the application of geospatial technology to urban ... · the application of geospatial technology...

Urban Morphology (2007) 11(2), 81-90 © International Seminar on Urban Form, 2007 ISSN 1027-4278

The application of geospatial technology to urbanmorphological research

C. P. Lo Department of Geography, University of Georgia, Athens, GA 30602, U.S.A.

E-mail: [email protected]

Revised version received 31 May 2007

Abstract. This paper reviews the development of geospatial technology inrecent years and its use and potential use in urban morphology. Suchtechnology has been enabled by the rapid advancement in computer technologyin both hardware and software. Remote sensing, photogrammetry and GIS areparticularly important and suitable for use in urban morphological research.The availability of geospatial data in digital form has facilitated town-plananalysis and metrological analysis. Three-dimensional GIS allows the formand function of a town or city to be realistically visualized. Urban simulationis also possible with geospatial data, thus enhancing understanding of theprocesses of growth of a city and the associated built forms.

Key Words: GIS, photogrammetry, urban simulation, urban morphologicalresearch, remote sensing, GPS

Urban morphological research has a longhistory of development that can be traced asfar back as 1832, when Quatremere de Quincyused the town plan to understand the history ofa town’s development (Gauthiez, 2004). Thescope for research in urban morphology wasenhanced greatly when reliable topographicmaps and plans became available in theeighteenth century. In the British tradition,urban morphology developed in close relationto the study of the geography of settlements, inwhich urban settlements constitute a major part(Hudson, 1970). M. R. G. Conzen’s classicwork on Alnwick (1960) is characteristic ofthis approach: through town-plan analysis hetraced the historical development of the town,exemplifying the urban morphogenetictradition (Larkham, 2006; Whitehand, 2001).

Studies of both the form and function of acity can benefit from the very rapid advancesin geospatial technology in the past decade.The term geospatial technology is used here to

include the following: photogrammetry,remote sensing, Global Positioning System(GPS), cartography and Geographical Infor-mation Systems (GIS), all of which can also besubsumed under the term GeographicalInformation Science (GISc). The developmentof GISc in recent years has been revolution-ized by rapid advances in computer hardwareand software. In this paper, the application ofeach of these types of geotechnology will beexplained, with the aim of demonstrating theiruse and potential use in urban morphologicalresearch.

Photogrammetry

Photogrammetry is the use of aerial photo-graphs to extract reliable metric and semanticinformation about a place. Traditionally, it isemployed to produce topographic maps, as isdone by such government agencies as the

82 The application of geospatial technology

Figure 1. (A) An orthophoto image of the University of Georgia Campus,which can be matched exactly with (B) a large-scale topographic map

of the same area

The application of geospatial technology 83

Ordnance Survey (OS) in the United Kingdomand the Geological Survey in the United States(USGS). In recent years, photogrammetry hasevolved from the manual analogue approach,using analogue stereoplotters, to the semi-automatic digital approach using computerworkstations. For the digital approach to bepossible, analogue aerial photographs have tobe digitized to become raster data (a regulararray of grid cells of a specific size thatrepresents the spatial resolution of the data).This is achieved normally using photogram-metric scanners. Today, digital mappingcameras, albeit expensive, are also availableand can produce high quality digital images ofthe Earth’s surface directly for use in digitalphotogrammetric workstations. The digitalapproach allows terrain heights to be extractedautomatically from a stereomodel using stereo-correlation or image matching techniques. Inaddition to producing conventional topo-graphic maps, it can also generate the DigitalTerrain Model (DTM), in which the heights ofthe terrain are stored together with theirlocational co-ordinates. The DTM can also beconverted into a Digital Elevation Model(DEM) using spatial interpolation techniques

to produce a regular array of terrain heights,which can also be employed to generateorthorectified photographic images. An ortho-rectified image can be treated like a mapbecause all the tilts and relief displacementshave been removed. With the same geo-referenced co-ordinate system, a town map (indigital form) can be superimposed on top ofthe orthorectified image for comparison sideby side (Figure 1). In addition, DEM data canbe used to generate the terrain relief of a townto help in urban morphological analysis.Figure 2 illustrates the draping of a topo-graphic map over a DEM of the same area,thus providing visually a 3D perspective of thecity and its environment. One furtheradvantage is that photogrammetry can helpextract accurate heights of buildings.

