents: specialized in non-predatory fish (NPS), specialized in predatory fish (PS) and unspecialized (US). The NPS group consisted of anglers who answered that they prefer to fish for non-predatory fish (4)(5) and, at the same time, spoke negatively (1-2) about catching predatory fish, or had a neutral opinion about catching them (3). Analogically, the PS group included anglers targeting predatory fish (4)(5) and that did not focus on non-predatory fish (1-2), or had a neutral opinion about catching them (3). The US group consisted of anglers without clear answers for specialization in this context, who answered neutral (3), positive (4)(5), or negative (1-2) to both questions, and those who answered negatively in one case and neutral in the other.
Classification to such groups seems to be justified by the already high specialization among anglers in catching specific fish species, which exists in angling reality since long time [34] and has also been described in the literature. The NPS anglers are most often identified as fishing for carp (Cyprinus carpio), grass carp (Ctenopharyngodon idella), bream (Abramis brama), roach (Rutilus rutilus), chub (Squalius cephalus), ide (Leuciscus ideus), and other popular and large cyprinids [16]. PS among which the most often fished mentioned in inland waters are large species, such as perch (Perca fluviatilis), pikeperch (Sander lucioperca), eel (Anguilla anguilla), catfish (Silurus Glanis) [35], pike (Esox Lucius) [36], and, especially, American muskie (Esox masquinongy) [37], as well as salmonid species [38]. It should be noted that the number of predatory and non-predatory fish species is very broad, and they are not limited to the above mentioned ones, but these represent the most common and major trends in the angling specialization in this area.
The questionnaire was anonymous and limited only by age (between 14 and 75 years old). It was targeted at people active on angling social media. Its dissemination was made throughout Polish and English language websites and social media links, e.g., angling associations and clubs, Facebook groups and fan pages, angler discussion groups, and Internet forums. The respondents were additionally encouraged to spread the questionnaire among their angling friends, with the restriction to complete it only once if it was noticed by someone in more than one source. The full questionnaire was widespread from August until November of 2019, and it is available in Table S1 (Supplementary Material). The time required to complete the questionnaire was about 4-8 min. In total, 718 respondents provided complete sets of answers.
The survey covered mainly Europeans residing in EU countries and the UK. Yet, determining the response rate seems impossible. Rough estimates based on the number of people in the groups, indicate a size in the range of 100-300 thousands. However, there was no chance to estimate exactly how many people actually saw it, because of the desire to create viral-like content to increase the spread of the survey, which is a modern and effective way of transmission of the content [39]. Nevertheless, our results appropriatelyreflect the entire population, given the margin of the sampling error, which was calculated at a 95% confidence level (MoE ± 3.66%, for each sociodemographic subgroup between ±1.6% and ±7.2%) ( Table S2). The smaller the error, the more confident the results are [40]. As a rule of thumb, most researchers accept a MoE at the 4-8% level to be adequate enough to consider the research results as representative of the general population [41].

Statistical Data Analysis
Complete sets of answers, characterizing the basic socioeconomic, demographic, and engagement characteristics, were carried out based on the frequency table characteristics. The percentages of the respondents at least agreeing with (Likert scale answers 4 and 5) and denying (Likert scale 1 and 2) with the statements about anglers' preferences and behaviors were analogically calculated. The assumption of a non-linear distribution of the Likert scale was made [42]. Non-parametric tests are found to provide better results when analyzing an ordinal scale such as the Likert [43]. With a large sample size, variation in the respondents' opinions were expressed not by the median, but as the mean with standard deviation. However, the statistical significance of the differences between anglers specialized groups preferences and behaviors were examined on the basis of the nonparametric Kruskal-Wallis ANOVA by ranks test for independent samples (p < 0.05). The relationship between the anglers' preferences for the C&R and C&K behavior was identified using cross-correlation analysis. All statistical significance tests were performed using STATISTICA version 13.3 software.
