Download Wildlife Populations in the Pacaya

Document related concepts

Maximum sustainable yield wikipedia , lookup

Conservation biology wikipedia , lookup

Theoretical ecology wikipedia , lookup

Bifrenaria wikipedia , lookup

Wildlife crossing wikipedia , lookup

Biodiversity action plan wikipedia , lookup

Island restoration wikipedia , lookup

Occupancy–abundance relationship wikipedia , lookup

Overexploitation wikipedia , lookup

Conservation movement wikipedia , lookup

Molecular ecology wikipedia , lookup

Habitat conservation wikipedia , lookup

Transcript
Wildlife Populations in the Pacaya-Samiria National Reserve and
Lago Preto, Peru
Report for Operation Wallacea
Prepared by Richard Bodmer, Tula Fang and Pablo Puertas
In collaboration with the Durrell Institute of Conservation and Ecology (DICE), World
Wildlife Fund – Peru Office, National University of the Peruvian Amazon (UNAP),
National Institute of Natural Resources – Peru (INRENA), and Pacaya-Samiria National
Reserve Authority
February 2009 1
Introduction The Amazonian forests of Loreto, Peru are situated in the western Amazon basin and harbor some of the greatest mammalian, avian, floral and fish diversity on Earth (Puertas and Bodmer 1993, Pitman et al. 2003). Indeed, these forests are one of the last remaining true wilderness areas left on the planet. However, these vast expanses of pristine forest will only remain intact if conservation programs are successfully implemented (Anderson 1990, Peres et al 2006). Wildlife conservation in Loreto must incorporate landscape features (in terms of major habitat types of both flooded and upland forests), the biology of landscape species and the socio‐economics of the rural people (Bodmer et al. 1998). Research and conservation activities should use an interdisciplinary approach to find a balance between the needs of the indigenous people and the conservation of the animals and plants (Bodmer and Puertas 2000, Peres 2000). For example, some animals like the primates, jaguars, manatees and tapirs are very vulnerable to hunting and their populations can rapidly be depleted by overhunting (Bodmer et al. 1997, Peres & Palacios 2007). Other animals are more appropriate as a source of meat for local people, such as peccaries, deer and large rodents (paca and agouti)(Bodmer & Robinson 2004, Fang et al. 2008). This project is helping to conserve wildlife, not only for saving the biodiversity of the Amazon, but also as a means of helping the poor indigenous people who rely on these resources for their food and shelter (Bodmer and Robinson 2004). The project is working together with the local people, because they are the true guardians of the forest, and information provided by this research can help the indigenous people make appropriate decisions on how best to save the Amazon (Halme & Bodmer 2007). The Amazon has been abused in the past, through deforestation for timber, overhunting of animals and overfishing (Dourojeanni 1990, Peres & Nascimento 2006). Local people are taking actions in places like the Pacaya‐Samiria Reserve and the Yavari River basin including the Lago Preto Conservation Concession. These areas are examples of how things are changing; they are examples of how conservation can work in collaboration with local people, governments and NGO’s (Bodmer et al 2008). Lago Preto: a Public‐Private Reserve The Lago Preto Conservation Concession is a public‐private Amazonian reserve, where the Peruvian government granted a 10,000 hectare block of rain forest to WCS (in collaboration with DICE) as a 40 year concession. Conservation concessions are a relatively new strategy in Peru with the first concession granted in 2004. The effectiveness of conservation concessions will be paramount to a diversified conservation landscape. The goal of the current research at Lago Preto is to investigate the effectiveness of a conservation concession for Amazonian biodiversity. Results will help determine the potential of public‐
private conservation strategies as a viable protected area category, thus increasing diversity in protected area systems. Pacaya‐Samiria: a National Reserve Incorporating Local People The Pacaya‐Samiria National Reserve has gone through a major shift in its management policies over the past two decades, from an area of strict protection where local people were excluded from the reserve to an area where the local indigenous people participate with the reserve management. This drastic shift in conservation policy has led to a reduction in hunting pressure and an increase in wildlife populations (Bodmer & Puertas 2007). When the park administration changed and the reserve began to incorporate the local communities in the management of the area, attitudes of the local people also changed (Puertas et al. 2000). Local management groups were given areas to manage and were no longer considered poachers. They were able to use a limited amount of resources legally and with reserve administration approval. Many of the local people changed their attitude 2
towards the reserve and began to see the long‐term benefits of the reserve for their future. The reserve became part of their future plans and there was increasing interest in getting involved with the reserve. Many local people can now see the socio‐economic benefits of the reserve and are themselves helping to conserve the area. Hunting has decreased substantially, both due to the poachers now becoming mangers, and because the local people are keeping the other poachers out of their management areas. Research Objectives The project is using a set of key wildlife species to evaluate the conservation success of the Lago Preto Conservation Concession and the Samiria River basin in the Pacaya‐Samiria National Reserve. 1) Ungulates are important bush meat species and were used to examine the impact of hunting by using sustainability models and long‐term monitoring. 2) River dolphins were used as indicator species for the aquatic systems by incorporating long‐term monitoring of the dolphin populations. 3) Macaws were used as general indicators of the terrestrial systems by evaluating changes in species numbers and composition. 4) Primate populations were monitored to determine the impact of hunting on arboreal mammals and to study ecological interactions between arboreal and terrestrial wildlife assemblages. 5) Game bird populations were monitored to determine their conservation status and impact of hunting. 6) Caiman populations were monitored to evaluate the recovery of black caiman populations and the ecological interactions between species. 7) River turtle populations were monitored to determine the success of the head‐
starting conservation programme in the Samiria River basin 8) Giant river otter populations were monitored to evaluate the recovery of this rare species. 9) Amazon manatee populations were monitored in the Samiria river to evaluate the conservation status of this vulnerable species. 10) Large cats, including jaguars, pumas and ocelots were monitored using camera traps to determine a baseline for long‐term population studies in the Lago Preto Concession. 11) The abundance, diversity and age structure of the large fish species were monitored to determine the impact of local fisheries and the effectiveness of fisheries management. 12) The project also worked with implementing and monitoring of community‐based conservation activities in indigenous communities. In these communities the project worked with sustainable use and community‐based management of natural resources. The Lago Preto Conservation Concession The Lago Preto Conservation Concession is a 10,000 hectare block of tropical forest located in the upper Yavari River just below the confluence of the Yavari and Yavari Miri Rivers. The Yavari River forms the boarder between Peru and Brazil in the western Amazon basin. The forests of Lago Preto are some of the most biodiverse habitats in the Amazon, with 13 sympatric species of primates, an extraordinary avian assemblage, a unique floral composition and one of the greatest diversities of fish and frog species found in the Neotropics. 3
The Lago Preto Conservation Concession adjoins the proposed Greater Yavari Protected area which covers over 1.5 million hectares of pristine Amazon forests. This area is one of the most remote and isolated regions of the Amazon. There are still uncontacted Indian tribes inhabiting the Greater Yavari landscape, and the overall human population is one of the lowest found anywhere in South America. The Lago Preto Conservation Concession has three major habitat types, including upland non‐flooded terra firma forests, seasonally flooded varzea forests and palm swamps. The upland forests consist of undulating hills broken by stream valleys and ravines. These forests have exceptionally high canopies often with emergent trees reaching 30 meters. The understorey is composed of sparse shade resistant vegetation. Seasonally flooded forests are inundated by the white waters of the Yavari between 4 to six months each year. These forests are situated in the Yavari floodplain and have softer alluvial soils and a lower canopy than upland forests. Levees, or slightly raised floodplain islands are interspersed by backswamps and low laying streams. The varzea forests have a lower floral diversity than upland forests. Palm swamps are usually situated at the boarder between the upland and flooded forests. These low laying backswamp habitats are dominated by the emergent mauritia palms. Palm swamps usually extended for several kilometers and have permanent water stands. Palm swamps have a high production of mauritia palm fruits, which are eaten by a great diversity of wildlife. This habitat supports many mammalian and avian species, such as uakari monkeys, squirrel monkeys, and most species of macaws. The Lago Preto Conservation Concession has seven major ox‐bow lakes. These lakes are protected under the concessions management plan and have an extraordinary diversity and abundance of fish, including large populations of peacock bass, the large Aripima paiche fish, arawana fish, sting rays, among many other species. In addition, the lakes have a recovering population of giant river otter and a healthy population of wading birds. The Lago Preto Conservation Concession has the greatest resident population of red uakari monkeys found to date (Salovaara et al 2003). Most of the red uakari monkeys are habituated and easy to observe. Large mammals are regularly sighted along the forest trails, including peccaries, the occasional tapir, many primate species, edentates such as armadillos, anteaters, and sloths, and large cat species.
4
Map of the Lago Preto Conservation Concession on the Yavari River, Peru. Pacaya‐Samiria National Reserve The Pacaya‐Samiria National Reserve extends over an area of 2,080,000 ha in the Department of Loreto, Peru. The reserve is dominated by white water flooded forests, known in Amazonia as Varzea forests. The reserve is comprised of two major drainage basins: the Pacaya River basin and the Samiria River basin. The Samiria River basin is the largest geological and ecological feature of the reserve. The principal habitat types of the Peruvian Amazon are a result of large scale geological changes that occurred during the tertiary and quaternary periods. The Samiria River basin is situated in the Pevas lake bed that formed after the uplifting of the Andes. When the Pevas Lake drained it left a geological depression in this area of western Amazonia, which is currently characterized by soft alluvial soils. The wildlife of the Samiria River lives in an ecosystem that is characterised by large seasonal fluctuations occurring between the high water and low water seasons. The ecology of the aquatic and terrestrial wildlife revolves around these seasonal changes in water level. The large seasonal inundations that annually occur along the immense floodplains of the Pacaya‐Samiria rivers are a result of rainfall in the foothills of the Andes. During the summer months of October to May precipitation off of the Pacific Ocean is pushed over the Andes by the strong uplifting winds originating from Southern Ocean currents. This results in heavy rainfall on the eastern Andes that runoff into the western Amazon basin. The result is large scale flooding along the major rivers situated in the old lake bed and the high water season. In contrast, during the winter months of June to September the precipitation off the Pacific Ocean decreases and the rains in the eastern Andes is greatly reduced, resulting in the drying up of the western Amazonian rivers and the low water season. The Pacaya‐Samiria National Reserve is situated between the confluence of the two largest tributaries to the Amazon in Peru, the Ucayali and Maranon Rivers. The meeting of these two rivers actually come together to form the Amazon River proper. The Ucayali and Maranon Rivers have snaked back and fourth across the Pacaya‐Samiria reserve over the millennia leaving behind an abundance of oxbow lakes, channels, levees, and other geological features. Indeed, both the Pacaya and Samiria Rivers are actually old channels. Indeed, the headwaters of the Samiria River originate principally from inflows of the Maranon River and the Samiria River drains back into the Maranon. Likewise, the Pacaya River originates from water of the Ucayali River that is drained back again at its mouth. The soils of the flooded forests in the Pacaya‐Samiria Reserve are rich in nutrients due to the white water rivers of the Ucayali and Maranon flowing through the forests and depositing sediments during the high water period. The Samiria River is characterised by a blackish colour during high water. This is a result of the white water from the Maranon entering the flooded forests, depositing the sediments as the currents slow during their movement through the forests, and the water picking up tannins from the leaf litter. Whilst the soil of these Varzea forests is rich, the ecological conditions of long periods of flooding, up to 6 months, is very harsh on much of the floral and faunal community. Many plant species can not withstand the long periods of inundation and the diversity of plants in the heavily flooded areas is lower than lightly and non‐flooded levees. Likewise, the terrestrial wildlife must seek out floodplain islands or levees during the high water season, which have increased competition and predation pressures. Even the arboreal wildlife is impacted by the flooding, since many of the fruit trees are quite seasonal in the Varzea forests, resulting in seasons with low food production. 5
The aquatic wildlife is equally affected by the large seasonal inundations. During the flooded periods the fish enter the flooded forests and feed on the abundance of vegetative and animal production, especially the abundance of fruits, invertebrates and other living organisms trapped in the annual floods. Indeed, many tree species fruit during this season and rely on the fish as their primary means of seed dispersal. During the flooded period many fish populations reproduce within the inundated forests. Other aquatic wildlife have a more difficult time during the floods, such as the dolphins, giant river otter and other fish predators, since their prey is more sparsely distributed throughout the large expanses of the flooded forests. When the waters recede during the dry months, fish populations become condensed in the reduced lakes, rivers and channels with ever increasing competition and predation. During this period many fish populations migrate out of the Varzea rivers and into the larger rivers. The dolphins and other fish predators have an abundance of prey during the low water season and even follow the fish migrations down the rivers and channels. The people who live in the flooded forests also have adapted to the seasonal fluctuations in both the use of the natural resources and their agriculture. During the high water season fishing is more difficult, since the fish are dispersed throughout the inundated forests. However, during this period hunting becomes easier with the large bushmeat species, such as deer, peccaries and tapir being trapped on the levees and islands. In contrast, during the low water season the bush meat species become difficult to hunt as they range throughout the entire forests, and the fish become easy prey being trapped in the reduced water bodies of the lakes, channels and rivers. The local indigenous people of the floodplain forests alter their hunting and fishing accordingly, with a greater emphasis on hunting during the high water season and a greater focus on fishing during the low water season. The agriculture of people inhabiting the flooded forests takes advantage of the rich soils from the annual deposits of sediments and the short growing period that needs to be harvested before the floods return. Traditionally, people of the flooded forests have relied on the manioc as their staple agricultural product. Manioc has a short growing period that can be planted and harvested within the low water season. Manioc flour (farainha) is produced by baking pounded. Manioc flour can be stored throughout the year and supply carbohydrates to the people during the flooded periods. 6
Partial view of the Samaria river basin. The area around PV Samiria is the mouth section. The area around PV Tachacocha is the mid‐section. The area around PV Ungurahui is the near up river section, and the area around PV Pithecia is the far up river section. Methods Terrestrial mammals Line censuses along transect trails were used to conduct terrestrial mammal and game bird censuses. Censues trails between 2‐5 km in length were surveyed repeatedly. Information registered on a census includes: day, site, species, number of individuals, and perpendicular distance from the individual to the transect line, habitat, time, distance travelled and weather conditions. The method assumes that all the animals that are on the center of the line transect (0 m perpendicular distance) will be observed. The technique is based on the notion that observers do not see all the animals that are off the center of the line, and that the probability of sighting an animal depends on the distance of the animal from the line. 7
Animals closer to the line have a higher probability of being seen than animals further from the line. The perpendicular distance of all solitary animals sighted, or the first animal sighted in a social species were recorded (Buckland et al. 1993). The DISTANCE estimation calculates the animals that you did not see, and includes these animals in the density estimate. The method relies on measuring the perpendicular distance of animals before they move as a consequence of seeing the observer. That means observers must try and see the animal before they sight the observer. It also means observers must measure the perpendicular distance of the first sighting. If animals move because of the observer than the estimate will be biased. With the DISTANCE programme trails do not have to be straight, but the perpendicular distances must be measured at the correct angle of the center line. The perpendicular distance will be measured directly from the point of first sighting (Buckland et al. 1993). The method assumes that animals are independently dispersed throughout the habitat. Since individual animals within a social group are not independent, but move dependant upon one another, animal groups in social species must be considered as the sampling unit. Thus, DISTANCE will calculate the density of animal groups (Buckland et al. 1993). The equipment used for line transects included: a map of the area, a compass, data sheets, pens and binoculars. Trails were not placed with any pre‐determined knowledge of the distribution of the animals. Censuses were done using small groups of three or four observers. Transects were walked slowly and quietly (500‐1,000 m/hr) between 7am and 3pm. Census information were analyzed using DISTANCE software (Thomas et al. 2002). This programme is regularly used in calculating individual or group densities (Buckland et al. 1993) and can estimate densities if the distribution of sightings within a transect line forms a clear probability function. When the number of sightings is deemed insufficient to determine a probability function, the method known as ‘fixed width’ was used to estimate the densities. 8
