Zoosyst. Evol. 96 (1) 2020, 237-262 | DOI 10.3897/zse.96.51997 zit Bab 0e0) ae TReP RE BRST 7B _________ape_. D> PENSOFT. Nites = NATURKUNDE BERLIN At the edge of extinction: a first herpetological assessment of the proposed Serra do Pingano Rainforest National Park in Uige Province, northern Angola Raffael Ernst', Thea Lautenschlager*, Makaya Futuro Branquima*’, Monique HO6lting!* Museum of Zoology, Senckenberg Natural History Collections Dresden, Konigsbriicker Landstr. 159, D-01109, Dresden, Germany Department of Biology, Institute of Botany, Faculty of Science, Technische Universitat Dresden, 01062, Dresden, Germany University of Kimpa Vita, Province of Uige, Rua Henrique Freitas No. 1, Bairro Popular, Uige, Angola BR WN Zoologisches Forschungsmuseum Alexander Koenig, Leibniz-Institut fiir Biodiversitdt der Tiere, Adenauerallee 160, 53113, Bonn, Germany http://zoobank.org/A 1670D77-0809-448 1-93B7-F5AA31D8&F4B5 Corresponding author: Raffael Ernst (raffael.ernst@senckenberg.de) Academic editor: Johannes Penner # Received 12 March 2020 # Accepted 5 May 2020 Published 5 June 2020 Abstract We systematically assess the herpetofaunal diversity of the Serra do Pingano Forest Ecosystem (SPFE) and additional localities throughout the northern Angolan province of Uige during four independent Rapid Assessment (RA) field campaigns held between 2013 and 2019. These assessments represent the first systematic surveys of amphibians and reptiles from the province, and thus we provide the first province-wide species list. We collected data on the status and current threats to amphibians and reptiles in the proposed Serra do Pingano Rainforest National Park and were able to document 33 species of reptiles from Uige province. Of the 33 species recorded from the province, 10 species are exclusively found in the SPFE. Amphibian surveys yielded 47 amphibian species from the province. These include 14 new country records and additional records that may represent undescribed species. This raises the amphibian count for Angola to at least 133 species, which includes 18 species exclusively found within the SPFE. Species-rich- ness estimators indicate that more species should be detected if survey efforts are intensified. The species composition in the SPFE is unique and consists of a high proportion of forest specialists with restricted ranges and species found nowhere else in the coun- try. This emphasizes today’s paramount importance of the SPFE, which is threatened by increasing agricultural encroachment and uncontrolled timber extraction and charcoal production. These principal factors need to be controlled and/or abandoned in already impacted areas. Conservation strategies should particularly consider the strict protection of remaining intact forests and both lentic and lotic aquatic systems. They are not only crucial for safeguarding a significant number of species that depend on these habitats for reproduction; they also provide key ecosystem services to the local population. Angola, and Uige province in particular, is at a crossroads concerning decisions and trade-offs among utilization, conservation, and preservation of its forests and, thus, substantial parts of the country’s biodiversity. The establishment of a National Protected Area in the Serra do Pingano Ecosystem is therefore a necessary and urgently needed first step towards protecting Angola’s national biodiversity heritage. Key Words Amphibia, biodiversity survey, conservation, new country records, rainforest fragments, Reptilia Introduction ready heavily fragmented rainforest. These forest ecosys- tems have been largely neglected in biodiversity surveys. Northern Angola, including the provinces of Cabinda, | However, due to an assumed link with the large Congo- Lunda Norte, Uige, and to a lesser extent Malanje and __lese Forest Bioregion and their unique position within Cuanza Norte, harbours the last remaining tracts of al- the range of Angolan ecosystems, they promise to har- Copyright Raffae/ Ernst et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 238 bour an exceptionally rich flora and fauna. Today, sizable and important patches can only be found in Cabinda and Uige. The latter province 1s particularly known for the Serra do Pingano Forest Ecosystem (SPFE), a site of re- cent herpetological explorations (Ernst et al. 2014, 2015). The amphibian and reptile fauna of this area was large- ly unknown until these first surveys were conducted. All species reported during these surveys, therefore, either represent first records for Uige province or for Angola as a whole (Marques et al. 2018). Amphibians and reptiles are amongst the world’s most threatened organisms (Stuart et al. 2004, 2008; Bohm et al. 2013). The reasons for their dramatic decline are multi- faceted, but unarguably the most important factor is land- use