A regression tree-based method for integrating land-cover and land-use data collected at multiple scales

As data sets of multiple types and scales proliferate, it will be increasingly important to be able to flexibly combine them in ways that retain relevant information. A case in point is Amazonia, a large, data-poor region where most whole-basin data sets are limited to understanding land cover interpreted through a variety of remote sensing techniques and sensors. A growing body of work, however, indicates that the future state of much of Amazonia depends on the land use to which converted areas are put, but land use in the tropics is difficult to assess from remotely sensed data alone. An earlier paper developed new snapshots of agricultural land use in this region using a statistical fusion of satellite data and agricultural census data, an underutilized ancillary data source available across Amazonia. The creation of these land-use maps, which have the spatial detail of a satellite image and the attribute information of an agricultural census, required the development of a new statistical technique for merging data sets at different scales and of fundamentally different data types. Here we describe and assess this nonlinear technique, which reinterprets existing land cover classifications by determining what categories are most highly related to the polygon land-use data across the study area. Although developed for this region, the technique appears to hold broad promise for the systematic fusion of multiple data sets that are closely related but of different origins. The figures in the printed version of this article appear in black and white. Color figures are available from the author upon request.     

Edible Invertebrates Among Amazonian Indians: A Critical Review of Disappearing Knowledge

For the indigenous populations of Amazonia, invertebrates constitute an important component of the diet. We have information on entomophagy for 39 ethnic groups, about 21,4 per cent of the 182 groups known in the Amazon Basin, but the use of this non conventional food resource is probably much more widespread. We present here a data-base of all the information available for each ethnic group regarding the species included in the diet, the scientific and the ethno name if known, the stage of life-cycle consumed, the manner of preparation and, when known, the host plant. This data-base lists 115 species scientifically identified and 131 ethno names. In addition, we have information about other 384 ethno names, with unsecure link to the Linnean taxonomy suggesting that local knowledge is very extensive. The data-base represents not only an easy to consult resource, but also a support for further research. The knowledge of the relations between indigenous populations and ecosystem is indeed the base for the natural and cultural biodiversity preservation.     

Written Accounts of an Amazonian Landscape Over the Last 450 Years

Abstract:  Books, articles, government documents, and other written accounts of tropical biology and conservation reach a tiny fraction of their potential audience. Some texts are inaccessible because of the language in which they are written. Others are only available to subscribers of developed-world journals, or distributed narrowly within tropical countries. To examine this dysfunction in the tropical literature—and what it means for conservation—we tried to compile everything ever written on the biology and conservation of the department of Madre de Dios, Peru, in southwestern Amazonia. Our search of libraries, databases, and existing bibliographies uncovered 2,202 texts totaling roughly 80,000 pages. Texts date from 1553 to 2004, but 93% were written after 1970. Since that year the publication rate has increased steadily from fewer than 10 texts/year to nearly 3 texts/week in 2004. Roughly half of the Madre de Dios bibliography is in Spanish-language texts written by Peruvian authors and mostly inaccessible outside Peru. Most of the remaining material is English-language texts written by foreign authors and largely inaccessible in Peru. Foreign authors tended to write about ecological studies with limited relevance to on-the-ground conservation challenges, whereas Peruvian authors were more likely to make specific management recommendations. The establishment of a Web-based digital library for Neotropical nature would help make the tropical literature a more efficient resource for science and conservation. Additional recommendations include investing in syntheses, translations, popular summaries, and peer-reviewed journals in tropical countries, providing incentives for management-relevant research in tropical protected areas, and reinforcing training of scientific reading and writing in tropical universities.     

