| Popis: |
In this work it is address the question of how certain climatic variables may be significant related to alterations of avian biodiversity in a semi-agricultural Natura wetland side in Northern Greece. Particularly, the current research highlights the effects of climate and land cover intensity on the Thermaikos gulf bird biodiversity and its importance for healthy ecosystem functioning. Also, the maintenance of a good state of conservation in the Thermaikos gulf has direct impacts on a larger scale since it benefits the rest of the Natura wetlands network considering the connectivity related to migratory birds. Furthermore, the methodology which is used is essential to help inform the science-based management of environments that support threatened and endangered wildlife and can be further applied to other wetlands in the Mediterranean with similar weather conditions and agricultural land use. The alteration in compositional diversity of bird abundances has been studied at the species level from 2012 to 2017 in one of the most important wetland Natura sites in Northern Greece and by using different biodiversity indices. Shannon Entropy was lower during 2012 (DH = 1.509) albeit remained in similar levels from 2013 and afterwards. The highest values of Shannon Entropy were recorded in 2014 (DH = 2.927) and 2016 (DH = 2.888) suggesting that there is a higher diversity compared to the other observation years and especially 2012. The yearly trends of the Simpson dominance index and the Gini-Simpson Index had quite similar patterns. The Berger-Parker index, DD, which represents the maximum proportion of any species estimated in the sample assemblage, had its highest values in 2012 (DD = 0.58) and 2017 (DD = 0.39) and its lowest in 2014 (DD = 0.13) and 2016 (DD = 0.15). A complete characterization of diversity was possible through the projection of Hill numbers and the Rényi entropy, parameterized by the order q in terms of an empirical curve. According to the Hills numbers pooled over the years, the mean species abundance (q = 0) was estimated at 31 species, the mean biodiversity (q = 1) was 13 species and the most dominant species (q = 2) were 8 species. The quantification of bird biodiversity in the particular research area patterns is a fundamental task to evaluate current management actions, improve conservation and design future management strategies. Moreover, the interplay between temperature, relative humidity and three different bird biodiversity indexes, including Shannon Entropy, Simpson’s dominance (evenness) index and the Berger-Parker index has been also examined. By using different modeling approaches, parametric and non- parametric multivariate models, we make effort to get a consensus on the interrelationships between climate and avian biodiversity. In particular, it is been shown that in most cases nonlinear models and surface-plot analysis methodology, are able to capture the relation of a considerable increase in the estimated biodiversity indexes with increased temperatures and rain levels. Thus, biodiversity is to a significant extent affected by the aforementioned climate factors at a proximate level involving synergies between the different climate factors. Finally, the combined effect of climate variables and remote sensing land cover indicators on bird richness has been also explored to detect any influence on bird diversity due to agricultural intensification. In particular the association between bird richness and environmental drivers, as well as remote sensed land cover indices was explored for seven successive seasons using correlation analysis and a Cox-Box transformed multivariate linear model. Three climate variables were tested: mean temperature, rain level and mean relative humidity and three land cover indices: the Normalized Difference Vegetation Index (NDVI), the Atmospheric Resistance Vegetation Index (ARVI) and an Agricultural Band Combination Index (ABCI). Among the environmental drivers explored, temperature, rain levels and ABCI were significantly correlated to bird richness in contrast to NDVI and ARVI which showed a lower correlation, while relative humidity displayed the poorest correlation. Additionally, the multivariable linear model indicates that temperature, rain levels and ABCI have a statistically significant effect (p |
