This study documents information on the composition, diversity, richness, and temporal occurrence of snakes at Sekayu’s lowland forest (SLF), Terengganu, Peninsular Malaysia for the first time. The snakes recorded within the SLF were sampled opportunistically from 2013 to 2019, employing the Visual Encounter Survey method (VES) and L-shape pitfall traps with drift fences. Forty-six snake species from 37 genera belonging to the nine families were recorded, of which 11 were new records to Terengganu. Individual-based rarefaction and extrapolation curves were not reaching asymptote, indicating that additional species can be recorded at the study area. Non-parametric species richness estimators estimated and produced a range between 51 and 57 species. ACE was the best estimator based on the quantitative evaluation. All species showed some variations of occurrence patterns across months. Fourteen species were only encountered once across the sampling years, and interestingly 11 of them were only detected during the rainy season (late October to January). In general, the number of species richness, abundance, and rare species were high during this season. Species richness of snakes is high at SLF but sampling effort should be intensified, especially during these rainy months, to obtain a robust estimated snake species richness in SLF. Terengganu harbor considerably high species richness of snakes with a total of 71 species to date (excluding marine snakes), but snake diversity is still underestimated as only a few localities were surveyed in the past years, primarily at the northern part. Future surveys should be commenced at the central and southern parts of Terengganu to complement the current investigation.

Monsoon, reptiles, species richness estimation, species turnover, tropical rainforest

Snakes are some of the most significant faunal components of an ecosystem. They play a crucial role in predator-prey relationships (Marshall et al. 2020; D’souza et al. 2021; Natusch et al. 2021), are highly potential bio-indicators for the ecosystem including for climate change (Bickford et al. 2010; Weatherhead et al. 2012; Böhm et al. 2016; Lourenço-de-Moraes et al. 2019; Zipkin et al. 2020) and habitat degradation monitoring (Todd and Andrews 2008; Beaupre and Douglas 2009; Pike et al. 2010; Shelton et al. 2020). Regrettably, snakes as well as lizards receive poor conservation attention compared to other reptile groups such as tortoises and turtles (Böhm et al. 2013; Shahirah-Ibrahim et al. 2018) and crocodiles and gharials (Martin 2008; Somaweera et al. 2019, 2020).

Malaysia is a tropical region with high endemism and richness of snakes (Roll et al. 2017). Currently, there are at least 191 species reported from Malaysia (MyBis 2021). Although snake species richness in this region is high, the distribution and genetic information are scarce due to limited sampling opportunities (Quah et al. 2018a, b, c; Chan and Grismer 2021a). This is likely because snakes are elusive fauna and notoriously difficult to sample due to their mobility (Barnes et al. 2017; Marshall et al. 2019; Fujishima et al. 2021; Jones et al. 2022), phenological idiosyncrasies (Brown and Shine 2002; Rahman et al. 2013), cryptic morphology and detection, and naturally occur in low densities (Chan and Ahmad 2009; Durso et al. 2011). Therefore, this hampers the conservation efforts and ecological studies of Malaysian snakes (Chan and Grismer 2021a, b).

Terengganu’s forests are still relatively understudied regarding snake diversity compared to other group of reptiles, and most of the information available for snakes are only from herpetofauna checklists, derived from short-term inventories (e.g., Dring 1979; Grismer et al. 2011; Sumarli et al. 2015; Badli-Sham et al. 2019; Zakaria et al. 2019; Fatihah-Syafiq et al. 2020; Komaruddin et al. 2020). In comparison to the lizards, freshwater turtles, and tortoises, ecological studies solely focusing on snakes in Terengganu is non-existent (Grismer and Chan 2008; Grismer et al. 2009, 2014a; Chan and Norhayati 2010; Chan and Chen 2011; Shahirah-Ibrahim et al. 2018).

Sekayu’s lowland forest (SLF), provide a potential site for conducting an ecological study on snake assemblages. A large part of this area resides within the Hulu Terengganu Tambahan Forest Reserve. The lowland forest includes a protected area, Sekayu Recreational Forest (SRF) and Sekayu Agricultural Park (SAP). The latter was developed for agro-based tourism and ecotourism purposes (Bhuiyan et al. 2012). The presence of visitors at these areas may induce human-wildlife conflict, between human and snakes. However, data on snake species richness in SLF is limited. Data to inform park’ managers to spread the awareness and information among the visitors are lacking. Only two snake species, Naja sumatrana Muller, 1887 and Tropidolaemus wagleri (Boie, 1827) are known from this locality based on Zakaria et al. (2019) checklist. Regrettably, this number is likely seriously underestimating the species richness of snakes from this area due to the short-term inventory executed by the study. In contrast, a few new discoveries in various faunal groups were made here such as crabs (Ng and Ahmad 2016; Ng 2020), frogs (Chan et al. 2020), and skinks (Grismer et al. 2014b, 2018; Sumarli et al. 2016), implying that the diversity of fauna is high in this underexplored forest.

