https://orcid.org/0009-0003-6396-8350
Figures
Abstract
Endemic fish species have long supported the livelihoods of local communities in the Rufiji River Basin (RRB). However, destructive fishing practices have led to a concerning decline in endemic fish stocks. To assess these changes, this study employed key informant interviews, focus group discussions (FGDs), and fishery surveys to assess the historical and contemporary distribution of endemic fishes within the RRB. DNA barcoding was also used to verify species identities. Out of 37 reported fish species, 33 species (54.55% endemic and 45.45% exotic to RRB) were confirmed through DNA barcoding and morphological characteristics. About 5 species including, Heterobranchus longifilis, Citharinus congicus, Labeo congoro, Mormyrus longirostris, and Labeobarbus leleupanus were rarely found in the field, despite being classified as Least Concern by IUCN. Additionally, five species that were reported to be present in the RRB by experienced fishers were not captured during sampling. This highlights the need for validation of the existence of such species through eDNA metabarcoding. Moreover, due to the rarity of some species in the area, their IUCN assessment should be revisited.
Citation: Saiperaki JL, Materu SF, Mkenda PA, Ligate EJ, Rumisha C (2024) Field and DNA-barcode based surveys reveal evidence of rare endemic fishes in the Rufiji River Basin. PLoS ONE 19(9): e0310387. https://doi.org/10.1371/journal.pone.0310387
Editor: Feng Zhang, Nanjing Agricultural University, CHINA
Received: March 21, 2024; Accepted: August 28, 2024; Published: September 19, 2024
Copyright: © 2024 Saiperaki et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data that support the finding of this study are available in NCBI databases under the accession number (OQ908874- OQ918545).
Funding: This study was funded by Sokoine University of Agriculture Research and Innovation Support (SUARIS) through the CoEF project (Conservation of Endemic Fishes project) under the grant number DPRTC/R/126/CoNAS/1/2022. The funder has no say on the design, data collection and analysis, writing of the manuscript and decision to submit a manuscript for publication consideration.
Competing interests: The author have declared that no competing interests exist.
1. Introduction
Freshwater fish have traditionally been a significant source of animal protein, income and employment to riparian communities globally [1]. In 2020 freshwater fisheries in Tanzania contributed for over 86% of total fish production and generated around two billion TZS [2]. However, unsustainable fishing practices driven by rapid population growth and high demand for fish protein [3] have resulted in a rapid decline of freshwater fish stocks, particularly in the Rufiji River Basin (RRB).
The decline in fish stocks in the RRB can be attributed to destructive fishing practices, such as poison fishing, dynamite fishing and the use of beach seine nets, as well as poor water quality from unsustainable agriculture and overgrazing [3, 4]. Additionally, the basin continues to shrink and the number of endemic fish species are declining due to land use change [5]. The native tribes of the RRB such as Ndamba and Pogoro have been engaged in fishing since time immemorial, but in the last 1–2 decades there is a great shift to crop farming as alternative source of food and livelihood support. Ndamba and Pogoro have a strong connection with endemic fish species, and thus their disappearance could have severe implications for household animal protein sources [6, 7].
Despite the designation of the Kilombero Valley Floodplain (KVFP) as a Ramsar site in 2002 and the establishment of the Nyerere National Park within the RRB [8, 9], the conservation of fish stocks in the RRB remains a critical issue. Unprotected areas within the RRB face significant fishing pressure, raising concerns about the potential disappearance of certain species from local catches [10]. Currently, the available information on the species composition in the region dates back over 20 years, originating from a study that identified 23 fish species in the RRB [11]. However, this study relied solely on morphological identification methods alone, which raise concerns about the potential existence of cryptic species and the possibility of misidentification [12]. Such inaccuracies can skew population assessments and conservation priorities, potentially leading to inadequate protection measures for vulnerable species. This lack of accurate and up-to-date information on species composition hinders conservation efforts, making it challenging to implement targeted interventions to protect vulnerable species and maintain ecosystem balance. Additionally, prevalent illegal fishing activities in the RRB [10] exacerbate these challenges, posing a direct threat to fish populations, especially rare and vulnerable species. Without accurate data on species composition and population dynamics, addressing and mitigating the impacts of illegal fishing activities become even more difficult. Hence, there is an urgent need to implement comprehensive monitoring programs that integrate advanced molecular techniques like DNA barcoding alongside traditional methods. This study integrated DNA barcoding and morphological identification techniques to reveal the composition of endemic fish species in the RRB. These approaches have been previously used in the country to uncover non-targeted tilapias among farmed fish and unveil protected elasmobranchs in Tanzanian fish markets [13, 14]. These approaches will provide more accurate assessments of species composition in the RRB, enabling better-informed conservation strategies and ensuring the long-term sustainability of its fish stocks.