Censuses of caimans To assess the population and ecology of caiman species in the ecosystem it is necessary to gain an understanding of their population size. Aquatic transects were used traveling upstream or downstream on the main river and in nearby channels or lakes. A GPS was used to determine the distance surveyed each night. All caimans seen were identified to the species level as best as possible and size of the caiman and location were noted. These data, along with data collected from captured caimans, were used to analyze the caiman population size. Caiman surveys and captures were conducted from a small boat fitted with a 15‐horsepower engine. Caimans were located by their eye reflections using a 12‐volt spotlight and approached to a distance where the engine was silenced and the boat paddled closer. Noosing was used to capture caimans. The noose was made of a long pole about 2 m in length with a loop of rope that can be pulled tight over the caiman’s neck. The caimans were secured with rope tied around the jaw behind the nostrils and around the neck. Total body length was measured from the tip of the snout to the tip of the tail, while head length was measured from the tip of the snout to the posterior edge of the orbital (the vent). The sex was also determined. Weight of the caiman was recorded in kilograms. A measuring tape and weighing scales was used. The population abundance of each species was calculated using the formula N/L, where N= the number of individuals and L= the distance travelled in kilometres. The results indicate the number of individuals per kilometre. 9
Censuses of macaws Point counts were used to monitor macaws. Between eight and nine points was established in each sampling unit separated by 500m. A GPS was used to measure the distance between points. Fifteen minutes was spent at each point. Censuses were carried out twice a day in the morning (5:30‐9:00h) and afternoon (16:00‐18:30). The censuses usually lasted longer in the morning than in the afternoon. Within the 15‐min counts, all macaw species either perched or flying were noted. The distances of the birds from the observer were estimated where possible. A motorized boat was used to travel to each point. Abundance data were calculated in each sampling zone. This was done by adding the total number of sightings and dividing this number by the number of points. Thus, abundance is expressed as the number of individuals per point. Census of turtles and the nesting programme The censuses of the river turtles, in particular Podocnemis uniflis, was carried out in the Samiria River basin. The method consists of travelling with the current of the river on a boat and registering the number of individuals sited, either sunbathing or swimming. The ‘fixed width’ method was used to estimate turtle abundance, where the fixed width was the width of the river. 10
The censuses was carried out using a boat and following an imaginary line across the middle of the river between the hours of 11:00 and 13:00, collecting data on: the perpendicular distance, the number of individuals, the location of the boat, the activity of the species and any other information deemed relevant. To facilitate the observations, binoculars were used when individuals were sunbathing more than 50 m away from the boat. The classification of the microhabitat was recorded. Beaches at were used to collect turtle eggs from nests as part of the head starting programme. Turtle eggs were collected from the nests and placed in an artificial beach constructed next to the park guard posts. After 60 days the turtles hatch and were released back into the main river. Preparation of the artificial beach The artificial beach was prepared by turning over the sand which had been carefully selected and free from organic remains and insects. The artificial beach was encircled by a barrier created from palm tree wood, in order to keep in the sand and to facilitate drainage whilst avoiding the loss of nests during the incubation process. The size of the palm frame is dependent upon the number of nests planted; in this case a frame of 5 by 6 metres was used. Collection of eggs The nests were located by gently probing the ground with a small stick; or by simply pressing the ground very gently with one’s heel when the footprints are not clear enough or had been washed away by rain. Transport of eggs 11
Once the nests are found, they were carefully excavated out of the soil using ones hands. Eggs were extracted, one by one, and placed very gently on a tray with a layer of sand. Care was taken to maintain the original position of the egg and not to turn them around during handling in order to make sure that the growth rate of the embryo remains intact. Every nest in the tray was labelled with the number of eggs collected and the number of eggs that were not viable. The latter were covered up with sand and separated with leaves. Selection of eggs The eggs were taken to the artificial beach built next to the guard post where they were incubated. Eggs considered viable are those that develop a whitish spot 24 hours after they have been laid. Cracked shells render eggs inviable. Eggs that were not considered viable due to cracks in their shells, fungus, abnormal size, or appearing flaccid, were discarded. Nesting A hole was dug by hand in the sand similar to that of a natural nest. The depth of the hole depends on the number of eggs that were placed in the nest. One by one, the eggs were placed inside the whole, in the same position in which they had been found, and the nest was then be covered up with sand and lightly patted to create a little mound of sand about 5cm high. The distance left between each nest was 20 cm, whilst 30 cm of space was left between each row. Information on the number of nests, the number of eggs, the percentage of inviable eggs as well as the date of collection and incubation was recorded. Censuses of dolphins and manatees Dolphin censuses were carried out at both sites. Five kilometres was travelled daily from 9:00 to 14:00 h along the centre of the river using a boat. Information collected included: species, group size, group composition, behaviour (travelling, fishing, playing), time, and any additional observations. N
Data were analysed using fixed width: D=
2 AL
Where: D= Density N= Number of individuals A= River width L= Distance travelled 2= Number of margins sampled Aquatic surveys were used to census dolphins. A GPS (Global Positioning System) was used to determine the length of each aquatic census. Dolphin transects usually take three to four hours to complete depending on the speed at which the river is flowing. A motorized boat was used to carry out the census. Any dolphins seen coming to the surface for air, swimming with their heads above water, sunbathing or swimming just below the surface of the water (i.e. no deeper than 5 cm) were recorded. Care was taken not to double count any dolphin sightings. Behavioral information on the dolphin activity was also recorded along with the size class of each dolphin sighting. For each transect the weather conditions and the start and finish times were recorded. 12
Censuses of fish Censuses were carried out at both sites. During the censuses green gill nets of 3.5’’ were used in lakes and channels with weak currents and white gill nets in the river. Fishing points were located on shores or banks where there is aquatic vegetation or shrubs, although meanders are the preferred areas. Individuals were identified, measured and weighed. Catch per unit effort was calculated by the number of individuals per species caught and the effort spent fishing at each zone. 13
Habitats were also compared (lake, channel, river) and diversity indices were used. Productivity of fish was shown in terms of catch per unit effort, using the ‘biomass captured per effort’ method. The CPUE method is a robust indicator over time for the level of abundance, density and pressure fishing in a given zone (Queiroz 2000). The length‐
frequency analysis helps to predict biological impacts of fisheries. A harvest focused on juveniles, for example, causes greater impact than a harvest of adult fish not in their breeding period. Camera Trapping These “traps” combine cameras with heat/motion sensors in order to “capture” photographs of cats and other rare species. The sampling design is based on other jaguar studies recently conducted in Latin America. Camera trap‐based sampling methods have emerged as a reliable technique for inventorying and monitoring medium‐ and large‐bodied mammals. A total of twenty paired MC2 GV StealthCam passive infrared camera trap stations were set covering an area of c. 50 km2 and distributed across four habitat types: varzea forest (five stations), Aguajal, Aguajal‐terra firme forest edge and terra firme forest. An additional single station (of six cameras) was placed around a natural salt lick known to be widely used by different species, producing a total of 21 camera trap placements. For the camera trap data, detection histories (H) were constructed for each species over one‐week sampling occasions (n = 15). For each species and for each occasion, ‘1’ indicated the detection (photograph) of a species, while ‘0’ indicated the non‐detection of a species. Species detection histories were then used to produce probabilities. For example, a detection history for species i (Hi) of 100011 would represent species detections on the first, fifth and sixth occasions over a single season and the probability of recording history Hi 14
would be, Pr(Hi = 100011) = ψp1(1 – p2) (1 – p3) (1 – p4) p5p6 eqn. 1
Trail systems of LPCC and Camera trap stations. White lines indicating trails used for line transects survey Where pj is the probability of detecting the species during sampling period j (= 1,…,6), conditioned upon the species being present. Thus, for the camera trapping dataset an X‐
matrix was constructed, with individual species representing the rows. These data were then entered into PRESENCE v2.1 software (Proteus Wildlife Research Consultants, New Zealand). Before importing the species specific taxonomic traits covariates into Presence, collinearity between covariates was tested within SPSS using a Spearman’s rank correlation. This test identified significant (P < 0.05) non‐independence between body mass and height, so, a data‐reduction technique (principal component analysis) was used to identify the components explaining most of the variance observed between the two indicators. From the final component matrix, a single covariate was created that explained 93% of this variance. This covariate was then imported into PRESENCE for use in the final analysis, along with the taxonomic trait covariates. The importance of covariates on species detection probabilities was explored by setting the occupancy parameter as constant, (ψ), and then allowing the detection probability parameter, p(.), to vary with covariates either individually, e.g. ψ(hunted), or in combination, e.g. ψ (hunted+nocturnal). Comparisons of the explanatory power of these candidate models were based on their delta Akaike Information Criterion (ΔAIC) values, adjusted for small sample sizes (ΔAICc), and their Akaike weights (wi). The influence of covariates from models that were within two ΔAICc units of the top ranked model, and exhibited a satisfactory goodness of fit (P > 0.05), were discussed. Giant River Otter Censuses were conducted for two months intervals. Sample counts were used to compare relative abundance between censuses rather than absolute counts which would have 15
required identification of all individuals. Censuses were conducted by by boat, scanning with binoculars and listening for otter calls. We recorded group sizes and locations using a handheld Global Position System (GPS), which we also used to calculate the length of each transect. Double counting was avoided by keeping a constant boat speed and where possible by identifying groups by the unique throat markings of individuals. The total length of each transect was recorded, along with the width and type of the body of water. Community work Data on people’s attitude to the wildlife management guidelines, on their knowledge of how to implement the guidelines, and on their knowledge of current wildlife agreements were collected by using a questionnaire survey, focus group discussions and interactive dialogues. As an independent measure of how well the guidelines were being implemented in relation to the community attitude was collected from the community hunting register. The hunting register also supplied information for the calculation of Catch per Unit Effort data to provide an index of the population responses of hunted species to hunting. As an independent measure of the impact of hunting on species populations, density data were also analyzed. Sample Sizes The following table presents the sample sizes of aquatic and terrestrial transects used during the three year study. Activity
Terrestrial censuses (km)
Dolphin censuses (km)
Caiman censuses (km)
Macaw counts (Points)
Fish surveys (hours)
Turtle censuses (km)
2006
540
217
189.23
331
307
101.95
Samiria
2007
820.87
259.47
124.62
622
175.3
24.95
2008
1068.65
442.15
252.8
383
135.45
193.69
2006
507.1
24
50.1
139
166
Lago Preto
2007
601.3
119.78
132.7
84
221.01
2008
196.2
137.76
189.6
130
120
Results Ungulates are important bush meat species and were used to examine the impact of hunting by using sustainability models and long‐term monitoring. Studies on the natural fluctuations of animal populations and how these can be incorporated into sustainability models A 12 year data set on wildlife populations in the Yavari Miri was used to determine how natural fluctuations in wildlife populations can be incorporated into sustainability analysis. Data were used for the five most socio‐economically important game animal species: Collared peccary (Tayassu tajacu), white‐lipped peccary (Tayassu pecari), lowland tapir (Tapirus terrestris), red brocket deer (Mazama americana) and woolly monkey (Lagothrix poeppigii). Tayassu tajacu “collared peccary” The population density of collared peccary increased from a mean of 1.58 ind/km2 in 1995 to 9.94 ind./km2 (ANOVA, F3,8 = 29.59, P < 0.001). However, mean population density did not differ significantly between 1999 and 2007 as indicated by Tukey’s test of homogenous subsets. This result indicates that the population of the source area is not declining under current levels of hunting. 16
Population density (ind./km2)
14
12
10
8
6
4
2
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Population density of Tayassu tajacu overtime (mean ± SD). CPUE (ind./hunting days)
The catch per unit effort analysis indicates that the abundance of this species was stable between 1999 to 2006 (Kruskal‐Wallis, H= 0.9912; df= 3; P= 0.8034). This result is similar to the population density analysis, strengthening the indication that the population is not declining under hunting pressure. Harvest model analysis indicate that the harvest is well within the suggested sustainability limits of 40% of production and it indicates that hunting is sustainable over time in Yavari Miri. 0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Catch per unit effort of Tayassu tajacu. (mean ± SE). 17
40% Sustainable yield
45
Harvested Yield (%)
40
35
30
25
20
15
10
5
0
1998
2000
2002
2004
2006
2008
Year
Evaluation of sustainability of hunting in Tayassu tajacu using harvest model analysis. (mean ± SD). The collared peccary population in the source area increased between 1995 and 1999, but thereafter remained stable until 2007. In the hunted area, CPUE analysis indicates that peccary abundance in the hunted area remained stable between 1999 and 2007. The harvest model analysis indicates that the harvest was consistently below the sustainability threshold. The relatively stable population of collared peccary was incorporated in the sustainability model demonstrating that stable populations are inherently incorporated into the models. Tayassu pecari “white lipped peccary” Density estimates for this species indicate a large fluctuation over a 12 year period, with a maximum mean density in 1999 and low densities in 1995, 2005 and 2007 (X2= 12.948; df= 3; P < 0.01). The population declined between 1999 and 2007. The catch per unit effort analysis shows a significant decrease in the number of individuals recorded per hunting day between 1999 to 2004 (Kruskal Wallis, H= 14.0216; df= 2; P < 0.001), that stabilised between 2004 and 2006. Meanwhile the number of animals harvested per km between 1995 and 2007 shows a similar pattern to the population density, with a peak in 1999 and low in 2004. The hunting pressure analysis was carried out in order to show that the local people are hunting less intensively, they have reduced their hunting from 2000 to 2005. The percentage harvested yield data indicated that the percentage was within the 40% limit of sustainability. Population density (ind./km2)
25
20
15
10
5
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Population density of Tayassu pecari overtime (mean ± SD). 18
1.2
CPUE (ind./hunting day)
1
0.8
0.6
0.4
0.2
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Catch per unit effort of Tayassu pecari. (mean ± SE). Hunting pressure (harvested
ind./km2)
2.5
2
1.5
1
0.5
0
1998
1999
2000
2001
2002
2003
2004
2005
2006
Year
Hunting pressure in sink area of Nueva Esperanza community. 40% Sustainable yield
45
Harvested Yield (%)
40
35
30
25
20
15
10
5
0
1998
2000
2002
2004
2006
2008
Year
Evaluation of sustainability of hunting in Tayassu pecari using harvest model analysis. (mean ± SD). A comparison of CPUE data on collared and white lipped peccary indicates that a high abundance of collared peccary is associated with a lower abundance of white lipped peccary with a significant negative correlation between these species. White lipped peccary (CPUE)
0.9
Pearson= -0.89334, P = 0.0411
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.2
0.22
0.24
0.26
0.28
0.3
0.32
Collared peccary (CPUE)
Relationship between white‐lipped and collared peccary using CPUE values which reflect abundance patterns. 19
The natural population fluctuation of white lipped peccary in the source area appears to have affected the population in the sink area since CPUE values have decreased since 1999. Hunting pressure has shown a similar decline suggesting that hunting is not responsible for the decline in the sink area and this is supported by the harvest model analysis which indicate that the harvest is within the sustainable yield threshold. Thus hunting appears sustainable in the sink area over time. The results suggest that natural fluctuations in the white‐lipped peccary population are inherently incorporated into the sustainability models. When the white‐lipped peccary went through a decline, the hunters took fewer animals, and the overall sustainability of the hunt was similar. Tapirus terrestris “lowland tapir” The population density analysis for the lowland tapir suggest that density may be decreasing from 0.21 ind./km2 in 1995 to 0.08 ind./km2 in 2007, however these values did not differ significantly over time (X2= 0.118; df= 3; P= 0.9896), likely because sample sizes are too small. The trend in mean catch per unit effort appears to be declining suggesting the same trend as the population density analysis, but due to small sample sizes this was not statistically significantly (Kruskal‐Wallis, H= 2.312; df= 3; P= 0.5102). The harvest model analysis shows that hunting is obviously not sustainable over time, because over 600% of annual yield is harvested in comparison with a threshold of 20 % of productivity. Population density (ind./km2)
0.5
0.4
0.3
0.2
0.1
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Population density of Tapirus terrestris overtime. 0.4
CPUE (ind./hunting days)
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
1994
1996