change, and to a lesser extent, climate change, direct exploitation, and pollution IPBES 2019). The SPFE, for- merly probably a vast and continuous area of intact rain- forest, is now under severe pressure from deforestation, encroachment, and land conversion for agriculture. Amphibians, due to their peculiar biology and diver- sity of reproductive strategies (Wells 2007), are affected by small changes in their microhabitats and are, therefore, prone to the effects of changes in land-use. Amphibians can be used to predict the response of other, less sensitive taxonomic groups (Roque et al. 2018), and they have been previously proven to be a suitable, sensitive organismal model system for analysing the impacts of environmental change on community structure, composition, and diver- sity at both taxonomic and functional levels (Ernst et al. 2006; Ernst and Rodel 2008; Holting et al. 2016). For these reasons, amphibians were specifically selected as a model indicator group to assess the status, potential, and perspec- tives of the larger Serra do Pingano Forest Ecosystem with respect to safeguarding and developing biological diversi- ty in the last remaining rainforest systems of northern An- gola within the context of a prospective National Protected Area Network Strategy. A feature film presenting current threats and perspectives developed during our activities can be accessed online: https://vimeo.com/405416258. Herpetological research in Angola Today’s knowledge of Angola’s national biodiversity re- mains poor compared to that of other Sub-Saharan coun- tries. This is particularly true for herpetofaunal diversity. Although herpetological research in Angola dates back to the late 19" century (e.g. Bocage 1867, 1879; Boulenger 1905, 1907; Ferreira 1906; Monard 1937; Laurent 1952, 1954, 1964: Hellmich 1957), it was only until recently that Bocage’s (1895) landmark book was updated by more recent research (see Baptista et al. 2019a for a summary). The war of independence from 1961-1974 and a subse- quent civil war (1975-2002) largely prevented any scien- tific exploration. Very few papers were published during this period (Poynton and Haacke 1993; Ruas 1996, 2002). After the war, numerous researchers initiated several in- dependent research projects and a constant flow of papers, zse.pensoft.net Raffael Ernst et al.: Herpetological diversity northern Angola reporting faunistic data (Ceriaco et al. 2014, 2016a, b, 2018a; Conradie et al. 2016; Baptista et al. 2019b; Butler et al. 2019), new country records (Ernst et al. 2014, 2015; Branch and Conradie 2015), and new species descriptions (Conradie et al. 2012, 2013; Stanley et al. 2016; Ceria- co et al. 2018b, 2020a, b; Branch et al. 2019a; Marques et al. 2019a, b) has been published. An updated atlas for the Angolan herpetofauna was recently published by Marques et al. (2018), and, together with Branch et al. (2019b) and Baptista et al. (2019a), these sources can now be used as a general reference and overview that is con- stantly being updated as new discoveries are published. Both rapid biodiversity surveys in neglected parts of the country and long-term monitoring routines in previously assessed sites are urgently needed to obtain a better un- derstanding of the largely underappreciated herpetofauna. This is particularly true for the fragile forest remnants of the northern provinces, as data established there can pro- vide the basis for a more comprehensive assessment of Angola’s biodiversity in general and is a prerequisite for the development of sound and sustainable conservation strategies of its most precious and unique ecosystems. Material and methods Study sites Herpetofaunal diversity of the Serra do Pingano Forest Ecosystem (SPFE) was systematically assessed during four independent Rapid Assessment (RA) field campaigns (1* campaign: 2-16 October 2013; 2"? campaign: 30 Oc- tober—17 November 2014; 3 campaign: 15 September-3 October 2018; 4" campaign: 11-22 November 2019) and opportunistically between 2013 and 2019. Sampling was conducted at three core localities forming the larger Serra do Pingano Forest Ecosystem (from north to south: 1. Serra do Uige (SU; core location: 7.6173S, 15.0658E), 2. Serra do Pingano (SP; core location: 7.6845S, 14.9298E), and 3. Serra Vamba (SV; core location: 7.8412S, 14.8422E), Fig. 1). Geographic coordinates use the WGS84 datum. Sampling was performed along an elevation gradient that covered both riparian as well as terrestrial forest habitats (for a selection of habitats, see Fig. 7). Additional surveys were conducted in different ecosystems (savannah, moor- land, urban, and agricultural land) throughout the larger Uige province to assess the Serra do Pingano herpetofau- na in a province-wide context. Additional localities in- clude the Planalto de Mucaba (wet savannah high plains with moorland) and