1. Introduction
Fish species richness has been an important component of ecological health status assessment and biodiversity conservation in river ecosystem management for many years [1,2]. Fish richness is partially affected by sampling effort, so it is important to prioritize sampling efficiency in survey designs [3]. Some studies have revealed that fish richness rises rapidly with increasing sampling effort, but the number of new species collected by further sampling gradually declines, which is known as the accumulation curve of species richness [4,5,6,7]. The smaller number of sampling locations greatly underestimates the fish species richness of a river or region, but a greater number of sampling locations does not obtain an infinite number of new species, resulting in unnecessary waste [8]. Therefore, an ideal balance needs to be struck between sampling effort and cost control, which can help the conservation entities to develop a reasonable scheme that is appropriate for understanding the fish assemblage for a particular river [9].
Previous studies have attempted to determine an efficient sampling effort based on sampling lengths [5,10], sampling gears [3,11], and sampling techniques (e.g., single pass or multiple pass [12]). Multiple-pass electrofishing has been recognized as an effective means of fish collection [12,13]. Therefore, most sampling effort studies have focused on sampling lengths, which is an important factor in reflecting the characteristics of fish assemblage among different regions or different types of rivers. Lyons [14] reported that a sampling length of 35 times the mean channel widths (MCWs) ensured the cumulative fish richness reached the asymptotic level in Wisconsin streams, whereas 40 MCWs in Nebraska and Kansas streams [5] and 87 MCWs in South Carolina coastal plain streams [15] were required to obtain an asymptote. However, a sampling length of 40 MCWs has been widely approved as the standardized fishing method [16], which is desperately needed to facilitate consistent data [9]. Furthermore, the number of sampling sites or sampling reaches has also been followed with interest.
In general, describing fish assemblage structure is often realized using the sample from a site to characterize a river. However, it is often impossible to collect all species of a river, which causes a knowledge gap of differences in community structure across spatial scales [17,18] and increases the need to understand the sampling effort (i.e., the number of sampling sites) required to characterize the river. Smith and Jones [10] found that 76–115 sampling sites were needed to collect the estimated species richness of first- to third-order streams in the Great Lakes watershed. This indicates that increasing the number of sampling sites may better represent the fish assemblage because fish assemblages distribute in patches in the river due to habitat heterogeneity [19,20]. Increasing the number of sampling sites will help in obtaining more species information, but this will inevitably reduce the sampling intensity (e.g., time spent sampling, distance sampled, etc.) as we ensure that the total cost of sampling inputs does not increase. Fischer and Paukert [5] suggested that more sites and shorter sampling lengths were needed to assess the estimated fish richness with a lower total sampling effort. Samarasin et al. [7] also determined that more sites were required to meet fish richness targets when decreasing sampling intensity (i.e., one seine haul replaced three hauls). Because the sampling intensity on a site is strictly regulated according to the fish protocols [16], the number of sampling sites is the key consideration to optimize sampling design in most cases.
Rare species or low-density species slow species accumulation and require more sampling effort to describe the species richness of an entire river or region [13]. Kanno et al. [21] determined that fish species-richness asymptotes were reached with shorter sampling lengths when rare species were excluded from statistical data. Pritt and Frimpong [22] also found that increasing sampling length led to an increased number of rare fish species. In addition, Sgarbi et al. [23] determined that the Procrustes correlation was higher with the removal of rare insect species than with the removal of common species, indicating that sampling designs could be optimized by reducing the sampling effort for rare species. Consequently, the presence/absence of rare species should be considered in determining a reasonable sampling protocol.
In recent years, the Chinese government has made great efforts to protect river fish diversity. A ten-year fishing ban has been in effect in the Yangtze River since 2021 [24]. In addition, the annual fishing moratorium on the Yellow River has been extended by one month to preserve fish diversity [25]. Moreover, fish species richness has been selected as an indicator for freshwater ecosystem assessment and performance appraisal management in key river basins in China, which needs more research suggestions on optimizing sampling scheme to meet the richness target (
2. Methods
2.1. Study Area and Sampling Scheme
The study area was confined to three forest rivers, including the Taizinan River (TN, length = 84.3 km), Lanhe River (LH, length = 49.9 km), and Tanghe River (TH, length = 77.4 km), in the temperate monsoon climate region of northeastern China [26] (Figure 1). TN, LH, and TH are third- to fourth-order rivers of the Taizi River basin (40°29′ N~40°39′ N, 122°25′ E~122°55′ E). Among them, TH and LH are located in the midland freshwater ecoregion, and TN is located in the highland freshwater ecoregion [27]. The water temperatures in the three sub-basins range from 8.8 to 17.9 °C, with an average of 13.2 °C. Water velocity in these sub-basins varies between 0.24 and 0.64 m s−1, with an average of 0.41 m s−1 (Appendix A). Furthermore, the bottom substrates of the three rivers are predominantly stoney. The land covers in the three sub-basins are mainly natural deciduous broadleaved forests, accounting for more than 80% [28]. The temperature and rainfall vary greatly during the year, with an annual average of approximately 6.2 °C and 778.1 mm, respectively [26].
