Amazon Pump Stream Amphibians & Benthic Report February 2009

Yachana Reserve Amphibian and Benthic Surveys Pump stream surveys 14th February 2009 (phase 091B)

Authors Christopher Beirne, Jonathan Escolar, Andrew Whitworth

Introduction: Amphibians are vital indicators of environmental quality as they are so susceptible to changes in the environment (Gardner et al, 2007; Lyaruu et al, 2000). According to Gardner et al (2007) it is essential that we learn more about the patterns of diversity and habitat preferences of individual species. This data can then be used to monitor population declines and inform effective conservation strategies, particularly where amphibians act as indicators of change in their environment. Over recent years there has been a growing concern that amphibian species are declining worldwide. According to the Global Amphibian Assessment (GAA) almost one-third (2,030 species) of the worlds 6,260 known amphibian species are listed as threatened or extinct. 1,533 are classed as data deficient so many of these may also be threatened. These declines are attributed to two major factors. Habitat loss and fungal disease with possible contributions from introduced species, climate change and pollution. According to the GAA there are 467 species of recorded amphibians found in Ecuador, 37% of which are endemic and 37% are listed as threatened or extinct under the IUCN Red List Criteria. Even where species richness has been estimated by fieldwork, it is thought that diversity estimates have drastically underestimated the true species numbers. Over 75% of amphibians depend on moist forest habitats so it is vital that further fieldwork is conducted so that the amphibian biodiversity can be properly assessed and potential threats elucidated. Over the past three years GVI (Global Vision International) have been surveying amphibians in the Yachana reserve (2000 ha) found in the Napo province of Amazonian Ecuador. This has allowed a collation of data to provide a catalogue of species found within the reserve. 61 species have been idientified across the reserve so far. Over the past twelve months GVI have also been contributing towards research at PUCE (Pontificia Universidad Catolica del Ecuador). This has involved swabbing amphibians to test for the presence/absence of chytridiomycosis, taking voucher specimens of particular species/species complexes and also taking skin, muscle and liver samples from voucher specimens for genetic testing. As part of our own research we have also collated data regarding locations around the reserve of where individuals were collected and the microhabitat where they were found. We now hope to expand upon our previous research into looking at localised areas of the reserve more specifically rearding the amphibian assemblages found within these areas. Problem Statement: The GVI base camp is situated in close proximity to a small stream running close to the camp in which water is regularly pumped out for use in daily camp life. Used water is then allowed to drain back into the local surrounding environment. Amphibians are particularly sensitive to changes in water quality and have frequently been described as excellent indicators of water quality and the stability of the surrounding environment. The impact that the base camp has on the surrounding environment should be assessed to determine any adverse affects that may be avoided in future work related to daily camp living.

© Global Vision International – 2007

2

Objectives and significance: The principal goal of the study is to investigate the amphibian assemblages along the pump stream close to the GVI base camp. Surveys will be conducted in areas before water is pumped from the stream and then further down stream after it has been pumped out. Through conducting benthic surveys along the stream we can also gain a good assessment of water quality related to invertebrate assemblages and detect any differences pre and post pumping from the stream in accordance with amphibian data. Hypotheses: The null hypothesis would be no difference in the species found in the upper and lower areas of the pump stream region with relation to amphibians or invertebrates. The testable hypothesis would be that the lower area of the pump stream would be lower in terms of species richness due to the environmental impacts on the surrounding environment, inflicted by the GVI base camp. Methods: Amphibian surveying Two main methods were implicated to assess the presence of different amphibian species along the lower and upper regions of the ‘pump stream’. The first of these was pitfall trapping and the second being visual encounter surveys: Pitfall trapping At two sites on both the Upper and Lower Pump Stream, Leaf-litter amphibian and lizards were sampled with pitfalls traps and drift fences. At each sample site, two 20-litre plastic buckets were installed parallel to the stream edge. The buckets were connected via an 8 meter long, 50 cm high plastic baffle. The Lower Pump Stream was sampled from the 2nd of March to the 9th of March and the Upper Pump Stream was sampled from the 9th of March to the 16th. This resulted in 28-nights of sampling effort in each stream. All amphibians captured were collected to remove the possibility of subsequent recapture at a later date. Visual encounter surveys Nocturnal and crepuscular visual encounter surveys were employed along 120m stream transects in the Upper and Lower Pump Stream. Where possible both transects were surveyed on the same day to minimise the effect of weather variation on the amphibian assemblage. The surveys took place the 23rd of February 2009 and the 12th of March 2009. The surveys took place at times that have been found to coincide with peak frog activity: crepuscular surveys took place between 5.30am and 8.30am, and nocturnal surveys took place between 7.30pm and 10.30pm. Transects were searched at a rate of roughly one meter per minute. All amphibians found within 3m of each side of the stream were captured and collected thus removing the possibility that they would be recaptured on a later survey. To minimise the effect of team size and transect duration variability, a measure of effort was calculated by determining the number of search