1998
2000
2002
2004
2006
2008
Catch per unit effort of Tapirus terrestris (mean ± SE). 20
1200
Harvested Yield (%)
1000
800
600
400
200
0
1998
2000
2002
2004
2006
2008
Year
Evaluation of sustainability of hunting in Tapirus terrestris using harvest model analysis. Tapirs occur at very low densities in the source area and results indicate that both density and CPUE appear to be decreasing over time. The sustainability models indicate that the harvest is not sustainable and reflect the decreasing population of lowland tapir. Thus, the sustainability model incorporates the decreasing tapir populations. Mazama americana “red brocket deer” The population density of red brocket deer in the source area is not stable over time (ANOVA, F3,8= 8.933; P < 0.01). Density increased from a low of 0.93 ind./km2 in 1995 and 0.72 ind./km2 in 1999 to a high of 2.29 ind./km2 in 2005, but then appeared to decrease to 1.17 ind./km2 in 2007, although this was not significant. Catch per unit effort analysis indicates that the abundance of red‐brocket deer is increasing (Kruskal‐Wallis, H= 9.154; df= 3; P = 0.0273). Whilst CPUE did not differ between 1999 and 2005 there was a significant increase in 2007. Harvest model analysis indicate that the harvest is below the suggested sustainability limits of 40% of production, and thus that hunting at this level is sustainable over time. Population density (ind./km2)
3
2.5
2
1.5
1
0.5
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Population density of Mazama americana overtime (mean ± SD). 21
0.35
CPUE (ind./hunting days)
0.3
0.25
0.2
0.15
0.1
0.05
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Catch per unit effort of Mazama americana. (mean ± SE). 45
40% Sustainable yield
Harvested Yield (%)
40
35
30
25
20
15
10
5
0
1998
2000
2002
2004
2006
2008
Year
Evaluation of sustainability of hunting in Mazama americana using harvest model analysis (mean ± SD). Red brocket deer in the source area appear to experience natural population fluctuations over a 10‐12 years cycle and this is reflected in the CPUE data from the sink area. The harvest model data suggest that harvest levels are sustainable being below the 40% harvest yield threshold. The red brocket deer analysis also demonstrates that the sustainability analysis inherently incorporates natural fluctuations in the species. In contrast to the white‐lipped peccary, red brocket deer populations are generally increasing in the Yavari‐Miri study site. The sustainability analysis captured these increases, and whilst hunting and CPUE increased, the sustainability of the hunting remained similar. Lagothrix lagothrica “woolly monkey” The population density of the woolly monkey increased over time in the source area (ANOVA, F3,8 = 8.0716; P < 0.01). Density increased significantly between 1995 and 1999, but then remained fairly stable between 1999 to 2007. Catch per unit effort analysis suggests that the abundances of this species was also stable in the sink area from 1999 to 2006 (Kruskal‐Wallis, H= 0.7349; df= 3; P = 0.865). 22
The harvest model analysis indicates that the harvest is within the suggested sustainability limits of 20% of production and it shows that hunting is sustainable over time. Population density (ind./km2)
45
40
35
30
25
20
15
10
5
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Population density of Lagothrix overtime (mean ± SD). CPUE (ind./hunting days)
0.25
0.2
0.15
0.1
0.05
0
1994
1996
1998
2000
2002
2004
2006
2008
Year
Catch per unit effort of Lagothrix (mean ± SE). Harvested Yield (%)
25
20% Sustainable yield
20
15
10
5
0
1998
2000
2002
2004
2006
2008
Year
Evaluation of sustainability of hunting in Lagothrix using harvest model analysis (mean ± SD). The woolly monkey population in the source area has been increasing. But the population trend stabilized between 1999 to 2007. This stability in population density coincided with a stable trend in catch per unit effort analysis in the sink area. Meanwhile, the harvest analysis indicated that the harvest level was below the sustainability threshold. The woolly monkey also shows that the sustainability models are robust to natural fluctuations in the populations. Woolly monkeys have increased in the Yavari‐Miri study area and hunting has decreased by local people implementing the wildlife management guidelines. This has resulted in a lower CPUE, even though the population has increased. This results indicate that the sustainability models inherently incorporated the natural increases in the woolly monkey populations. 23
The 12 year data set for the Yavari‐Miri shows that natural fluctuations in wildlife populations are inherently incorporated into the sustainability models. Indeed, these models seem particularly robust in incorporating the natural population fluctuations. The sustainability models therefore can continue to be used and do not require any refinement in including natural fluctuations. The relatively random nature of hunting by Amazonian people is probably an important factor that affects the robust nature of the sustainability models to natural fluctuations. Random hunting should reflect natural fluctuations. Thus, decreases in animal populations should reflect decreases in hunting pressure, and visa versa. In addition, the sustainability models incorporate density as an important variable. Thus, increasing densities raise the level of individuals that can be hunted sustainably, and visa versa. Both of these factors appear to influence the robust nature of the sustainability analysis. Studies on determining the size of source (non‐hunted) and sink (hunted) areas There needs to be a method to determine the size of source areas. Source areas are a key aspect of the cross cutting program, since they can be used to increase the number of hectares under protection using a wildlife management strategy. Source‐sink areas can provide a buffer against the biological and socio‐economic uncertainty of bush meat hunting. Source areas are non‐hunted areas adjacent to hunted (sink) areas. Wildlife emigrates out of source areas and immigrates into sink areas, if sink areas have been overhunted. Indeed, implementing the guidelines incorporates community‐based protected areas as an integral part of the process through the use of source areas—the establishment of non‐hunting areas adjacent to hunting zones that allow animal populations to flourish in an undisturbed environment.The cross cutting wildlife program has developed a method to estimate the size of source areas using the collared peccary (Tayassu tajacu) as the landscape species determining the size of areas needed. The results of the analysis showed the following. To determine the size of a source area needed to ensure the sustainability of collared peccary hunting, a PVA was used to determine the population size required to support the current level of harvest with zero probability of extinction. For the baseline PVA model, the stochastic population growth rate is 0.054 ± 0.138 SD with zero probability of extinction (PE) in 50 years, and a mean population size of 4790, having set carrying capacity equal to the initial population size of 7000. The minimum population size with zero PE in 50 years is 3250 with a population growth rate of 0.042 ± 0.162 SD and a mean population size in 50 years of 4621 individuals. 24
Initial Population size
PE = 0
0 > PE ≤ 0.1
0.1 ≥ PE < 0.2
0.2 ≥ PE < 0.5
0.5 ≥ PE < 0.7
PE ≥ 0.7
7000
6750
6500
6250
6000
5750
5500
5250
5000
4750
4500
4250
4000
3750
3500
3250
3000
2750
2500
2250
2000
1750
1500
1250
500 600 700 800 900 1000
Number of harvested individual/ year
Results obtained from modeling different harvesting levels against populations of different sizes. PE (probability of extinction). At current densities observed in the study, a population of 7136 individuals at carrying capacity would inhabit an area of 800 km2. Thus, 3250 individuals (minimum population size with zero PE) would require an area of about 364.35 km2. Therefore, the source area size required to conserve the species over 50 years (with conditions established in the baseline) with a population of 7136 individuals (carrying capacity) is 800 km2 and for 3250 individuals (minimum population size with zero PE) is 364.35 km2. The pairwise obtained in PVA (initial population transformed in population density vs harvested individual) allow us to obtain a formula to calculate the maximum harvested individual with zero probability of extinction in 50 years through population density data (Figure below). A regression analysis showed that geometric regression was the best fitted (R2= 98.90, df= 3, P= 0.0005) and the equation is the following: Y = aX b Where: Y= Maximum harvested individual with zero probability of extinction a = 0.187 b = 0.8699 X= population density (initial population size measured in km2) 25
Hunting pressure (ind./km2) with
zero probability of extinction in 50
years
1.20
1.00
0.80
0.60
0.40
0.20
0.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
Population density (ind./km2)
Figure. Relationship between hunting pressure and population density data obtained in PVA with zero probability of extinction in 50 years. Sensitivity analysis shows that many variables negatively or positively affect the growth rate in the population. The variables that increase the growth rate in the population are: 30 or 40% sex ratio at birth in males, 55% females in breeding and 18 years as the maximum age of reproduction. The variables that affect growth rate negatively are: 10 years of harvest, 1 catastrophe, 60% sex ratio at birth in males, 8 years as the maximum age of reproduction, 30% females in breeding, density dependent reproduction and 70% sex ratio at birth as males. Of these variables, sex ratio at birth in males affects the growth rate of the population the most since an increase in the male‐biased sex ratio decreases the growth rate while a female‐biased sex ratio at birth produces the inverse effect. There are also variables with neutral effect in the growth rate such as percentage of males in breeding pool (20 ‐ 50%) which is close to the baseline with 10% of males in the breeding pool. 26
r (growth rate)
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0
-0.02
70 sex ration at birht in males
Density dependece
30% of females in breeding
8 years of maximum age of
reproduction
60 sex ration at birht in males
1 catastrophe
10 years of harvest each year
50% males in breeding pool
20% of males in breeding pool
Baseline
non harvest
55% females in breding
18 years of maximum age of
reproduction
40 sex ratio at birht in males
30 sex ration at birht in males
-0.04
-0.06
variable
Sensitivity analysis showing variables that affect growth rate positively (square above of horizontal line), negatively (below of the line) and have no effect (on the line). Studies on the acceptable leeway of hunting species vulnerable to overhunting A regression between female density and number of individuals harvested was developed as a means of determining the harvest limits of wildlife species. This regression can be used to determine the harvest limits of vulnerable species if they are occasionally hunted. This will allow for a leeway of hunting of vulnerable species, and would not jeopardize community based wildlife management if vulnerable species are occasionally taken. The regression between female density and number of individuals harvested was calculated for collared peccary, white‐lipped peccary, red brocket, lowland tapir and woolly monkeys. Collared peccary Using the harvest model to calculate the maximum sustainable harvest quota (SMHQ) indicated a maximum quota for collared peccary of 403 ± 75 ind/160 km2/year. The SD of the quota reflects the variation in the population density values. The mean quota of 403 individuals harvested within an area of 160 km2 translates into 2.53 ± 0.45 ind./km2/year or 31.39% ± 1.83 of production. The variation in the density estimates means that the sustainable yield is below the theoretical maximum sustainable yield of 40%. Variables Gross productivity Gestation per year Density (ind./km2) Hunting area (km2) Theoretical Sustainable Yield % 1999 1.02 1.88 6.5829 ± 1.2962 160 2005 1.02 1.88 10.268 ± 2.7282 160 2007 1.02 1.88 9.9388 ± 3.2937 160 40 40 40 27
Mean Maximum harvested individuals (harvested ind./160km2/ year) Maximum harvested individual per km2 Sustainable yield threshold (%) 327 470 417 2.0437 2.9375 2.6062 33.25 ± 6.67 31.34 ± 8.63 29.59 ± 10.37 404.7± 72.3 2.53± 0.45 31.39± 1.83 Variables used to obtain the sustainable maximum number of harvested individuals and sustainable yield in collared peccary A geometric regression between female density and number of individuals harvested provided the best fit (R2= 70.05%; P = 0.0377; GL = 4) and is described by the following equation: Variables Equation X Y density 3.29 3.94 5.13 6.50 4.97 6.62 Sustainably harvested individuals 327 327 470 470 417 417 Y = aX b ; ( R2= 70.05%; P = 0.0377; GL = 4) Y = Maximum harvested individuals a = 179.4342 X = female density b = 0.5033 In future, if the population density of collared peccary is known, the maximum harvest quota (maximum harvested individuals) can be calculated from the regression equation. White‐lipped peccary The harvest model indicated a maximum quota for white‐lipped peccary of 147 ± 108.25 ind./km2/year. The SD of the quota reflects the variation in the population density values. The mean quota of 147 individuals harvested within 160 km2 translates into 0.92 ± 0.68 ind./km2/year or 29.91% ± 1.16 of production. The variation in the density estimates means that the sustainable yield is below the theoretical maximum sustainable yield of 40% Variables Gross productivity Gestation per year Density (ind./km2) 2
Hunting area (km ) Sustainable Yield % Maximum harvested individuals (harvested ind./160km2/ year) Maximum harvested individual per km2 Sustainable yield threshold (%) 1999 0.51 1.69 15.174 ± 5.5734 160 40 2005 0.51 1.69 3.9607 160 40 2007 0.51 1.69 4.06 160 40 272 84 85 1.7 0.525 0.5313 28.5894± 11.3663 30.7636 30.3631 Mean 147± 108.25 0.92± 0.68 29.91± 1.16 Variables used to obtain the sustainable maximum number of harvested individuals and sustainable yield in white‐lipped peccary A geometric regression between female density and number of individuals harvested provided the best fit (R2 = 97.83; P= 0.0109; GL= 2) and is described by the following 28
equation: Variables Equation X Y density 7.587 10.37371 1.98 2.03 Sustainably harvested individuals 272 272 84 85 Y = aX b ; (R2 = 97.83; P= 0.0109; GL= 2 Y = Maximum harvested individuals a = 50.1257 X = female density b = 0.769 In future, if the population density of white‐lipped peccary is known, the harvest quota can be calculated from the regression equation. Red brocket deer The harvest model indicated a maximum quota for red brocket deer of 30.67 ± 19.66 ind./km2/year. The SD of the quota reflects the variation in the population density values. The mean quota of 30.67 individuals harvested in 160 km2 translates into 0.19 ± 0.12 ind./km2/year or 31.94% ± 1.86 of production. The variation in the density estimates means that the sustainable yield is below the theoretical maximum sustainable yield of 40%. Variables Gross productivity Gestation per year Density (ind./km2) Hunting area (km2) Sustainable Yield % Maximum harvested individuals (harvested ind./160km2/ year) Maximum harvested individual per km2 Sustainable yield threshold (%) 1999 0.44 2 0.7266 ± 0.1534 160 40 2005 0.44 2 2.2895 ± 0.4135 160 40 2007 0.44 2 1.1693 ± 0.3640 160 40 16 53 23 0.1 0.3312 0.1437 32.25 ± 6.96 33.62 ± 6.17 29.94 ± 9.79 Mean 30.67± 19.66 0.19± 0.12 31.94± 1.86 Variables used to obtain the sustainable maximum number of harvested individuals and sustainable yield in red brocket deer An exponential regression between female density and number of individuals harvested provided the best fit (R2= 94.71; P= 0.0002; GL= 5). The equation is as following the next table: Variables X Y density 0.29 0.36 0.44 1.14 1.35 0.58 0.77 Sustainably harvested individuals 16 16 16 53 53 23 23 Equation Y = ae bX ; ( R2= 94.71%; P = 0.0002; GL = 5) Y = Maximum harvested individuals a = 10.065 X = female density b = 1.2893 e = 2.7183 29
In future, if the population density of red brocket deer is known, the harvest quota can be calculated from the regression equation. Lowland tapir The harvest model indicated a maximum quota for lowland tapir of 0.53 ± 0.25 ind./km2/year. The SD of the quota reflects the variation in the population density values. The mean quota of 0.53 individuals harvested within 160 km2 translates into 0.0033 ± 0.0016 ind./hunted/km2/year or 16.06 ± 1.88 of production. The variation in the density estimates means that the sustainable yield is below the theoretical maximum sustainable yield of 20%. Variables Gross productivity Gestation per year Density (ind./km2) Hunting area (km2) Sustainable Yield % Maximum harvested individuals (harvested ind./160km2/ year) Maximum harvested individual per km2 Sustainable yield threshold (%) 1999 0.54 0.5 0.2719 160 20 2005 0.54 0.5 0.1493 160 20 2007 0.54 0.5 0.0765 160 20 0.8 0.5 0.3 0.005 0.0031 0.0018 14.53 ± 4.58 15.5044 18.1663 Mean 0.53± 0.25 0.0033± 0.0016 16.06± 1.88 Variables used to obtain the sustainable maximum number of harvested individuals and sustainable yield in lowland tapir A geometric regression between female density and number of individuals harvested provided the best fit (R2 = 97.33; P= 0.0134; GL= 2) and the equation is: Variables X Y density 0.13593 0.1768 0.0747 0.0382 Sustainably harvested individuals 0.8 0.8 0.5 0.3 Equation Y = aX b ; (R2 = 97.33; P= 0.0134; GL= 2 Y = Maximum harvested individuals a = 2.821 X = female density b = 0.6771 In future, if the population density of lowland tapir is known, the harvest quota can be calculated from the regression equation. Woolly Monkey The harvest model indicated a maximum quota for woolly monkey of 33.33 ± 2.52 ind./km2/year. The SD of the quota reflects the variation in the population density values. The mean quota of 33.33 individuals harvested within 160 km2 translates into 0.23 ± 0.21 ind./km2/year or 16.04 ± 0.65 of production. The variation in the density estimates means that the sustainable yield is below the theoretical maximum sustainable yield of 20%. 30
Species and variables litter size number of gestations per year adult female proportion actively reproductive adult female proportion Density (ind./km2) Hunting area (km2) Sustainable Yield % Maximum harvested individuals (harvested ind./160km2/ year) Maximum harvested individual per km2 1999 1 0.5 0.3 2005 1 0.5 0.3 2007 1 0.5 0.3 0.3 27.95 ± 4.67 160 20 0.3 27.87 ± 6.22 160 20 0.3 34.05± 8.99 160 20 33 31 36 0.20625 0.19375 0.225 Sustainable yield threshold (%) 16.71 ± 2.83 15.99 ± 3. 66 15.42 ± 4.21 Mean 33.33± 2.52 0.23±0.21 16.04± 0.65 Variables used to obtain the sustainable maximum number of harvested individuals and sustainable yield in woolly monkey A geometric regression between female density and number of individuals harvested provided the best fit (R2 = 73.22; P= 0.0297; GL= 4) and is described by the following equation: Variables Equation X Y density 27.955 32.623 27.873 34.047 43.032 21.655 Sustainably harvested individuals 33 33 31 36 36 31 Y = aX b ; (R2 = 73.22; P= 0.0297; GL= 2 Y = Maximum harvested individuals a = 14.2434 X = female density b = 0.2482 In future, if the population density of woolly monkey is known, the harvest quota can be calculated from the regression equation. Harvest quotas were calculated for all species. Harvest quota per
km2
ind/km2/year
2.53 ± 0.45
MSY limit
%
Equation
T. tajacu
Harvest quota in
160 km2
ind/160 km2/year
403 ± 75
31.34 - 33.25
Y = aX b
T. pecari
147 ± 108.25
0.92 ± 0.68
28.59 - 30.36
M. americana
30.67 ± 19.66
0.19 ± 0.12
29.94 - 33.62
T. terrestris
0.53 ± 0.25
0.0033 ± 0.0016
14.53 - 18.17
L. lagothrica
33.33 ± 2.52
0.23 ± 0.21
15.42 - 16.71
Y
Y
Y
Y
Species
= aX b
= ae bX
= aX b
= aX b
Harvest quota, MSY limit and regression equation of species for the Yavari Miri 31
River dolphins were used as indicator species for the aquatic systems by
incorporating long-term monitoring of the dolphin populations.
