Endemic bird areas of the world pdf

We'll do our best to fix them. Check all that apply - Please note that only the first page is available if you have not selected a reading option after clicking "Read Article". Include any more information that will help us locate the issue and fix it faster for you.

Endemic Bird Areas of the World by A. Statterfield, M. Crosby, A. Long and D. ISBN 0——— This massive tome—it has pages—is certain to become one of conservation's major reference books.

It reflects the growing development of conservation as a serious science, with a serious purpose: in this case, helping to set priorities. In the world as a whole, it identifies endemic bird areas. Each of these EBAs contains at least two bird species whose global distribution is less than 50, km 2. The idea that we should focus on species with restricted ranges is becoming widely accepted and the 50, km 2 criterion seems to make practical sense as a definition of endemism that is free from political connotation.

Although more than a quarter of all the world's birds of which there are about have these restricted ranges, the total area of all of the EBAs is only a tiny fraction of the earth's surface.

They thus provide realistically small areas in which to concentrate conservation efforts, at least for birds. However, it is often the case that places with high numbers. African Journal of Ecology — Wiley.

We maximized the representation and persistence of each endemic bird species considering different conservation targets and minimized co-occurrence with densely populated areas, using counties as the unit of analysis.

We also simulated the impact of more clumped solutions. Finally, we contrasted the most cost-effective solution in under each climatic condition, current and future, using both county and regional level strategy i. We used the geographic boundaries of the Atlantic Forest established under the Brazilian legislation Law No.

We looked at every occurrence record, excluding those with will-defined geographic coordinates or that clearly fell outside species range as defined by the scientific literature, the digital maps of BirdLife International, or the range description provided the IUCN Red List of Threatened Species. We excluded species with fewer than 12 reliable occurrences. We used bioclimatic variables Hijmans et al. We selected six variables that show relatively low correlation among each other in the Atlantic Forest see Souza et al.

The data had 2. For the latter, we used a threshold value, maximizing the number of occurrence records under suitable areas sensitivity and pseudo-absences under unsuitable areas specificity. Ensemble models were then clipped to the Atlantic Forest remnants, excluding areas where forest no longer exists. We did not use a prediction of forest configuration in because: i there is not one available at a meaningful spatial resolution e.

Specifically, cvi1,i2 represents a penalty for selecting planning unit I but not selecting neighboring planning unit j. In equation 2, t j is the amount of each target j that must be selected.

We used the Atlantic Forest counties as the unity of analysis i. We included different conservation targets i. We performed a sensitivity analyses, halving or increasing twofold these values, to evaluate the robustness of our results. We used human population density per county as a surrogate for socio-economic development in our cost data.

Thus, we considered counties with denser populations a lesser priority in selecting areas for the establishment of protected areas. For the future scenario, we assumed that the human population density among counties will maintain the same relative ranking, i.

The planning considering BLM aggregates counties in the final solution, and therefore, provides regional level solutions that are applicable at the state or federal administrative spheres hereafter, called regional level strategy. We found the best BLM for each scenario by plotting total cost versus total edge of selected planning units BLM for the best solutions, and identifying BLM values where total cost and total edge intersects.

We calculated the percentage of overlap between the selected planning units in the most cost-effective solution under current and future conditions, planning both county and regional level strategies, in order to quantify the differences in the configuration between these protected area networks. Additionally, we calculated, for each species, the amount of environmentally suitable area protected under the solution of both scenarios i. All analyses were carried out in ArcGIS Prioritizing counties for protected area creation under climate change An ad hoc prioritization scheme was created for counties that were selected in the most cost-effective solution under both current and future scenarios.

The scheme places higher priority on the counties that were particularly relevant for the final proposed protected area network, had the greatest amount of forest remnant, and lowest area already within protected areas. The maximum possible Priority Score, therefore, is i. For birds endemic to the Brazilian Atlantic Forest, as measured by True Skill Statistics, for the five modeling algorithms used. Most species spp. The models predicted a contraction of environmentally suitable areas under climate change conditions for the vast majority of the species modeled Fig.

Planning protected areas network under current and future conditions Most species achieved their specific conservation targets, both under current and future conditions, independent of the use of county or regional level strategy Table 2.

Only three species did not achieve their specific conservation target in a specific combination of climatic conditions or strategy level: Acrobatornis fonsecai, Phylloscartes beckeri and Tangara fastuosa. The final solutions i. Under current conditions, ca. Under future conditions, however, and counties were required to achieve that representativeness under the county Vale et al. C Difference in between richness under current and future conditions showing areas that lose positive numbers and areas that gain negative numbers species in the future.