Mercury conversion processes in Amazon soils evaluated by thermodesorption analysis

This paper reports on the speciation study and the Hg redox behavior in Amazon soils not influenced by gold mining and collected near Manaus, AM, Brazil. The samples were incubated by adding Hg(0) and HgCl₂ to dry soil. Solid phase Hg speciation analysis was carried out using a Hg thermodesorption technique with the aim of distinguishing elemental Hg(0) from Hg(II) binding forms. In the first case, we observed the conversion of Hg(0) to Hg(II) binding forms in the range of 28–68% and a correlation between the percent of oxidation and OM content. Samples incubated with Hg(II) showed the formation of Hg(I) and/or Hg(0) in the range of 19–69%. The lowest values corresponded to the samples with the lowest clay contents. The kinetics of conversion of Hg(0) as well as HgCl₂ were roughly fitted to the two first order reactions, a fast one and a slow one. It was not possible to evaluate differences between sampling sites and types of soils, but the mean half-life of the first order reaction obtained by the addition of Hg(II) was slower (t_1/2 = 365 d) than the one obtained by the addition of Hg(0) (t_1/2 = 148 d). Previous studies have shown the predominance of organically bound Hg in these samples. Thus, the kinetic difference between Hg oxidation and reduction in combination with the efficient retention processes by OM may explain the high background values found in Amazon soils.     

Swidden agriculture in a protected area: the Matsigenka native communities of Manu National Park, Peru

In this study, we investigate whether traditional swidden agriculture on nutrient-poor tropical soils can sustainably support a growing indigenous population within Manu National Park, Peru. Based on interviews with 50 Matsigenka farmers, as well as GPS mapping and field visits, we established the location and size of all 124 swidden gardens cultivated in 2000 and 2001. Using a GIS-based soil map of Manu Park, we identify the total extent of potential arable land (1) throughout the park zones currently available for Matsigenka habitation, and (2) within the vicinity of the two study communities. Taking into account current per capita garden size, cultivation and fallow practices, we calculate the maximum human carrying capacity of swidden agriculture for (1) all available park zones and (2) the immediate vicinity of the two current communities. Even given the second, highly limited scenario, the sustainable human carrying capacity is estimated to be 2,138 individuals, which is far higher than the current population of about 420 people. Thus, arable land does not appear to be a limiting factor for the growth of the indigenous population in Manu Park at the current time. Other possible limiting factors for population growth are identified and discussed. We conclude that a conservation policy of population stabilization around the current settlements is viable in that it will not result in reduced resource availability over the next few decades, and that the challenge is to identify and mitigate the forces causing population fissioning and spread. Readers should send their comments on this paper to BhaskarNath@aol.com within 3 months of publication of this issue.     

Uses of Fish and Game by Inhabitants of an Extractive Reserve (Upper Juruá, Acre, Brazil)

The fishing and hunting performed by inhabitants of the Extractive Reserve of the Upper Juruá, Acre, Brazil was studied, and an ecological analysis on the use of fish and game, including preferences, taboos and medicinal uses, and of fish abundance was made. Research was carried out at the Upper Juruá Extractive Reserve in three visits (1993 and 1994). A total of 143 individuals (101 families) living along the banks of the rivers Juruá, Tejo, Bagé, S. João and Breu were interviewed, and 115 species of fish with set gillnets were collected. Local economic activities included rubber-tapping and small-scale agriculture. Fishing was especially important in the dry season, when cast net and hook and line were used. Hunting and game consumption were intensive in the wet season. Some local rules may be useful as starting points for local management, such as the use of fish diversity through lake management, the use of water level seasonality, hunting periods or quotas, and food taboos. The integration of academic research with indigenous knowledge and demands is important to develop management plans which protect both biological and cultural diversity.     

