Search

Experimental approaches to measure territory defence can be used to interpret the relative magnitude of intraspecific aggression in different ecological contexts. The aim of this study is to examine the response of Fijian forest bird species to conspecific intruders using playback of previously recorded songs along an elevational gradient. We address two questions: Do different foraging guilds in Fiji's forest birds show different patterns of territorial defence behaviour towards conspecific intruders? Secondly, do Fiji's forest birds show stronger territorial defence at higher elevations? We explore these questions across two foraging guilds (insectivore and omnivore) on Viti Levu Island, Fiji. All four focal species had a stronger response at the highest elevation, and the endemic insectivore had the strongest response. Perhaps high elevation insects are less abundant and hence increased aggressiveness is favoured to defend the territory, and/or perhaps in general more aggressive birds colonise higher elevation areas, both ideas that require further testing. This study provides first insights into response to conspecifics in Fiji's forest birds that can be explored in relation to avian behaviour also in the context of human activity, human disturbance and threats to the persistence of birds across elevational gradients. [ABSTRACT FROM AUTHOR]

Islands in the tropical Pacific Ocean are renowned for high biodiversity and endemism despite having relatively small landmasses. However, our knowledge of how this biodiversity is formed remains limited. The taxon cycle, where well‐dispersed, earlier colonizers become displaced from coastal to inland habitats by new waves of colonizers, producing isolated, range‐restricted species, has been proposed to explain current biodiversity patterns. Here, we integrate the outcomes of phylogenetic studies in the region to investigate the sources, age, number of colonizations, and diversification of 16 archipelagos in the tropical and subtropical South Pacific. We then evaluate whether the results support the taxon cycle as a plausible mechanism for these observations. We find that most species in the Pacific arrived less than 5 Mya from geographically close sources, suggesting that colonization by new taxa is a frequent and ongoing process. Therefore, our findings are broadly consistent with the theory of the Taxon Cycle, which posits that ongoing colonization results in the gradual displacement of established lineages. Only the oldest archipelagos, New Caledonia and Fiji, do not conform to this trend, having proportionally less recent colonization events, suggesting that the taxon cycle may slow on older islands. This conclusion is further validated by New Caledonia having lower diversification rate estimates than younger islands. We found that diversification rates across archipelagos are negatively correlated with area and age. Therefore, a taxon cycle that slows with island age appears to be a suitable concept for understanding the dynamic nature and biodiversity patterns of the Pacific Islands. [ABSTRACT FROM AUTHOR]

Landslides, resulting from disturbances in slope equilibrium, pose a significant threat to landscapes, infrastructure, and human life. Triggered by factors such as intense precipitation, seismic activities, or volcanic eruptions, these events can cause extensive damage and endanger nearby communities. A comprehensive understanding of landslide characteristics, including spatio-temporal patterns, dimensions, and morphology, is vital for effective landslide disaster management. Existing remote sensing approaches mostly use either optical or synthetic aperture radar sensors. Integrating information from both these types of sensors promises greater accuracy for identifying and locating landslides. This study proposes a novel approach, the ML-LaDeCORsat (Machine Learning-based coseismic Landslide Detection using Combined Optical and Radar Satellite Imagery), that integrates freely available Sentinel-1, Palsar-2, and Sentinel-2 imagery data in Google Earth Engine (GEE). The approach also integrates relevant spectral indices and suitable bands used in a machine learning-based classification of coseismic landslides. The approach includes a robust and reproducible training and validation strategy and allows one to choose between five classifiers (CART, Random Forest, GTB, SVM, and Naive Bayes). Using landslides from four different earthquake case studies, we demonstrate the superiority of our approach over existing solutions in coseismic landslide identification and localization, providing a GTB-based detection accuracy of 87–92%. ML-LaDeCORsat can be adapted to other landslide events (GEE script is provided). Transfer learning experiments proved that our model can be applied to other coseismic landslide events without the need for additional training data. Our novel approach therefore facilitates quick and reliable identification of coseismic landslides, highlighting its potential to contribute towards more effective disaster management. [ABSTRACT FROM AUTHOR]