2. Material and methods
2.1 Study site
The present study was conducted in the RRB which consists of the Kilombero River, the Great Ruaha, the Rufiji River and other small rivers [15]. Six landing sites in the RRB including Kidatu, Kivukoni, Mofu, Dinari, Ngalimila and Zombe were selected based on the availability and accessibility of landing sites (Fig 1). The RRB is the largest river basin in East Africa, which is rich in fish biodiversity [16]. It lies between 5.7° to 10.5°S and 33.5° to 39°E, covering an area of about 177,429 km2, which accounts for 20% of the total land area of Tanzania [17]. The RRB includes the KVFP, the largest seasonal freshwater lowland floodplain in East Africa [8]. It contains the Kilombero Valley Ramsar site (KVRS), an internationally recognized site of local and international importance. The Ramsar site covers an area of 796,735ha with the wetland catchment area of 40,000 km2 [8]. The RRB also contains Nyerere National Park, which is the largest National Park in Africa covering an area of over 30,000 km2 [9, 18]. The main economic activities in the RRB are fishing, crop production and livestock keeping [19]. The climatic condition of the RRB varies from tropical humid in the east to temperate in the southern highlands. In the east, the mean daily annual temperature is around 39°C while it is around 23°C in southern highlands [17]. The rainfall ranges from 250 mm in some areas to over 1800 mm on the east of the Udzungwa Mountain [17].
The map was created using ArcGIS software and shapefiles from National Bureau of Statistics (NBS) https://www.nbs.go.tz/index.php/en/census-surveys/gis/385-2012-phc-shapefiles-level-one-and-two. Accessed 10 April 2024.
2.2 Data collection
Fish sampling was conducted during two sampling seasons, between July 2022 (the onset of the dry season) and January 2023 (the onset of the wet season). A total of 46 different species were collected at six landing site of the RRB. Fish were initially identified using the available fish identification keys [20, 21]. Fish species that showed potential differences from those already sampled were specifically collected from each landing sites. For every landing site, where samples were taken (Table 1), coordinate points were recorded using Geographical Positioning System (GPS) device. Fin clip tissues of about 0.05 grams were cut from each fish, stored in 1.5 ml micro centrifuges and preserved using 99.9% ethanol until further analysis. Additionally, three focus group discussions were conducted to gather information about species composition, local fish identification techniques, fishing trends and fish management strategies. In-depth interviews were conducted to 4 groups of key informants including village elders, environmental management officers, fisheries officers and village chairpersons to gather information about the composition of fish in the RRB, local fish identification techniques.
2.3 Ethical statement
The fish sampled in this study were obtained from landing sites in the study area where they had already been caught by local fishers for human consumption. Therefore, no additional methods of sacrifice, anesthesia, or analgesia were required or administered by the researchers. The sampling process involving collecting fin clips from deceased fish only, ensuring that nor further suffering was inflicted. Authorizations for sampling were obtained from the Sokoine University of Agriculture and the Tanzania Ministry of Regional Administration and Local Government under permit number AB.307/323/01/24.
2.4 DNA extraction, COI amplification, and sequencing
Genomic DNA was extracted from each sample using the TIANamp Genomic DNA kit (TIANGEN Biotech, Beijing) according to the manufacturer’s protocol. Then the quality of each DNA extract was evaluated on 1% agarose gel before further analysis [22]. Thereafter, fragments (620 base pairs) of the cytochrome oxidase subunit I gene (COI) were amplified from the DNA extracts of each sample in a T100TM Thermal cycler machine (Bio-Lab Inc, GA, USA) using the Forward primer FishFI (5’-TCAACCAACCACAAAGACATTGGCAC-3’) and the reverse primer FishR1 (5’-TAGACTTCTGGGTGGCCAAAGAATCA-3’) [23]. Amplification reactions were done in a total volume of 35 μL consisting of 2 μL template DNA, 1 x OneTaq 2X Master Mix with Standard Buffer (New England BioLabs Inc., MA, USA), 5 mg bovine serum albumin and 0.3 μM of each primer. Each reaction was initially denatured at 94°C for 5 min, followed by 35 cycles of 94°C for 40 s, 54°C for 45 s and 72°C for 60 s. The final extension of 72°C for 15 min was added to ensure complete elongation. The quality of each PCR product was checked on a 1.5% agarose gels. The successful PCR amplicons were Sanger sequenced by Macrogen Europe Laboratory in the ABI 3730XL automated sequencer (Applied Bio systems, Foster City, USA).