Scale dependency of conservation outcomes in a forest-offsetting scheme

Offset schemes help avoid or revert habitat loss through protection of existing habitat (avoided deforestation), through the restoration of degraded areas (natural regrowth), or both. The spatial scale of an offset scheme may influence which of these 2 outcomes is favored and is an important aspect of the scheme's design. However, how spatial scale influences the trade-offs between the preservation of existing habitat and restoration of degraded areas is poorly understood. We used the largest forest offset scheme in the world, which is part of the Brazilian Forest Code, to explore how implementation at different spatial scales may affect the outcome in terms of the area of avoided deforestation and area of regrowth. We employed a numerical simulation of trade between buyers (i.e., those who need to offset past deforestation) and sellers (i.e., landowners with exceeding native vegetation) in the Brazilian Amazon to estimate potential avoided deforestation and regrowth at different spatial scales of implementation. Allowing offsets over large spatial scales led to an area of avoided deforestation 12 times greater than regrowth, whereas restricting offsets to small spatial scales led to an area of regrowth twice as large as avoided deforestation. The greatest total area (avoided deforestation and regrowth combined) was conserved when the spatial scale of the scheme was small, especially in locations that were highly deforested. To maximize conservation gains from avoided deforestation and regrowth, the design of the Brazilian forest-offset scheme should focus on restricting the spatial scale in which offsets occur. Such a strategy could help ensure conservation benefits are localized and promote the recovery of degraded areas in the most threatened forest landscapes.     

How Well Does Brazil's Environmental Law Work in Practice? Environmental Impact Assessment and the Case of the Itapiranga Private Sustainable Logging Plan

/ The Itapiranga Sustainable Logging Plan provides an example of how Brazil's licensing system functions for logging companies in the state of Amazonas. Two questions need to be dealt with: "How sustainable can logging in the Amazon be?" and "What and how effective are existing legal mechanisms to deal with logging projects?" The environmental impact assessment (EIA) and environmental impact statement (EIS, known as the RIMA in Brazil), present relatively detailed accounts of biodiversity and the need to adopt conservation strategies to protect it. However, social and health impacts are only superficially addressed. The economic sustainability of the operation over multiple cycles is not demonstrated. The multidisciplinary teams responsible for the EIA and EIS (RIMA) reports are hired by the project proponent, an arrangement inherently carrying the risk of biasing the result. Logging reduces biodiversity, releases greenhouse gases and inflicts social and health costs. These impacts reduce the ability of Amazonian forests to provide environmental services and to supply food and livelihood security to local populations. The reports inflate positive effects such as employment: the estimated number of jobs was cut by more than half in a revision made after the EIA and EIS (RIMA) had been approved. Not only do the reports need to be more realistic in assessing both positive and negative consequences of proposed projects, but better means are needed to ensure that promised mitigatory measures are enforced in practice. Many of the lessons that can be drawn from the Itapiranga Plan are not unique to logging projects and apply to licensing of development activites generally in Brazil and elsewhere.     

Assessment of mammal reproduction for hunting sustainability through community‐based sampling of species in the wild

Wildlife subsistence hunting is a major source of protein for tropical rural populations and a prominent conservation issue. The intrinsic rate of natural increase. (r_max) of populations is a key reproductive parameter in the most used assessments of hunting sustainability. However, researchers face severe difficulties in obtaining reproductive data in the wild, so these assessments often rely on classic reproductive rates calculated mostly from studies of captive animals conducted 30 years ago. The result is a flaw in almost 50% of studies, which hampers management decision making. We conducted a 15‐year study in the Amazon in which we used reproductive data from the genitalia of 950 hunted female mammals. Genitalia were collected by local hunters. We examined tissue from these samples to estimate birthrates for wild populations of the 10 most hunted mammals. We compared our estimates with classic measures and considered the utility of the use of r_max in sustainability assessments. For woolly monkey (Lagothrix poeppigii) and tapir (Tapirus terrestris), wild birthrates were similar to those from captive populations, whereas birthrates for other ungulates and lowland‐paca (Cuniculus paca) were significantly lower than previous estimates. Conversely, for capuchin monkeys (Sapajus macrocephalus), agoutis (Dasyprocta sp.), and coatis (Nasua nasua), our calculated reproductive rates greatly exceeded often‐used values. Researchers could keep applying classic measures compatible with our estimates, but for other species previous estimates of r_max may not be appropriate. We suggest that data from local studies be used to set hunting quotas. Our maximum rates of population growth in the wild correlated with body weight, which suggests that our method is consistent and reliable. Integration of this method into community‐based wildlife management and the training of local hunters to record pregnancies in hunted animals could efficiently generate useful information of life histories of wild species and thus improve management of natural resources.