New Generation of GIS

A new GIS paradigm is emerging in lieu of the development of cloud computing, big data, AI, Internet of Things, Smart Cities, etc.

Data Structures

  • New geographic coordinate system(s) – better integration between 2d and 3d, such as the Global grid systems. Compatible with OGC DGGS digital global grid standards abstract model.
  • Need to anticipate National Reference Frame transitions in 2021 that will build time (GPS epoch) into geographic coordinate specifications, led by NOAA.
  • Global temporal gazetteer – crowd source, integrate with existing regional gazetteers; Support streaming place name tagging for unstructured text
  • Place-based protocols or standards for machine learning
  • Need unified data structures (e.g. a triple store- for raster, vector and attribute data) that includes time stamps and temporal relations.

Computing Methodology

  • Integrated tools for ingesting multitemporal layers into GIS, handling time stamps and multiple image dates (4-5D)
  • Automatic creation of change statistics, tables, Markov matrices, etc.
  • Mining hidden information from big spatiotemporal data through a GIS tool
  • Workflows portable from desktop to cloud
  • Features s/b emergent, not prespecified, at least in some cases
  • Spatiotemporal interpolation in both space and time, scalable autocorrelation analysis
  • Event simulation and forecasting
  • Spatiotemporal analysis and visualization of block-chain events and systems.
  • Use of spatiotemporal blockchain events to verify location c.f. “Cryptospatial”. Could solve geomasking problem.

Tools

  • More robust tools for LiDAR, e.g. point-cloud to DSM/DTM/DEM; LIDAR tools need accurate coordinate registration, segmentation and classification/labeling tools. These are available across platforms, but not in any unified way.
  • Better handling of 3D features, follow 3D standards such as X3D, be able to define fully 3D objects in time, not just time slices. Link 3D objects to databases, so that objects can be searched and managed.
  • Add GIS and spatiotemporal analysis tools to open source platforms such as R
  • With ever increasing access to big data, retool GIS to allow users to do Python scripting without having to go offline to manage data.
  • Let me choose from a list of spatiotemporal statistics function to investigate my data from global weather observation stations.
  • Smoother platforms for syncing across multiple users, ArcGIS Online is ok, but still needs work.

Computational Efficiency

  • Robust capabilities for linking clouds, sharing resources (compute and/or data and/or code) between clouds
  • Parallel or distributed versions of the GIS geoprocessing or algorithms, such as merging, union, intersection, etc.
  • Put all my computers to be used by a GIS so it can both run faster and reuse my past investment
  • Sharing data as reusable services – enhance current WMS, WFS, etc.
  • Geoprocessing tools as a service (WPS), putting tools where data are, sharing analysis capabilities as service as well, such as spatial analysis methods, numerical models, simulation & prediction tools.
  • Make use of GPU resources to handle GIS various operations, particularly the rendering of large numbers of features quickly on a map

Time GIS (or GIS with scalable dimensions)

  • Robust capabilities for handling large continuous data streams – harvesting, database storage and processing, analytics, visualization, archiving.
  • Develop a typology of geospatial processes, e.g. diffusion, outward spread, infill, hierarchical diffusion, movement of static pattern, movement of developing pattern, etc.
  • Time integration into space and temporal dimensionality refined

Visualization

  • Spatial display tools, e.g. for creating Hagerstrand diagrams, running animations, selecting and displaying changes, etc.
  • Better scale-dependent displays, adaptive to human vision system and virtual devices