2.5 Data analysis
A total of 46 samples were successfully analysed. The obtained sequences were edited to trim the ends and aligned using ClustalW algorithm as implemented in the program MEGA ver. 11 [24] to obtain sequences with equal length of 600 base pairs. Each sequence was then compared with COI sequences in the GenBank Nucleotide Database using the BLAST (Basic Local Alignment Search Tool) and BOLD (Barcode of Life Data System). The sequences were then submitted to GenBank and accession numbers (OQ908874- OQ918545) were provided. At least 90.91% of the unknown fish were identified to species level and the samples were classified to family, genus and species following the Linnaean taxonomy.
The Bayesian phylogenetic tree was constructed using BEAST ver 2.5 [25] to assess the evolutionary relationships among species. The analysis employed a relaxed uncorrelated log-normal molecular clock and a general time- reversible evolutionary model, running for 10 million generations. The tree was annotated using TreeAnnotator ver 1.10 and visualized using FigTree ver 1.4. The COI sequence of Leopard whip ray Himantura leoparda with the accession number MK422130 was retrieved from GenBank and included in the dataset as outgroup.
3. Results
3.1 Fish diversity
Fishers and the key informants mentioned a total of 37 different fishes found within the RRB. About 5 fish species were not verified during fishery survey suggesting that they are either no longer abundant in the wild or they are present in a very low numbers (Table 2). About 5 species including H. longifilis, C. congicus, L. congoro, M. longirostris, and L. leleupanus were rarely found in the field. Moreover, fishers in the RRB used the local identification techniques such as fish morphology including the size of the fish and number or structure of fins to identify fish. This identification knowledge was obtained from village elders and the experienced fishermen. The provided local names, however, do not reflect the Linnaean taxonomy and the DNA barcoding results. For example, Synodontis multipuctatus was named as ngogo ng’andu and ngogo mwanajeshi while Labeo congoro was named as mtuku and ningu depending on morphological characteristics and stage of development. Additionally, one local name was given to more than one species, particularly those with similar morphologies. For example two different species of tilapia Oreochromis korogwe and Oreochromis urolepis were reported as perege, while Glossogobius giuris and Eleotris klunzingerii were reported as bubu mchanga while Hippopotamyrus spp. and Petrocephalus affinis were reported as ndipi (Table 2). Furthermore, although fishers could distinguish matured bula Schilbe moebiusii and luepe Eutropiellus longifilis, they could not distinguish juveniles of these species due to their similar morphologies.
3.2 Confirmation of morphologically identified species through DNA barcoding
A total of 46 COI barcode sequences representing 33 different species belonging to 24 different genera, 11 different families and 8 different orders were obtained from the sampled specimens. About 18 (54.55%) out of 33 species were endemic while 15 (45.45%) species were exotic to RRB (Fig 2). These endemic species included; njuju Brycinus affinis, mtuku or ningu Labeo congoro, bula S. moebiusii, ndipi mdomo mrefu Marcusenius macrolepidatus, ndungu D. petersii, njege H. tanzaniae, kitoga B. orientalis, ngogo ng’andu Synodontis rufigiensis, ngogo mweusi Synodontis rukwaensis, mbala Citharinus congicus, tilapia O. urolepis, benasongo Enteromius apleurogramma, mbewe Brycinus spp, luepe E. longifilis, mgundu Alestes stuhlmanni, ndipi P. affinis, ndipi Hippopotamyrus spp and ndipi mdomo mfupi Marcusenius livingstonii. About 13 different fish species were identified using GenBank and BOLD databases. However, higher identities (98.87%-99.84%) failed to confirm ndipi P. affinis and sulusulu Mormyrus longirostris while low identities confirmed mbala C. congicus (93.96%) and ndipi kongwe Pollimyrus nigrican (96.73%) in GenBank database (Table 3). The taxonomic identity of 20 different fish species were not confirmed using DNA barcode alone due to lack of reference barcodes in the GenBank and BOLD databases. Therefore, the integration of DNA barcode results and morphological identification was used to confirm the identity of the 20 fish species. Yet, the identification sheets were poor for ndipi Hippopotamyrus spp, mbewe Brycinus spp and gugutuu Ctenopoma spp.
