Mammals and Macaws at Refugio Amazonas

Introduction. The results presented here are from field work conducted by the Tambopata Macaw Project at the Amazonian jungle lodge Refugio Amazonas, owned and operated by Rainforest Expeditions SAC (RFE). The research was carried out on properties owned by RFE, the buffer zone and the Tambopata National Reserve with permission from RFE and the Peruvian Institute for Natural Resources (INRENA) under the permit No 025 S/C-2005-INRENA-IANP issued to Dr Donald Brightsmith (RFE Research Director) and permit No 11 S/C-2006-INRENA-IANP issued to researcher Alan Tristram Kenneth Lee. Work on the construction of Refugio Amazonas was started at the end of 2004 and started accepting tourists towards the end of 2005. The surrounding forest has a history of timber extraction, with up to 600 Tornillo (Cedrelinga spp) extracted. An active Brazil nut concession has been operational for at least the past 50 years, with associated low scale hunting. However, a “meat for bullets” agreement has been signed by the operator of this concession and the lodge management whereby the lodge operators supply an agreed quantity of commercial meat to the concession operators in exchange for a cessation in hunting activities. Tourism operators use areas of land to the north of the lodge, but this is the first time that tourists will be visiting the forests surrounding the lodge. Implications are that wildlife will be encountering a higher number of people. It is predicted that the short term impact of this activity will be a reduction in the observed number of mammal species traditionally targeted by hunters in relation to increased tourist traffic along key routes, but that long term monitoring will reveal an overall increase in the abundance of these species as observed at other lodge sites (e.g. Posada Amazonas and Piedras Biodiversity Research Station (Tatum Hume et al. 2005). The primary goal of this study is to establish baseline population indices comparable to studies conducted in other parts of the Amazonia and Tambopata. Methodology Variable Distance Line Transects Mammal populations were monitored using standardized line-transect methodology as described by Peres (1999). Line transects are also the preferred methodology for monitoring parrot and macaw abundance following recommendations that this method is suitable for large, colourful and sparsely encountered species like parrots and hornbills (Bibby et al. 1998) and after a trial on the methodologies was conducted to compare the methodologies (Tatum-Hume et al. 2005). Transect surveys have proven to be the most reliable method of producing relative abundance data in rainforest environments. Previous studies (Perez 1999, Emmons 1984) recorded primates, caviomorph rodents, sciurids, ungulates, cracids, trumpeters, tinamous, wood quails and a number of species of avian canopy frugivores. In this study the only bird species included were those that are commonly hunted (Spix’s guan, razor-billed currasow). All mammal species encountered on the transects were recorded. The transects were walked daily between 05:30 – 10:30 to avoid the hottest part of the day when animals tend to be less active (Peres 1999) and again in the late afternoon from 15:30 – 17:30. If it rained whilst on the transect for more than ten minutes, the transect was abandoned as rain decreases the observer’s ability to detect species. Surveys were conducted along 3 transects, although most sample effort occurred on one 5km transect (ARA). 2 other transects (GATO and SACHA) were initiated later in the study, both of 4km in length. The transects were cut in the direction of a fixed bearing but avoiding natural obstacles such as lakes and “aguajales” (swamp areas). In most cases transects followed the line of existing extraction trails which are abundant in the area and to avoid further perturbation to the forest. The ARA transect runs through a mixture of floodplain forest and terra firme high forest. The transect GATO runs through terra firme only and SACHA runs through floodplain forest only. All transects were marked with tape every 50 m to aid distance measurements. It took a week to prepare and was then cleared of debris before each subsequent monitoring event in order to reduce the chance of mammals detecting the observer before data were collected. The transects were walked at an average speed of 1 km/h and took between two and four hours to complete as due to a rotation system to avoid biasing certain sections of the ARA transect subsections were walked in rotation. The transect was walked by a minimum of one and maximum of three observers. One observer was always a local guide was primarily responsible for detecting all mammals and birds. For each detection event the observers recorded the time, distance along the transect, species, number of individuals, the perpendicular distance from the trail, demography, visibility and weather conditions. Where possible the perpendicular distance was measured using a tape measure or rangefinder. Diagram 1 A map of the study area showing the location of Refugio Amazonas and the location of the 3 transects used for surveys. Results - Mammals Sample Effort A total of 279km of morning and afternoon transects have been completed during the 8 month survey period. Transects were predominantly conducted on the 5km transect called ARA. Surveys started on the GATO transect in July and on the SACHA transect in August. Table 1: Transect names and total number of kilometres sampled per transect from February to September 2006.

Table 2: Common names of the species encountered during transects, with the total number of groups encountered during the entire survey on all transects, together with the number of individuals and relative abundance in terms of individuals per 10km