Herein, we compiled a checklist of snakes from Sekayu’s lowland forest (SLF) from our study and included information on snakes from previous studies at other localities in Terengganu. We also examined the snake composition using our data at SLF, Terengganu, delivering information about snakes’ diversity, richness, and their temporal occurrence from this locality.

This study elevates the current knowledge of snakes in Terengganu regarding new records, species richness, and temporal occurrence of species at Sekayu’s lowland forest (SLF). The number of snake species found in SLF represents 64% of the total recorded snake species found in Terengganu (Table 1). To date, SLF is regarded as the locality with the highest species richness of snakes in Terengganu. Tembat Forest Reserve, a site comparable with SLF due to the deployment of similar methods and efforts, had two species fewer than SLF, for which Nur Amalina et al. (2017) obtained a total of 44 species. Despite that, while we observed 46 species of snakes in SLF, we did not observe 16 of the species that were documented by Nur Amalina et al. (2017). In contrast, their study did not record 22 species that were recorded in our study. In total, 65 species of snakes were documented from both sites. The marked difference in species composition of snakes between both studies warrants future studies that compare the reptile diversity (snakes in particular) between these sites to elucidate the species distribution and diversity pattern and intersite similarity of the two locations.

Eleven new records were obtained in this study. These species are widespread and distributed around Peninsular Malaysia (MyBis 2021). However, these species were not found previously in other localities in Terengganu including the offshore islands (Table 1). In general, snakes are naturally rare and elusive due to their cryptic morphology, phenological idiosyncrasies, and climate and habitat-sensitivity, causing low detectability of snake species during inventories (Durso et al. 2011; Ward et al. 2017; Frazao et al. 2020; Asad et al. 2021). This factor may explain why these 11 species were previously not observed in other localities in Terengganu’s forested area.

Two species ranked as the topmost abundant snakes in SLF, Tropidolaemus wagleri (ten individuals) and Boiga drapiezii (nine individuals). The widespread distribution and many occurrence records (Vogel et al. 2007; HerpMapper 2023) of the former species may indicate that the species is more readily detected. Despite that, natural history information such as prey items and movement of most of the recorded snakes, including these two topmost abundant species, is limited to ascertain and explain their occurrence at the study area. However, multiple records at other sites in Terengganu (Table 1) imply that T. wagleri can inhabit a wide range of habitats including offshore islands while B. drapiezii can be found in most lowland to hilly areas near water bodies such as streams, which is the geographical characteristic of the study area.

The low number of dominant species (two species) with large proportions of rare species (14 singletons) results in high unevenness of SLF’s snake assemblage (Fig. 5). The geometric series model was chosen as the best model to describe species abundance distribution for the snake assemblage of SLF. The large proportions of the singletons resulted from the rarity and elusiveness of these 14 species clearly shaped the species distribution pattern. These species only occurred once across the sampling years (Fig. 8). Despite being sampled relatively well, these species were really difficult to spot and are highly elusive species.

The individual-based rarefaction and extrapolation curves demonstrate that the sampling of snake species in SLF is not yet complete (Fig. 6). The species diversities showed an increasing trend that was not yet stabilized. Hence, the observed species richness from this study may not represent the true species richness of snakes in SLF. This may also be true for the abundance and evenness of the snake assemblages. This study attempted to estimate the species richness of snakes in SLF. Based on Table 2, an additional 5–27 species could be discovered with continuous sampling in the future. However, some of these estimates could be over-estimated due to the biasness, precision, and accuracy of the estimators used. This study evaluated the utilized estimators and found that the ACE estimator performed the best among the estimators (Table 2). The ACE estimator was moderately precise and had relatively low biasness (Table 2). The performance of low biasness and high precision of the ACE estimator in this study is also shown by the performance of this estimator in the study by Hortal et al. (2006). According to that study, the ACE estimator’s advantage is that it is non-sensitive to the grain sizes (sampling effort units). Snake richness and abundance in the snake inventories can vary due to different methods used, sites, and times (e.g., Sumarli et al. 2015; Nur Amalina et al. 2017; Fiorillo et al. 2021). Hence, an estimator with such an advantage is crucial to estimate the snake species richness. According to this estimator, 54 snakes were estimated to be discovered at SLF, adding six species to the observed species richness in this study.

A study by Kery (2002) in Europe demonstrated that the probability of finding snake species might vary depending on habitat, year, season, the area surveyed, the population size of the species, and the observer. Hence, the occurrence of the snake species in the respective months in this study may or may not also apply to the same snake assemblages in other localities in Terengganu. However, we provided essential information on which months the respective snake species can be detected in this study. For instance, some species were repeatedly found in the same month over the sampling years (Fig. 7).