The ordinal regression analysis with link a function logit was used to assess the potential impact of the sociodemographic, economic, and engagement indicators (independent variables) on the specialization in angling (dependent variables). The fitting of the regression model was explained by the Nagelkerke Pseudo R 2 coefficient and the statistical significance of the test, χ 2 . The statistical significance of the b-coefficient ordinal regression with a link function logit was tested using Wald's statistics for non-linear models. For each sociodemographic, economic, and engagement predictors, the reference level of one of the subgroups was determined (parameter b has been set to zero because it was redundant). Independent variables, such as "gender", "age", "earnings per month", "marital status", "employment status", "costs", and "distance traveled to fishing grounds" proved to be statistically insignificant components of the ordinal regression models. The ordinal logit regression analysis and the multidimensional visualization of the differences between the three specialization groups were performed in IMAGO PRO SPSS Statistics version 27.

Summary of Responses:Sociodemographic, Economic, and Engagement Differences in Specialization Groups
Among all the respondents (N = 718), the three specialization groups had the following quantities: NPS = 276, PS = 252, and US = 190. The vast majority of respondents (N = 685) were male (Table S2). The largest disproportion can be seen in the PS group, where the female-to-male ratio is about 1/50 (it is about 1/15 in the other groups). The non-predatory specialized anglers were the youngest, with 19.2% younger than 26 years of age. This was also the group with the lowest earnings (84.8% with income less than EUR 1000/month). The group with the highest level of education was found to be the PS group (the ratio of those with tertiary education to those with secondary and primary education) and the NPS group had the lowest level. Similar proportions of marital status are seen in all groups (between 56.2-61.6%). The highest share of unemployed anglers is in the NPS group (28.6%). The PS group has the largest representation of people from urban areas (81.7%), while the NPS group had the largest from villages (28.3%).
The least inexperienced group was the PS group, with the smallest number of people with 10 or fewer years of experience (16.6%). The level of the most experienced anglers (more than 30 years) is comparable in the PS (29.8%) and US (29.5%) groups. The predatory specialized group also has the most avid anglers, with the largest proportion of anglers angling more than once a week (35.3%). Generally, only 7% of the surveyed anglers should be classified as occasional, and over 80% are considered avid anglers (there were no noticeable differences between groups in that matter). Members of the PS group spent more money on their hobby (63.9% spent more than EUR 250 per year). The group traveling the least frequently on long fishing trips is the NPS group with 33.3% of the journeys being more than 20 km from the place of residence.
Socioeconomic predictors and angler engagement indicators only partially explain the tendencies of anglers to specialize in fishing. The ordinal logit regression models for both specialized angler groups were characterized by high statistical significance (Table 2). However, no such significance was found for US anglers. Predictive factors contributed significantly to the explanation of 13.1% of the variability for the specialization in fishing for predatory fish, and 12.1% of the variability for anglers specialized in non-predatory fish. Table 2. Results of ordinal regression, with a link function logit of 11 socioeconomic and engagement factors for angling predictors (independent variables) on the specialization (dependent variables) of anglers (N = 718). Only three predictors, i.e., education, experience, and activity, were statistically significant for explaining the variation within the two specialized groups. A higher inclination to specialize in non-predatory fishing was characteristic of anglers with the lowest level of education (W = 7.598, p < 0.01), and among those engaged in angling for no more than 20 years, with particular attention to anglers with between 10 and 20 years of experience (W = 8.033, p < 0.01). In addition, the inclination to fish at least once a week (W = 5.538, p < 0.05), and not less than 2-3 times a month (W = 7.678, p < 0.01), was significant for the NPS group. The significant predictors for the model explaining the inclination to catch predatory fish are characterized by a negative value of the estimation parameter b. This means that the lowest level of education is statistically the least characteristic for the PS group (W = 11.056, p < 0.001). Moreover, significant for the model is the lowest association of these anglers with less than 10 years engagement in angling (W = 4.106, p < 0.05), and between 10 and 20 years (W = 4.573, p < 0.05). Fishing 2-3 times per month (W = 17.893, p < 0.001) and once a week (W = 9.004, p < 0.01) are statistically the least characteristic for the PS group. Furthermore, such specialization is statistically the least characteristic of anglers living in villages (W = 4.550, p < 0.05).