Samiria
The pink river dolphin (Inia geoffrensis) and the grey river dolphin (Sotalia fluviatilis) were used as indicator species for the aquatic hydroscape. These species are appropriate as indicator species because 1) they are top predators of the hydroscape, 2) they are not intentially killed by people due to strong taboos, 3) they can move in and out of river systems over short periods of time, and 4) they are easy to count and observe. The dolphin’s ability to move widely means that changes in dolphin populations within a river system will be caused more by dolphins leaving an area or immigrating into an area rather than a result of mortality or reproduction. Thus, if a hyrdoscape is going through negative changes, such as pollution or overfishing, the dolphin numbers in the system will be observed rapidly. Likewise, if a river system becomes healthier than surrounding hydroscopes dolphin numbers would increase from dolphins moving into the area. Dolphin densities in the Samiria River basin are some of the greatest recorded throughout Amazonia, with densities often well above 100 ind/km2. Overall, pink river dolphins in the Samiria River are greater than the grey river dolphin. Between 2004 and 2008 there have been variations in the density of both species, with the greatest densities occurring in 2008. In the Samiria River, pink river dolphins have greater densities at the mouth section and lower densities in the mid‐river and up‐river sections. There is a strong correlation between the abundance of commercial fish, measured as CPUE, and pink river dolphins. This indicates that pink river dolphins focus their feeding on similar sized fish as local fishermen. The mouth of the Samiria River has a concentration of commercial fish, especially during the low water periods when fish migrate out of the Samiria River and into the main channel of the larger Maranon River. In contrast, grey river dolphins have generally greater densities in the mid‐river and up‐river sections and lower densities at the mouth. Grey river dolphins do not show a correlation between commercial fish abundance. Grey river dolphins appear to feed on the smaller, less commercial fish species that are more abundant up river from the mouth of the Samiria River. There appeared to be an influx of dolphins into the Samiria River in 2008. This might be due to the increase in petroleum exploitation and increased fishing activities outside the reserve. The increase in oil prices and commodities led to greater resource extraction in areas outside the reserve. The conservation management inside the reserve might have led to a generally healthier hyroscape and an influx of dolphins in 2008, compared to areas outside the reserve. 32
Density of pink river dolphins in the Samiria river Densities of grey river dolphins in the Samiria River
Inia geoffrensis
Density (ind/Km2)
140
120
100
80
60
40
20
0
2004
2005
2006
Up River
Mid Section
2007
2008
2009
Mouth Section
Densities of pink river dolphins in the different sections of the Samiria River 33
Sotalia fluviatilis
Density (ind/km2)
60
50
40
30
20
10
0
2004
2005
2006
Up River
2007
Mid Section
2008
2009
Mouth Section
Densities of grey river dolphins in different sections of the Samiria River Samiria River's Mouth
Density ind/km2
140
120
100
80
60
40
20
0
2005
2006
Inia geoffrensis
2007
2008
Sotalia fluviatilis
Densities of river dolphins at the mouth of the Samiria River Correlation between river dolphins and commercial fish abundance, measured as CPUE in different sections of the Samiria River 34
Lago Preto The dolphin densities at the Lago Preto Conservation concession have also shown a upward trend since 2005 with both the pink and the grey river dolphins increasing in density. The increasing density of the dolphins is correlated with an increase in the commercial fish population, measured as CPUE. The Lago Preto Conservation Concession has implemented a fisheries management plan for the lakes within the hydroscape. These lakes have had an increase in the abundance of fish stocks, presumably due to setting up management actions. 12
Density (Ind/Km)
10
8
6
4
2
0
2004
2005
2006
2007
2008
2009
Density of pink dolphin at Lago Preto (ind/10km2) 30
Density (Ind/Km2)
25
20
15
10
5
0
2004
2005
2006
2007
2008
2009
Density of gray dolphin at Lago Preto 35
Fish populations at the Lago Preto Conservation Concession measured as CPUE. The pink dolphins (botos) and grey river dolphins (tucuxi) use the hydroscape at Lago Preto differently. Botos were shown to use the lake habitats more frequently, whereas the tucuxi were more common in the river systems. Total Observations for both Dolphin Species
Frequency
40
35
Boto
30
Tucuxi
25
20
15
10
5
0
River
Tapishca
Ipiranga
Abundance of pink and grey river dolphins in the river and lakes of Tapishca and Ipiranga at Lago Preto. The pink river dolphins have adaptations to the flooded forests of the Amazon basin and are a more ancient species than the grey dolphins. Indeed, the botos appear to be using the lakes as nurseries. A greater frequency of young botos was found in the lakes of Tapishca and Ipiranga at Lago Preto than in the main channel of the Yavari River. Preliminary results suggest that the botos use the lakes to teach their young how to fish in the more isolated and calm waters of the lakes, were they are better adapted to the interface between forests and water bodies. 36
Boto: Number of Observations in each Area
Frequency
35
30
Adult
25
Young
20
15
10
5
0
River
Tapishca
Ipiranga
Abundance of adult and young pink river dolphins in the hydroscape of Lago Preto. In contrast, the grey river dolphin is a more recent species to the Amazon basin and is better adapted to the larger and faster flowing rivers. The grey river dolphin does not use the lakes as nurseries, but raises their young in the main river channels. At Lago Preto a greater frequency of young grey dolphins were observed in the main river channels, with very few young observed in the lakes. Tucuxi: Number of Observations in each Area
35
Frequency
30
Adult
25
Young
20
15
10
5
0
River
Tapishca
Ipiranga
Abundance of adult and young grey river dolphins in the hydroscape of Lago Preto. Macaws were used as general indicators of the terrestrial systems by
evaluating changes in species numbers and composition. Samiria The macaw populations were used as indicator species of the terrestrial forest landscape. However, it appears that competitive interactions with other frugivore assemblages, especially the primates, might be influencing macaw abundance. Results in both the Samiria River basin and the Lago Preto Conservation Concession show that macaw populations decrease when primate populations increase. 37
In the Samiria River basin the macaw populations were compared between the moderately hunted zone (caza moderada) and the lightly hunted and non hunted zones (caza ligera). The moderately hunted zone had substantially greater abundances of macaws than the slightly hunted zone. In 2008, the dominate species in the moderately hunted zone was Ara ararauna. Howver, there is considerable variation between years, and in 2007 Orthopsittaca manilata was dominant in the moderately hunted zone. Macaws are very mobile species and these variations might reflect movement patterns. In the lightly hunted zone, Orthopsittaca manilata was dominant in 2008. Similar to the moderately hunted zone there is considerable variation between years. Previous to 2007, Ara ararauna was consistently dominant in the lightly hunted zone. Overall macaw abundance in the Samiria River between the moderately hunted zone (caza moderada) and the lightly hunted zone (caza ligera), data are from 2008. Macaw abundance by species in the Samiria River between the moderately hunted zone (caza moderada) and the lightly hunted zone (caza ligera), data are from 2008. 38
Macaw abundance in the moderately hunted zone of the Samiria River basin. Macaw abundance in the moderately hunted zone of the Samiria River basin. Relationship between mammal biomass and macaw abundance in the Samiria River basin. 39
Lago Preto The macaw population at Lago Preto has shown a cyclical pattern with the overall population peaking in 2007. Ara ararauna was the species that has the overall greatest abundance and showed the greatest increase in 2007. Ara chloroptera, Ara severa and Orthopsittaca manilata also show similar trends in their abundance at Lago Preto. Ara macao generally shows a more stable population trend at the concession. Overall Macaw abundance at Lago Preto per 10 points. 40
Yearly trends in abundance of macaws at Lago Preto Primate and terrestrial mammal populations were monitored to determine the
impact of hunting on arboreal mammals and to study ecological interactions
between arboreal and terrestrial wildlife assemblages.
Samiria The terrestrial mammals of the Samiria River have shown healthy populations during the past three years. Overall, the heavily hunted zone has shown the greatest increase, measured as biomass (kg/km2), which was largely due to increased sightings of white‐lipped peccary. The lightly hunted and moderately hunted zones have shown some variation, and overall have shown relatively constant mammalian biomass. Cumulative biomass of mammals from 2006‐2008 in the Samiria River Overall Biomass of mammals from 2006‐2008 in the lightly hunted zone 41
Overall Biomass of mammals from 2006‐2008 in the moderately hunted zone Overall Biomass of mammals from 2006‐2008 in the heavily hunted zone The primates have generally maintained their population levels in the Samiria River. The densities in 2008 were slightly lower overall than 2007, except for the woolly monkey which showed an increase for 2008. The greatest decline was in the populations of saddled‐back tamarin, Saguinus fuscicollis, which is a small‐bodied primate not used by hunters. It is likely that competition from the larger primates is driving down the populations of the smaller tamarins. 42
Density of Saimiri boliviensis Density of Cebus apella Density of Saguinus fuscicollis Density of Pithecia monachus Density of Alouatta seniculus Density of Lagothrix poeppigii The densities of the terrestrial mammals showed considerably greater variation than the arboreal primates. White‐lipped peccary(Tayassu pecari), , lowland tapir (Tapirus terrestris), and coati (Nasua nasua ) showed the greatest increases during the three year period. In contrast, red brocket deer (Mazama Americana), collared peccary (Tayassu tajacu)and sloth (Bradypus variegates ) showed the largest decreases. The agouti (Dasyprocta fuliginosa) and tamandua (Tamandua tetradactyla ) showed relatively stable populations. 43
2.5
Density (ind./km2)
2
1.5
1
0.5
0
2005
2006
2007
Years
Density of Tayassu pecari Density of Tayassu tajacu Density of Tapirus terrestris Density of Mazama americana Density of Dasyprocta fuliginosa Density of Nasua nasua Density of Bradypus variegates Density of Tamandua tetradactyla 44
2008
Densities of arboreal and terrestrial mammals were compared between the lightly hunted (caza ligera), moderately hunted (caza moderada) and heavily hunted (caza persistente) sites of the Samiria River basin. While sampling variance may account for some of the fluctuations, the overall trends are an important indication of management actions. The squirrel monkey (Saimiri boliviensis ) showed similar densities between the three hunting zones. This is the most abundant primate in the Samiria River basin. Density of Saimiri boliviensis by year from 2006‐2008 for the Samiria river The brown capuchin monkey also showed similar densities and population trends between the three hunting zones over the past three years. The lightly hunted site showed the greatest decline in 2008, which suggests that this was due more to natural fluctuations than by hunting. Density of Cebus apella by year from 2006‐2008 for the Samiria river The howler monkey showed considerable variation in sightings between the three hunting zones. Over the past two years the howler monkey was most abundant in the lightly hunted sites. It also showed a decrease in sightings between the moderately hunted sites between 2006 and 2007. 45
Density of Alouatta seniculus by year from 2006‐2008 for the Samiria river The saddled‐back tamarin has also shown considerable variation over the past three years. The heavily hunted site has shown the greatest decline. This might be due to competition from larger‐bodied primates as their populations increase. Tamarins are not hunted and it is unlikely that this decline was caused by hunting. Density of Saguinus fuscicollis by year from 2006‐2008 for all the Samiria River The woolly monkey has generally shown healthy populations in the lightly hunted and moderately hunted zones, and is still very uncommon in the heavily hunted sites. There was a dip in sightings in the lightly hunted site in 2007, but this might have been due to sampling variations. Density of Lagothrix poeppigii by year from 2006‐2008 for the Samiria river 46
The saki monkey had a greater number of sightings in the heavily hunted sites, compared to the lightly hunted and moderately hunted sites. This species is used for bush meat in the Peruvian Amazon and its relative abundance in the heavily hunted site is encouraging. The sightings in the lightly hunted and moderately hunted sites are similar. Density of Pithecia monachus by year from 2006‐2008 for the Samiria river Competition appears to be a major factor determining the primate communities along the Samiria river basin. Whilst squirrel monkeys are the most abundant species in all three sections of the Samiria basin, biomass reflects ecological dominance. Howler monkeys and capuchin monkeys have similar biomasses in the mouth region, whereas howler monkeys dominate the mid‐section and woolly monkeys (Lagothrix lagotricha) dominate up river (Fig. 2). Using censuses it is clear that the howler monkeys and woolly monkeys are in a Lokta‐
Volterra competitive interaction, with woolly monkeys winning up river and howler monkeys winning in the mid section (r= 0.98, p<0.001)(Fig. 3). The total K of both species appears to be around 30 ind/km2. Other species also appear to have Lokta‐Volterra interactions, but sample sizes do not currently show significant relationships. These results have important implications for conservation strategies and management policies. The conservation actions of reducing hunting clearly are helping to increase the populations of large bodied primates, but at the expense of small bodied species. Also, the dominance of one species will impact the population of other species. For example, if woolly monkeys or howler monkeys begin to dominate an area of the Samiria basin, this will decrease the population of the competing species. 47
Lotka‐Volterra relationship between the woolly monkey and howler monkey in the Samiria river basin. Sightings of the white‐lipped peccary have increased substantially in the lightly hunted site and remained at relatively low levels in the moderately hunted and lightly hunted sites. The white‐lipped peccary is one of the most important bush meat species and increased sightings in the lightly hunted site demonstrates the positive impact of wildlife management in the reserve. The white‐lipped peccary ranges over wide areas of forest and can move into the heavily hunted areas over short time periods. Density of Tayassu pecari by year from 2006‐2008 for the Samiria river The collared peccary is sighted consistently more in the lightly hunted sites. In the heavily hunted and moderately hunted sites the collared peccary has very few sightings. The collared peccary is more sedentary that the white‐lipped peccary and does not generally move over large distances, which would make its repopulation of the heavily hunted sites take long periods of time. 48
Density of Tayassu tajacu by year from 2001‐2008 for the Samiria river The red brocket deer has shown decreases in sightings in the lightly hunted and moderately hunted sites. Sightings in the heavily hunted site has been consistently low. It is unclear what might be causing these declines. Density of Mazama 49mericana by year from 2006‐2008 for the Samiria river The lowland tapir has shown increase in sightings in the lightly hunted sites, especially during 2008. This species has been generally low in the Samiria River basin due to overhunting in the past. Lowland tapir abundance in the moderately hunted and heavily hunted sites is still very low. 49
Density of Tapirus terrestris by year from 2006‐2008 for the Samiria river The agouti has shown considerable variation in sightings over the past three years between the sites. Overall, sightings in all three sites were most similar during 2008, with all three zones increasing from 2007. Density of Dasyprocta fuliginosa by year from 2006‐2008 for the Samiria river combined The coati has shown substantial increases in sightings in the lightly hunted and moderately hunted sites during 2008. The abundance of coati in the heavily hunted site has been consistently low. Density of Nasua nasua by year from 2006‐2008 for the Samiria river The tamandua also has shown considerable variance in its sightings between the three hunting zones. The greatest abundance during 2008 were in the moderately and lightly hunted zones, with a general decrease in the heavily hunted zone. This is not a common 50
bush meat species, so it is unlikely that this decrease was a result of hunting, but is more likely a natural fluctuation or a result of sampling variance. Density of Tamandua tetradactyla by year from 2006‐2008 for the Samiria river The populations of terrestrial mammals in the Samiria basin is confounded by both hunting and habitat types. The area around the mouth of the Samiria is flooded more intensively than the forests in the mid section, and up river is flooded the least. There is also greater hunting of terrestrial mammals closer to the mouth and less hunting up river. The terrestrial mammals show a clear relationship with distance from the mouth, and species have greater densities and biomasses further up river. For example, terrestrial mammals had a total density of 2.4 ind/km2 at the mouth, 4.2 ind/km2 at the mid section and 10.2 ind/km2 up river and their populations in an 100km2 area are greater in the up river sections of the basin (X2=595, p<0.001). Similar to previous surveys very few ungulates occur in the mouth region and have increasing populations further up the basin. The peccary biomass dominates the up river sections of the Samiria river. In contrast, the aboreal Amazon squirrels (Sciurus spp.) have a greater density at the mouth (8.5 ind/km2), compared to the mid and upper sections (2.2 and 2.5 ind/km2 respectively). Likewise, the arboreal sloth (Bradypus variegatus) has greater densities at the mouth (1.1 ind/km2) compared to the mid section (0.5 ind/km2). This species was not observed up river. These results suggest that the extensive flooding around the mouth of the Samiria river basin is limiting the populations of terrestrial mammals. During high water periods terrestrial mammals retreat to levees or floodplain islands to wait out the floods. Levees are more common up river and in turn these areas support greater populations of terrestrial mammals. 51
Densities of terrestrial mammals in different sections of the Samiria river basin. Lago Preto The red uakari monkey (Cacajao calvus) dominates the density of mammals at the Lago Preto Conservation Concession. In terms of densities, primates are by far the most common mammal assemblage at the concession, with other important species being the moustached and saddled back tamarins (Saguinus mystax and S. fuscicollis), the squirrel monkey (Saimiri sciureus), the woolly monkey (Lagothriz lagothricha), the brown capuchin monkey (Cebus apella) and the monk saki monkey (Pithecia monachus). 52
Density of the most abundant mammals of mammals at the Lago Preto Conservation Concession The biomass of mammals at the concession is also dominated by the red uakari monkey, followed by the squirrel monkey and the woolly monkey. The collared peccary, with its large body size is also important in terms of its biomass at the concession. Biomass and metabolic biomass of mammals at the Lago Preto Conservation Concession The red uakari monkeys are the flagship species of the Lago Preto Conservation Concession. The PI’s, R. Bodmer and P. Puertas, first explored the area in 1992 and observed healthy numbers of red uakari monkey. In subsequent years the PI’s realised that the Lago Preto area had a resident population of this rare primate, and unlike other areas the uakaris ranged year round in a predictable manner. Censuses conducted in other areas of the Peruvian Amazon do not have the high densities or predictable ranging of the uakaris found at Lago Preto. The PI’s realised that Lago Preto was a unique habitat for the red uakari monkey and began to set up official protection of the site. Preliminary research at Lago Preto was followed by more in depth studies, not only of the uakari monkeys, but also on a wide range of species and species assemblages. 53
Conservation actions at Lago Preto began in 2001, firstly by working with the nearby community of Carolina on community‐based conservation activities. The PI’s then set up permanent research activities with a more constant presence at the site. The PI’s also worked with WCS and DICE to obtain official protection of the area and in 2006 the Lago Preto Conservation Concession was decreed as a type of public‐private partnership. Prior to the conservation actions red uakari monkey were occasionally hunted for bush meat by local hunters. Similar to other primates, red uakari are vulnerable to overhunting and their populations decrease rapidly when harvested. Density surveys have been conducted on a regular basis at Lago Preto since 1999, when the red uakari population was at around 14 ind./km2. This was a particularly large population compared to other sites on the Yavari and Yavari Miri Rivers, which had populations between 0.63 – 4.9 individuals per km2. Density estimates of red uakari monkeys at different sites along the Yavari and Yavari Miri rivers in individuals per km2.