© Global Vision International – 2007

3

minutes per transect (see table 1). These values were calculated by multiplying the search time for each transect by the number of observers on each transect.

Table 1 Sampling Effort (sampling time x num. Observers) Total sampling time (mins) Upper Lower 23-Feb-09 896 25-Feb-09 833 03-Mar-09 575 642 06-Mar-09 810 400 09-Mar-09 575 565 12-Mar-09 420

Total Effort

2793

2923

Surveying rainforest habitat is a privileged opportunity; however there is the potential to negatively affect the ecosystem by passing infections between sites and species. Good practices will be strictly adhered to so as to ensure transmissions are not possible. This will be achieved by systematic cleaning of tools, equipment, and sterile gloves will be changed when handling different individuals. All expedition members were fully briefed regarding precautionary measures and effective surveying techniques. For sampling, representatives of each taxa were anaesthetised, fixed in formalin, and transferred to ethanol, precise locality data was taken for each individual (including habitat information). Preserved specimens will be transferred to the PUCE. In addition, tissue samples of sacrificed specimens will be taken, stored in 70% ethanol and deposited in the tissue storage facility at PUCE. This will allow DNA analysis to be undertaken for any taxonomically problematic species, and will also enable identifications to be reassessed in the future in light of ongoing changes in amphibian taxonomy. Benthic surveying Benthic surveys were conducted in both the upper and lower transects of the pump stream to assess water quality based upon invertebrate species found within the stream. The methods and details of this work can be viewed in the corresponding benthic survey report of the GVI Yachana reserve pump stream. Weather Data Weather data was collecting using a Sun-Moon Radio Controlled Weather Station at the GVI Base Camp. Temperature, pressure, humidity, cloud cover and rainfall data was recorded at 6am, 12am and 6pm daily.

© Global Vision International – 2007

4

Results: In total, three species of amphibian were found in the Lower Pump Stream region (see Table 2) whilst in the upper pump stream region we found seven species (see Table 3). Eight different species were found in total. Rhinella marinus was the only species unique to the lower pump area, whilst Allobates bilinguis, Hypnodactylus nigrovittatus, Pristimantis martiae and Pristimantis acuminatus were all species found only within the upper areas of the pump stream. Pristimantis ockendeni and Bolitoglossa peruviana were the two species that were common to both the upper and lower pump stream transects. In total 18 individuals were caught throughout the sampling period, six of which were caught within the pitfall traps and the other twelve being caught on visual encounter surveys. Five of the nine P.ockendeni caught were captured in pitfall traps with only a single H.nigrovittatus being the only other pitfall capture. No significant correlation was found between the number of frogs encountered and; temperature (R2=0.229), rainfall (R2=0.036) or pressure (R2=0.010).