savannah and forest ecosystems of Maquela do Zombo and Quimbele in the far north, bor- dering the Democratic Republic of Congo (DRC). Data acquisition and processing We used opportunistic visual and acoustic encounter surveys (VES and AES) both during the day and at Zoosyst. Evol. 96 (1) 2020, 237-262 T10'0"S 15.00000"E L6.00000"E 7°20'0"S 6.000005 7°30'0"S TOO" S 7400'S FOREST-SAVANNA MOSAIC, mosaic of denseforests, gallery forests (Mimutos), Peniguinense, Zambezi- Guinense in alluvial plains 7500'S THICKET-SAVANNA MOSAIC CLOSED FOREST FORMATIONS 8°0'0"S FOREST-SAVANNA MOSAIC, mosaic of Zambezi savarmnaandcofteeforests FOREST-SAVANNWA MOSAIC, Periguinese mosaic THICKET-SAVANNA MOSAIC, mosaic of scrublands and savanna 14°30'0"E 14°30'0"E 0 4,258,5 239 14°40'0"E 14°50'0"E 15°0'0"E 15°10'0°E 0 o a | id oaures devi, Diultinds, Gece, morborbw Geonyrasitiess, Cou! Ee, LE arerd, 1 a et, LS Peer, eA, nel The GS | Sait 14°40°0"E 14°50'0"E 15°0'0"E * GPS Points AO! 15°10°0°E 17 25,5 Kilometers Figure 1. Study area depicting the larger Serra do Pingano Forest Ecosystem (SPFE) with core herpetological survey sites (white rectangle) and sampling localities (blue dots, 50 independent localities). From north to south: SU: Serra do Uige, SP: Serra do Pingano, SV: Serra Vamba. The inlet left provides a coarse overview on major vegetation formations in Uige province according to Barbosa 1970. Carta fitogeografica de Angola. Instituto de Investiga¢gao Cientifica de Angola, Luanda. night, supplemented by arrays of pitfall traps placed along drift fences in combination with funnel traps (Rodel and Ernst 2004; Ernst et al. 2005), to obtain ba- sic faunistic data. Additionally, sound recordings of an- uran calls were taken using a Marantz PMD661 digital field recorder and a Sennheiser ME66 directional shot- gun microphone. Frog calls are species specific and can be used to identify species and qualitatively and quanti- tatively assess frog populations in an area. Recordings were processed following standardized routines using Raven Pro 1.5 software (Cornell Lab of Ornithology, Bioacoustics Research Program). Amphibian larvae were assessed by dip-netting in all potentially avail- able breeding sites, both lentic and lotic. Our sampling strategy only provided qualitative and semi-quantitative data. For calculations of estimated species richness and sampling efficiency using species accumulation curves and statistical estimators, we exclusively used incidence based non-parametric procedures and estimators (1*' and 2™ order Jackknife, Chao 2, and bootstrap) on presence/ absence data for the combined SPFE and in comparison for the entire Uige province. Estimators were correct- ed for small sample sizes to yield more realistic results. We performed all statistical analyses using R statistical software, v. 3.4.4 (R Core Team 2018), with the “veg- an” (Oksanen et al. 2012) and “BiodiversityR” (Kindt 2016) packages. As sample sizes for reptiles were too small, statistical analyses are restricted to the amphibian data set. Thus, reptile data are presented as occurrence data for respective core sites. The nomenclature for am- phibians follows Frost (2020) and for reptiles Uetz et al. (2019). Voucher specimens and tissue samples (stored in 99.9% ethanol) were collected for ex situ identification of problematic taxa that cannot unambiguously be iden- tified in the field or based on morphological and acoustic data, using a mt 16S rRNA-barcoding approach (Vences et al. 2005). Available sequences will be published else- where. All collected specimens were euthanized using commercially available toothache pain relief gel con- taining 20% benzocaine and subsequently preserved in zse.pensoft.net 240 70% ethanol (denatured with 10% isopropanol). Speci- mens are stored in the herpetological collection of the Museum fiir Tierkunde, Senckenberg Natural History Collections Dresden (MTD). Tissue samples used in molecular analyses are additionally stored in the tissue bank of the MTD. The combination of RA methods and lab-based molecular identification techniques is cru- cial to resolve basic taxonomic questions and provide baseline data on actual species richness and distribution within and among the focus localities and the wider na- tional and cross-regional context. Results While we were able to obtain a very solid overview of the amphibian fauna of the region, reptiles are largely underrepresented in our sampling and reptile data must be considered too incomplete to allow comprehensive conclusions. Nonetheless, we recorded several unique species and species of conservation concern in both tax- onomic groups. Reptiles Of the more than 280 species of reptiles reported for Angola today, we were able to record 33 species from Uige province. Although this is a significant increase compared to the previously reported