Nearly 40 years ago, this author used ananalogue stereoplotter coupled with a panto-graph, called Persktomat, to produce manuallya three-dimensional map of a part of the citycentre of Glasgow to study its urban form(Figure 3) (Lo, 1970). Such a map can now beproduced much easier with the use of 3D GISand digital photogrammetry, to be discussedlater in the section on GIS.

Figure 2. A topographic map of the city of Athens, Georgia, USA, draped on the DEM.


Remote sensing

Remote sensing technology has advanced tomeet the increasing availability of digitalimage data from polar orbiting earth resourcessatellites launched since 1972, notably theLandsat series from the USGS. These imagedata have been used to produce land use andland cover maps with specific imageprocessing software, using computers.Initially, the spatial resolution of thesesatellites was rather low, ranging from 79 mfor Landsat MSS data to 30 m for Landsat TMdata. However, in 1986, a French companylaunched the first commercial satellite, calledSPOT, which has a much higher spatialresolution of 10 m for its panchromatic bandand 20 m for its multi-spectral bands, and hasthe capability of producing stereoscopicimages for photogrammetric application intopographic mapping. In 1999, an Americancompany launched a satellite called IKONOS,which has a spatial resolution of 1 m for itspanchromatic band, and 4 m for its multi-spectral bands. In 2001, another Americancompany launched another satellite, calledQuickBird, which is capable of imaging theearth at a spatial resolution of 0.61 m in the

panchromatic band and 2.44 m in its multi-spectral bands. Even SPOT has launched anew satellite in 2002 to compete with a spatialresolution of 2.5 m for its panchromatic bandand 10 m for its multi-spectral bands. Thespatial resolution of these new satellite data isso high that it is comparable to that of aerialphotography, but it has the advantage ofcovering a much larger area of the earth. Inaddition, like aerial photography, all thesesatellites can produce 3-D stereoscopic imagesso that digital photogrammetry can be appliedto produce topographic maps or DEMs. Thesevery high spatial resolution satellite data canhelp in research on urban form and town-plananalysis (Mesev, 2005). However, the cost ofthese ultra-high spatial resolution data fromthese commercial companies is very high, andmany researchers have to resort to the use ofLandsat data, which are much cheaper toacquire. The National Aeronautics and SpaceAdministration has also launched a relativelyhigh spatial resolution instrument on the Terrasatellite. The sensor known as ASTER, whichis capable of producing a 3-D stereomodel ata spatial resolution of 15 m for its three visibleand near infra-red bands, can be acquired atlow cost and is particularly suitable for urban

Figure 3. Three-dimensional map of part of the city centre of Glasgow.


form study (Hirano et al., 2003; Welch et al.,1998). Finally, satellite data have theadvantage of a good temporal resolutionbecause they record the earth at regularintervals through their orbital cycle. Thisallows changes in the city or town to bemapped. All these satellite images have beenused mainly for land use and land covermapping. A land use and land cover map of atown gives insights into the functional zones inthe town and their changes over time, allowingchanges in building forms and functions to berelated.

Despite the costs of the ultra-high spatialresolution satellite data from commercialvendors, the abundance of satellite images anddigital aerial photographs from governmentshas benefited the public through the WorldWide Web. This has allowed free access,notably, through Google Earth, which displaysany cities in the world on demand with quitedetailed images.

In recent years, a new form of remotesensing, known as LIDAR (Light Detectionand Ranging) has taken on an important role in3D terrain modelling. LIDAR is an activeremote sensing system that can collectthousands of elevation values per second withunprecedented detail and accuracy. It is anairborne system, which can be mounted onboard a helicopter or a fixed-wing aeroplane.In conjunction with the use of GlobalPositioning Systems (GPS – to be discussed inthe next section), accurately positioned data ofthe terrain in X, Y, and Z co-ordinates can beobtained to produce a Digital Surface Model(DSM). An accuracy of 15 cm can beachieved for bare-earth surface. LIDAR hasbeen successfully applied to capture data overcities with high building densities, andresearch has been conducted to extractbuildings in 3D form automatically from thedata (Madhavan et al., 2005; Vogtle andSteinle, 2005). These data provide valuableinsights to the third dimension of the city withan accurate record of the heights of thebuildings. An example of applying LIDAR toa city in the United States is shown in Figure4. The data can also be presented in the formof contours.