Specialization Versus Need to Keep Caught Fish
Groups showed different behavioral preferences towards the fish they caught. Unspecialized anglers showed more neutral attitudes towards both C&R and C&K models ( Table 3). The US group was significantly less willing to release fish, and more willing to keep it than those in both the specialized groups. In this group, 27% reported that keeping fish is important to them, while 55% reported compliance with the C&R rules. The group of anglers who prefer releasing fish, in total, represented, respectively, 72% in the NPS group, and 69% in the PS group. These anglers are also very similar in their approaches to the second model. In the US group, very interesting is the share of undecided anglers on both questions, which is noticeably higher (5-15%) than in both the specialized groups. Greater differences between all groups can be seen in the C&K model. Unspecialized anglers strongly supporting the C&R model showed a greater willingness to keep fish by about 1 (NPS: 0.95; PS: 0.96) on a 5-point scale ( Figure 1). The relationship between the models for fish release and the need to keep fish is linear, but the cross-correlations are not fully symmetrical. Anglers who do not prefer releasing fish ( Figure 2) are also less common in specialized groups (11-12%) than in unspecialized groups (16%).

Anglers' Fishing Preferences and Behaviors
All groups are very keen on angling in the lakes (Table 4), with a mean score from 3.94 to 4.04, with 65.9-72.1% who agree that they usually fish in lakes (Table 5). However, the intensity of this desire for PS anglers is very similar for both lakes and rivers. For US and NPS anglers, the preference for fishing in rivers is significantly lower and rather neutral (46.3% and 49.6%, respectively). Predatory specialized anglers are also the least likely to fish in one preferred location, although all groups expressed neutral opinions (mean score from 2.95 to 3.41, with 35.7-46.4% agreeing). Nevertheless, it is worth adding that the preferences for fishing in rivers and one favorite fishery have quite a considerable divergence (highest SD from 1.29 to 1.45). not fully symmetrical. Anglers who do not prefer releasing fish (Figure 2) are also less common in specialized groups (11-12%) than in unspecialized groups (16%).

Anglers' Fishing Preferences and Behaviors
All groups are very keen on angling in the lakes (Table 4), with a mean score from 3.94 to 4.04, with 65.9-72.1% who agree that they usually fish in lakes (Table 5). However, the intensity of this desire for PS anglers is very similar for both lakes and rivers. For US   As expected, the use of ground baits turned out to be the highest among NPS anglers. The preference for their use was expressed by 75.7% of these anglers. Only 6.2% of the NPS group denied using it, with a mean score of 4.12. An inverse relationship between both specialized groups can be seen for the preference for fishing with artificial bait. The predatory group (with a mean score of 4.65 and 92.5% of anglers agreeing) are the highest in this regard (with only 3.6% of the PS group denying usage). The unspecialized anglers indicated a more neutral attitude towards both of these behaviorss with means of 3.56 and 3.69, respectively, and 55.5-60.5% agreeing. Although each group rarely used live fish as bait (mean 1.49-1.89), the US group was significantly less negative about using this type of bait. In this group's population, 9.5% confirmed using it, with the proportion of the undecideds higher than in the specialized groups (15.8%). The most common use of boats during angling is by the PS group (mean score of 3.17, and 43.7% agreeing), while the NPS group use boats very rarely, with a mean score of 1.56, SD 0.88, and only 4.7% agreeing on that (without responses for "strongly agree"). The unspecialized group was significantly different to both the specialized groups, positioning itself between them with a mean score of 2.50 (SD 1.42), and 23.7% agreeing.
