Tuesday, 16 December 2014

CHURA: Conserving Hyperendemic Udzungwa Restricted Amphibians

Posted by Elena Tonelli & Simon Valle

CHURA means frog in Swahili (Tanzania’s official language), and is also the acronym chosen for our project 'Conserving Hyperendemic Udzungwa Restricted Amphibians'.

The project focuses on the Udzungwa Scarp Forest Reserve (USFR; 200 km2 ), a global biodiversity hotspot at the south-western end of the Eastern Arc Mountain (EAM) range, Tanzania. The great altitude range  (300–2068 m) and variety of vegetation types have contributed to an extraordinary number of strictly endemic species at the site. Four frog species (Kihansi spray toad Nectophrynoides asperginis, Poyntoni forest toad N. poyntoni, Wendy forest toad N. wendyae, and Kihanga reed frog Hyperolius kihangensis) are known only from specific sites on the scarp, and are regarded as ’hyper-endemics’. However, large portions of the forest remain unexplored and basic genetic and ecological data are lacking. 

The Kilombero Valley visible from the top of a waterfall around 1100 m (Photo E. Tonelli) 

Currently, we are refining amphibian inventories of the area, and examining species distribution, ecology and community dynamics throughout the USFR using ecological and molecular approaches. Given the occurrence of morphologically similar anuran species in the EAM range and that Eastern Arc herpetofauna shows distinctive changes in assemblages with small increases in altitude, genetic analysis is crucial to enable correct taxonomic classification, detection of potential sibling species and proper estimation of distributions. By adequately surveying the area and revising the conservation status of its amphibian species, my PhD project aims to support conservation management of this proposed Nature Reserve. 

Pitfall traps with drift fences 
(Photo E. Tonelli)
Three field seasons have been planned for the whole PhD project, each one lasting from November to March and coinciding with the short rains and the reproductive period, when amphibians’ detectability is higher. Field methods comprise pitfall traps with drift fences, night transects, day search with timed digging, opportunistic surveys, and audio recording. Since different species have different needs, all habitats are sampled, from closed canopy forest to open grasslands, from large rivers with waterfalls to shallow streams or ponds. The USFR is an exciting but demanding study area, with very steep, muddy paths connecting the Kilombero valley (about 200 m ) to the villages on the plateau at the top (about 2100 m). In many areas there are no paths at all, making movement slow.
In a previous study Menegon and Salvidio (2005), recorded 36 amphibian species in just five sites within USFR. We suspect this forest will surprise us again with some species still unknown to science and possibly a few more highly range-restricted amphibians.My team surveyed the historical sites of two hyper-endemics, with a rewarding outcome for the Kihanga reed frog H. kihangensis, which we found both in its historical site and in a new area over 7 km far from that previously known. This is potentially good news for the species, but it also suggests that some of the frogs we refer to as hyper-endemics are possibly more widespread in the scarp than first thought.

New sites have been discovered for the range restricted Kihanga reed frog (Photo E. Tonelli) 

This said, we weren’t as lucky with Poynton’s forest toad N. poyntoni, which we unsuccessfully tried to locate both in November and February. This is worrying since the species has not been recorded in its only known location in the past 10 years. We plan to survey the historical site again in the coming field season, hopefully with more encouraging results. Along with previously studied sites, three new sites at different altitudes were investigated. A total of 8 genera and 18 species were recorded, at least one of these species is new to science.

Callulina sp.,a new record for the USFR and possibly a new species (Photo E. Tonelli) 

 Table 1. Anuran species recorded along with altitudinal range.

    Recorded species                                                                             Altitude range m

    Afrixalus cf. uluguruensis                                                                     1650 - 1760

    Amietia angolensis                                                                                 1650 - 1740

    Arthroleptides yakusini                                                                          900 - 1200

    Arthroleptis affinis                                                                                  900 - 1760

    Arthroleptis reichei                                                                                 900 - 1760

    Arthroleptis stenodactylus                                                                    800 - 1760

   Arthroleptis xenodactyloides                                                                900 - 1760

   Callulina sp. (new record for USFR and possibly a new species)       1200

   Hyperolius kihangensis                                                                                1750

   Hyperolius minutissimus                                                                     1650 - 1740

   Hyperolius sp.                                                                                                1750

   Hyperolius substriatus                                                                          1750 - 1740

   L. cf. grandiceps                                                                                      1200  - 1760

   Leptopelis parkeri                                                                                         1650

   Leptopelis uluguruensis                                                                        900 - 1200

  Nectophrynoides sp. (new species)                                                          1650

  Nectophrynoides tornieri                                                                           1200

Molecular analysis on museum samples showed a remarkable genetic variety in the Scarp, especially within the genus Arthroleptis. We aim to disentangle the relationship between many cryptic sibling-species which occur in this forest using multiple genes.

Afrixalus cf. Uluguruensis (adult male, eggs and juvenile)

Next steps
In the two field seasons ahead I plan to survey 15 more sites, one of which will be the historical location of the hyper-endemic Wendy’s forest toad Nectoprhynoides wendyae at the south-western end of the reserve. Skin swabbing will be used to screen encountered individuals for Batrachochytrium dendrobatidis (Bd), a fungal pathogen which has been responsible for local amphibian extinction worldwide. 

