Mediterranean semi-arid landscapes are currently experiencing accelerated land use changes which are affecting species distributions. An important fraction of the European short-toed eagle (Circaetus gallicus) population, a specialised snake predator, breeds in this kind of landscapes. Information on the habitat characteristics that may affect the occurrence of this species is of paramount importance to try to evaluate how this species could be affected by ongoing territory changes. In this study, we use hierarchical partitioning analysis to identify environmental variables (natural vegetation, agricultural uses, hydrological web, topography, human pressure and climate) at two resolution scales (1 and 4 km 2 ) that may influence the presence of the short-toed eagle in a populated Mediterranean region in southeastern Spain. Results were highly consistent between scales. Forest was the most important variable determining the presence of the short-toed eagle in the study area, showing a positive effect, particularly those inland forests with nearby presence of sunny open scrublands where the species may find its prey. On the contrary, herbaceous crops and all variables related to human pressure, including urbanisations, dispersed houses and roads, are negatively affecting the occurrence of the species. These variables are strongly related to the main driving forces that change landscapes in our study area, including the abandonment of traditional territory uses and the expansion of urbanised areas. These results indicate that the increasing number of housing developments projected in inland areas would have a detrimental effect on the shorttoed eagle, especially if they are located close to pine forests and rivers.

Landscape metrics are used in a wide range of environmental studies such as land use change and land degradation studies, soil erosion and runoff predictions, management of hunting communities, and strategic planning for environmental management, to name but a few. Due to their utility for a variety of applications, there are many indices and software packages that have been designed to provide calculations and analysis of landscape structure patterns in categorical maps. With the purpose of making a profound comparison between the most used tools (Fragstats, V-Late, PA4…), we examined their advantages and disadvantages in order to create a list of common features that need to be incorporated into this type of software. We believe that an API without limitations on data input is necessary, capable of calculating vector or raster metrics and very extensible. This API should make it possible not only to build third party applications in easily, but would also make it possible to add new metrics and research into new paradigms related to traditional landscape metrics. We have started to develop a proposal based on open standards, which is FOSS. We have called this API Land-metrics DIY (Do It Yourself). It can calculate almost 40 landscape metrics from geometry provided by an ESRI Shapefile, but we are working to complete its contents as we explain in this article.

Hunting bags provide important information for conservation measures and wildlife management. This study is to assess relationships between landscape structure and game species. The community parameters (abundance, richness and diversity) and landscape/land use indices have been related, using GIS and statistical analysis, in the South-East of Spain (Marina Baja, Alicante). Game species richness (S) is determined by the presence of fruit groves (p = 0.001, R = 0.714) and landscape shape. The total density of species (TD) is influenced positively by fruit groves (p = 0.001, R = 0.783) and wooded shrublands (p = 0.002, R = 0.911), but is influenced negatively by urban areas (p < 0.001, R = 0.844). Small game communities correlate to irrigated fruit (p = 0.002, R = 0.754) and dry vineyard (p = 0.021, R = 0.839) and also with the diversity landscape index (p = 0.029, R = 0.708). Big game density is positively related to holm oak (p = 0.018, R = 0.812) and dense pine forests (p = 0.001, R = 0.849) and also with the total area landscape index (p = 0.011, R = 0.921). Population control species prefer irrigated fruit (p < 0.001, R = 0.775), fruit groves (p < 0.001, R = 0.857) and irrigated vineyard (p = 0.017, R = 0.833) land uses. Our conclusion is that most game species presents a positive relation with landscape structure, such as fractal dimension and shape index, and traditional agriculture based on irrigated and dry fruit crops.Additional key words: game community; GIS; hunting bags; Mediterranean agrosystem; semi-arid climate. ResumenPrincipales indicadores del paisaje que afectan a la comunidad de especies cinegéticas en un agroecosistema semiárido en la región mediterránea Los estadísticos de caza proporcionan información fundamental para implementar medidas de conservación y manejo de fauna. Este estudio pretende evaluar las relaciones entre la estructura del paisaje y las especies de caza. Se han relacionado los parámetros de la comunidad (abundancia, riqueza y diversidad), el paisaje y los índices de uso del suelo, usando SIG y análisis estadísticos, en el sureste de España (Marina Baja, Alicante). La riqueza de especies (S) está correlacionada positivamente con los frutales (p = 0,001, R = 0,714) y la forma del paisaje. La densidad total de especies (TD) está influenciada positivamente por los frutales (p = 0,001, R = 0,783) y el matorral arbolado (p = 0,002, R = 0,911), aunque influida negativamente por las zonas urbanas (p < 0,001, R = 0,844). Las especies de caza menor se correlacionan con el frutal de regadío (p = 0,002, R = 0,754), el viñedo de secano (p = 0,021, R = 0,839) y con el índice de diversidad del paisaje (p = 0,029, R = 0,708). La densidad de especies de caza mayor se relaciona positivamente con el encinar (p = 0,018, R = 0,812), el pinar denso (p = 0,001, R = 0,849) y con el índice de área total del paisaje (p = 0,011, R = 0,921). Las especies que requieren control de la población prefieren el frutal (p < 0,001, R = 0,775), la viña de regadío (p = 0,017, R = 0,833) y frutales de sec...

No abstract

No abstract