Interestingly, all groups expressed, at best, neutral opinions about the water quality and fish stocks. Anglers from the PS group were the most negative in this regard, with a mean score of 2.09, and only 14.7% claiming that the fish stocks were good. Specialized groups were significantly different from each other, with the US group having features of both. Water quality was the most positively evaluated by the NPS group, with a mean score of 2.99, and 34.1% saying that the water condition was good. Depending on the group, 52.9-65.5% expressed disagreement with the statement that the fish stocks of the waters they fish are good. There is also a noticeably lower rating for fish stocks than water quality in all groups (about 0.55 on a 5-point scale, with about 14-16% fewer anglers agreeing on it). In summary, the three extracted groups showed different environmental preferences and behaviors.

Determinants of Specialization among Anglers
The anglers in the NPS and PS groups are specialized in their recreational activity or are definitely heading towards specialization because of their overall engagement. Clearly, this is a representation of a specialization among men. Although the women in the study mostly preferred catching non-predatory fish (48%), their share in the group of respondents was only 4.5%. Nevertheless, this rate reflects the overall participation of women in angling, as confirmed by the results of recent studies, e.g., [44,45]. There are multiple reasons for this, often linked to sociocultural backgrounds, less leisure time and different prioritizations for it, a lowered sense of personal security, and simply less interest in this type of activity [46,47].
The current study found that demographics and socioeconomic indicators do not identify the specialization propensities of anglers. More significant relationships were found between the specialization choices and the years of engagement in fishingactivities. The anglers who were the least specialized (US) were found to be the most heterogeneous. This group combines the characteristics and preferences of anglers catching predatory and non-predatory fish; thus, it shows resistance to modeling. More differences were found between the specialized groups of anglers. The NPS anglers are relatively less experienced, with lower levels of education, and they come from smaller communities. By contrast, predator fishing is more characteristic for anglers who are urban residents with higher levels of education, longer engagement in fishing and the highest levels of activity. From this, it may be assumed that the specialization for predatory fishing comes with increasing years of engagement and higher activity. For the model explaining the specialization preferences, "education," "experience" and "activity" were found to be the most significant. Skrzypczak and Karpiński [21] show that these same predictors (as well as earnings) made the most significant contribution when explaining motivational needs in angling. When contrasting these with the education and experience of anglers, it indicates that for the NPS group, "Reputation", "Social ties" and "Excitement & Self-fulfilment" are significantly more important needs. On the basis of the above results, it can be seen that both specialized groups have their strengths and weaknesses in relation to the environment. Managers can try to capitalize on this. The NPS group, on the basis of their characteristics, could do well as local fisheries guards, while the PS group, could have a predisposition to be pro-environmental practitioners. However, this should be verified by managers at all times on the basis of local conditions and other factors related to the sustainable development. Arlinghaus and Mehner [16,48] have similarly characterized anglers highly specialized in carp fishing. They identified their motivational-ethical dissimilarities and their strong dependence on trophies. Carp-specialized anglers also have different expectations of the fishing grounds and are more C&R focused [49]. Beardmore et al. [50] linked the specializations of anglers to their motivational needs. They included trophy-seeking anglers as one of the most specialized groups. Their needs have been demonstrated to be satisfied by predatory and non-predatory fish species that reach large body sizes [50]. However, these studies did not find a link between specializations and angler experience. Moreover, Beardmore et al. [50] did not identify the fishing preferences of anglers for different ecological formations of fish. Specialization in catching them requires different fishing techniques and a specific knowledge of fish biology. Therefore, experience and practical engagement is an attribute of specialized anglers.

Angler Specialization Versus Fish Handling
Specialized anglers showed more willingness to follow the rules of the C&R model and were less likely to consider it, as important to keep the fish they caught. This behavior is consistent with recreational specialization theory, which states that the support for rules and regulations increases with increasing specialization [13,14,18]. However, it should be noted that fishing, according to the guidelines of the C&R does not automatically cause the full rejection of the C&K, because its cross-relations are not fully symmetrical.