seven species (Marques et al. 2018), this figure is still comparatively moderate. This is partially because more arid and open environments generally tend to support higher reptile diversity than closed forest ecosystems. This is particu- larly true for lizards. However, with respect to snakes, more intensified surveys in other seasons will likely yield additional species that are expected to occur in the region. Of the 33 species recorded from the province, eight species were exclusively found in the SPFE. Even though the recorded reptile diversity 1s moderate, a sig- nificant proportion of the recorded species are unique for the country and can only be found in the forested areas of the northern provinces and in adjacent regions of West-Central Africa. For a summary of records, see Table 1. As most species have not yet been formally assessed by the IUCN, we refrain from presenting re- spective categories here. Annotated species list (Reptilia) Numbers listed in parentheses refer to voucher specimens deposited in the MTD herpetological collection. Table 1 summarizes all species recorded throughout Uige prov- ince (UP) and in three core localities within SPFE. These include the Serra do Uige (SU), Serra do Pingano (SP), and Serra Vamba (SV). zse.pensoft.net Raffael Ernst et al.: Herpetological diversity northern Angola Reptilia Testudines Pelomedusidae Pelusios rhodesianus Hewitt, 1927; (MTD 49953, 49954) Only two specimens were collected by local villagers in a pond, and this species was not frequently observed in aquatic habitats. Densities are probably low due to hunt- ing pressure. Testudinidae Kinixys belliana Gray, 1831 Carapaces of unconfirmed origin were for sale in local markets or in the possession of villagers. We never observed this species alive during our surveys. This species 1s apparently heavily hunted. Kinixys belliana is associated with open savannah habitats and is unlikely to occur in the closed-forest systems of the SPFE, where it might be replaced by the forest-dwelling Kinixys erosa. However, we have no indication yet of the occurrence of K. erosa in Uige province. More field surveys and interviews with local hunters are required to confirm the presence of K. belliana within the wider SPFE area. Crocodylia Crocodylidae Crocodylus niloticus Laurenti, 1768 A single skull of unclear origin was found at a local vil- lage, but occurrence of this species in Uige province is unconfirmed. Despite intensive searches and interviews with the local population, we were unable to unambig- uously establish the existence of any crocodilian in the SPFE. However, it is likely that up to three species, Croc- odylus niloticus, Mecistops leptorhynchus, and Osteolae- mus tetraspis, occur or have occurred within the bound- aries of the SPFE. Sauria Agamidae Agama agama-complex This complex of species is mainly found in populated areas, such as within the provincial capital of Uige, and largely associated with anthropogenic habitats, such as junkyards and buildings. Because of the unresolved tax- onomy of this group in northern Angola, we refrain from Zoosyst. Evol. 96 (1) 2020, 237-262 241 Table 1. Reptile species recorded in the Serra do Pingano Forest Ecosystem (SPFE) and the wider province of Uige (in alphabetic order) with occurrence in respective core areas (SU: Serra do Uige, SP: Serra do Pingano, SV: Serra Vamba, UP: Uige province outside SPFE), general range (ANG: Angola including Cabinda, WCA: West-Central Africa, SSA: Sub-Saharan Africa (any range beyond WCA)) and general habitat type (PF: primary forest, DF: degraded forest, OH: open habitats, including natural savannahs as well as anthropogenic habitats and peri-urban sites). * = species listed under CITES II. ? = status questionable/not yet confirmed. Taxa Reptilia Testudines Pelomedusidae Pelusios rhodesianus Testudinidae Kinixys belliana* Crocodylia Crocodylidae Crocodylus niloticus Sauria Agamidae Agama agama-complex Chamaeleonidae Chamaeleo gracilis etiennei Gekkonidae Hemidactylus paivae Hemidactylus longicephalus Gerrhosauridae Gerrhosaurus cf. nigrolineatus Lacertidae Holaspis guentheri Scincidae Panaspis cabindae Trachylepis cf. affinis Trachylepis maculilabris Varanidae Varanus niloticus * Serpentes Colubridae Crotaphopeltis hotamboeia Dispholidus typus punctatus Grayia ornata Philothamnus carinatus Philothamnus dorsalis Thelotornis kirtlandii Thrasops jacksonii Toxicodryas blandingii Elapidae Dendroaspis jamesoni Elapsoidea semiannulata Naja melanoleuca Lamprophiidae Boaedon angolensis Boaedon olivaceus Lycophidion cf. multimaculatum Natricidae Natriciteres olivacea Pythonidae Python sebae* Viperidae Atheris squamigera Bitis gabonica Bitis nasicornis Causus maculatus SU SP SV UP x? Restricted to ANG WCA x into DRC x into DRC x into DRC x into Zambia & Namibia X x into East A x into East A Habitat PF DF OH