The largest megalake in the geological record formed in Eurasia during the late Miocene, when the epicontinental Paratethys Sea became tectonically-trapped and disconnected from the global ocean. The megalake was characterized by several episodes of hydrological instability and partial desiccation, but the chronology, magnitude and impacts of these paleoenvironmental crises are poorly known. Our integrated stratigraphic study shows that the main desiccation episodes occurred between 9.75 and 7.65 million years ago. We identify four major regressions that correlate with aridification events, vegetation changes and faunal turnovers in large parts of Europe. Our paleogeographic reconstructions reveal that the Paratethys was profoundly transformed during regression episodes, losing ~ 1/3 of the water volume and ~ 70% of its surface during the most extreme events. The remaining water was stored in a central salt-lake and peripheral desalinated basins while vast regions (up to 1.75 million km2) became emergent land, suitable for development of forest-steppe landscapes. The partial megalake desiccations match with climate, food-web and landscape changes throughout Eurasia, although the exact triggers and mechanisms remain to be resolved.
Traditional methods for individual tree-crown (ITC) detection (image classification, segmentation, template matching, etc.) applied to very high-resolution remote sensing imagery have been shown to struggle in disparate landscape types or image resolutions due to scale problems and information complexity. Deep learning promised to overcome these shortcomings due to its superior performance and versatility, proven with reported detection rates of ~90%. However, such models still find their limits in transferability across study areas, because of different tree conditions (e.g., isolated trees vs. compact forests) and/or resolutions of the input data. This study introduces a highly replicable deep learning ensemble design for ITC detection and species classification based on the established single shot detector (SSD) model. The ensemble model design is based on varying the input data for the SSD models, coupled with a voting strategy for the output predictions. Very high-resolution unmanned aerial vehicles (UAV), aerial remote sensing imagery and elevation data are used in different combinations to test the performance of the ensemble models in three study sites with highly contrasting spatial patterns. The results show that ensemble models perform better than any single SSD model, regardless of the local tree conditions or image resolution. The detection performance and the accuracy rates improved by 3–18% with only as few as two participant single models, regardless of the study site. However, when more than two models were included, the performance of the ensemble models only improved slightly and even dropped.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.