For this study, we sampled 36 river sites once from May to July 2012 during the base-flow period. At each river, we set 12 sites equidistantly along the longitudinal gradient. The interval between the adjacent sites was 3 km, which was suggested as the minimum sampling distance to avoid the excessive spatial autocorrelation of fish assemblages [29]. To compare the effects of sampling schemes on fish species richness with different space intervals, we set space intervals of 3 km, 6 km, and 9 km for sampling scheme 1 (SS1), sampling scheme 2 (SS2), and sampling scheme 3 (SS3), respectively (Figure 2). Actually, the data for the latter schemes were the subsets of SS1.
2.2. Fish Sampling Procedure
Initially, we measured the wetted channel widths using a diastimeter at five random sections and obtained the mean channel widths (MCWs) for each site. We then collected fish within a 40 times MCW distance range by the electrofishing method [30]. All habitat types within the survey range were sampled using double round-trip collections. The actual collection operation time was recorded and controlled to 1 h to ensure the amount of effort was the same at each site. The sampling period was restricted to the time frame between 9 a.m. and 11 a.m. for each site. All fish were identified to the species level and counted in the field. Afterward, all fish were released back into the river according to the local guidelines (i.e., The Fishery Administration Regulations of Liaoning Province).
2.3. Data Analysis
Rarity can be defined at a spatial scale based on the number of site occurrences. The terms singleton and doubleton refer to the species that only occur at one site or two sites in each river and have previously been used as an indicator of rarity in fish assemblages [9]. For the study, singleton and doubleton fish were considered rare species and identified using the data from three rivers.
To compare fish species richness among different sampling schemes, species accumulation curves (SACs, also known as sample-based rarefaction) were obtained by Mao Tau’s method [31,32]. In this analysis, we used the presence/absence data for fish species occurrence at each study site to calculate cumulative species richness versus the number of sites (i.e., sampling effort) [33]. In order to avoid any potential influence of site selection on cumulative richness, we applied a randomization procedure to reduce the random error [34] by using 100 permutations. The curves were smoothed by averaging 100 curves randomized for each sampling effort [35].
It is a laborious task to obtain all species for a region, and this requires infinite effort [36]. Additionally, the basic data of fish species spatial distribution and community composition in our study basin are still lacking. Thus, the extrapolated fish species richness (i.e., total number of fish species expected to be present in a particular location) was used as regional true richness and estimated by using the Jackknife 2 estimator [37,38], which showed a better performance in terms of bias and accuracy for estimating true regional richness than other estimators [39]. We plotted the cumulative percentage of species-richness curves versus the number of sites and compared the final percentage observed at the maximum sampling effort with 80% and 90% of the Jackknife 2 estimator, which were used as the acceptable level of understanding for regional fish richness [9]. In addition, to evaluate the differences in sampling effort inputs and outputs between three sampling schemes, we calculated the cumulative richness per unit effort (CRPUE) as follows:
where Si is the cumulative species richness for the ith replicate, Ti is the cumulative sampling time and n is the number of replicates. The differences in CRPUE between different schemes were analyzed by using one-way ANOVA with Duncan’s multiple tests after passing the homogeneity of variance test.3. Results
We collected a total of eighteen fish species in our study (Table 1). For the total richness of each river, ten species, twelve species, and fifteen species were collected in TN, LH and TH, respectively. Among them, seven species (i.e., Zacco platypus, Phoxinus lagowskii, Pseudorasbora parva, Abbottina liaoningensis, Barbatula nuda, Cobitis granoei, Misgurnus anguillicaudatus, Odontobutis obscurus) were common species that occurred at three rivers. Inversely, two species (i.e., S. wolterstorffi and Odontobutis obscurus) were found only in TN, and Gobio cynocephalus was endemic to LH, and four species (i.e., L. waleckii waleckii, S. chankaensis, Lefua costata, S. asotus) were just collected in TH. In addition, Silurus asotus and Squalidus chankaensis were identified as singletons, while Leuciscus waleckii and S. wolterstorffi were identified as doubletons in this region.