Location
Red Uakari Density
Upper Yavari
1.5
Upper Yavari Miri
0.6
Lower Yavari Miri
4.9
Lago Preto
88.0
Since conservation actions have been set up at the Lago Preto concession the red uakari population has increased steadily and dramatically. Censuses conducted in 2006 and 2007 have shown the greatest densities of red uakari at Lago Preto with numbers around 116 ind./km2. The current red uakari population at the Lago Preto conservation Concession is estimated at around 500 individuals, correcting for variance in densities between habitat types. Population growth curve for Cacajao calvus ucayalii in the Lago Preto Conservation Concession. The rate of increase in the red uakari population appears to be at or close to the maximum. Indeed, average annual increase of red uakari at Lago Preto spread over all years between 1999 and 2007 is approximately 20%. The absolute maximum rate of increase (rmax) of the red uakari is 33%, if all reproductively active females reproduced each year. The predicted normal rate of increase would be 17%, if all reproductively active females reproduced every 54
other year, and the current rate of increase falls somewhere between the absolute maximum and more normal rates. The other large bodied primates found at Lago Preto appear to be impacted by the high numbers red uakari population. Competitive interactions between large bodied primates have been shown in the research expeditions to the Samiria River. Likewise, competition and lotka‐volterra interactions also seem to be occurring at Lago Preto. The woolly monkey (Lagothrix lagotricha) and howler monkey (Alouatta seniculus) populations at Lago Preto have fluctuated in Lotka‐Volterra fashion, with increase in woolly monkeys resulting in decrease in howler monkeys and visa versa (Figure 3). However, when the red uakari reached its peak population in 2007 both woolly monkeys and howler monkeys fell to low densities. Woolly monkey numbers at Lago Preto are lower than other sites in the Yavari valley, which may reflect competition with the large population of red uakari monkeys. Woolly monkeys in other regions of the Yavari and Yavari Miri are some of the greatest reported in the Amazon and in these sites red uakari populations are very small. Population trends of woolly monkey and howler monkey at Lago Preto. Density estimates of woolly monkeys at different sites along the Yavari and Yavari Miri rivers in individuals per km2. Location
Woolly Monkey Density
Upper Yavari
32.7
Upper Yavari Miri
24.5
Lower Yavari Miri
27.6
Lago Preto
1.5
The medium and small bodied primates at Lago Preto show similar population numbers as other areas of the Yavari valley, with squirrel monkey populations dominating the assemblage. The tamarins have shown an increase in their population at the concession over the past decade. These small bodied primates are not hunted by local people, so the increase in their population is a result of natural events. 55
Population trends of tamarins at Lago Preto. The White‐lipped peccary has shown a general decrease in its population at the concession. This follows the general trend in White‐lipped peccary populations in the Yavari region. Indeed, white‐lipped peccaries were very common in the Yavari until 2003, when their populations crashed throughout the entire region. This appears to be caused by a natural fluctuation in the white‐lipped peccary populations, since un hunted areas of the region also showed population crashes of this species. The collared peccary has shown a relatively constant population in the concession. Population trends of white‐lipped and collared peccary at Lago Preto. Game bird populations were monitored to determine their conservation status
and impact of hunting.
Samiria
Game birds become increasingly more common from the mouth of the Samria to the up river sections. Game birds are generally rare at the mouth, with only the Tinamus spp. sighted along the transects. In the mid‐section the Tinamus spp. became even more abundant and the Pipile cumanensis were also sighted. The up river section has the greatest density and species richness for the game birds. The game birds have not recovered to the same extent as the primates, and are still at relatively low numbers, especially around the mouth of the basin. 56
Density of game birds in different sections of the Samiria river basin. Density of Tinamus sp . in the Samiria River Density of Crypturellus sp. In the Samiria River Density of Pipile cumanensis in the Samiria River Density of Mitu tuberosum in the Samiria river Yearly fluctuations in the sightings of game birds in the Samiria River basin 57
Lago Preto The game birds at Lago Preto are dominated by Penelope jaquacu and Psophia leucopterra. There has been considerable variation in the sightings of game birds at the concession. The spike in Tinamus sp. n 2007 appears to be due sampling, and further analysis of the data will be necessary before drawing conclusions. Otherwise, the populations have been relatively stable over the years. Density and biomass of game birds at Lago Preto Yearly fluctuations in game bird sightings at Lago Preto Caiman populations were monitored to evaluate the recovery of black caiman
populations and the ecological interactions between species.
Samiria Three species of caimans occur in the Pacaya‐Samiria National Reserve, the black caiman (Caiman niger), the common caiman (Caiman crocodylus) and the dwarf caiman (Paleosuchus trigonatus). The black caiman was intensively overhunted during the 1950’s – 58
1970’s and has been recovering in the Samiria river over the past decades. The abundance of caimans varies along the Samiria river basin. All three species were rare around the mouth of the Samiria river (0.4 ind/km) and were more common in the intermediate zone between the mouth and the mid‐section (1.4 ind/km). Caimans were most abundant at the mid section (2.7 ind/km) and the beginning of the up river section (2.2 ind/km), and then decreased at the upper reaches of the Samiria (1.4 ind/km)(X2=18, p<0.001). 1.8
1.6
Abundance (ind/km)
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Mouth
Mouth/Mid-section
Caiman niger
Mid-section
Caiman crocodylus
Near Up River
Far Up River
Paleosuchus trigonatus
The abundance of caimans in different sections of the Samiria river. The population of black caiman increased slowly between 1981 to 1996 with an er of 1.02 and an r of 0.02. Between 1996 and 2004 the population increased slightly faster with an er of 1.08 and an r of 0.08. The black caiman population has increased considerably faster since 2004 er of 1.64 and an r of 0.49. The black and common caiman appear to be in a Lotka‐Voltera competitive interaction, with the black caiman winning over the common caiman. Indeed, the common caiman was very abundant in 1981 with over 8 ind/km, but has been decreasing as the black caiman populations have increased (r=.55, p=0.03). Therefore, the conservation actions to recover the black caiman populations have led to a decrease in common caiman numbers over the years. 59
Population trends of black caiman in the Samiria River 9
Abundance (ind/km)
8
7
6
5
4
3
2
1
0
1981
2004
2005
2006
2007
2008
Caiman crocodilus
Population trends of common caiman in the Samiria River Lotka‐Volterra relationship between black and common caiman in the Samiria River basin. 60
Abundancia (Ind. / Km)
The age structure of the common and black caiman differs considerably in the Samiria River basin. The common caiman has a greater proportion of the juvenile age class, whilst the black caiman has a greater proportion of adults. The change in the populations of common and black caiman may be related to these differences in demography. The competitive interactions of the black caiman appear to be decreasing the adult common caimans, and decreasing the number of juvenile common caiman surviving to the adult stage. 1.40
1.20
0.95
1.00
0.80
0.60
0.40
0.41
0.40
0.20
0.00
Cría ≤ 50
Juvenil < 120
Adulto ≥ 120
Age structure of common caiman Caiman crocodilus in the moderately hunted zone of the Samiria River. Abundancia (Indiv. / km.)
3.00
2.50
2.00
1.28
1.50
1.00
0.50
0.57
0.13
0.00
Cría ≤ 70
Juvenil < 200
Adulto ≥ 200
Age structure of black caiman Melanosuchus niger in the moderately hunted zone of the Samiria River. Lago Preto The common caiman populations at Lago Preto between 2001 to 2008 showed a general decrease in their populations (Kruskal‐Wallis, H= 10.777; gl= 4; P= 0.029). This may be in response to an increasing black caiman population, which would follow the same trends as in the Samiria River basin. 61
Abundance (Ind/Km)
7
6
5
Figure 11. Populations trends of common caiman at Lago Preto
4
3
2
1
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Abundance of black caiman at the Lago Preto Concession The black caiman populations showed a slight increasing trend at Lago Preto between 2001
and 2008.
0.8
Abundancia (Ind/Km)
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Abundance of black caiman at Lago Preto The smooth fronted caiman population also showed an increasing trend at Lago Preto in 2008. The smooth fronted caiman is not hunted by people and its population fluctuations is probably linked to ecological interactions. 62
Abundancia (Ind/Km)
0.5
0.4
0.3
0.2
0.1
0
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
Abundance of smooth fronted caiman at Lago Preto River turtle populations were monitored to determine the success of the headstarting conservation programme in the Samiria River basin
Samiria The Pacaya‐Samiria National Reserve has implemented a turtle conservation program based on head starting for a number of years, where eggs of Yellow‐spotted Amazon River Turtle (Podocnemis unifilis) and Giant Amazonian River Turtle (Podocnemis expansa) have been removed from their wild nests, replanted at guard stations, hatched, and released back into the river. This conservation strategy has been set up to overcome the intensive poaching of turtle eggs during the laying season. River turtle numbers are recovering along the Samiria river basin as a result of the head starting program. The abundance of Podocnemis unifilis was determined by turtles sunning along the riverbanks, whereas Podocnemis expansa does not sun along the banks and was not censused. Very few turtles were sighted around the mouth of the Samiria river basin. The density of Podocnemis unifilis ind/km was greatest at the Tascha Cocha and Ungurahui zones, which are in the mid section of the Samiria River. Wishto had the lowest densities of river turtles. This difference appears to be related to park guard vigilance, since both the Tascha and Unguurahui sites have permanent park guard stations, which are not present at the Wishto site. During turtle egg collection at the Wishto site there were numerous eggs poachers observed in the area. The numbers of turtle sightings has increased substantially since 2006 and clearly shows the success of the turtle programme in the Samiria River. The expectation is that it takes between 8 to 10 years for turtle hatchlings to become adults and start sunning themselves. Since 1996 a total of around 865,864 turtle hatchlings have been released in the head starting program, with an annual average of 73,507 Podocnemis unifili and 16,348 Podocnemis expansa. The hatchling numbers are in accord with the increasing turtles sunning along the Samiria River. 63
Abundance of Podocnemis unifilis in the Samiria River. Density of Podocnemis unifilis in the Samiria River. Beaches used by turtles along the mid section of the Samiria River were identified and the number of eggs collected at the beaches were recorded. This information will be used to set up vigilance programmes for the head starting initiative and to evaluate fluctuations in turtle nests and egg numbers between years, especially as the turtle populations grow. 64
Map of the turtle egg laying beaches at the Tascha site. Map of turtle egg laying beaches at the Wishto and Ungurahui sites. 65
Percent of eggs collected from the different turtle egg laying beaches The number of hatchling turtles resealed during the head starting program. 66
Giant river otter populations were monitored to evaluate the recovery of this
rare species.
Samiria The giant otter is endemic to South America and has shown a marked decline due to excessive pelt hunting during the 1940’s to 1970’s with many pollutions becoming extirpated. By the end of the 1970s, the giant otter was nearly extinct. Studies of giant otter are a high priority for the IUCN and long‐term conservation efforts for this critically endangered flagship species are needed. Today, the giant otter is beginning to show a slow recovery in population size in many areas of its former range in the Amazon, including the Samiria River basin and the Lago Preto Conservation Concession. The sightings of giant river otter have increased substantially in the Samiria River basin in 2008. Abundance of giant river otter along the Samiria River. Lago Preto At Lago Preto research is being conducted to monitor giant otter populations at the different lakes. At the Buen Fin Lake a group of otters is being studied to determine resource characteristics needed for recovering populations. Behaviour, diet, macro habitat and micro habitat around the lake are being studied to help better understanding the resources needed to help assist the recovery of the giant otter and to bolster conservation and management efforts of the species. 67
Distribution of giant river otter groups at the Lago Preto Conservation Concession.
40
35
30
25
20
15
10
5
0
1992
1997
2001
2004
2005
2006
2007
2008
Population trend of giant river otter at the Lago Preto Concession
Between 2005 and 2007 there was a steady increase in the giant otters population on Buen Fin. This suggests that Buen Fin provides a good habitat for the giant otters. 68
The growth of the Buen Fin giant river otter group over four years. Throat
Name
Markings
Rico
Sex
Year of Birth
Male
?
Years Seen
on Buen Fin
2005
Mimo
Male
?
2006 & 2007
Isla
Female
?
2006 & 2007
Stitch
?
2006
2007
Sisi
?
2006
2007
Identification table for the individuals present on Buen Fin The diet of the otters at Buen Fin was dominated by the fish of the order Characiformes. Perciformes were the second most abundant order eaten with Osteoglossiformes and Siluriformes being least represented in the diet and also the least abundant in Buen Fin. 69
Diet
Abundance
Buen Fin
in
63.97
61
1.049
3.68
4
0.92
Perciformes
30.88
33
0.94
Siluriformes
1.47
2
0.74
Preference
Order
Characiformes
Osteoglossiformes
Diet and preference of fish eaten by Giant River Otter at the Buen Fin Lake. Amazon manatee populations were monitored in the Samiria river to evaluate
the conservation status of this vulnerable species.
The manatee (Trichechus inunguis ) population along the Samiria River has been relatively stable over the years. This species is still occasionally hunted by local people, often killed when mistaken for Piache fish. The Samiria River basin continues to be a stronghold for the species in Loreto, but further conservation efforts are required. 2.5
2
1.5
1
0.5
2008
2007
2006
2005
2004
0
2003
Average Observations 3
Abundance of manatees along the Samiria River 70
Large cats, including jaguars, pumas and ocelots were monitored using
camera traps to determine a baseline for long-term population studies in the
Lago Preto Concession.
