Cartography and Digital Mapping

Cartography has evolved from manual engraving of geospatial features to a fully digital discipline that integrates surveying, geographic information systems (GIS), remote sensing and computer graphics. At its core, traditional cartography used systematic field surveys and map‐making conventions to represent terrain, boundaries and thematic data on paper. Digital mapping extends these foundations by encoding spatial features as vector geometries and raster grids, managing them in databases and automating symbolisation, generalisation and labelling. Advances in satellite and aerial imagery, unmanned aerial vehicles and global navigation satellite systems have furnished cartographers with vast, up‐to‐date data sources. Modern workflows employ spatial analytics and machine‐learning methods to extract roads, buildings and land‐cover classes, while multi‐scale generalisation algorithms ensure legible representations across zoom levels. Web‐based mapping platforms and interoperable geospatial standards now permit real‐time map updates and custom map services. Together, these developments have transformed cartography from a manual craft into a flexible, data‐driven science capable of delivering tailored maps for navigation, planning, environmental monitoring and emergency response.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Graph-based deep‐learning approaches have been shown to excel in map generalisation tasks for vector data. One study formulated building footprint simplification as a joint node‐classification and node‐movement regression problem and proposed a multi‐task graph convolutional neural network (MT_GCNN) that learns both operations simultaneously. Applied to medium-scale generalisation of urban blocks, the method removed redundant vertices and adjusted remaining nodes to preserve shape and positional accuracy, achieving over 95% of buildings with sub-metre errors in urban test areas.

Ensuring visual continuity across zoom levels has motivated research on seamless multi‐scale map rendering. A recent algorithm precomputes optimal merging sequences of area objects using a greedy space–scale cube (SSC) framework. By recording the spatio-temporal progression of merges in a three-dimensional SSC and streaming it to the client GPU, the method delivers smooth animations between scales without discrete level-of-detail jumps. User studies report improved cognitive mapping and a natural transition experience compared with traditional tiling approaches.

Cartography and Digital Mapping publication trend

The graph below shows the total number of articles in cartography and digital mapping across all publications each year (not limited to Nature Index journals).

Technical terms

Cartography: The discipline of designing and producing maps, encompassing data acquisition, symbolisation and projection.

Digital Cartography: Cartographic processes conducted in a digital environment, involving spatial databases, automated symbolisation and computer graphics.

Geographic Information System (GIS): A software framework for storing, manipulating and analysing spatial and attribute data in map form.

Map Generalisation: The process of reducing map detail and simplifying features to maintain legibility at varying scales.

Graph Convolutional Neural Network (GCNN): A deep-learning model that applies convolution operations to graph-structured data, useful for vector geometry processing.

Space–Scale Cube (SSC): A precomputed data structure encoding spatial features across multiple scales to facilitate smooth zoom transitions.

References

1. Principles and Techniques of Cartography.
2. Move and remove: Multi-task learning for building simplification in vector maps with a graph convolutional neural network. ISPRS Journal of Photogrammetry and Remote Sensing (2023).
3. Generalizing Simultaneously to Support Smooth Zooming: Case Study of Merging Area Objects. Journal of Geovisualization and Spatial Analysis (2023).