For different sampling schemes, we collected the following number of fish species: eighteen in SS1, seventeen in SS2 and thirteen in SS3. The number of individuals collected ranged from 1107 (SS3) to 3300 (SS1). The most abundantly collected species (>1000 individuals) across all sites was P. lagowskii, which has been identified as a dominant species in this region by Zhang et al. [26].
In the study, the capture of common species (occurring at three rivers) with more than five individuals was almost unaffected by the different sampling schemes. In contrast, the capture of common species with less than five individuals was affected by the sampling schemes, such as A. liaoningensis in TN, P. parva in TH, etc. (Table 1). Similar results were found in the capture of localized species (occurring at one river). Four endemic species with a few individuals were not found in SS3, whereas only one endemic species (Cottus poecilopus) was collected in all three sampling schemes due to the large number of individuals.
For each river, the SACs from 100 permutations showed a sharp rise in the front and an asymptotic trend in SS1 (Figure 3), indicating that the number of sites was sufficient for obtaining the whole picture of regional species composition. For the whole region, three SACs had an obvious asymptotic trend, especially in SS3. The cumulative species richness increased from SS1 to SS3. The results indicated that the potential maximum number of fish species can be obtained with sufficient samplings but decreases with the increase in space interval between adjacent sites.
We used 90% of the Jackknife 2 estimator as the first criteria to compare the final accumulative number of fish species richness among three sampling schemes. For SS1, 90% criteria were only achieved in LH. On the contrary, the final richness did not reach the criteria for each river or the whole region in SS2 and SS3 (Figure 4). Compared with 80% of the estimated true richness, the final richness collected from each river or whole region exceeded the criteria in SS1 and SS2 (Figure 4). The average of CRPUE in SS1, SS2, and SS3 increased gradually, and there were significant differences between SS1 and SS2, and SS1 and SS3, respectively (Figure 5), which indicated that SS2 and SS3 have more cost advantages in monitoring fish species richness than SS1.
4. Discussion
Species richness is a common estimator in biodiversity evaluation [3]. In fact, it is very hard to obtain a complete species list of fish in a river, so researchers try to find a sufficient sampling effort to obtain it [10]. Our results show that the largest number of fish species richness was collected in SS1 (i.e., minimum interval of sampling sites) for each river and overall area, while the lowest species richness was obtained in SS3 (Table 1). This indicates that the greater the sampling effort, the more species are collected. Similar research has demonstrated that an increased sampling effort helps to collect more new species, but the chances of acquiring new species will gradually decrease [4,7,40]. For each river, the SACs of SS2 and SS3 had an asymptotic trend but did not reach the asymptotic level (Figure 1). The reason is related to the small number of sampling sites in SS2 and SS3. However, for the overall area, the SACs of SS2 and SS3 showed an obvious asymptote (Figure 1), indicating that the sampling effort was sufficient. Based on the above, we propose that the large interval of sampling sites may miss some species, especially in SS3. Furthermore, the amount of fish species richness that can be obtained may have been determined when the monitoring scheme was developed.
Rare species are important to the bioassessment and the focus of conservation efforts. Due to the discontinuous distribution of rare species, the shape of SACs is greatly affected by the collection of rare species, which has been recognized by many studies [9,21,22,23]. In the present study, four rare species (i.e., singletons and doubletons) were not obtained in SS3, one rare species (S. chankaensis) was not obtained in SS2, and they were just collected in SS1 (Table 1), indicating that the smaller sampling intervals tend to increase the collection efficiency of rare species. Kanno et al. [21] reported that singletons and doubletons significantly increase the sampling effort needed to reach the species-richness target. Hughes et al. [9] also reported that the sites that had three or four rare species required 200 or more MCWs to reach the richness target, whereas the sites that had one rare species required <100 MCWs. Singletons and doubletons are extremely rare in their spatial distribution, so the distribution pattern increases their omission risk for large interval sampling schemes. In other words, a small sampling interval scheme (i.e., more sampling effort) has efficient detectability for rare species in the present study. In addition, our results showed little difference in the asymptotic levels among three sampling schemes in LH, where no rare species occurred (Figure 1). This indicates that the species-richness target is less affected by different sampling schemes when rare species are absent.