Lago Preto From 870 trap nights, 69 animal photographs and 57 events of 12 mammals and three bird species were recorded, which represents respectively, 24% and 48% of the “potential” and “observed” terrestrial mammal species for the Yavari river valley. Three new mammal records can be added to the “observed” list: puma (Puma concolor), giant anteater (Priodontes maximus), and spiny rat (Proechymis sp.). The most frequently photographed mammal species were the lowland tapir (Tapirus terrestris) and the red brocket deer (Mazama americana), followed by puma and collared peccary. Three terrestrial‐arboreal bird species were also recorded, with the pale‐winged trumpeter (Pshophia leucoptera) being the most commonly photographed. All the mammal species recorded were terrestrial with the only exception of a squirrel (Sciurus sp.) catalogued as terrestrial‐arboreal. Four of the 15 species were non‐herbivores and eight were known to have nocturnal habits. Two mammal species were listed in the IUCN’s Red List as Vulnerable (Tapirus terrestris and Priodontes maximus), two as Near Threatened (Jaguar, Panthera onca and Puma concolor) and two as Data Deficient (Mazama sp.). Those species that were hunted by the local communities had a higher encounter rate than those not hunted (Mann‐Whitney U = 9.5; z = ‐2.08; P < 0.05). Encounter rates for individual species showed no relationship with mean body mass (Spearman: rs= 0.357, P>0.05, n=15), but a positive and significant relationship with mean species’ height (Spearman: rs=0.582, P<0.05, n=15). Comparing the results from this study with those from Tobler et al.(2008), we obtained on average, more pictures of red brocket deer (9%) puma (6%) and tapir (4%) and less of the pale‐winged trumpeter (13%), ocelot (5%) and jaguar (4%). The rest of the species in this study were encountered at similiar rates. 71
Species characteristics, number of captures and encounter rates (ER = number of photographs/1000 trap nights) from this study in Lago Preto and from Tobler et al. (2008) in
Southern Peruvian Amazon (ER-T).
B = mean body mass (kg); H = mean shoulder height (m); Habits = T, terrestrial; T/A, terrestrial‐arboreal; D, diurnal; D/N, diurnal/nocturnal; Nh, non‐herbivore; H, herbivore; Red List Status = VU, vulnerable; NT, near threatened; DD, Data deficient, LC, Least concern. The grey colour indicates new “observations” for the study area. Mean encounter rates (ER pictures/1000 trap nights), with SD bars, between hunted and non‐hunted species (n = 15; Mann‐Whitney U = 9.5; z = ‐2.08; P < 0.05). 72
Spearman’s rank correlation coefficient (rs) associations between mean encounter rate (ER, photographs/1000 trap nights, n = 15) and (a) mean body mass of species (rs = 0.357, P < 0.05) and (b) mean height (rs = 0.582, P < 0.05). Camera trap detection probabilities Investigating the covariates that best explained species detection probability, found five candidate models as being within two ΔAICc units of the top ranked model (Models 1.1‐1.5). From these models, the summed model weights for each covariate with respect to detection probability were explained by: hunted or not (74%), height/body mass (35%), nocturnal/diurnal (17%) and threat status (13%). Species that were hunted by the local communities had a higher detection probability. The remaining covariates also had positive relationships with detection probability, indicating that larger animals, nocturnal and threatened species are more likely to be detected by cameras traps. Camera trap detection probabilities explained by mammal species characteristics Notes: ψ is the occupancy parameter which remains as constant and p is the species detection probability parameter.K is the number of parameters in the model, ΔAICc is the difference in AICc values between each model with the low AICc model, and wi is the AICc model weight. Model beta coefficients are: B1= hunted; B2 = height‐body mass, HB; B3 = nocturnal; and, B4 = threatened). 73
Detection probability generated from the model, psi(.),p(hunted), for ‘hunted’ and ‘non‐
hunted’ species Overall, the camera trap method performed better at detecting species of conservation management concern, e.g. hunted species and threatened species, than the line transect method. For the Carnivora order, three of the five felids present in the area were recorded by camera traps. The South American coati and the tayra were, on the other hand, only recorded by the line transect technique. These species are medium sized and known to have terrestrial/arboreal habits. The puma, recorded by the camera trap survey, was another confirmed species and was recorded in five events at three different locations with a separation of 3km and 2.5km between station number 9 and stations 13 and 21, respectively. Although an individual identification was not definitive, these events could constitute at least two different individuals, one male (station number 9) and a female (stations 13 and 21). The puma, together with the Jaguar, are classified as near threatened species on the IUCN Red List. Although individual a jaguar photo‐identification was not possible, both records were in the same station suggesting that it could account for just one individual. The separation between the Jaguar (station 8) and the puma record were 3km, 3.3Km and 1.3km (stations 13, 21, 9 respectively). They were both recorded in a relatively small area and account for at least three individuals. Stations numbers 8 (Jaguar) and 13 are located in the Aguajal‐terra firme forest edge; station number 21 is the natural salt lick and station 9 is terra firme forest habitat. The third felid recorded was the ocelot, with a single event. For the these three felids previous population assessment by means of capture‐recapture methodology have proven to be successful. Camera traps, again constitute a potential survey technique for monitoring these predator species. These preliminary results constitute the first evidence of puma and jaguar as sympatric species occurring together in LPCC. The monitoring and the ecology study of these landscape species are important for LPCC. For management strategies, it is important to account for the status of predators which rely on game that undergo high hunting pressure by local communities. On the other hand, top predators, specially the jaguar are flagship species, that can be used to attract funding for wider conservation in an area or to establish a new protected area. The jaguar together with the black caiman (Melanosuchus niger) and the giant river otter (Pteronura brasiliensis) were heavily hunted in the Yavari until 1972 (Bodmer and Puertas 2003). For this reason, it is important to document their conservation 74
status and whether, or not, their populations have recovered from overhunting. The abundance, diversity and age structure of the large fish species were
monitored to determine the impact of local fisheries and the effectiveness of
fisheries management.
Samiria There were 56 species belonging to 14 captured during the surveys in the Samiria River basin. Most common species surveyed in the Samiria River basin Familia Nombre científico Nombre común Potamotrygonidae Potamotrygon motoro (Naterrer, 1841) Raya Osteoglossidae Osteoglossum bicirrhosum (Vandelli, 1829) Arahuana Curimatidae Potamorhina latior (Spix, 1829) Llambina Semaprochilodus amazonensis (Fowler, 1906) Yaraqui Prochilodontidae Prochilodus nigricans (Agassiz, 1829) Boquichico Anostomidae Schizodon fasciatum (Spix, 1929) Lisa Erytrhinidae Hoplias malabaricus (Bloch, 1794) Fasaco Cynodontidae Hydrolicus scomberoides (Cuvier, 1817) Chambira Characidae Triportheus elongatus (Spix, 1829) Sardina Piraña Captopryon sp. Piraña alargada Serrasalmus elongatus (Kner, 1860) Piraña Serrasalmus sp. Piraña roja Pygocentrus nattereri (Linnaeus, 1766) Piraña blanca Serrasalmus rhombeus (Linnaeus, 1766) Palometa Mylossoma duriventris (Cuvier, 1818) Serrasalmidae Palometa Mylossoma sp. Curuhuara Myleus rubripinnis (Müller y Troschel, 1845) Gamitana Colossoma macropomum (Cuvier 1818) Piaractus brachypomus (Cuvier 1818) Paco Doradidae Oxiydoras níger (Valenciennes, 1833) Turushuqui Pimelodidae Pseudoplatystoma fasciatum (Linaeus, 1766) Doncella Ancistrus sp. Carachama ancistrus Hypostomus emarginatus (Valenciennes, 1840) Carachama Carachama Liposarcus pardalis (Castelnau, 1855) Loricariidae Carachama Liposarcus sp. Shitari Loricariia sp. Acarahuazú Astronotus ocellatus (Agassiz, 1831) Tucunaré Cichla monoculus (Spix, 1829) Bujurqui Aequidens tetramerus (Heckel, 1840) Cichlidae Bujurqui azul Heros appendiculatus (Castelnau, 1855) Bujurqui vaso Chaetobranchus flavescens (Heckel, 1940) 75
The most common fish species in the Samiria River basin was Common Pleco (Liposarcus pardalis), followed by Red‐Bellied Piranha (Pygocentrus nattereri ) and the White Piranha (Serrasalmus elongatus). Likewise, the biomass of fish was dominated by the Common Pleco, followed by Red‐Bellied Piranha and the Lisa (Schizodon fasciatum). Percent of captures of fish in the Samiria River basin Biomass of fish in the Samiria River basin In the varzea of the Samaria River basin fish move into the flooded forests during the high water period from December to June. This is a very productive period for the fishery with abundant food in the forests, including insects, fruits and forest debris. During the low water period from July to November the fish move into the lakes, channels and main river. Competition is greater during low water and many species migrate out of the Samiria and into the larger Maranon River. This difference is reflected in the CPUE of high and low water periods. 76
Variation in the capture, biomass and CPUE between high and low water seasons in the Samiria River basin. High Water Low Water Number of Captures 128 1988 Time (hours) 26.15 109.30 Biomass (kg) 18.261 328.400 CPUE 4.89 18.18 The diversity of fish was greatest in the lower reaches of the Samiria River and decreased further up river. Tascha Cocha had the greatest diversity and is situated in the lower mid section of the river. Diversity at the mouth was was slightly less than Tascha Cocha, due to a greater dominance of several species. The diversity of fish decreased at the Wishto site in the upper mid section, and further decreased at the Pithecia site in the upriver section. Overall, the diversity between the main river, channels and lakes was similar, with the channels having the greatest diversity, followed by the lakes. The main river channel of the Samiria River had the lowest diversity. Diversity of fish in the Samiria River basin. Habitat River Sections Upriver‐Mid river‐Mouth Río Canal Lago Pithecia Wishto Tacshacocha Boca Diversity 2.5403 2.8874 2.6485 1.3168 1.6933
3.1443 2.2942
Index (H’) CPUE of biomass of fish along the Samiria River basin. 77
The frequency distribution of body lenth was used to compare commercial fish surveyed with standars developed by the Peruvian Ministry of Fisheries. The results showed that Astronotus ocellatus and Prochilodus nigricans in the Samiria basin had healthy age structures that correspond to a strong fishery. However, Colossoma macropomum showed a younger age structure than found in lightly fished areas. Further analysis is required, especially for Colossoma macropomum. Legal limits of size classes of three commercially important species in the Samiria River basin Length(cm) Species N Mín.
Máx. Media Astronotus ocellatus 60 Colossoma macropomum 130
Prochilodus 136
nigricans 13.7 27 20.3 15 42 28.5 10 36 23 Legal Reference 20 cm INDERENA, Colombia.1987 45 cm. R.M Nº 147‐2001‐
Pe 25 cm R.M Nº 147‐2001‐Pe Frequency distribution of body length of boquichico (Prochilodus nigricans). 78
Frequency distribution of body length of Astronotus ocellatus 46 - 51
41 - 46
36 - 41
31 - 36
26 - 31
21 - 26
16 - 21
90
80
70
60
50
40
30
20
10
0
11_16
Frecuencia de capturas
Frequency distribution of body length of Colossoma macropomum. Lago Preto Ninety four different species were found in total in Lago Preto, Ipiranga, Tipishca, Huapo and San Jorge over the three years of dry season sampling from 2005 to 2007. Of these, 64% were found in 2007, 59% in 2006 and 43% in 2005, thus implying a promising and steady increase in species richness in the area. There appeared to be no significant trend of higher species richness in black‐water rather than white‐water lakes, as was the case in other studies. There were, however, greater numbers of species belonging to the order Characiformes than other studies (49% of the total). The second most abundant species belonged to the order Siluriformes (19% of the total), which coincides with the Rapid Biological Inventory studies of lakes on the Yavarí River. Species richness of fish populations in the oxbow lakes of Lago Preto has steadily increased from 2005 to 2007. The most abundant species in 2007 was Potamorhina latior, followed by Serrasalmus rhombeus and Pellona castaeneana. 79
Number of fish species captured by order in the lakes of Lago Preto. Order
Species
Individuals
%
Rajiformes
Osteoglossiformes
Clupeiformes
Characiformes
Siluriformes
Perciformes
Soleidae
2
1
2
33
16
8
1
2
21
35
494
67
106
4
0.2
2.9
4.8
69.2
8.9
14.4
0.5
Total
63
729
100
Catch‐per–unit‐effort (CPUE) was used as a measure of fish population abundance. The CPUE showed that the abundance of fish is increasing in the managed lakes of Tipishca and Ipiranga where subsistence fishing is permitted by members of a local Brazilian fishing community. In contrast the CPUE showed a decrease in the fully protected lake of Lago Preto. CPUE of fish at three lakes of LP The paiche fish (Arapima gigas) is the largest freshwater fish on earth. It is also the most valuable commercial fish for fishermen and its populations can be decimated rapidly, because adults and juveniles surface every 15 minutes for air making it easy to harpoon. The paich is a key species for the Lago Preto Concession and its populations have been increasing in the protected lakes of Preto and Buen Fin. In the fished lakes of the concession the paiche population is considerably lower. 80
Density of Paiche in lakes at the Lago Preto Conservation Concession The project also worked with implementing and monitoring of communitybased conservation activities in indigenous communities. In these
communities the project worked with sustainable use and community-based
management of natural resources.