Identifying a more appropriate fishing model has been a source of dispute for many years. From an environmental perspective, C&R (e.g., more fish in the lake) is more preferred. However, from the perspective of fish health and the ethics towards them, C&K is better because it does not cause unjustified abuse, pain, suffering, or damage to a vertebrate, which, in some countries (e.g., Germany, Austria and Switzerland), is prohibited by law [49,51]. The interpretation of the differences in the data depends on which view is taken.
In the C&R context, NPS anglers were considered to be the least invasive; however, they were not significantly different from the PS group. The US anglers are considered to have the most negative impact on fish stocks. They can cause the overfishing of reservoirs to a greater extent than other groups.
Our results (more frequent desire to take fish), as well as the above-mentioned theory (e.g., less desire to enjoy simple recreation and to escape boredom) suggest that unspecialized anglers are presumably less oriented towards the fun aspectof angling [14,18]. They may feel that pursuing a hobby only for enjoyment is insufficient and may choose to compensate for the expense of the fishing trip in this way. Specialized individuals indicate a greater need for experiencing emotions and relaxation [19] than those with lower specialization rates. The implication is that US anglers will reach for other methods of satisfying their overall needs sooner than those who specialize. A possible reason is the greater awareness of the environmental consequences of angling among the specialized groups [36]. This has also been noted by Salz et al. [19]. It is also due to a greater willingness to learn more about angling and also to improve skills, as more skill is required to catch the right fish [26,50].

The Potential Impact of Specialization Groups on the Environment
Specialization groups demonstrate many different preferences and behaviors in terms of fishing techniques and the equipment used, which have potential impacts on the environment. These impacts are diverse, and the scale of an impact depends on its intensity [52]. In terms of the proven negative impacts of angling on the water quality, the preferred activity in the lake environment is unfavorable. Lakes are ecosystems that are particularly prone to the adverse effects of eutrophication [53].
Using ground bait is a common practice in freshwater non-predatory recreational angling. It is used to attract fish, mainly Cyprinidae, to the fishing ground [16]. The authors' own research shows that this practice is also common among US anglers. Anglers who specialize in catching predatory fish have the least impact on the environment in this respect (nine times less than NPS anglers and five times less than US anglers). The frequency with which they use bait may seem high, but it should be remembered that the PS anglers do not give up carp fishing altogether, nor does the NPS group give up the use of artificial lures entirely. The impact of ground baits has been described extensively, taking into account various aspects of the phenomenon, from the impact of bait on fishing success and their composition [54][55][56], to the impact on the environment [57][58][59], or even changes in fish behavior [59]. The issue seems to be serious on the grounds of sustainable development. Too much ground bait has a negative impact on the environment (it causes a decrease in water quality through its fertilization) and it does not need to be positively correlated with an increase in the biomass of fish caught [54,55,57]. However, banning them would be harmful to the economy (revenue from licensing, production, etc.) and to the non-predatory angling community, who will underperform and become unsatisfied when fishing. As a result, they may abandon their hobby, which may have negative consequences for the whole fishing industry.
The current study shows that anglers who specialize in predator fishing are the most likely to use artificial lures, while live bait is marginally used. As many as 13 out of 14 anglers in this group use artificial baits, compared to only 1 in 5 anglers in the NPS group. However, artificial lures are also very popular among less specialized anglers. It has been proven in research that artificial lures and flies could be less mortal for fish than natural baits, for example, for salmonid fish [60][61][62][63], and that fewer fish of sub-legal sizes are caught [64]. Unfortunately, it is quite common for lures to remain in the fish orto be left in the environment as a result of a fish attack or the angler's inexperience. Artificial lures can be potentially harmful as a result of the line snaps, whether because of hooking onto the ground or surrounding elements or because of remaining in a fish that has swallowed such bait. The vast majority of such lures, because of the material they are made of, are not biodegradable. Even those that boast such a designation (such as some soft rubber lures) have not shown satisfactory biodegradability results in studies [65]. It should be assumed that specialized anglers using such equipment have sufficient skills and ecological awareness (i.e., they do not litter with such lures or their parts on purpose, and they also know how to professionally place such a lure in the set).