Full-size DOI: For the proposed protected area network, under current and future climate change scenarios. Results using a county level strategy ig- noring the boundary length modifier - BLM and regional level strategy considering BLM. Overlap refers to the counties selected in both current and future scenarios, with the number of and the species repre- sented in the future.

Current Future Overlap Counties Species Counties Species Counties Species County level Regional level and regional level strategies, respectively Table 2. Such increase in the number of counties selected in the future for the regional level strategy resulted in a higher overlap of the most cost-effective solutions for current and future scenarios, from under the county level strategy to We concluded, therefore, that for Atlantic Forest birds, there is a high congruence in protected areas network when planned under current and future conditions.

Although there is almost no numeric difference between county and regional level strategies, the spatial configuration of the two protected areas network was quite different Vale et al.

We used a county level strategy ignoring the boundary length modifier - BLM and regional level strategy considering BLM. Black counties already have protected areas and were always included in the final solution; red counties were selected in both current and future scenarios, while green counties were selected in either current or future scenario. A Selected counties using a county level strategy under current climatic conditions.

B Selected counties using a county level strategy under future climatic conditions. C Selected counties using a regional level strategy under current climatic conditions. D Selected counties using a regional level strategy under future climatic conditions. As expected, the county level strategy produced a more fragmented selection of counties across the Atlantic Forest, while the regional level strategy produced a greater clumping of selected counties Fig.

The sensitivity analyses, halving or increasing twofold the conservation targets i. S2, Table S3. For the standard conservation strategy, for example, when halving the conservation target, there is an overlap of counties between the solution under current and future conditions, representing species. When doubling the conservation target, Vale et al. Only counties that were selected both under current and future conditions are shown.

Higher priority is placed on counties that were particularly relevant for the final proposed protected area network, had the greatest amount of forest remnant, and lowest area already within protected areas. The greatest losses in the future are predicted in the highly biodiverse central portion of the Atlantic Forest Serra do Mar and Interior Forest subregions Jenkins et al.

At the same time, we predicted a moderate increase in species richness towards the south, as a result of shifting distribution Fig. The general pattern of greater species richness in the central portion of the Atlantic Forest, however, persists under the future conditions Fig. Another reason for the relatively large overlap in selected counties under current and future conditions is the use of counties as planning units. Most systematic conservation planning studies use regular units, such as squares or hexagons, which are typically smaller than Brazilian counties e.

This broad planning unit can also help to guide, with high return on investment, the allocation of limited resources for conservation. Our proposed protected areas network requires only ca. Studies have shown that planning protected areas network under current environmental conditions tends to become less effective in species representation under future climatic change conditions e.

Unfortunately, however, they do not provide results at the county level, hindering a comparison with our study and a direct guidance to policy makers. Our proposed protected areas network using endemic birds as the target biological model has counties selected throughout the Atlantic Forest, even when using the regional level conservation strategy Fig. The availability of proprietary counties throughout the entire Atlantic Forest may be interesting, in practical terms, because it creates a flexible conservation portfolio.

Conservation actions in Brazil are still often driven by Vale et al. Having priority counties spread throughout the biome, therefore, provides specific guidance wherever the opportunity of creating new protected areas arises.

Our results are intended to help decision makers by identifying priority counties in which conservation efforts should focus, i. After this first needed step, the specific location where each protected area should be created is contingent on other criteria and needs, such as landscape connectivity e.

The administrative level in which decision occurs county, state or federal may directly affect the effectiveness of protected areas network. Our networks were quite similar numerically at either county or regional level strategies, meaning that our results are useful at different administrative spheres. Most states within the Atlantic Forest, however, have yet to join the mosaic concept, and for those, the results of our county level strategy can be more relevant.

The greatest threat is habit loss and excessive hunting for meat. The Ethiopian siskin is locally referred to as the black-headed siskin or the Abyssinian Siskin.

It is a small bird measuring less than four inches long with a thin pointed beak. The males have the entire head and neck sooty black. Their natural habitat is montane forests and open grasslands. There are two spawning periods in a year; May-June and August-October.

They are territorial during the breeding period. The natural habitat of this bird is tropical and subtropical shrubland.

It favors dry areas, especially on rocky hillsides with thick patches of scrubs. It is in the Fringillidae family and prefers living in highland areas as opposed to lowlands. It is listed as a threatened species by the IUCN with habit loss being the leading threat.

The natural habitat of this bird is subtropical dry forests. It was first discovered by Prince Ruspoli in and later examined by T. Salvadori who named the bird Ruspoli in honor of its founder.