Except for spatial rarity (i.e., singletons and doubletons in the present study), abundance rarity also affects the collection efficiency of fish species [22]. In the present study, we found that the species with a low individual abundance were less likely to be collected. For example, A. liaoningensis, with one individual in TN, was not obtained in SS3 (Table 1), indicating that this species runs the risk of being missed in a single river but not the overall area. Four rare species had a low number of individuals and could not be obtained in SS3. Therefore, we speculate that species with low abundance are more susceptible to sampling efforts, especially those whose spatial distribution is also rare (i.e., localized species). As sampling efforts decrease, the chances of these species being missed increase. In addition, species missed due to low abundance may be found in other river collections to supplement this species information, but localized species information is difficult to supplement.
The adequacy of site numbers is often addressed in terms of sampling goals. In previous studies, sampling effort was quantified at different levels, typically ranging from 75% to 95% of the species target [7,10,41]. We quantified the number of sites required to capture 80% to 90% of species targets. Our study showed that SS1 reached 90% of the species target in LH; SS1 and SS2 reached 80% of the species target in each river and the overall region; and SS3 did not reach any targets (Figure 3). This indicates that two sampling schemes (i.e., minimum interval and medium interval) can meet 80% of the species target and have difficulty meeting 90% of the species target, except in LH, where no rare species occur. In addition, our study also showed that CRPUE in SS1 was significantly lower than that in SS2 and SS3 (Figure 4), indicating that the latter two sampling schemes were more advantageous in terms of inputs and outputs. Consequently, combining the requirements of the species target and the cost-effectiveness of the inputs and outputs, we propose that SS2 is a better sampling design for each river and the overall region.
Identifying fish species-richness patterns at local and regional scales is the first key step toward successful conservation [42]. Thus, if monitoring objectives involve estimating species richness at the regional scale, sampling designs must balance the effort spent against the site number that can be sampled [5,10]. In terms of monitoring inputs and outputs, the total CRPUE is about half the average CRPUE of a single river (Figure 4), indicating that monitoring efficiency on a regional basis to obtain an overall picture of fish species richness is superior to working independently on individual rivers. Future studies should test these different sampling schemes in non-wadeable rivers, which have higher levels of fish diversity.
5. Conclusions
For the wadeable rivers we studied, the sampling schemes with smaller intervals increased the number of fish species acquired. Rare species (i.e., singletons and doubletons) are more vulnerable to monitoring schemes, and as the investment in monitoring efforts decreases, the rare species show a greater chance of being missed. For a single river, the species with a low number of individuals are hard to obtain in sampling schemes with large site intervals. For the overall region, the species omitted due to low numbers of individuals may be found in other rivers. In terms of CRPUE, the overall monitoring inputs and outputs are half the average monitoring inputs and outputs of a single river. In addition, the monitoring of species richness on the watershed scale is more economical than working independently on single rivers. Therefore, the results of this study can inform the development of fish monitoring plans in other small and medium-sized wadeable rivers during the low-flow period.
Conceptualization, S.D.; methodology, Z.L.; software, M.Y.; investigation, Q.Z. and M.Y.; writing, M.Y.; supervision, S.D.; project administration, S.D. All authors have read and agreed to the published version of the manuscript.
Not applicable.
The datasets presented in this article are not readily available because (the policies and confidentiality agreements of our laboratory, we regretfully cannot furnish the original data).
The authors declare no conflicts of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 1. Geographical location of the Taizinan, Lanhe and Tanghe streams in the Taizi River basin. Circles represent sampling sites.
Enlarge this image.Figure 2. Conceptual model of sampling schemes. Circles represent sampling sites. SS1: 3 km equidistant sampling scheme. SS2: 6 km equidistant sampling scheme. SS3: 9 km equidistant sampling scheme.
Enlarge this image.Figure 3. Cumulative fish species richness obtained by Mao’s tau method with standard deviation versus number of sampling sites. White: SS1. Gray: SS2. Black: SS3.
