Physical barriers such as dams and waterfalls may isolate stream branches, and potentially interrupt longitudinal gradients of fish distribution15. The influence of a single barrier on fish assemblages is well documented, but the influence of multiple barriers in small streams is not well characterized. Studies on the combination of natural barriers (e.g., waterfalls) and the man-made barriers (e.g., small dams), give an opportunity to better understanding how these interuptions in waterflow affect fish assemblages in headwater streams.
The increasing number of fish species from upstream to downstream in Sungai Enam and Sungai Kooi indicates that species profile is affected by the presence of physical barriers. This effect is more evident when barriers have a steep slope and larger height such as the W0 waterfall in Sungai Enam and the 65 m high waterfall in Sungai Kooi. A similar effect was also recorded by previous studies16,17,18. In Yakushima Island, Japan, a waterfall beyond 5 m in height had a significant negative effect on fish distribution19.
Besides the barrier itself, the barrier’s slope also plays a role in fish distribution. A steeper slope increases water velocity, discharge volume and momentum, thus, impeding the migratory passage of fish species due to physical resistance to movement they offer20. Water velocity controls the distribution of substrate and food as well as exerting a direct influence on the aquatic organism21. In terms of morphological or behavioural aspects, fish species, which lack capabilities or adaptations to transverse physical barriers in watercourses, will remain below the waterfall. Above the waterfall therefore, the abundance and biomass of fishes are significantly lower than in all other sites8.
The existence of fish species below a barrier, may be due to the fish species being washed away8 by strong water current from upstream that is known as stream drift22. Due to stream drift, fish assemblages in downstream is replenished by eggs and larvae7 and the abundance of drifting fish of particular size, age, and taxa may increase as water velocity increases23. On the other hand, the existence of P. smedleyi, D. regina and C. striata above the waterfalls is still unresolved.
The ability to invade pools above a waterfall is of a tremendous importance to several fish species, allowing them to penetrate or repopulate habitats that otherwise could be impossible to access. For example, guppies (Balitoridae) had been recorded above waterfalls19 and balitorids have special adaptations that enables them to resist strong water current8. These adaptations include the use of pectoral and pelvic fins which form a ‘suction-cup’ or the pre-valvular cavity24 and enable fish to adhere to substrates amongst rapids and hence enable them to transverse waterfalls.
Even though most cyprinids do not have such organs that facilitate attachment to a substrate, it has been suggested that cyprinids have evolved from their descendants in terms of their evolution of body shape and mouth orientation enabling them to resist downstream displacement25. Such cyprinids are able to move against the water current due to their evolutionary convergence on a pelagic, large-eyed, upturned-mouth morphology and many have a fusiform body that helps to increase their swimming efficiency25. However, these adaptations are still not enough to completely explain how cyprinids transverse waterfalls.
In our investiagtion P. smedleyi, D. regina and C. striata were noted in watercourses upstream of waterfalls, and we propose three hypotheses which may explain this distribution: (1) the world was once flooded and all lands were submerged under water which gave fish freedom to extend their distribution. When the water subsided, fish got trapped in their respective habitat. This hypothesis also means that all freshwater species existed prior to the phenomena and may have been banished due to salt water intrusion, (2) some fish species which had an extensive distribution in freshwater, became isolated by the land uplifting phenomena that created the mountains and waterfalls, and (3) respective streams may have had another watercourse branch with a less gradient of slope which enabled fish passage to higher elevation, and thus bypass the stream branch which had a steeper slope or high waterfalls.
Although the first two hypotheses require extensive investigation of local geological evolution, a fossil-based comparative analysis has revealed that freshwater fish ancestor lineage goes back to marine species26. In addition, the uplifting of land in Asia, which occurred million years ago is thought to occur rapidly i.e over decades or over an extensive time period e.g over a million years27. The rate of land uplifting may have caused the destruction or alteration of the original stream. Mountains uplifting can either trigger the speciation process, or even reduce biodiversity28. However, the distribution of P. smedleyi throughout most states in Peninsular Malaysia and even in Thailand, Cambodia, Laos and Vietnam29 highlights that if fish in each of these localities originated from the same ancestor species, then species evolution follows a certain embedded evolution template such as DNA rather than following natural selection that would probably had created different species.