Systematic conservation planning for the Paranã River Basin, Brazil, under climate change

Sevilha, Anderson Cassio (2016) Systematic conservation planning for the Paranã River Basin, Brazil, under climate change. PhD thesis, James Cook University.

PDF (Thesis)
Download (5Mb) | Preview


Planning for effective conservation begins with precise information about species occurrences, accurate species identifications, knowledge of the distributions of species and communities and threats to biodiversity posed by anthropogenic activities and resultant climate change. In the Paranã River Basin, a key priority area for conservation within a global biodiversity hotspot – the Cerrado biome in Brazil – the areas previously highlighted as priorities for conservation by the Brazilian Government are too large and roughly delineated to be translated directly into actions on the ground. Which areas should be prioritized for biodiversity conservation within the Basin and why those areas should be prioritized are questions that are still not answered. These questions underpin the aims of this thesis. Answering those questions requires specific information, including: (a) identification of species occurrences within the Basin and gaps in biological information; (b) determination of patterns of species occurrences, species richness, and distribution of communities to support area prioritization for nature conservation; (c) evaluation of different biodiversity features as inputs for area prioritization, considering, at least, distributions of species and communities, richness patterns, threats and endemism of species, connectivity, complementarity, irreplaceability and vulnerability to anthropogenic activities; and (d) evaluation of the likelihood of changes in species ranges, habitat resilience and connectivity to identify priority areas for conservation considering the impacts of climate change on biodiversity.

To answer those questions, firstly, we compiled a database drawn from our intensive surveys of plants, birds, mammals, reptiles, amphibians, termites, drosophilids, and saturnids. Additionally, we exhaustively searched for biodiversity records in museums, herbaria, literature, and online databases. Having exhaustive biological information for the study area, we standardized taxonomic names and data formats compiled from diverse and disparate sources and applied an innovative method to automate identification of duplicate records. Secondly, we evaluated biodiversity patterns considering the distributions of species and communities. We analyzed richness and community patterns using, respectively, Maxent (a maximum entropy machinelearning algorithm) and GDM (Generalized Dissimilarity Modelling). We modelled distributions for 2159 species and modelled dissimilarities for each of the terrestrial vertebrate groups as well as insects and vascular plants. Thirdly, we evaluated the use of species and community distributions as biodiversity features for area prioritization using the Zonation conservation planning software. Here, we evaluated the separate contributions to area priorities of species distributions and community turnover as well as the influence on priorities of combined data on species and communities and the addition of other biodiversity features. We also compared the conservation solutions we generated with a previous prioritization of conservation units by the Brazilian Government and made inferences about the use of different approaches to area prioritization. Finally, we used Maxent to project species distributions for current and future climate and then applied Zonation again to identify areas for conservation that are resilient to the effects of climate change on species distributions.

Of ~140,000 records initially compiled for the entire Basin, only ~20,000 reliable records were retained. The 5,130 species currently known constitute between 35% and 54% of the larger Cerrado biome’s terrestrial vertebrate species, 34% of its plant species, and 11–43% of its endemics. Even with 200 years of data collection, spatial information gaps still exist for all taxa sampled in the Basin. The biodiversity information generated, however, was sufficient to build species and community distribution models for some important biological groups. Species distribution models showed differences in richness patterns between vertebrate groups (amphibians, birds, mammals, and reptiles) and plants. Plants are more diverse at higher elevations, contrasting with vertebrates that had higher richness at lower elevations. On the elevational gradient, environmental heterogeneity – represented by differences in climate and substrate – is related to more differentiation of vegetation types. At lower elevations, climate conditions are more stable, mean temperatures are higher, and the environment – dominated by dry forests and arboreal savanna – is vertically more structured, i.e. showing up to five strata or vertical layers (herbaceous, shrub, understory, canopy and emergent). Reliable models for community patterns could be obtained only for birds, mammals and plants. With few exceptions, community groups based on birds and mammals followed the same pattern as those based on plants. The community patterns indicate three major bioregions within the Basin: the mountain ranges on the altitudinal gradient and the areas occupied by dry forest and savanna formations at lower elevations. Although the results produced by models of species distribution and community turnover are different in their details in depicting biodiversity patterns, they are complementary in highlighting areas of importance for biodiversity conservation. Additionally, we concluded that the use of richness patterns of some biological groups as proxies for other groups is unreliable. Different groups respond differently to variations in abiotic and biological heterogeneity across the Basin. These differences should be considered in conservation design.

The choice of whether species- or community-level patterns should be used in area prioritization using Zonation depends on the objectives of conservation and data availability. Considering the most important areas identified for biodiversity conservation (the top 20% fraction of the landscape), all Zonation solutions retained a similar amount of the total distributions of each biodiversity feature [4% on average (2.37 SD) across species, communities, and threatened ecosystems]. Additionally, the priority areas in these different solutions were strongly congruentities, whether the biodiversity features were populations, communities or ecosystems. Those most important areas were located along the mountain ranges of the Basin. Differences between species- and community-level approaches in allocating priorities are evident only in relatively small areas at lower elevations.

Prioritization of areas for the Paranã River Basin considering the impacts of climate change reveals some congruence between current and future areas indicated as important for biodiversity conservation. Besides the mountains ranges, the overlap of present and future priorities includes elevational gradients, some flat terrain that surrounds the ranges, and limestone outcrops. Additional important areas are those prioritized only for the future scenario, which emphasised the connecting zones between current and future priority areas in different elevation gradients of the Basin.

The broadly defined conservation planning done for the Basin in 2007 must be revisited to maintain resilience under climate change and facilitate species persistence into the future. There is a very low overlap between priority areas for biodiversity conservation revealed by this study and the areas identified for conservation action by Brazilian Government. Of the total area of the Basin officially identified for conservation measures in 2007 (some 82% of the Basin), the establishment of protected areas overlaps only 6.95% with the most important areas for biodiversity conservation identified by this study, considering the intersection of current and future priorities (~10% of the Basin). The enforcement of sustainable use overlaps 5.23%, while existing protected areas overlap 0.09%. However, a few small priority areas identified in this study (0.3%) are outside those identified in 2007. This thesis is a guide to the establishment of protected areas, spanning gradients from the flat areas of the lowlands to the tops of mountains, as a core part of a strategy to promote the persistence of species in the face of future shifts in climate.

Item ID: 46300
Item Type: Thesis (PhD)
Keywords: assemblage distribution, biodiversity assessment, biodiversity conservation, biodiversity patterns, Cerrado hotspot, Cerrado, climate change, conservation planning, conservation prioritization, endangered species, GDM, Maxent, Paranã River Basin, Paranã, RecordLinkage, SDM, species distribution, zonation
Date Deposited: 10 Nov 2016 04:16
FoR Codes: 05 ENVIRONMENTAL SCIENCES > 0502 Environmental Science and Management > 050202 Conservation and Biodiversity @ 100%
SEO Codes: 96 ENVIRONMENT > 9608 Flora, Fauna and Biodiversity > 960805 Flora, Fauna and Biodiversity at Regional or Larger Scales @ 100%
Downloads: Total: 8
Last 12 Months: 3
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page