Enlarge this image.Figure 4. The cumulative percentage of species richness in three sampling schemes. White: SS1; Gray: SS2; Black: SS3. Blue line: 90% of Jackknife 2 estimator; Red line: 80% of Jackknife 2 estimator.
Enlarge this image.Figure 5. Histogram of CRPUE (mean ± S.D.) in three sampling schemes. Asterisk: significant difference between different sampling schemes. Blue line: total CRPUE.
List of fish species observed in three sampling schemes. #: singleton. ##: doubleton.
| Name | SS1 | SS2 | SS3 | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| TN | LH | TH | Total | TN | LH | TH | Total | TN | LH | TH | Total | |
| Abbottina rivularis | - | 140 | 57 | 197 | - | 19 | 16 | 35 | - | 36 | 2 | 38 |
| Abbottina liaoningensis | 1 | 27 | 15 | 43 | 1 | 19 | 6 | 26 | - | 6 | 3 | 9 |
| Barbatula nuda | 42 | 361 | 38 | 441 | 26 | 180 | 9 | 215 | 5 | 123 | 5 | 133 |
| Carassius auratus | - | 3 | 3 | 6 | - | 1 | 1 | 2 | - | - | - | 0 |
| Cobitis granoei | 5 | 30 | 9 | 44 | 3 | 15 | 5 | 23 | 1 | 7 | 1 | 9 |
| Cottus poecilopus | 95 | - | - | 95 | 52 | - | - | 52 | 37 | - | - | 37 |
| Gobio cynocephalus | - | 49 | - | 49 | - | 35 | - | 35 | - | 27 | - | 27 |
| Lampetra mori | - | 3 | 6 | 9 | - | - | 2 | 2 | - | 3 | 4 | 7 |
| Lefua costata | - | - | 4 | 4 | - | - | 1 | 1 | - | - | 1 | 1 |
| Leuciscus waleckii ## | - | - | 2 | 2 | - | - | 1 | 1 | - | - | - | 0 |
| Misgurnus anguillicaudatus | 4 | 17 | 19 | 40 | 3 | 2 | 18 | 23 | - | 2 | 6 | 8 |
| Odontobutis obscurus | 178 | 7 | 11 | 196 | 69 | 3 | 8 | 80 | 48 | 1 | 6 | 55 |
| Phoxinus lagowskii | 1304 | 604 | 226 | 2134 | 672 | 179 | 134 | 985 | 547 | 142 | 86 | 775 |
| Pseudorasbora parva | 4 | 5 | 3 | 12 | 4 | 3 | 1 | 8 | - | 2 | - | 2 |
| Silurus asotus # | - | - | 1 | 1 | - | - | 1 | 1 | - | - | - | 0 |
| Squalidus chankaensis # | - | - | 5 | 5 | - | - | - | 0 | - | - | - | 0 |
| Squalidus wolterstorffi ## | 8 | - | - | 8 | 5 | - | - | 5 | - | - | - | 0 |
| Zacco platypus | 10 | 4 | - | 14 | 4 | 2 | - | 6 | 6 | - | - | 6 |
Appendix A. List of Environment Variables
| Environmental Variables | Minimum | Maximum | Mean |
| Water temperature (°C) | 8.8 | 17.9 | 13.2 |
| Mean channel depth (cm) | 12 | 36.67 | 17.3767 |
| Mean channel width (m) | 5 | 60 | 16.5333 |
| Mean velocity (m·s−1) | 0.24 | 0.64 | 0.41 |
References
1. Simon, T.P.; Morris, C.C.; Robb, J.R.; McCoy, W. Biological diversity, ecological health and condition of aquatic assemblages at national wildlife refuges in southern Indiana, USA. Biodivers. Data J.; 2015; 3, e4300. [DOI: https://dx.doi.org/10.3897/BDJ.3.e4300] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25632261]
2. Anas, M.U.M.; Mandrak, N.E. Patterns and drivers of native, non-native, and at-risk freshwater fish richness in Canada. Can. J. Fish. Aquat. Sci.; 2020; 79, pp. 724-737. [DOI: https://dx.doi.org/10.1139/cjfas-2020-0396]
3. Dunn, C.G.; Paukert, C.P. A flexible survey design for monitoring spatiotemporal fish richness in nonwadeable rivers: Optimizing efficiency by integrating gears. Can. J. Fish Aquat. Sci.; 2020; 77, pp. 978-990. [DOI: https://dx.doi.org/10.1139/cjfas-2019-0315]
4. Reynolds, L.; Herlihy, A.T.; Kaufmann, P.R.; Gregory, S.V.; Hughes, R.M. Electrofishing effort requirements for assessing species richness and biotic integrity in western Oregon streams. N. Am. J. Fish Manag.; 2003; 23, pp. 450-461. [DOI: https://dx.doi.org/10.1577/1548-8675(2003)023<0450:EERFAS>2.0.CO;2]