3.3 Phylogenetic analysis of experimental fish species
The bayesian phylogenetic analysis performed from 46 nucleotide sequences (Fig 3) provided additional confirmation to the identified fish species. Closely related species were clustered under the same node implying that the amplified barcodes correctly identified the species.
3.4 Conservation status
It was revealed that 90.91% (30 different fish species) of the identified species are categorized by IUCN as least concern (LC), 3.03% as near threatened (NT), and 6.06% as vulnerable (VU) (Table 3). Hence, none of the sampled fish species is either endangered or critically endangered. Similarly, none of the sampled fish species is either CITES protected or protected by Tanzanian laws.
4. Discussion
The present study revealed 33 different fish species in the RRB. This number is higher than the number reported in a previous study [11] which showed that there was only 23 different fish species. The variation in results can be attributed to differences in sampling techniques employed, limited sampling sites and shorter duration of sampling. Therefore, a total of 10 fish species identified in this study were not reported in the earlier studies. These newly identified species include, bubu mchanga G. giuris, E. klunzingerii, gugutuu Ctenopoma spp, luepe E. longifilis, mkuyu L. leleupanus, ndipi P. affinis, ndipi kongwe P. nigrican, ngogo mwanajeshi S. multipuctatus, ngogo ng’andu S. rufigiensis and ndipi mdomo mrefu M. macrolepidatus. Eighteen out of 33 species were endemic to RRB while 15 were exotic. The presence of a high number of exotic fish species poses a serious threat to the endemic fish populations. Some of these exotic species can act as competitors, predators, or even hybridize with the endemic species, further exacerbating the risk of extinction [26]. The present study also confirmed the presence of H. longifilis, C. congicus and L. coubie contrary to study conducted by [27] which revealed that the species have disappeared in the RRB. However, the fact that these species were rare in the catch suggests that the current IUCN assessment of them as Least Concern should be revisited. This is particularly critical for H. longifilis because it was found at only one site and was reported by experienced fishers to be among the fishes that were highly abundant in the past but are currently rare.
The local fish identification techniques used was found to be inaccurate, leading to numerous contradictions, especially when distinguishing closely related species. Despite using the Field guide for freshwater fishes of Tanzania [20], there were limitations in the identification sheets, particularly for certain fish species. This is similar to the study conducted in the study area [11] which showed the limitation of the identification sheets in identifying Mbewe Brycinus spp and Sheta C. werneri.
DNA barcoding alone confirmed identities of 13 species. However, low identities were used to confirm some species in the GenBank database while higher identities failed to confirm the identity of ndipi P. affinis and sulusulu M. longirostris, suggesting a high probability of tentative, incorrect or low-quality sequences being submitted to the database [28]. BOLD database confirmed less species than GenBank. However, most of the confirmed species were identified with 99–100% identities. This reveals that BOLD database has greater resolution than GenBank database [29]. The COI sequences of 21 fish species have not been recorded in the GenBank database, and the COI sequences of 17 fish species do not match any sequence in the BOLD database. Thus, this study added COI sequences for 21 fish species to the GenBank database and introduced sequences for 17 fish species that did not previously exist in the BOLD database. Furthermore, fish species identified from this study would help to solve the problem of unidentified species from the previous studies [11, 27]. Some fish species were however, not verified through DNA barcoding alone due to absence of corresponding COI sequences in the GenBank and BOLD Database. The integration of DNA sequencing information with the morphological traits of the fish showed great efficiency.