This project seeks to assess the potential of wildlife management as a tool for increasing the area that is under protection by 1) enhancing the sustainable use of wildlife and in turn conserving wildlife populations, 2) implementing wildlife habitat conservation and in turn conserving the entire array of biodiversity, and 3) setting non‐hunted source areas that are in fact a type of fully protected area which concurs with the cultural and socio‐economics of the rural people. The objectives are to 1) assist in setting up wildlife management programs in the sites, 2) test and refine the wildlife management model, 3) demonstrate the model’s usefulness as a conservation strategy, and 4) disseminate the wildlife management model throughout Amazonia. A study was conducted to assess the process for developing the implementation of the wildlife management model and the wildlife management guidelines in local communities of the Peruvian Amazon. The methods used in the study included consultation with communities of the Tahuayo – Blanco and Yavari ‐ Miri and examination of independent data on species population density and catch per unit effort. The study had the following objectives: 1. Determine community attitudes toward implementing the wildlife management guidelines 2. Determine the limitations and difficulties of setting up the guidelines in local communities 3. Determine the impact of wildlife management on wildlife numbers 4. Revise the wildlife management guidelines in accordance with the realities of local communities. All respondents acknowledged that their community had established wildlife agreements, but not all respondents were able to actually name aspects of the guidelines. The proportion 81
of respondents who acknowledged having wildlife agreements in their communities differed across the five communities (X2=30.466, df=4, P<0.001). % of Respondents
100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Yes
Diamante 7 de Nueva Esperanza
Julio
- Carolina
No
Percentage of respondents who acknowledged having wildlife agreements in their communities Although, the respondents were not able to name specific aspects of the certification guidelines such as species that are either vulnerable to or less vulnerable to over‐hunting, establishing of the hunting register either hunting quota, establishing source and sink areas and also management of habitat conservation, they were able to name some aspects of the wildlife agreements connected with the guidelines, such as whether or not the village had a hunting quota, had wildlife vigilance patrols etc. 100
Hunting quota
Primate hunting ban
Vigilance of wildlife
Source - Sink area
Tree felling ban
Do not know
90
% of Respondents
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Diamante 7
Nueva
de Julio
Esperanza Carolina
Local knowledge of wildlife agreements Three factors, age, gender and level of education were found to explain respondent knowledge of the wildlife agreements. Older respondents above 41 years old who were male and had secondary school education were more likely to know about the wildlife agreements. 82
Results of a logistic regression to identify which factors best explain local knowledge of wildlife agreements Variable B 1.082 2.642 0 ‐3.799 ‐6.860 Age (>41year) Gender (man) Education Secondary school education Constant *p < 0.05; ** p < 0.01 SE 0.444 1.288 0 1.384 2.525 df 1 1 3 1 1 Significance 0.015* 0.040* 0.056 0.006** 0.007** 60% (n=86) of interviewees correctly identified 5 species as resistant to overhunting including the two peccaries, collared peccary (Tayassu tajacu) and white lipped peccary (Tayassu pecari) and the rodents such as paca (Agouti paca), agouti (Dasyprocta fuliginosa) and capybara (Hydrochaerus hydrochaerus). Panthera onca (V)
43
59
Pteronura brasiliensis (V)
Priodontes maximus (V)
Cebus apella (V)
43
51
55
Cebus albifrons (V)
Lagothrix poepiggi (V)
51
Cacajao calvus (V)
47
Alouatta seniculus (V)
44
Ateles sp (V)
42
49
Tapirus terrestris (V)
Hydrochaerus hydrochaeris (NV)
Agouti paca (NV)
60
64
Dasyprocta fuliginosa (NV)
69
Mazama goauzoubira (NV)
50
Mazama americana (NV)
49
Tayassu tajacu (NV)
72
Tayassu pecari (NV)
73
0
20
40
60
80
100
120
% of Respondents
No vulnerable
Vulnerable
Percentage of respondents who identified particular species as vulnerable or not to overhunting in Tahuayo – Blanco and Yavari – Miri (Sample size is indicated in the bars) In group discussions communities indicated that they understood non‐vulnerable species to 83
be those with high annual reproductive rates and large population size. Communities recognized that species with low annual reproductive rate and low population size were identified as vulnerable to overhunting and identified eight species as vulnerable to overhunting: jaguar (Panthera onca), giant armadillo (Priodontes maximus), tapir (Tapirus terrestris), red uakari monkey (Cacajao calvus), spider monkey (Ateles paniscus), howler monkey (Alouatta seniculus) and both deer, red broket deer (Mazama americana) and gray deer (M. goauzoubira). The solitary behaviour of the both deer raised uncertainty in the communities about the vulnerability of these species to overhunting as was mentioned in the group discussions. In addition they mentioned that species such as the giant river otter, capuchin monkey and woolly monkey had increased in population size in recent years when they were not hunted. In the five communities, the majority of respondents were agreed about including a list of species that are either vulnerable to or less vulnerable to over‐harvest as a part of wildlife agreement. 100
90
80
% of Respondents
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Agree
Diamante 7 de
Julio
Nueva Esperanza
- Carolina
Disagree
Percentage of respondents who supported the inclusion of a list of vulnerable and non‐
vulnerable species as a part of the wildlife agreement % of Respondents
Overall, the proportion of respondents who were knowledgeable about hunting registers, ranged from around 60% to 100% of respondents, but differed across the five communities (X2=20.850, df=4, P<0.001). 100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Yes
Diamante 7 de
Julio
Nueva Esperanza Carolina
No
Proportion of respondents who had knowledge of the hunting register in Tahuayo – Blanco and Yavari – Miri 84
The majority of respondents in each community (ranging from 75% to 100%) agreed with including the hunting register as a part of the wildlife agreement. % of Respondents
100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Agree
Diamante 7 de
Julio
Nueva Esperanza
- Carolina
Disagree
Proportion of respondents who agreed with the establishment of hunting register as a part of wildlife agreements. % of Respondents
Similarly, the communities also were positive about including quotas for hunted species in the agreement. 100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Agree
Diamante 7 de
Julio
Nueva Esperanza
- Carolina
Disagree
Proportion of respondents who agreed with the establishment of hunting quota as a part of wildlife agreements. % of Respondents
The communities were generally in agreement about designating hunting and non‐hunting areas (source‐sink areas) as a part of the wildlife agreements. 100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Agree
Diamante 7 de Nueva Esperanza
Julio
- Carolina
Disagree
Proportion of respondents who agreed with the establishment of hunting and non‐hunting areas (source – sink) as a part of the wildlife agreement 85
Overall, 80% of the respondents agreed that habitat conservation should be part of the wildlife agreement. In addition when was asked about some measures to conserve the palm trees, the respondents mentioned palm reforestation; a ban on felling trees and use of a climbing device to allow the harvesting of fruits and leaves without harming the tree. % of Respondents
100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Agree
Diamante 7 de
Julio
Nueva Esperanza Carolina
Disagree
Proportion of respondents who agreed with the establishment of habitat conservation as part of the wildlife agreement During the discussion groups, the five communities made several suggestions to improve implementation of the wildlife management guidelines, including requests for more information. Each community had different suggestions, for instance Diamante 7 de Julio requested information concerning the process of the program; whereas Buena Vista and Nueva Esperanza – Carolina requested information about future earnings. The majority of respondents in each community mentioned that the wildlife would increase as a result of wildlife management. 100
90
80
% of respondents
70
60
50
40
30
20
10
0
Buena Vista
El Chino
San Pedro
Wildlife increase
Wildlife decrease
Diamante 7 de
Julio
Nueva Esperanza
- Carolina
Do not know
Proportion of respondents who mentioned conservation benefits from the wildlife management In addition, respondents were asked whether they had an economic interest in supporting wildlife management and respondents with primary school education tended to be more interested to support wildlife management if economic incentives are provided from the program. 86
Group of respondents who were interested to wildlife management with salary B Variable Education SE df Significance 0 0 3 0.013* Primary school education ‐4.110 1.461 1 0.005** Constant ‐1.417 2.356 1 0.548 *p < 0.05; **p < 0.01 Respondents expected to gain increased income from the meat sales as result of the wildlife management; in fact they mentioned that the income would help in buying basic products such as rice, salt and sugar. Indeed, the majority of respondents in each community (ranging from 70% to 10%) wished to buy food and clothes as result of the meat sales. A smaller proportion (30%) of respondents did not expected any economic benefit from the wildlife management. % of respondents
100
90
80
70
60
50
40
30
20
10
0
Buena Vista
El Chino
Buying food
San Pedro
Buying clothes
Diamante 7 de
Julio
Nueva
Esperanza Carolina
Do not expected nothing
Proportion of respondents who mentioned economic benefits from the wildlife management with economic incentives Although there were some gaps in knowledge of the wildlife management program most respondents had a positive attitude about the implementation of the guidelines in their wildlife agreements established in the communities. A comparison of the main results from the different communities shows clearly where more work may be required to improve knowledge and manage expectations. Comparison of the CPUE and density data The catch per unit effort and density results for the different species and communities show that the density of all species has remained stable since the wildlife management were initiated, with some species showing increasing abundances in different areas. 87
A summary of the questionnaire results showing how the communities varied in levels of knowledge and attitudes to the peccary pelt certification program and wildlife management in general. Area/community
Tahuayo – Blanco
Indicators
1. Local knowledge concerning wildlife agreements
a. Hunting quota
b. Primate hunting ban
c. Vigilance of wildlife
d. Source – sink area
e. Tree felling ban
2.Local knowledge and attitude about the
certification guidelines
a Correct identification of species that are vulnerable or
less vulnerable to over-hunting
b1.Hunting register and hunting quota
b2.Number of hunting registers participation from 2006 –
2008
c. Source – Sink area
d. Habitat conservation
3. Percentage of respondents who were aware of the
peccary pelt certification program (PPCP)
4. Respondents expectations of the PPCP and
wildlife management
Buena Vista
El Chino
San Pedro
*
*
*
*
*
*
*
*
*
*
*
*
80% **
73%**
100% **
90%**
90% of respondent
agree **
90%**
6 to 8 register
4 to 2register
Yavari - Miri
Diamante
7 de Julio
Nueva Esperanza
- Carolina
*
*
*
*
86% **
92%**
4 to 3 register
*
100% **
100%**
4 to 3 register
12 to 2 register
90% **
90%**
60% **
80%**
100% **
100%**
90% **
80%**
100% **
100%**
60%
Increased of wildlife
and more income to
buy basic products
and clothes
40%
80%
55%
35%
Increased of wildlife and
more income to buy basic
products and clothes
Increased of wildlife
and more income to
buy basic products
Increased of wildlife and
more income to buy basic
products
Increased of wildlife and
more income to buy basic
products
Legend: (*) wildlife agreements identified in each community; (**) percentage of respondents that agreed with the certification guidelines; increasing or decreasing of number of registered hunter
88
Summary of the Catch per Unit Effort and density studies. Tahuayo – Blanco
Species
Buena Vista
(2005 - 2007)
El Chino
(2005 – 2007)
San Pedro
(2004 -2006)
Yavari - Miri
Diamante 7
de Julio
(2004 – 2007)
Density
(Indv/Km2)
(2004 – 2008)
CPUE (Indv/hunting days)
Nueva Esperanza
- Carolina
(2004 – 2007)
CPUE
(Indv/hunting days)
Density
(Indv/Km2)
(2005 – 2007)
CPUE constant
CPUE constant
CPUE constant
CPUE increased
CPUE
increased
CPUE
decreased
CPUE
increased
No data
recorded
No data
recorded
CPUE
decreased
Agouti paca
CPUE constant
No sightings
Density remained
stable
CPUE increased
No sightings
Density remained
stable
Tayassu tajacu
CPUE
increased
Tayassu pecari
CPUE constant
that tend to
increase
No sightings
Density remained
stable
CPUE remained stable from
2004 – 2005, then catches
were not recorded
No sightings
Density decreased
CPUE increased
Mazama americana
Density remained
stable
Lagothrix poepiggi
89
CPUE constant
Density remained
stable
Discussion
The Samiria River Basin Populations of the terrestrial mammals are at healthy levels in the Samiria River basin. The primates were, by far, the animals with the greatest population numbers, followed by the rodents, birds, and ungulates. Twelve species of primates occur in the reserve although only 8 species were actually recorded during this study. The red howler monkey was the species with the highest density in the basin. White‐lipped peccary was the ungulate with the highest population, and it can live in large groups of up to about 200 individuals. Piping guans, spix guans and the razor‐billed curazao were the most common game birds in the river basin. The abundance of dolphins is very healthy in the Samiria River with the Pink River Dolphin having a higher density than the Grey River Dolphin. The Samiria River basin has one of the greatest densities of dolphins found anywhere in the Amazon basin. Results showed that both species are often sighted alone, or in pairs, and sometimes in pods larger than four. The Amazonian Manatee is difficult to sight, but with the limited sightings it is clear that the Samiria River is one of the last refuges of manatees in the Peruvian Amazon. The lower basin had very few sightings possibly as a result of human disturbance. The local people say that they often see them close to their villages. The Pacaya‐Samiria Reserve has one of the greatest populations of black caimans in the Peruvian Amazon. This species was overhunted in the 1940’s‐1970’s because of the value of its leather. The conservation of the Samiria River basin has allowed the black caiman to recover. However, the common caiman is decreasing in the basin, apparently from competition from the black caiman. The smooth‐fronted caiman is generally rare in the Samiria, but can occasionally be sighted. Amongst the macaws, the blue & yellow and the red‐bellied macaws were the most abundant species in the study zones. Both species make their nests in dead palms, although the former also takes advantage of hollows made by other birds. The other less abundant species are scarlet macaw and red & green macaw that nest in relatively soft trees. In the study areas there are large patches of palm swamps which are key habitats of the macaw species. Conservation of palm swamp will be very important and changes in macaw abundance are likely to correlate to changes in the fruit production within the palm habitats. The density of the yellow spotted river turtles in the study zone is at a healthy level. The turtle head starting recovery programme appears to be successfully helping to increase the turtle populations. The incubation of eggs in artificial nests is directly correlated to the increase in turtle numbers. The time between infant turtles to adult turtles is estimated at around eight years and the turtle nesting programme started 10 years ago. Therefore, it is predicted that turtle numbers should increase dramatically during the coming years. The two most abundant species of fish in the study sites were Common Pleco, followed by RedBellied Piranha and the White Piranha. The Piranha feeds on bleeding or sick individuals, and helps to clean the ecosystem and serve an important ecological function. The Common Pleco is abundant in its preferred habitat of shallow waters. In general the diversity of fish recorded during in the Samiria River basin is high because of the variety of habitats. In conclusion, conservation activities produce the best results when professionals, institutions, local communities and the reserve authorities work together. The monitoring activities conducted by the Op Wall expedition helped to collect information on a variety of wildlife populations, which can help to determine whether the current conservation strategies are being successful and if the best 90
decisions are being taken. Understanding fluctuations in wildlife populations over time is extremely important in relation to conservation and management strategies of the reserve. These fluctuations can be influenced by natural events or induced by human impacts. Thus, only through long‐term monitoring can the true impact of conservation be determined. It is therefore very important that monitoring of wildlife in the Pacaya‐Samiria National Reserve is continued and research records fluctuations in population densities. This information is extremely useful to the reserve authorities and local communities and for the conservation of natural resources in the Pacaya Samiria National Reserve. The Lago Preto Conservation Concession The public‐private conservation concession of Lago Preto is helping to conserve the wildlife, landscapes and hydroscapes of the Yavari River basin. The goals of the concession are to conserve the rare red uakari monkey, giant river otter and other important species including the paiche fish and black caiman. These goals are being realised and the populations of the key species are being conserved. The red uakari monkey has increased significantly since the implementation of the Lago Preto concession. The population of red uakari is roughly an order of magnitude greater than it was prior to conservation actions. However, the increase in the red uakari monkey is impacting other primate species. Both the woolly monkey and howler monkey populations have decreased, presumably due to competition from the large red uakari population. Indeed, in other areas of the Yavari basin woolly monkeys attain very high densities, but these areas have few red uakari monkeys. There are clear trade offs with conserving large‐bodied primates, since the competitive interactions do not permit all species to be simultaneously at their maximum populations. The Lago Preto site is an unique area for the red uakari monkeys and should be conserved as a site for this rare species. However, alternative site should be conserved for the woolly monkeys and other large‐bodied primates of the Yavari basin in areas with naturally low numbers of red uakari monkey. Macaw populations have also shown changes since the implementation of the concession. The blue and yellow macaws have increased significantly, and continue to do so, which is a good sign. However, the scarlet macaws have not increased in the same manner and are slightly lower in numbers. It is unclear why the macaws show this variance in their populations and further research will be conducted to discriminate the variables that impact their populations. The giant river otter population at Lago Preto is increasing each year. They were severely overhunted between the 1950’s to 1970’s for the international pelt trade. The pelt trade was closed at the end of the 1970’s and the giant river otter populations have been increasing slowly in remote areas, such as Lago Preto. One of the giant river otter groups is semi habituated and studies of this group indicated the conditions of the lake and their use of fish resources. These studies will help determine the suitability of other lakes at Lago Preto for the species and management recommendations that can help with the recovery of the giant river otter populations The dolphins have a healthy population at the Lago Preto concession and are important species for the hydroscape. Research at Lago Preto shows that the pink river dolphins prefer the oxbow lakes and are using the lakes as nurseries for raising their young. Pink river dolphins are ancient species that have evolved in the Amazon basin and are well adapted to the flooded forests and oxbow lakes. The lakes are excellent fishing grounds for the dolphins and are obvious places to teach their young how to fish. In contrast, the grey river dolphins are a much more recent species to the Amazon and prefer the main channels of the rivers. Grey river dolphins are not adapted to the flooded forests and oxbow lakes to the same degree as the pink river dolphins and they raise their young in the main channels of the river. The black caiman is also a key species for conservation at the concession. However, unlike the red uakari, giant river otter, and paiche fish, the black caiman have not recovered. Similar to the giant 91