5. Fischer, J.R.; Paukert, C.P. Effects of sampling effort, assemblage similarity, and habitat heterogeneity on estimates of species richness and relative abundance of stream fishes. Can. J. Fish Aquat. Sci.; 2009; 66, pp. 277-290. [DOI: https://dx.doi.org/10.1139/F08-209]
6. Terra, B.D.F.; Hughes, R.M.; Araujo, F.G. Sampling sufficiency for fish assemblage surveys of tropical Atlantic Forest streams, southeastern Brazil. Fisheries; 2013; 38, pp. 150-158. [DOI: https://dx.doi.org/10.1080/03632415.2013.775572]
7. Samarasin, P.; Reid, S.M.; Mandrak, N.E. Optimal sampling effort required to characterize wetland fish communities. Can. J. Fish Aquat. Sci.; 2017; 74, pp. 1251-1259. [DOI: https://dx.doi.org/10.1139/cjfas-2016-0424]
8. Flotemersch, J.E.; Stribling, J.B.; Hughes, R.M.; Reynolds, L.; Paul, M.J.; Wolter, C. Site length for biological assessment of boatable rivers. Riv. Resear. Appl.; 2011; 27, pp. 520-535. [DOI: https://dx.doi.org/10.1002/rra.1367]
9. Hughes, R.M.; Herlihy, A.T.; Peck, D.V. Sampling efforts for estimating fish species richness in western USA river sites. Limnologica; 2021; 87, 125859. [DOI: https://dx.doi.org/10.1016/j.limno.2021.125859]
10. Smith, K.L.; Jones, M.L. Watershed-level sampling effort requirements for determining riverine fish species composition. Can. J. Fish Aquat. Sci.; 2005; 62, pp. 1580-1588. [DOI: https://dx.doi.org/10.1139/f05-098]
11. Robinson, W.A.; Lintermans, M.; Harris, J.H.; Guarino, F. A landscape-scale electrofishing monitoring program can evaluate fish responses to climatic conditions in the Murray-Darling River system, Australia. Advances in Understanding Landscape Influences on Freshwater Habitats and Biological Assemblages; Symposium 90 Hughes, R.M.; Infante, D.M.; Wang, L.; Chen, L.; Terra, B.F. American Fisheries Society: Bethesda, Maryland, 2019; pp. 179-201.
12. Vehanen, T.; Sutela, T.; Jounela, P.; Huusko, A.; Mäki-Petäys, A. Assessing electric fishing sampling effort to estimate stream fish assemblage attributes. Fish. Manag. Ecol.; 2013; 20, pp. 10-20. [DOI: https://dx.doi.org/10.1111/j.1365-2400.2012.00859.x]
13. Cao, Y.; Larsen, D.P.; Hughes, R.M. Evaluating sampling sufficiency in fish assemblage survey: A similarity based approach. Can. J. Fish. Aquat. Sci.; 2001; 58, pp. 1782-1793. [DOI: https://dx.doi.org/10.1139/f01-120]
14. Lyons, J. The length of stream to sample with a towed electrofishing unit when fish species richness is estimated. N. Am. J. Fish. Manag.; 1992; 12, pp. 198-203. [DOI: https://dx.doi.org/10.1577/1548-8675(1992)012<0198:TLOSTS>2.3.CO;2]
15. Paller, M.H. Relationships among number of fish species sampled, reach length surveyed, and sampling effort in South Carolina coastal plain streams. N. Am. J. Fish. Manag.; 1995; 15, pp. 110-120. [DOI: https://dx.doi.org/10.1577/1548-8675(1995)015<0110:RANOFS>2.3.CO;2]
16. Peck, D.V.; Herlihy, A.T.; Hill, B.H.; Hughes, R.M.; Kaufmann, P.R.; Klemm, D.J.; Lazorchak, J.M.; McCormick, F.H.; Peterson, S.A.; Ringold, P.L. et al. Environmental Monitoring and Assessment Program–Surface Waters: Western Pilot Study Field Operations Manual for Wadeable Streams; U.S. Environmental Protection Agency: Washinton, DC, USA, 2006.