The constructed phylogenetic tree provided similar classification concerning taxonomy and morphological traits of the fishes. All closely related species were clustered under the same nodes revealing that the amplified barcodes correctly identified the species. The results of the present study indicate that none of the sampled fish in the RRB are classified as endangered or critically endangered according to the IUCN. However, due to the rarity of some species in the catch, their IUCN assessment should be revisited. This is critical for species such as mjongwa H. longifilis, mbala C. congicus, ningu L. congoro, sulusulu M. longirostris, and mkuyu L. leleupanus because they were particularly rare. These rare species require reassessment and reclassification as their current IUCN criteria does not accurately reflect their actual status on the ground. Additionally, because none of the rare species are listed in either CITES Appendices or the Third Schedule of the Tanzania Fisheries (Amendment) Regulations of 2009. This implies that there are currently no specific legal measures in place to regulate or protect these fish species from overexploitation or illegal trade. This highlights the need to update CITES Appendices and the Third Schedule of the Tanzania Fisheries (Amendment) Regulations of 2009 to include the above-mentioned rare species if they are to be protected from extinction. Furthermore, the absence of some reported species during sampling does not conclusively indicate their complete disappearance in the RRB; instead, it calls for further studies employing environmental DNA (eDNA) to confirm the presence of these species.
5. Conclusion
The present study confirmed 33 different species in the RRB, including species that were reported to have disappeared. However, some species were rarely found in the field despite being classified as Least Concern by the IUCN, suggesting the need for their IUCN Red List status to be reevaluated. Additionally, the presence of rare species suggests the need to protect them in the RRB to prevent further decline in fish populations. This can be achieved through promoting sustainable fishing practices by raising awareness among local fishers about techniques that minimize harm to fish populations and their habitats. Furthermore, the expansion of protected areas within the RRB could provide safe havens for rare species, potentially reversing the observed declining trends. Moreover, the findings of this study should be validated using environmental DNA (eDNA) to confirm the existence of species reported to have disappeared.
Acknowledgments
We would like to express our gratitude to the local community in the RRB who helped us during fieldwork. We extend our appreciation to the Tanzanian Ministry of Regional Administration and Local Government for issuing the necessary permits. Lastly, we acknowledge the contributions of the reviewers, whose constructive criticism greatly improved this work.
References
- 1. Sayer CA, Máiz-Tomé L, Darwall WRT. The importance of freshwater species to livelihoods in the Lake Victoria Basin. Freshwater biodiversity in the Lake Victoria Basin: Guidance for species conservation, site protection, climate resilience and sustainable livelihoods. 2018. 136–151 p.
- 2.
URT (United Republic of Tanzania). Livestock and Fisheries Commodity Value Chain Briefs. Dodoma; 2020.
- 3. Gebrekidan BH, Heckelei T, Rasch S. Characterizing farmers and farming system in Kilombero Valley Floodplain, Tanzania. Sustainability. 2020;12(17):7114.
- 4. BTC. Technical advisory services for biodiversity conservation and wetland management in Kilombero valley RAMSAR site, Tanzania. 2015.
- 5. Msofe NK, Sheng L, Li Z, Lyimo J. Impact of land use/cover change on ecosystem service values in the Kilombero valley floodplain, southeastern Tanzania. Forests. 2020;11(1):1–17.
- 6. Turpie JK. The Valuation of Riparian Fisheries in Southern and Eastern Africa. Trop River Fish Valuat Backgr Pap to a Glob Synth [Internet]. 2008;107–46. Available from: http://dlc.dlib.indiana.edu/dlc/handle/10535/43
- 7. Saiperaki JL, Materu SF, Mkenda PA, Ligate EJ, Rumisha C. Restricted genetic connectivity and conservation prospects of Bagrid catfish, Bagrus orientalis, populations in the Rufiji River basin, Tanzania. Fish Manag Ecol. 2024;1–9.
- 8. Ramsar. Information Sheet on Ramsar Wetlands, the Kilombero Valley Floodplain, United Republic of Tanzania. 2002.
- 9. Kijazi A. Tanzania National Parks. 2020. 58 p.
- 10. Nindi S, Maliti H, Bakari S, Kija H, Machoke M. Conflicts over Land and Water Resources in the Kilombero Valley Floodplain, Tanzania. Afr Study Monogr. 2014;50(1):173–90.
- 11. Utzinger J, Charlwood JD. Fishery methods and fish diversity in the Kilombero River in south-eastern Tanzania. African J Trop Hydrobiol Fish. 1996;7(1):55–64.