river otter, the black caiman was excessively overhunted for its pelt during the professional pelt period. In the late 1970’s black caiman hides were prohibited for export. However, the bush meat and fisheries trade has continued to impact the black caiman populations. Indeed, during a three year period between 2001‐2003 black caiman meat was used as a copy cat meat of the valuable paiche fish meat. This resulted in a recent and intensive harvest of the balck caiman along the Yavari River and has presumably been an important reason for its lack of recovery. This meat trade is now better regulated by the authorities and hopefully over the coming years the black caiman population will recover at Lago Preto. The conservation and management of the fish populations at Lago Preto have also shown considerable success since the implementation of conservation actions. The diversity of fish is increasing overall in the lakes of the concession. In addition, the conservation actions at the lake of Lago Preto has seen a significant increase in the populations of the world’s largest fresh water fish, the paiche , which often reaches over 3m in body length. However, the increase in paiche fish has resulted in a decrease in the abundance of other fish at this lake. The paiche is a top predator and its large populations are presumably consuming large numbers of smaller fish. Again, this shows how conservation actions on one species have an impact on other species. The management of the fisheries at the Lago Preto concession allows local Brazilians to fish at certain lakes, while others are fully protected. This diversity in management strategies has allowed for different lakes to conserve varying abundances of the fish assemblage. While fished lakes have greater abundances of the smaller species, the protected lakes have greater numbers of the large paiche fish. Conservation concessions are clearly a positive form of protected area. The funds for their management are raised through the private sector and allow for an additional means of conservation. The Lago Preto Conservation Concession is clearly resulting in positive wildlife conservation as a result of the public‐private partnership. Conservation impacts
The Samiria River Basin The Pacaya‐Samiria National Reserve is a refuge for the high diversity assemblages of aquatic and terrestrial wildlife, including large populations of river dolphins, recovering populations of manatees and giant river otter, 12 species of primates, and a wide range of terrestrial rainforest mammals (Aquino et al. 2001). Macaws, wading birds, and game birds are abundant (Gonzalez 2004), as are populations of caimans and river turtles. The hyrdoscape of the reserve has a great diversity of fish and abounds with economically important species. The Pacaya‐Samiria National Reserve has approximately 95,000 people living in villages and towns along its boundary (INRENA 2000). Some of the villages lie just inside the reserve, however there are no human settlements within the core area. Most of the inhabitants are Cocama‐Cocamilla Indians (Puertas et al. 2000). The Cocama‐Cocamilla people are renowned for their mobility with families continually moving between villages (Newing & Bodmer 2004). They have always adapted well to other societies and integrated well with the influx of European customs beginning with the missionaries, and then through the rubber boom and twentieth century. The Pacaya‐Samiria National Reserve has gone through a major shift in its management policies over the past two decades, from an area of strict protection where local people were excluded from the reserve to an area where the local indigenous people participate with the reserve management. This drastic shift in conservation policy has led to a reduction in hunting pressure and an increase in wildlife populations (Bodmer & Puertas 2007). When the park administration changed and the 92
reserve began to incorporate the local communities in the management of the area, attitudes of the local people also changed (Puertas et al. 2000). Local management groups were given areas to manage and were no longer considered poachers. They were able to use a limited amount of resources legally and with reserve administration approval. Many of the local people changed their attitude towards the reserve and began to see the long‐term benefits of the reserve for their future. The reserve became part of their future plans and there was increasing interest in getting involved with the reserve. Many local people can now see the socio‐economic benefits of the reserve and are themselves helping to conserve the area. Hunting has decreased substantially, both due to the poachers now becoming mangers, and because the local people are keeping the other poachers out of their management areas. Many animal populations are recovering in the Samiria river basin during the past decade. The results of this report document how the population dynamics, community structure and land/hyroscape use of the wildlife is being affected by the successful conservation actions of the reserve. A range of terrestrial and aquatic wildlife taxa were monitored to understand the changes, including mammalian, avian, reptilian and fish species. The results of the research show that many animal populations have recovered along the Samiria river basin over the past decades. This has been due in large part to the change in management policy of the Pacaya‐Samiria National reserve, from an area of strict protection to an area that incorporates local indigenous people in the reserve management. The level of poaching was much greater during the period of strict control and hunting pressure was considerably greater. Local people saw no future in the reserve and considered the reserve administration as an enemy who took away their access to natural resources in their traditional lands. In the late 1990’s the strict policies of the reserve were changed and local people became involved with reserve management. This has permitted them to change their views of the reserve and move from being poachers to managers. With the decrease in poaching activities and an increased involvement of local people in the management of the reserve the animal populations have increased. There are now larger populations of most of the wildlife species than in the recent past. The success of the reserve has seen recovering populations of key wildlife species, such as the large‐
bodied primates, the blue and gold macaws, the Amazonian river turtles, and the black caiman. However, the research clearly shows that species are often in Lotka‐Volterra competitive interactions, and recovery of one species will actually impact the population of other species. For example, large‐bodies primates appear in competition with each other, and outcompete smaller species. As the blue and gold macaw numbers increase, the chestnut fronted macaw is in decline. Likewise, the recovery of the black caiman is resulting in a decline of common caiman. The local management policies of the reserve need to take into account the results of being successful. It may lead to an increase in the species being conserved, and in turn lead to a change in the animal community structure and a decrease in competing species. The Pacaya‐Samiria National Reserve is the largest protected area in the north eastern Peruvian Amazon and is key to biodiversity conservation of the western Amazon basin. The results of this study clearly show that the reserve is conserving biodiversity. The results also show that having a reserve administration that involves local people in park management has a positive effect on biodiversity conservation. The Lago Preto Conservation Concession Conservation of wildlife and wildlands will only be successful if a diversity of strategies are used that buffer against unpredictabilities in the biological and human dimensions. Biological uncertainties 93
include unpredictable changes in habitats, climate, wildlife populations, and ecological interactions, whilst human uncertainties include unpredictable changes in human demograhies, macro and micro economics, political turns, among others. Conservation of the Amazon rainforests and its immense biodiversity will require systems that incorporate diversified strategies. Currently, biodiversity conservation in Amazonia is using strategies such as protected areas, sustainable wildlife use, sustainable timber extraction, carbon offsetting, community‐based conservation, compensation by oil and agricultural industries, and environmental education. Protected areas have been traditionally, and remain, a major conservation goal of many governments and NGO’s. Protected areas use a diversity of management and land right categories ranging from fully protected parks, to national reserves, sanctuaries, community reserves, wildlife refuges, hunting reserves, private lands and public‐private reserves. Public‐private reserves are protected areas that are concessioned by governments to private concerns, such as individuals, companies or NGO’s. In developing countries funds available for maintaining and managing protected areas are often limited, and many protected areas have legal status without corresponding financial support, and are known as ‘paper parks’. This had led to the development of alternative strategies to insure funding for biodiverse areas requiring conservation protection. Public‐private reserves are a potential strategy to help conserve areas of special conservation interest in economically poor regions, such as the Peruvian Amazon. The Lago Preto Conservation Concession is a public‐private Amazonian reserve, where the Peruvian government granted a 10,000 hectare block of rain forest to WCS (in collaboration with DICE) as a 40 year concession. Conservation concessions are a relatively new strategy in Peru with the first concession granted in 2004. The effectiveness of conservation concessions will be paramount to a diversified conservation landscape. The goal of the current research at Lago Preto is to investigate the effectiveness of a conservation concession for Amazonian biodiversity. Results will help determine the potential of public‐private conservation strategies as a viable protected area category, thus increasing diversity in protected area systems. The red uakari monkey population at Lago Preto has successfully increased since conservation actions were implemented. The other large bodied primates found at Lago Preto appear to be impacted by the high numbers red uakari population. Competition and lotka‐volterra interactions seem to be occurring at Lago Preto. The woolly monkey (Lagothrix lagotricha) and howler monkey (Alouatta seniculus) populations at Lago Preto have fluctuated in Lotka‐Volterra fashion, with increase in woolly monkeys resulting in decrease in howler monkeys and visa versa. However, when the red uakari reached its peak population in 2007 both woolly monkeys and howler monkeys fell to low densities. Woolly monkey numbers at Lago Preto are lower than other sites in the Yavari valley, which may reflect competition with the large population of red uakari monkeys. Woolly monkeys in other regions of the Yavari and Yavari Miri are some of the greatest reported in the Amazon and in these sites red uakari populations are very small. The terrestrial mammal fauna at Lago Preto shows similar patterns to other areas of the Yavari valley. The artiodactyls are dominated by white‐lipped peccary, both in density and biomass, followed by collared peccary, red brocket deer and grey brocket deer being the least abundant. The populations of carnivores and edentates are dominated by the coati, followed by similar numbers of tayra and tamandua. The macaws were monitored along river point counts. Species have shown a significant change in abundance since the implementation of the conservation concession. The abundance of the blue 94
and yellow macaw (Ara ararauna) has increased steadily. In contrast the scarlet macaw (Ara macao) has decreased slightly Between 2005 and 2008 there was a steady increase in the giant otters population on Buen Fin Lake from 2 individuals to 8 individuals. This suggests that Buen Fin provides a good habitat for the giant otters. The diet of the otters at Buen Fin was dominated by the fish of the order Characiformes. The pink dolphins (botos) and grey river dolphins (tucuxi) use the hydroscape at Lago Preto differently. Botos were shown to use the lake habitats more frequently, whereas the tucuxi were more common in the river systems. The pink river dolphins appear to be using the lakes as nurseries. A greater frequency of young botos was found in the lakes of Tapishca and Ipiranga at Lago Preto than in the main channel of the Yavari River. Preliminary results suggest that the botos use the lakes to teach their young how to fish in the more isolated and calm waters of the lakes, were they are better adapted to the interface between forests and water bodies. The grey river dolphin does not use the lakes as nurseries, but raises their young in the main river channels. At Lago Preto a greater frequency of young grey dolphins were observed in the main river channels, with very few young observed in the lakes. The caimans at Lago Preto are dominated by the common caiman, Caiman crocodiles. The black caiman, Melanosuchus níger, was exploited extensively between the 1950´s to 1970´s for their pelts, and again between 2002‐2004 for their meat. The populations of black caimans have not recovered at the Lago Preto site, unlike other areas of the Peruvian Amazon, such as the Saimiria River. The smooth fronted caiman, Paleosuchus trigonatus, is generally rare in the rivers of the Peruvian Amazon and populations at Lago Preto compare to other river systems. Ninety four different species of fish were found in total in Lago Preto, Ipiranga, Tipishca, Huapo and San Jorge Lakes. There appeared to be no significant trend of higher species richness in black‐water rather than white‐water lakes, as was the case in other studies. Species richness of fish populations in the oxbow lakes of Lago Preto has steadily increased. The most abundant species was Potamorhina latior, followed by Serrasalmus rhombeus and Pellona castaeneana. Catch‐per–unit‐
effort (CPUE) was used as a measure of fish population abundance. The CPUE showed that the abundance of fish is increasing in the managed lakes of Tipishca and Ipiranga where subsistence fishing is permitted by members of a local Brazilian fishing community. In contrast the CPUE showed a decrease in the fully protected lake of Lago Preto. The CPUE results correlate with the increasing numbers of the largest fish in the world, the Arapaima gigas, or paiche . The fully protected lake of Lago Preto has seen a significant increase in the population of the paiche fish over the past three years. This large predatory fish appears to have impacted the abundance of the smaller fish species and explains the decrease in CPUE of the fish at Lago Preto. The public‐private conservation concession of Lago Preto is helping to conserve the wildlife, landscapes and hydroscapes of the Yavari River basin. The goals of the concession are to conserve the rare red uakari monkey, giant river otter and other important species including the paiche fish and black caiman. These goals are being realized and the populations of the key species are being conserved. Significance/ benefits of the research
95
The Peruvian Amazon is a mega‐diverse rainforest that harbors one of the greatest, if not the greatest, assemblages of biodiversity on Earth. The project in Loreto has been developing and implementing a conservation strategy for the mega‐diversity of the Samiria‐Yavari landscape by showing the national government, regional government, international community and other actors that protected areas in Loreto can be successful both socially and for biodiversity conservation, and by increasing the on‐the‐ground capacity of local professional to insure that protected areas in Loreto will be successfully implemented in the long‐term. The project worked with the National and Regional Government (INRENA & GOREL) to officially decree the Lago Preto Conservation Concession. The project worked with the National Government and Regional Government (INRENA & GOREL) to begin the process of creating a Conservation Corridor linking the Lago Preto Conservation Concession with the Regional Tamshiyacu‐Tahuayo Community Reserve. The project worked with the National Government and Regional Government (INRENA & GOREL) to begin the process of creating an extended buffer zone around the Lago Preto Conservation Concession based on traditional rubber extraction. Research in the Pacaya‐Samiria National Reserve continues to show increases in key wildlife species as a result of successful conservation actions. As a result of the conservation successes of the reserve the Regional Government of Loreto (GOREL) considers the Pacaya‐
Samiria National Reserve as a cornerstone for conservation in Loreto and is taking a more positive view of new protected areas. The project worked with CITES, INRENA, local communities, and other stakeholders to implement a wildlife certification program as a way to catalyze community‐based wildlife management and increase the number of hectares under protection through the creation of community protected non‐hunting areas. The project is working closely with implementing community based wildlife management in numerous local communities in the Samiria‐Yavari landscape. Four communities have implemented wildlife management plans that incorporate source (non‐hunted protected) areas as part of their wildlife management systems. Literature Cited in the Proposal
Anderson, A.B. 1990. Alternatives to deforestation: Steps toward sustainable use of the Amazon rainforest. Columbia University Press. Aquino, R., Bodmer,R.E., and Gil, G. 2001. Mamíferos de la cuenca del rio Samiria: Ecología poblacional y sustentabilidad de la caza. Publicación Junglevagt for Amazonas, AIF‐WWF/DK. WCS. Lima, Perú. 108 p. Bodmer, R.E., Penn, J. W., Puertas, P., Moya, L. and Fang, T. 1997. Linking conservation and local people through sustainable use of natural resources: community‐based management in the Peruvian Amazon. In: F. Curtis (ed) Harvesting wild species. The John Hopkins University Press. Baltimore y London. Pp. 315‐358. Bodmer, R. and P. Puertas. 2007. Impacts of Displacement in the Pacaya‐Samiria National Reserve, Peru – Richard. In: Kent H. Redford and Eva Fearn (eds). Protected areas and human displacement: a conservation perspective. WCS Working Papers: Nº 29. WCS, pp. 29‐33 96
Bodmer, R.E., Puertas, P.E., Garcia, J.E., Diaz, D.R. and C. Reyes. 1998. Game animals, palms and people of the flooded forests: Management considerations for the Pacaya‐Samiria National Reserve, Peru. Advances in Economic Botany, 13: 217‐232 Bodmer, R. and Puertas, P. 2000. Community‐based co‐management of wildlife in the Peruvian Amazon. In: J. Robinson y E. Bennett (eds) Hunting for Sustainability in Tropical Forests. Columbia University Press, New York. Pp.395‐409. Bodmer, R., Puertas, P. and Fang, T.G. 2008. Co‐managing Wildlife in the Amazon and the Salvation of the Pacaya‐Samiria National Reserve in Peru. In: Wildlife and Society. The Science of Human Dimensions. In: Michael Manfredo, Jerry Vaske, Perry Brown, Daniel Decker and Esther Duke (eds). Island press. Washington DC, pp 104‐116. Bodmer, R.E. and Robinson, J.G. 2004. Evaluating the sustainability of hunting in the Neotropics. In: Silvius, K., Bodmer, R. and Fragoso, J. (eds.) People in Nature: Wildlife Conservation in South and Central America, Columbia University Press, New York. Buckland, S.T., Anderson, D., Burnham, K. and J. Laake. 1993. Distance Sampling: Estimating the Abundance of Biological Populations. Chapman & Hall, London. Dourojeanni, M.J. 1990. Amazonía, qué hacer?. CETA. Iquitos, Perú. Bodmer & Puertas 2007 Fang, T., Bodmer, R., Puertas, P., Mayor, P., Perez, P., Acero, R. and D. Hayman. 2008. Certificación de pieles de pecaries (Tayassu tajacu y t. pecari) en la Amazonía peruana: Una estrategia para la conservación y manejo de fauna Silvestre en la Amazonia peruana. Wust Editions‐Darwin Institute, Lima, Peru. Halme, K.J., and R.E. Bodmer. 2007. Correspondence between scientific and traditional ecological knowledge: Rain forest classification by the non‐indigenous ribereños in Peruvian Amazonia. Biodiversity and Conservation, 16, 1785‐1801. Little, P.D. 1994. The link between local participation and improved conservation: a review of issues and experiences. In: D. Western y R.M. Wright (eds) Natural connections: perspectives in community based conservation. Washington D.C. Covelo, California. Pp. 347‐371. Newing, H. and R. Bodmer. 2004. Collaborative wildlife management and adaptation to change: The Tamshiyacu‐Tahuayo Communal Reserve, Peru. Nomads 7. Pp 12. INRENA. 2000. Plan Maestro para la Conservación de la Diversidad Biológica y el desarrollo sostenible de la Reserva Nacional Pacaya Samiria y su zona de Amortiguamiento. Lima. 152 pp. Queiroz, H. 2000. Natural history and conservation of pirarucú, Arapaima gigas, at the Amazonian giants in muddy waters. Thesis doctor of Philosophy, University of St Andrews. 225 pp. Peres, C.A. and E. Palacios 2007. Basin‐Wide Effects of Game Harvest on Vertebrate Population Densities in Amazonian Forests: Implications for Animal‐Mediated Seed Dispersal. Biotropica 39: 304‐315. Peres, C.A. and Nascimento, H.S. 2006. Impact of game hunting by the Kayapó of southeastern Amazonia: implications for wildlife conservation in tropical forest 97
indigenous reserves, Biodiversity and Conservation 15: 2627‐2653. Peres, C.A., Barlow, J. and Laurance, W. 2006. Detecting anthropogenic disturbance in tropical forests, Trends in Ecology and Evolution 21: 227‐229. Peres, C.A. 2000. Effects of subsistence hunting on vertebrate community structure in Amazonian forests, Conservation Biology 14: 1, 240‐253. Pitman, N et al, 2003. Peru: Yavari: Rapid biological inventory report No.11. The field museum, Chicago Puertas. P., R. Bodmer, J. López, J. del Aguila & A. Calle. 2000. La importancia de la participación comunitaria en los planes de manejo de fauna silvestre en el nor oriente del Perú. Folia Amazónica 11(1‐2): 159‐179. Puertas, P.E. and Bodmer, R.E. 2004. Hunting effort as a tool for community‐based wildlife management in Amazonia. In: Silvius, K., Bodmer, R. and Fragoso, J. (eds.) People in Nature: Wildlife Conservation in South and Central America, Columbia University Press, New York, pp. 123‐135. Salovaara, K et al, 2003. Diversity and abundance of mammals in Rio Yavari and Rio Yavari‐Mirin. In: Pitman, N et al, 2003. Peru: Yavari: Rapid biological inventory report No.11. The field museum, Chicago Thomas L, S. T. Buckland, K. Burnham, D. Anderson, J. L. Laake, D Boechers y S. Strindberg. 2002. In: Encyclopedia of Envirometrics. A. H. Shaarawi & W. W. Piegorsch (Eds). 1: 544‐552. 98