MODELLING: GIS
making this capability accessible directly online. It’s fascinating to see the kinds of GIS applications people in all fields have developed (see [1]). The figure from the Amazon Initiative (see page 32), for example, allows the user to select a region and then it predicts deforestation over time.
Statistical importance
Statistics as a science has also become an important aspect of GIS, and packages such as ArcGIS offer a wide arsenal of tools [2]. Traditional statistics don’t include both space and time, so they don’t tell the entire story. Spatial statistics use area, proximity and direction to identify structure and trends in the data. The nature of geographic data contradicts the basic assumptions of statistical analysis that each observation is equally likely to occur in a sample and that observations don’t influence each other. However, most geographic data exhibits both regional and local trends, so features or values are not equally likely to occur at any location. Sometimes, drawing conclusions from a map
isn’t easy. Statistics tools can cut through the map display and get directly to the patterns and relationships in the data – where is something happening and how widespread is it? Spatial statistics opens up a new set of questions you can answer: how sure are you that the pattern isn’t simply due to a random occurrence? To what extent does the value of a feature depend on the values of surrounding features? How well does the value of one attribute predict the value of another? What are the trends? Another aspect of the IT part of GIS is transforming GIS information from one state to another. Examples are reprojections, raster/ vector conversions and polygon overlays, but in any such transformation information is lost. The fitness for purpose of information can only be determined if the history of the information is known or if appropriate measures of accuracy are recorded.
Looking for differentiators
Besides ESRI, there are a number of other suppliers of GIS software. Given the dominance of ESRI, these companies generally focus on niche markets such as utilities or telecom applications or are addressing the market by driving down prices. One of the very first desktop GIS packages
came from MapInfo, which was acquired in 2007 and now operates under the name Pitney
34
are being collected and made available, and more detailed analysis is required. Satellite photos once available only to intelligence analysts are now viewable by anyone with an internet connection. In addition, satellites now image the Earth using multiple sensors that collect data in multiple wavelengths and resolutions and from many different angles. Without GPU acceleration, adds Rostov, GIS users would need to settle for lower-resolution data, which can miss subtle terrain artifacts – for an archaeologist, for example, this lower resolution could mean finding the outline of a buried fortress, but not the individual walls, making a dig problematic and costly. Within the GIS area, users can also choose
An Oracle Spatial mapping of the KEGG Pathway database.
Bowes Business Insight. The firm specialises in commercial applications in what it calls location intelligence to help companies identify growth opportunities, such as choosing profitable locations for expansion and streamlining delivery routes. Very aggressive with pricing is Manifold,
whose base package costs $295; the Professional Edition adds IMS (internet map service), while the Enterprise Edition adds spatial DBMS capability. Product manager Dimitri Rotow explains that Manifold is an integrated system with its own database, and he compares it to combining a database program with Illustrator, AutoCAD, Excel, Visual Studio and a web facility, all with a slant on GIS functionality. You can think of it as a ‘word processor for maps’ for creating new maps or editing existing ones; meanwhile, it’s also a ‘database system for maps’, so a map can be used as a visual interface into a large dataset. Besides price, another interesting aspect of
Manifold is that it’s the first GIS product to leverage the speed of GPGPUs from Nvidia. The current release of Manifold provides almost three dozen CUDA-enabled functions for computations within the optional Surface Tools extension. Rostow reports that with the CUDA configuration, calculations that previously took 20 minutes are now done in 30 seconds, and calculations that previously took 30 to 40 seconds are now real-time. One reason for the increasing complexity
of GIS datasets that can benefit from GPU acceleration is that massive amounts of data
SCIENTIFIC COMPUTING WORLD APRIL/MAY 2010
from many open-source packages. Professor Maguire suggests GRASS (Geographic Resources Analysis Support System). Developed by the US Army, it has evolved into a utility with a wide range of applications in scientific research. GRASS contains more than 350 utilities and tools to render maps and images; manipulate raster, vector, and site data; process multi spectral image data; and create, manage, and store spatial data. It has both a command line and a graphical user interface.
Unconventional maps, too
Maps based on spatial data need not be geographically related. Oracle Spatial product manager Xavier Lopez notes that it has broken boundaries with traditional GIS technologies, and that the product is not limited to earth mappings. You can also work with non- geographical systems, whether it be celestial or cellular mapping, or biological pathway analysis as in the figure, above left. It is an Oracle Spatial mapping of the KEGG Pathway database, which is a collection of maps representing molecular pathways for metabolism, genetic information processing, environmental information processing, other cellular processes, human diseases and drug development. This bioinformatics resource is part of projects of the Kanehisa Laboratories in the Bioinformatics Center of Kyoto University and the Human Genome Center of the University of Tokyo.
References
1. A sampling of user-submitted GIS applications: www.esri.com/software/arcgis/ arcgiasserver/live-user-sites.html
2. Mitchell, A., The ESRI Guide to GIS Analysis Volume 2: Spatial Measurements and Statistics,
ESRI Press, 2009.
www.scientific-computing.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44