Specialization in predator fishing is associated with a more intensive use of boats. PS anglers use boats with nine times greater frequency than NPS anglers, who use them extremely rarely (only 5%). However, boats are also used relatively often by less specialized anglers. This supports the thesis by Beardmore et al. [50] that specialized anglers are more likely to fish from a boat, indicating, however, that this frequency is more the domain of anglers fishing for predatory fish. The consequences of using a boat while fishing vary widely [66].
The magnitude of the potentially negative environmental impacts is most often related to the type of boat propulsion. Even rowboats have a negative impact on fish [67]. However the greatest negative impact is associated with internal or external combustion engines. These include fish scaring and nesting disturbances [68], plant fragmentation [69,70], and environmental contamination by operating fluids [70].
There is an ecological and ethical dimension to using live fish as bait. The current research shows that it is a rather rarely used technique, but it is worth mentioning that the US group use live bait the most often (twice more than in the specialized groups). The environmental consequences of this phenomenon could be dangerous, and the role of US anglers is significant in this context. When live bait comes from another body of water, it is a potential threat to biodiversity and native species. This transfers not only vertebrates (which can be invasive), but also any potential diseases, parasites and viruses they may contain, e.g., as in the study of [71][72][73]. As a result, laws in some countries (e.g., Poland) prohibit such practices as they are potentially negative for biodiversity and many policies, for example, in the EU, are trying to prevent this phenomenon [74].
The quality of the water environment and fish resources for angling is very important to both anglers and managers. The results show that most anglers, regardless of their specialization, are not satisfied with the fish stocks or the environmental quality. The PS anglers turned out to be the most critical. The opinion of this group is particularly important because they demonstrate the greatest preference for fishing in rivers and atach the least importance to a favorite fishing ground. Most predatory fish species show less tolerance to the effects of eutrophication or climate change [75]. Therefore, PS anglers are potentially the most quickly affected by deteriorating water quality.

Management Recommendations
It can be concluded that more frequent fishing is a cause of the decline in fish stocks and, thus, environmental quality. This leads to a desire to limit fishing opportunities for anglers so that fish stocks can recover and be as attractive as possible. However, it is important to consider that the high frequency of fishing may also be the result of restrictions already in force. Daily bag limits, and other restrictions, are often already high in many places, e.g., [48], driving anglers to fish more frequently to meet their needs. These needs, in many cases, are crucial to fishing preferences.
Managers have an interest in gathering and promoting less environmentally damaging (specialized) anglers in their fisheries and discouraging those with the greatest potential negative impact, whose needs seem more utilitarian. The proposed ways of limiting negative impacts while seeking to maximize profits could be based on demand control on the basis of the number of anglers.
One of the solutions might be the implementation of a strict system for controlling the number of anglers in the fishery. This effect can be achieved by a bivariant solution. The first involves a limited number of seasonal permits, with no option to purchase short-term permits. This allows the "elite" anglers gathering around the fishery, for whom the element of prestige may be a decisive factor when choosing a fishery. In this case, the high price of licenses may compensate for the small number of anglers. The obvious necessary condition is extremely good fishing conditions supported by the fish stock.
The second system is to limit the permits to short-term only, preferably online. An electronic system can effectively limit the number of licenses for each day (or, for example, a weekend, or a week), and anglers purchase licenses until the limit is reached. In such a case, with a properly designed system, showing managers the number of people willing to buy licenses, the price could be set in the fairest market way, i.e., by analyzing the supply and demand.