17. Fausch, K.D.; Torgersen, C.E.; Baxter, C.V.; Li, H.W. Landscapes to riverscapes: Bridging the gap between research and conservation of stream fishes. Bioscience; 2002; 52, pp. 483-498. [DOI: https://dx.doi.org/10.1641/0006-3568(2002)052[0483:LTRBTG]2.0.CO;2]
18. Kennard, M.J.; Pusey, B.J.; Harch, B.D.; Dore, E.; Arthington, A.H. Estimating local stream fish assemblage attributes: Sampling effort and efficiency at two spatial scales. Mar. Freshw. Res.; 2006; 57, pp. 635-653. [DOI: https://dx.doi.org/10.1071/MF06062]
19. Cheek, B.D.; Grabowski, T.B.; Bean, P.T.; Groeschel, J.R.; Magnelia, S.J. Evaluating habitat associations of a fish assemblage at multiple spatial scales in a minimally disturbed stream using low-cost remote sensing. Aquat. Conserv.; 2015; 26, pp. 20-34. [DOI: https://dx.doi.org/10.1002/aqc.2569]
20. Spurgeon, J.J.; Pegg, M.A.; Parasiewicz, P.; Rogers, J. Diversity of river fishes influenced by habitat heterogeneity across hydrogeomorphic divisions. River Res. Appl.; 2018; 34, pp. 797-806. [DOI: https://dx.doi.org/10.1002/rra.3306]
21. Kanno, Y.; Vokoun, J.C.; Dauwalter, D.C.; Hughes, R.M.; Herlihy, A.T.; Maret, T.R.; Patton, T.M. Influence of rare species on electrofishing distance when estimating species richness of stream and river reaches. Trans. Am. Fish. Soc.; 2009; 138, pp. 1240-1251. [DOI: https://dx.doi.org/10.1577/T08-210.1]
22. Pritt, J.J.; Frimpong, E.A. The effect of sampling intensity on patterns of rarity and community assessment metrics in stream fish samples. Ecol. Indic.; 2014; 39, pp. 169-178. [DOI: https://dx.doi.org/10.1016/j.ecolind.2013.12.016]
23. Sgarbi, L.F.; Bini, L.M.; Heino, J.; Jyrkänkallio-Mikkola, J.; Landeiro, V.L.; Santos, E.P.; Schneck, F.; Siqueira, T.; Soininen, J.; Tolonen, K.T. et al. Sampling effort and information quality provided by rare and common species in estimating assemblage structure. Ecol. Indic.; 2020; 110, 105937. [DOI: https://dx.doi.org/10.1016/j.ecolind.2019.105937]
24. Wang, H.; Wang, P.; Xu, C.; Sun, Y.; Shi, L.; Zhou, L.; Jeppesen, E.; Chen, J.; Xie, P. Can the “10-year fishing ban” rescue biodiversity of the Yangtze River?. Innovation; 2022; 3, 100235. [DOI: https://dx.doi.org/10.1016/j.xinn.2022.100235]
25. Guo, X.; Lin, Q.; Zheng, X.; Wang, S.; Li, Q.; Shi, C. Protect native fish in China’s Yellow River. Science; 2024; 383, 598. [DOI: https://dx.doi.org/10.1126/science.adn7432]
26. Zhang, Y.; Ding, S.; Bentsen, C.N.; Ma, S.; Jia, X.; Meng, W. Differences in stream fish assemblages subjected to different levels of anthropogenic pressure in the Taizi River catchment, China. Ichthyol. Res.; 2015; 62, pp. 450-462. [DOI: https://dx.doi.org/10.1007/s10228-015-0459-6]
27. Kong, W.; Meng, W.; Zhang, Y.; Gipple, C.; Qu, X. A freshwater ecoregion delineation approach based on freshwater macroinvertebrate community features and spatial environmental data in Taizi River Basin, northeastern China. Ecol. Res.; 2013; 28, pp. 581-592. [DOI: https://dx.doi.org/10.1007/s11284-013-1048-7]
28. Bu, H.; Meng, W.; Zhang, Y.; Wan, J. Relationships between land use patterns and water quality in the Taizi River basin, China. Ecol. Indic.; 2014; 41, pp. 187-197. [DOI: https://dx.doi.org/10.1016/j.ecolind.2014.02.003]
29. Scrimgeour, G.J.; Hvenegaard, P.J.; Tchir, J. Cumulative industrial activity alters lotic fish assemblages in two boreal forest watersheds of Alberta, Canada. Environ. Manag.; 2008; 42, pp. 957-970. [DOI: https://dx.doi.org/10.1007/s00267-008-9207-2]
30. Lazorchak, J.M.; Klemm, D.J.; Peck, D.V. Environmental Monitoring and Assessment Program—Surface Water: Field Operations and Methods for Measuring the Ecological Conditions of Wadeable Streams; EPA/620/R-94/004 Environmental Monitoring Systems Laboratory, Office of Research and Development, U.S. Environmental Protection Agency: Cincinnati, OH, USA, 1995.