John Lyakurva, a Tanzanian student from Dar es Salaam University will join the project next month. He will be trained in herpetological survey techniques to eventually lead one more team thereby covering a larger sampling area. Along with data on amphibians however, he will collect  data on reptile species occurrence and distribution in relation to altitude.

Menegon,  M.  and  Salvidio, S.  (2005)  Amphibians  and  reptiles  diversity  in  the  southern Udzungwa Scarp Forest Reserve, south-eastern Tanzania. In: Huber, B.A., B.J. Sinclair, K.-H.  Lampe  (eds).  African  biodiversity:  molecules,  organisms,  ecosystems.  Springer,  New York, pp 205–212.


Elena's work is done in collaboration with University of Basel and has been generously supported by

Tuesday, 9 December 2014

Is 'irregular forestry' the way forward for UK's woodland biodiversity?

Posted by Danny Alder
A new approach to managing broadleaf woodland in lowland Britain is providing a unique opportunity for research into how well irregular forestry stands conserve woodland biodiversity as compared to traditional coppice systems and neglected stands. Danny Alder, a PhD student working on birds and bats, along with Dr Phil Sterling and Mike Jeffes (moth diversity) and Bryan Edwards (plant diversity), have embarked on a five year research programme at Rushmore estate in Cranborne Chase on the Dorset/Wiltshire border.

A well developed irregular stand with a mixture of understorey and canopy trees. The irregularity of both creates a patchy appearance with high complexity (Photo: Danny Alder).

Unlike even-aged high forest management, irregular forestry creates a mix of permanently irregular aged stands of trees that provide an economic return. Tree-felling is selective and stand development relies on natural woodland processes (Susse et al 2011). Continuous cover, irregular forestry enjoys something of a reputation in continental Europe as a ‘close to nature’ management type, yet, in this country, it’s almost unheard of in semi-natural woodlands. In UK, more often than not, coppice with or without standards, and high-forest are the most widely known and practiced. High forest relies on thinning regularly producing a rather even-aged structure of trees with little regard to the understorey, while coppicing is reliant on traditional niche markets or targeted grant aid for its viability.

Traditional coppicing at the Rushmore Estate (Photo:Danny Alder).

Much of the 440 hectares of semi-natural woodland at Rushmore was, until recently, managed as coppice. Now, only enough of a market for coppice products survives to keep one or two people employed. Since the 1980s, many stands have been converted to the irregular management system. As a result, and because many stands have been neglected, there are few places in Britain that provide the opportunity to compare the effects of each management type. Simple coppice, primarily of birch, is used to produce faggots used for either brooms or horse-racing jumps. This is cut on a 5-7 year rotation and produces very dense thickets between 3-5 years and is frequented by Willow Warbler Phylloscopus trochilus and Garden warbler Sylvia borin.

Don Taylor prepares hazel rods which he splits to form ‘sails’ (uprights) which he weaves to 
make hurdles for fencing (Photo:Danny Alder) .

Early stage irregular woodland stand (Photo:Danny Alder)
The above stand is in the early stages of transformation to irregular with a range of younger and only a few old growth trees. The canopy has been manipulated to allow the understorey to regenerate which protects tree seedlings which were evident. However, it is important not to ‘lose control’ of the understorey vegetation which can smother newly developing seedlings and patchiness is encouraged by cutting swathes through the bramble where necessary. 

One of 310 plots laid out across the woodland management types. During 2014, Danny 
conducted two rounds of timed bird point counts covering the early and late parts of breeding 
season (Photo:Danny Alder).

The project will examine the usage of the different woodland types by key birds - both in terms of overall abundance in the summer and winter, and during detailed investigation of birds foraging. The figure above shows that there has been a general decline in woodland birds with specialists, those reliant on all their resources within woodland suffering the most. In the first year, birds and habitat structure have been recorded at a total of 310 plots. Plants, bats, and moths will be recorded at a subset of these plots in 2015. 

Different age classes of trees and saplings were counted in each plot along with understorey 
density scores using a chequer board to estimate % obscuration (left), and canopy openness 
values using a spherical densitometer (right; Photos:Danny Alder).
By examining in detail, for the first time in British woodland, the congruence between vegetation structure, woodland plants, Lepidoptera, bats and birds we will be able to identify the strongest influences resulting from differing management types in broadleaf woodland.

The retention of standing and fallen deadwood within irregular management is encouraged 
and here the work of great spotted woodpecker Dendrocopos major can be seen in the ash 
Fraxinus excelsior stump (Photo: Danny Alder).


Susse, R, Allegrini, C., Bruciamacchie, M, and Burrus, R. (2011) Management of Irregular Forests: developing the full potential of the forest. Association Futaie Irreguliere. English translation P.Morgan 144pp.