The snake species richness was the highest in Oct (Fig. 7). Consequently, species appearances were the highest during the transition of Sep to Oct (Fig. 9). The month of Sep marked the beginning of the monsoon season. Asad et al. (2021) demonstrated that snake species occurrence in Borneo was positively associated with humidity and rainfall. Although we did not statistically test the relationship between the rainfall and the snake assemblage, we postulate that the increase of humidity and the rainy period in Oct might also influence species richness and abundance (Figs 7, 9). Some of the species were found twice to four times during this month over the sampling years (Fig. 7): their breeding phenology may explain their high numbers of occurrences at certain months (Cox et al. 1998; Das 2010). Additionally, the snake species’ high occurrence may also coincide with increased prey activity during the rainy period (Brown and Shine 2002; Natusch et al. 2021) and signal the onset of hunting period for the snake species (Natusch et al. 2022). This interaction has been demonstrated by previous studies on scrub python, Simalia amethistina (Schneider, 1801) in Australia, that has seasonal prey items (Natusch et al. 2021; 2022). Another example was a study in Thailand that radio-tracked Boiga cyanea (Dumeril, Bibron & Dumeril, 1854), and found an increase in their movements, space use, and activity during the nesting season of its prey, a songbird (D’souza et al. 2021). These results indicate that snakes have seasonal prey items, and their increased activity may be associated with prey nesting and the wet season. The previous studies used radio-tracking and yielded novel insights into the natural history, movement, and behavioral ecology of their snake species (D’souza et al. 2021; Natusch et al. 2021, 2022); hence this method also should be applied to the snakes recorded in this study area in the future to gain such information.

The heavy rain might interrupt the visual of the search parties hence causing low species richness detected in these months. Overall, these results elucidate that effort to sample snake species in SLF could be maximized during these rainy months to improve snake detection. Three other rare species, Calliophis intestinalis, Chrysopelea ornata, and Oligodon purpurascens, were found during the dry period. Sperry and Weatherhead (2008) found that the loss of and shifts in water availability increased the activity of the terrestrial snake Pantherophis obosletus (Say, 1823) in the USA. This might also be the case for the occurrence of terrestrial species of the first and last species above. Terrestrial snakes may prefer to be close to the riparian areas during the dry period (Asad et al. 2021), hence increasing the detectability of these species at our sampling sites (Figs 7, 9). Study that investigates and correlate snake species richness and abundance with rainfall and sampling sites (i.e., riparian versus hilly areas) in Terengganu should be conducted to test this hypothesis. We hope the information from our study can stimulate such study.

Herpetofaunal studies in riparian forests in Peninsular Malaysia have demonstrated that this forest type harbor significant number of species richness not only limited to reptiles but also amphibians, with new records and species (e.g., Chan et al. 2020; Badli-Sham et al. 2021; Fatihah-Syafiq et al. 2021; Quah et al. 2021). Previous studies suggested that riparian habitats should be preserved to reduce the extinction risk of many snake species as this habitat support high species richness (Todd et al. 2017; Guzy et al. 2019). Todd et al. (2017) discovered that human-dominated landscapes exacerbated snake species richness for those species that consume small vertebrates and species associated with aquatic habitats, and that species with these traits occurred more frequently in a natural landscape. Many of the observed species in Sekayu Lowland Forest (SLF) have similar traits, thus explaining the high species richness of snakes in this area. This number reflects the need for sustainable management of SLF particularly of the remaining undisturbed habitats of this area to safeguard the snake species.

Despite the SLF location within the forest reserve, many of the riparian forests within such reserves in Terengganu have been transformed into anthropogenic recreational areas (e.g., Lata Belatan and Lata Tembakah). This is worrying because unsustainable development and other anthropogenic activities affect reptile species richness, particularly of snakes (Gillespie et al. 2015; Bauder et al. 2020; Doherty et al. 2020; Mohd Izam et al. 2021). Sekayu’s lowland forest has become the major source of new reptile species discoveries in Terengganu (Grismer et al. 2014b, 2018; Sumarli et al. 2016), implying that the remaining intact forests in SLF and other riparian forested areas in Terengganu should be preserved so that their yet unknown species are not lost before they are officially described (e.g., Grismer et al. 2016; Nur Amalina et al. 2017). The fact that SLF has two frequented localities by local visitors (Bhuiyan et al. 2012) increases the possibility of human-wildlife conflict between human and snakes. We hope the information available from this study is used to inform to park’s authorities in SLF to spread awareness among the visitors to reduce such human-wildlife conflict. Overall, the results of this study echo the SLF’s paramount importance as a potential conservation area for snakes of the Terengganu.

We are grateful to our field associates namely Associate Professor Dr. Evan S. H. Quah and Mr. Daicus Belabut for their assistance and guidance in completing this article. We also wish to express thanks to all Universiti Malaysia Terengganu staff and students for their participation in the field sampling. In particular, we thank Syed Ahmad Rizal, Mohd Luqman Hakim bin Ismail, Mohamad Aqmal bin Mohd Naser, Muhammad Fahmi bin Ahmad, Nur Izwan Anas, and Muhamad Izzathusna. Specifically, we thank Universiti Malaysia Terengganu for providing research equipment used during this study. We are also deeply grateful to the Department of Wildlife and National Parks Malaysia (DWNP) (research permit T-00563-16-17), Department of Forestry Peninsular Malaysia (JPSM), Forestry Department of Terengganu, Sekayu Recreational Forest staff, Agriculture Department of Terengganu, and Sekayu Agricultural Park authorities for permission to conduct the study. We thank anonymous reviewers for the insightful comments.