In addition, the managers of specific fisheries, with a predominance of specific types of fish species, can use this knowledge to better manage the fishery, i.e., tailor it to the preferences of specific anglers while limiting its negative impact on the environment. The perceptions of anglers towards the environment and the fish stocks are important and, generally, not positive. According to responses, anglers will most likelygather at those types of fisheries where managers provide changes improving the attractiveness of the fish stocks and the condition of the environment. Thus, managers should put more effort into attractive stocking, especially in predatory fish sites. The attractiveness of such stocking should be based not on quantity, but on quality, which is important in the case of predatory fish. Therefore, such stocking could be based on trophy fish, which would have to be released after being caught. There is also an apparent need to improve the water quality. In such conditions, it would be worth promoting anglers who discard behaviors that drive the eutrophication of waters, i.e., those who do not use ground bait by offering them cheaper licenses (especially NPS and US anglers). Another solution that may provide more attractive fishing conditions is zoning, where a segment of the water body is excluded from commercial, or even recreational, use. In such a case, fish have a kind of refuge where they can rebuild their population. Such a solution should be especially important in ecologically valuable water bodies and spawn areas.
Managers should make good use of these indicators, for example, by creating dedicated communities around their fisheries on the basis of their attractiveness. In contrast, conservation agencies, as well as policymakers, should consider the appropriateness of limits on the basis of sustainability, rather than just conservation. In each case, such decisions should be made on the basis of the local conditions, taking into account the specific environment, the cost-effectiveness of such decisions, and community support assessed on the basis of proper insight. Not following these rules may cause the opposite effect in different communities. Pursuing the specializations of fisheries for particular species may eliminate the anglers with potentially the most negative environmental impact (US).
However, it should be noted that over-interference in waters, especially natural, may be contrary to the protection of biodiversity and the preservation of ecological balance.

Conclusions
This article provides a deeper insight into the problem of specialization and may support further research identifying anglers with the greatest potential negative impacts on the environment. This identification should translate into appropriate strategies for both marketing and fishing site creation.
The least specialized groups are more likely to take fish out of the environment. Using live fish as bait indicates that they show a greater tendency to use ethically questionable methods, although this topic needs to be explored in further studies. The attitudes of the specialized groups toward fish handling were found to be similar and appeared to be less oriented towards keeping fish (less consumptive) than the US group. It was shown that the largest proportion of anglers who fish most often are in PS group. They are also less attached to particular fishing grounds. There is, therefore, a greater challenge for managers over encouraging predatory anglers to stay in their particular fishery. The PS anglers are also more characteristic of urban areas and better educated and, therefore, may have a greater awareness of the need for self-education.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/su132313167/s1. Table S1: The full survey questionnaire. Table S2: Sociodemographic, economic and engagement characteristics of the surveyed anglers' specialized groups (NPS: nonpredatory specialized; PS: predatory specialized; US: unspecialized) with the numbers and percentages of respondents (N = 718).  Institutional Review Board Statement: The study was designed according to the guidelines of the Declaration of Helsinki and the guidelines of the Research Ethics Committee of the University of Warmia and Mazury in Olsztyn. The study was not a clinical trial and did not involve sensitive data and as such did not need approval from a bioethics and ethics committee. The e-survey did not contain questions about data considered sensitive (concerning health, ethics, religion, etc.). Completion of the e-survey itself unambiguously indicated the respondents' voluntary consent to participate in the study, of which the respondents were informed at the beginning. Therefore, the ethical evaluation and obtaining consent for this study was waived due to the clear information at the beginning of the study that participation in the study is voluntary, fully anonymous (no technical possibility of linking the respondent with the answers in any way, even by the authors -a survey without contact with the respondent, online) and can be cancelled at any time for any reason, and this cancellation does not imply any consequences. No forms of pressure were used and no material benefit was given to participants for taking part in the study. The respondents were informed that data included in the questionnaire will be used only for scientific purposes and it will not be shared with third parties.

Informed Consent Statement:
The need for consent to participate in the study was waived due to the nature of the study, which does not (and is not capable of) identifying any of the participants. The submission of the questionnaire itself implied a clear consent of the respondent to participate. The nature of the study (anonymous group survey) according to the guidelines of the Research Ethics Committee of the University of Warmia and Mazury does not require written consent of the participants.

Data Availability Statement:
The data that support the findings of this study are available upon request from the corresponding author, and are in the Polish language.