Danny is a PhD student working with Stu and Huw Lloyd in the CEB group. Phil Sterling works with Dorset County Council, Mike Jeffes and Bryan Edwards with Dorset Environmental Records Centre.The project runs under the auspices of Dorset Environmental Records Centre, and is funded by the Golden Bottle Trust. Acknowledgements to Rhiannon Rogers DERC, Ian Burt Rushmore Estate and Andy Poore Rushmore Estate, plus Andy Taylor Freelance birch coppice worker and deer manager.


Monday, 13 October 2014

Reconstructing Atheris viper evolution in Africa

Posted by Michele Menegon and Stuart Marsden

A paper published recently in Molecular Phylogenetics & Evolution by Michele Menegon, PhD student in the group, Simon Loader and
Sylvain Ursenbacher (University of Basel), William Branch (Nelson Mandela Metropolitan University), Tim Davenport (Wildlife Conservation Society), and Stu, reconstructs the evolutionary history of Atheris vipers across Africa.

The team analysed mitochondrial DNA to uncover phylogenetic relationships among Africa's wonderful Atheris viper species, including Matilda’s Horned Viper (Atheris matildae), a species recently discovered by Michele, Tim Davenport and Kim Howell (Menegon et al. 2011).
The recently discovered Matilda's Horned Viper Atheris matildae (Photo: Michele Menegon)

The study used molecular clock estimates to age taxa, and clades within ten of the genus’ 15  species. This was done to investigate whether the timing of the divergence within Atheris corresponds to specific geographic and climatic events. In other words, where did the forest-associated species we find today in the mountains of the EAM hotspot come from, and how does their current pattern of distribution and endemism across mountains fit with geological changes such as the formation of the Rift Valley, and climatic changes over the past 20 million years?

Africa's dry lowlands form barriers to dispersal in forest-associated reptiles and amphibians (Photo: Michele Menegon)

Analyses indicate that the early diversification of the genus probably took place in lowland Central/West Africa around 15-18 mya, potentially coinciding with the final phase of the existence of the ‘pan-African forest’ (a period when moist forest stretched right across the continent). The descendants from this first colonization from Central Africa to Eastern Africa are species currently confined to high mountains of Mozambique (A. mabuensis) and Kenya (A. desaixi). The presence of these species on distant mountain tops, with no known sister species from intervening mountains, may indicate that there have been many extinctions of species since the initial colonisation and divergence.

Atheris rungweensis (Photo: Michele Menegon)

More recently, between seven and nine million years, we highlighted divergences of extant species from the fragmented montane forests of the Albertine Rift, Southern Highlands and Eastern Arc Mountains. This phylogenetic pattern suggests a more recent colonization into these areas, delayed by around 7 million years from the first appearance of the genus in East Africa (A. nitschei and A. rungweensis in the Albertine Rift and A. ceratophora, A. matildae and A. barbouri in the Eastern Arc). 

Atheris nitschei (Photo: Michele Menegon)

A possible reason for this late colonization is the physical and climatic barrier of the Great Rift Valley. Such narrowing of dispersal routes might have promoted a ‘C-shaped’ dispersal route with a northern and a southern front – where species’ ‘relatedness’ tracks the species along these routes. Interestingly, despite the late colonization of the mountain ranges east of the Albertine Rift, a high proportion of current Atheris diversity occurs there, highlighting the likely role of mountains in promoting speciation and species accumulation.

The exquisite Uzungwe viper Atheris barbouri, considered one of the rarest snakes in the world (Photo: Silvia Ceppi)

Of the 15 recognised species, only one (A. squamigera) has a particularly wide distribution ranging from Nigeria in the west, across the Congolian forest belt to western Kenya in the east and Angola in the south (Phelps 2010). All other Atheris species have small known ranges. For example, A. desaixi is found in just two localities in the Kenyan highlands, A. mabuensis in two mountain blocks in Mozambique, A. acuminata is known for a single locality in Uganda and the newly described A. matildae in a single forest fragment in the Southern Highlands of Tanzania. Of course, these restricted-range vipers, like other herps of the Eastern Arc such as Nectophrynoides toads, will face intense battles to avoid extinction in the future. Michele’s PhD will further examine the protection offered to these herps across the region. 

Maintaining intact Afromontane forests holds the key to conserving restricted-range Atheris vipers (Photo:Michele Menegon)


Menegon M, Davenport TRB, Howell KM. (2011). Description of a new and critically endangered species of Atheris (Serpentes: Viperidae) from the Southern Highlands of Tanzania, with an overview of the country's tree viper fauna. Zootaxa 3120: 43-54. (Atheris matildae sp. nov.)

Menegon, M., Loader, S.P., Marsden, S.J. Branch, W., Davenport, T. & Ursenbacher, S. (2014). The genus Atheris (Serpentes: Viperidae) in East Africa: phylogeny and the role of rifting and paleoclimate in shaping the current pattern of species diversity. Molecular Phylogenetics & Evolution 79: 12-22.

Phelps, T. (2010). Old World Vipers, A Natural History of the Azemiopinae and Viperinae. Edition Chimaira, Frankfurt am Main, Germany, 558 pp.