Table 2 - Bar chart to show how many of each species were found in the lower pump stream region and which methods were successful in finding them. 6 Pitfall Trapping Visual Encounter Survey

Number of Individuals Encountered

5

4

3

2

1

0 Prystimantis okendeni

Bolitoglossa peruviana

Rhinella marinus

Species Encountered on Lower Pump Stream

© Global Vision International – 2007

5

Table 3 - A bar chart to show how many of each species were found in the upper pump stream region and which methods were successful in finding them. 6

Pitfall Trapping Visual Encounter Survey 5

Number of Individuals Encountered

4

3

2

1

0 Prystimantis okendeni

Allobates biliguis

Bolitoglossa peruviana

Hypnodactylus nigrovittatus

Pristimantis martiae

Oreobates quixensis

Pristimantis acuminatus

Species Encountered on Upper Pump Stream

Discussion: The results from this phase do appear to indicate that there is a greater species richness of amphibians to be found within the upper areas of the pump stream than the lower areas. This may indicate that the GVI base camp is having an adverse effect on the quality of the lower area by frequently pumping out water and allowing used water to drain off into the area. For this reason we suggest that this work be repeated throughout future GVI phases in order to increase the robustness of the data collected here, enough to carry out statistical analysis. The number of individuals actually caught within the study was very small to be able to conduct any reliable statistical analysis. It may be a factor that must be taken into consideration for collection and removal of water for the GVI base camp in the near future. At this stage we can not simply assess the differences in species found due to any effects of the GVI camp. It could simply be that there are differences in microhabitats at these different transects of the stream. Vegetation mapping could be used to investigate this in a future surveying period on these sites. The study was too short term to see any significant relationships between weather variables and amphibian abundances were present. This data was taken recorded throughout the study however for any comparison in other research phases to see if any trends between weather data sets and amphibian abundances become apparent in the future.

© Global Vision International – 2007

6

The project was conducted not only to assess the effects of GVI camp on the pump stream but also to trial sampling methods on amphibians within the area for their suitability and effectiveness. This was very useful in allowing us to determine the effectiveness off both methods. The pit fall traps are labour intensive to prepare and install but once up are very quick and easy to check. They seemed particularly useful in detecting P.ockendeni which is a leaf litter egg laying species. The visual encounter surveys are particularly important in detecting a variety of different species which otherwise would not have been found. Overall we can see how a combination of methods is essential and will be used within our future GVI surveys. We would also like to trial acoustic surveys in future method combinations too.

Research staff: Christopher Beirne graduated in Biological Sciences from the University of Edinburgh in the summer of 2008. Having conducted research on selective logging practises in Bolivia and chameleon population dynamics in Madagascar, he is keen to apply his knowledge to amphibian conservation in the Yachana Reserve. Jon Escolar is the field manager for the GVI Amazon Expedition in Ecuador. His education includes an undergraduate degree in zoology and an MSc in sustainable development and human geography. He has worked as a naturalist in Peru and for an environmental consultancy in Spain. Andrew Whitworth completed his MSc in Conservation Biology at Manchester Metropolitan University and currently works as a field officer for GVI. After visiting Tanzania in 2008 he developed a particular interest in the current amphibian crisis and a need for amphibian conservation and research. He now hopes to extend his knowledge towards Latin American amphibian assemblages based at the Yachana reserve.

Thanks: Thanks to all the EMs who have put their time and effort into surveying for amphibians with us on a few late evenings and early mornings.

References: Bartlett RD, Bartlett P. 2003. Reptiles and Amphibians of the Amazon - An Ecotourist’s Guide. Heyer WR, Donnelly MA, McDiarmid RW, Hayek LAC, Foster MS. 1994. Measuring and Monitoring Biological Diversity - Standard Methods for Amphibians. Gardner TA, Fitzherbert EB, Drewes RC, Howell KM, Caro T. 2007. Spatial and temporal patterns of abundance and diversity of an east African leaf litter amphibian fauna. Biotropica 39(1):105113. Lyaruu HV, Eliapenda S, Backeus I. 2000. Floristic, structural and seed bank diversity of a dry Afromontane forest at Mafai, central Tanzania. Biodiversity and Conservation 9(2):241-263.

© Global Vision International – 2007

7