geomorphometry.org

Multiscale mapping provides an advanced method for visualizing and studying geographic phenomena via automated scale-dependent adjustment of image detailization. This kind of interactive flexibility makes analysis of complex multilevel objects and their relationships handier. Despite the fact that Earth topography has strongly pronounced hierarchical nature, its representation on multiscale maps has not been elaborated. Most of allied scientific knowledge has been worked out in multiscale cartographic representations of non-surface objects, such as hydrography, city points, buildings, roads and so on. A lot of closely related research provided valuable results in digital (non-visual) multiscale surface representations, i.e. hierarchical/pyramidal models derivation, DTM generalization and so on. In this work a methodology of multiscale hypsometric mapping is proposed, which covers digital and visual contexts of multiple relief representations to an equal degree. Initially, scale-dependent features of relief are analyzed and three main rules of its multiscale cartographic representation are formulated, which include scale-correspondent representation of relief forms, intensive generalization in scales of smaller than 1:100 000 and gradual change of symbology and detailisation with scale. Map scales and projection are defined with respect to levels of detail needed and area of interest. A scale sequence of 2-2.5-times multiplication is substantiated. Database logical structure is developed to supply Levels of Detail (LoDs) for scales of the sequence chosen. A new algorithm for DEM generalization is proposed which is based upon stream network and watersheds generalization and subsequent triangulation of simplified framework. The resulting DEM is then processed via high and low quartile filters to widen valleys and watersheds for their better readability. Testing of algorithm in different levels of detail and upon various morphological relief types is made and sample images are provided with their logical analysis revealing some advances and shortcomings of new algorithm. Then multiscale map preparation using hypsometric tints, contours and hill shading is discussed. The proposed methodology is demonstrated on creating multiscale hypsometric map of European part of Russia with scale range of 1:25 000 – 1:50 000 000. Results and perspectives of future developments are outlined in the concluding part.