31. Colwell, R.K.; Mao, C.X.; Chang, J. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology; 2004; 85, pp. 2717-2727. [DOI: https://dx.doi.org/10.1890/03-0557]
32. Colwell, R.K. EstimateS: Statistical Estimation of Species Richness and Shared Species from Samples Version 9. 2013. Available online: https://purl.oclc.org/estimates (accessed on 3 May 2024).
33. Moreno, C.E.; Halffter, G. Assessing the completeness of bat biodiversity inventories using species accumulation curves. J. Appl. Ecol.; 2000; 37, pp. 149-158. [DOI: https://dx.doi.org/10.1046/j.1365-2664.2000.00483.x]
34. Thompson, G.G.; Thompson, S.A. Using species accumulation curves to estimate trapping effort in fauna surveys and species richness. Austral Ecol.; 2007; 32, pp. 564-569. [DOI: https://dx.doi.org/10.1111/j.1442-9993.2007.01728.x]
35. Willott, S.J. Species accumulation curves and the measure of sampling effort. J. Appl. Ecol.; 2001; 38, pp. 484-486. [DOI: https://dx.doi.org/10.1046/j.1365-2664.2001.00589.x]
36. Soberón, M.J.; Llorente, B.J. The use of species accumulation functions for the prediction of species richness. Conserv. Biol.; 1993; 7, pp. 480-488. [DOI: https://dx.doi.org/10.1046/j.1523-1739.1993.07030480.x]
37. Palmer, M.W. The estimation of species richness: The second-order jackknife reconsidered. Ecology; 1991; 72, pp. 1512-1513. [DOI: https://dx.doi.org/10.2307/1941127]
38. Colwell, R.K.; Coddington, J.A. Estimating terrestrial biodiversity through extrapolation. Philos. Trans. R. Soc. Lond. B; 1994; 345, pp. 101-118.
39. João, C.C.; Nelson, V.; Markus, M.; Jeremy, C.T.; Martin, W. Estimation of regional richness in marine benthic communities: Quantifying the error. Limnol. Oceanogr.-Meth.; 2008; 6, pp. 580-590.
40. Pompeu, P.S.; de Carvalho, D.R.; Leal, C.G.; Leitão, R.P.; Alves, C.B.M.; Braga, D.F.; Castro, M.A.; Junqueira, N.T.; Hughes, R.M. Sampling efforts for determining fish species richness in megadiverse tropical regions. Environ. Biol. Fish; 2021; 104, pp. 1487-1499. [DOI: https://dx.doi.org/10.1007/s10641-021-01184-7]
41. Ross, J.E.; Mayer, C.M.; Tyson, J.T.; Weimer, E.J. Comparison of electrofishing techniques and effort allocation across diel time periods, seasons, sites, and habitat in the Ohio coastal waters of western Lake Erie. N. Am. J. Fish. Manag.; 2016; 36, pp. 85-95. [DOI: https://dx.doi.org/10.1080/02755947.2015.1111275]
42. Blackman, R.C.; Osathanunkul, M.; Brantschen, J.; Di Muri, C.; Harper, L.R.; Mächler, E.; Hänfling, B.; Altermatt, F. Mapping biodiversity hotspots of fish communities in subtropical streams through environmental DNA. Sci. Rep.; 2021; 11, 10375. [DOI: https://dx.doi.org/10.1038/s41598-021-89942-6]
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.