- 12. Dudgeon CL, Blower DC, Broderick D, Giles JL, Holmes BJ, Kashiwagi T, et al. A review of the application of molecular genetics for fisheries management and conservation of sharks and rays. J Fish Biol. 2012;80:1789–843. pmid:22497408
- 13. Mbilinyi WL, Rumisha C, Mwandya AW, George M. Genetic Analysis Reveals Substantial Proportion of Non-targeted Tilapias among Farmed Stocks in Kilosa and Kibaha, Tanzania. Tanzania J Sci. 2023;(July):559–67.
- 14. Rumisha C, Simwanza R, Nehemia A, Mdegela R, Kochzius M. DNA barcoding reveals endangered and protected elasmobranchs in Tanzanian fish markets Open access Citation. WIO J Mar Sci [Internet]. 2023;22(1):2023–57. Available from: http://dx.doi.org/10.4314/wiojms.v22i1.6
- 15. Mcclain M, Williams K. Environmental Flows in Rufiji River Basin Assessed from the Perspective of Planned Development in Kilombero and Lower Rufiji Sub-Basins. 2016.
- 16. Hamerlynck O, Doody K. Biodiversity of Rufiji District-A Summary Woodlands, and Wetlands of the Rufiji Delta and Floodplain Biodiversity. 2019.
- 17. Mwalyosi R. Resource Potentials of the Rufiji River. springer. 1990;19(1):16–20.
- 18. Mkwizu KH. Experiences and enjoyment of national parks: study of Nyerere National Park in Tanzania. Int Hosp Rev. 2022;
- 19. Wilson E, McInnes R, Mbaga DP, Ouedaogo P. Kilombero Valley, United Republic of Tanzania. In: Ramsar Advisory Mission Report [Internet]. 2017. p. 78. Available from: https://www.ramsar.org/document/ramsar-advisory-mission-report-83-united-republic-of-tanzania-2016
- 20.
Eccles DH. Field guide to the fresh water fisheries of Tanzania. Rome, Italy: United Nations Development Programme; 1992. 145 p.
- 21. Genner M, Turner G, Ngatunga B. A guide to the tilapia fishes of Tanzania. 2018;36.
- 22. Rumisha C, Barongo D, Saiperaki JL, Materu SF, Mdegela RH. Threatening the endangered: Uncovering endangered elasmobranchs and factors perpetuating the Tanzanian shark- - fin trade. 2024;(April):1–12.
- 23. Rumisha C, Barongo DK, Saiperaki JL, Materu SF, Mdegela RH. Devils in distress: Insights into the composition and genetic population structure of devil rays landed in the Southwestern Indian Ocean. Reg Stud Mar Sci [Internet]. 2024;77(July):103695. Available from: https://doi.org/10.1016/j.rsma.2024.103695
- 24. Tamura K, Stecher G, Kumar S. MEGA11: Molecular Evolutionary Genetics Analysis Version 11. Mol Biol Evol. 2021;38(7):3022–7. pmid:33892491
- 25. Bouckaert R, Vaughan TG, Barido-Sottani J, Duchêne S, Fourment , Gavryushkina M, A Drummond AJ. BEAST 2.5: An advanced software platform for Bayesian evolutionary analysis. PLoS Comput Biol. 2019;15(4). pmid:30958812
- 26. Marine AC, Castaldelli G, Pluchinotta A, Milardi M, Lanzoni M. Introduction of exotic fish species and decline of native species in the lower Po basin, north-eastern Italy. Aquat Conserv Mar Freshw Ecosyst. 2018;(June 2013):1–14.
- 27. Mombo F, Speelman S, Huylenbroeck G Van, Hella J. Ratification of the Ramsar convention and sustainable wetlands management: Situation analysis of the Kilombero Valley wetlands in Tanzania. 2011;(February 2016).
- 28. Wong LL, Peatman E, Lu J, Kucuktas H, He S, Zhou C, et al. DNA Barcoding of Catfish: Species Authentication and Phylogenetic Assessment. PLoS One. 2011;6(3):1–7. pmid:21423623
- 29. Panprommin D, Soontornprasit K, Tuncharoen S, Pithakpol S, Keereelang J. DNA barcodes for the identification of species diversity in fish from Kwan Phayao, Thailand. J Asia-Pacific Biodivers. 2019;12(3):382–9.