Patterns of coastal dune vegetation are closely related with soil conditions controlled by geomorphic forms and processes. This study developed a conceptual model integrating these relationships in a spatially explicit manner. A rectangle of 180 9 280 m containing 126 grids of 20 9 20 m was established in the Sindu coastal dunefield in west Korean Peninsula. Sampling from each grid determined 11 soil properties and identified percent cover of 21 woody and herbaceous plant species. Digital elevation models were generated by topographic survey and used to derive eight topographic parameters. Redundancy analysis and canonical correspondence analysis examined the effect of geomorphic factors on edaphic characteristics and the edaphic influence on spatial distribution of vegetation, respectively. The spatial pattern of soil properties and plant species were inferred from spatial interpolation techniques. In the foredune area, distance from the coastline was a significant indicator of soil nutrients derived from the marine sources by aeolian processes. This favored the dominance by Elymus mollis. Moisture-tolerant species (e.g., Calamagrostis epigeios) had high cover in the acidic soils of dune slacks, which covaried with wetness index, an indirect measure of the depth to the freshwater table. Vegetation-soil interactions (e.g., nitrogen fixation by legumes) were important in secondary dune areas, with topographic effects less significant. Vegetation, soil, and geomorphic factors are closely connected in a causal chain across a whole dune area. Our model thus addresses the importance of integrating foredune, dune slack, and secondary dune into one continuous system.
The efficient extraction of road information is increasingly important with the rapid growth of road-related services, such as car navigation systems, telematics, and location-based services. Conventional methods of creating and updating road information are expensive and time consuming. Therefore, a set of processes is required that collects the same information more efficiently. We propose a new method for collecting road information in complex urban areas from road pavement markings located on aerial images. This information includes lane and symbol markings that guide direction; the geometric properties of the pavement markings and their spatial relationships are analyzed. Road construction manuals and a series of cutting-edge techniques, including template matching, are used in our analysis. To validate our approach, the accuracy of our results was evaluated by comparing the data with manually extracted ground truth data. Our approach demonstrates that road information can be extracted efficiently to an extent in a complex urban area.Résumé : Il est de plus en plus important d'extraire efficacement des informations routières lorsque l'on tient compte de la croissance rapide des services touchant les routes, soit les systèmes de navigation automobile, la télématique et les services offerts par téléphone mobile. Les méthodes conventionnelles de création et de mise à jour de l'information routière sont dispendieuses et gruge-temps. Un ensemble de procédés est donc requis pour colliger la même information plus efficacement. Nous proposons une nouvelle méthode de colliger l'information routière dans les zones urbaines complexes à partir de marques sur la chaussée visibles sur des images aériennes. Cette information comprend des marques des voies et de symboles qui guident la direction; les propriétés géométriques des marques de chaussée et leurs relations spatiales sont analysées. Notre analyse utilise des manuels de construction routière ainsi qu'une série de techniques d'avant-garde, dont l'appariement par forme de référence. Afin de valider notre approche, la précision de nos résultats a été évaluée en comparant les données aux données extraites manuellement sur le terrain. Notre approche dé-montre que l'information routière peut être extraite efficacement jusqu'à un certain point dans un contexte de zone urbaine complexe.Mots clés : image aérienne, extraction automatique, marques sur la chaussée, information routière, système de navigation automobile.[Traduit par la Rédaction] Kim et al. 1331
Orbital‐ and millennial‐scale climate and vegetation changes between 32.5 and 6.9k cal a BP from Hanon Maar paleolake on Jeju Island, Korea, were demonstrated by high‐resolution multi‐proxy data (pollen, magnetic susceptibility, total organic carbon, carbon/nitrogen ratios and sediment grain size). High‐resolution pollen records provided detailed information on the vegetation response to climate change. The Thalictrum pollen percentage, which corresponded to the magnetic susceptibility data, and the first principal component of principal‐component analysis of pollen data were good indicators of the orbital‐ and millennial‐scale dry/wet cycle in the study area. Hanon pollen and magnetic susceptibility records suggested that colder and drier conditions were prevalent between 25.2 and 17.6k cal a BP in the study area due to a precession‐induced decrease in summer insolation. Hanon multi‐proxy data were correlated with oxygen isotope records from Greenland and Hulu cave, suggesting the possible teleconnection of millennial‐scale climate oscillations between North Atlantic and East Asian regions. According to our results, Heinrich‐like cold events probably occurred around 31.1, 25.2 and 16.7k cal a BP, and the warm intervals observed around 30.7 and 29.9k cal a BP may have corresponded to Dansgaard–Oeschger warm events 4 and 3, respectively.
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.