Global Positioning Systems

GPS is another major advancement in landsurveying that has facilitated topographicmapping. By receiving signals from four ormore satellites known as NAVSTAR(Navigation System with Time and Ranging)in space, the GPS receiver fixes the co-ordinates (X, Y and Z) of the position.NAVSTAR consists of 24 high-altitude c.26 600 km) satellites revolving around theearth. GPS makes use of the time arrival ofthe satellite signal to determine the distancebetween the satellite and the GPS unit, havingknown the speed of travel of the signal (a formof electromagnetic energy). GPS can beconducted using static, kinematic anddifferential approaches. The differentialapproach, which requires the use of two GPSunits, produces the most accurate results.

GPS has been used in conjunction with GISin town-plan analysis and metrologicalanalysis. Lilley et al. (2005) used differentialGPS for the field survey of the medievalEnglish town of Winchelsea. Observationswere taken at selected street intersections, plotboundaries, buildings, and earthworks, givingan accuracy of ±2 cm. From the field surveyand the measurements of plot and streetpatterns, units of measure used to plan amedieval town could be established. Thecollected GPS data were input to a GIS (forexample, ArcGIS – a standard GIS software),which was used to store, manipulate andanalyse the spatial data. The GPS elevationpoints were spatially interpolated using anearest neighbour algorithm, thus producing a3-D terrain surface. The GIS allows the thirddimension of the town to be incorporated. Byincorporating a 3-D function of GIS (notablythe 3D Analyst feature of ArcScene), digitizedurban features can be draped on the terrainsurface model. Thus the town can be relatedto its underlying topography, in a mannersimilar to that shown in Figure 2 using DEM.

Much of the work done with the GPS canbe done also by using stereo-photogrammetryand LIDAR, both of which are less labourintensive.


Cartography and Geographical Infor-mation Systems

Since the advent of computers, particularly inthe form of personal computers (PC),cartography has changed tremendously. Mapproduction, a tedious and time-consuming taskin the past, is now replaced by a much more

efficient digital approach by which the mapsare produced by computers. There are at leasttwo classes of maps: topographic and thematic.Topographic maps are no longer produced insuch large numbers in hard copy form. Thespatial data required for the production oftopographic maps are usually stored in thecomputer in a GIS, so that a topographic map

Figure 5. A 3-D map of the University of Georgia campus, produced using digitalphotogrammetry and the function of 3-D Analyst in ArcView GIS.

Figure 4. A LIDAR image of a city in California, USA. The heights of the buildings areshown as contours. The highest contours are shown in red and the lowest (at street level)

in blue. Reproduced by permission of Optimal Geomatics.


can be produced of a particular place ondemand. The availability of a digital base fora town or city is therefore very useful to urbanmorphologists. If old town plans can bedigitized, comparison between old and newplans can be easily achieved and some form ofquantitative analysis is possible (as in themetrological approach).

Cartography has evolved from being adiscipline producing static maps into adynamic computer-based GIS, capable ofstoring and manipulating geographical data tosolve spatial problems, in addition toproducing maps in hardcopy or softcopy form(Lo and Yeung, 2007). It consists of a suite ofspatial analytical functions, which can dealwith data in raster and vector formats. Thetopological function allows GIS to conductnetwork analysis and overlay, which are themost useful functions of GIS. Networkanalysis allows connectivity between featuresto be identified and when applied to a townplan will reveal how planners design the streetpatterns. Smeed (1968) developed an index tocompare the efficiency of the road networks.For example, a coarse grid pattern is moreefficient than a dense grid pattern, and a gridpattern is more efficient than an irregularpattern. The network analysis approach hasalso been modified to produce a space syntaxmeasure. This has been employed to studystreet patterns and pedestrian flows,characterizing the form of the city by thetopology of its street network (Batty, 2004;Jiang and Claramunt, 2002).

The overlay function in GIS through theapplication of intersect and union logics hasbeen used to combine different spatial layerstogether to pinpoint a specific location, such asan ideal site to develop a residential estate inthe city. In this way, different elements of thecity can also be separated. In a three-dimensional city, a polygon overlay can beused to represent different layers of thebuildings in the city, as is demonstrated by thework of Holtier et al. (2000), who used a GISsoftware known as Smallworld GIS torepresent the 3-D built forms of buildings infour English towns: Manchester, Swindon,

Tamworth, and Bury St Edmunds. All non-domestic buildings were surveyed within adefined sample area for this research.Ordnance Survey base maps were also used.The buildings were broken down into ‘floorpolygons’ with attribute information on floorlevel and storey height. Once this was set upin GIS, a number of data could be computedfor each floor polygon: size data andtopological data. The topological data refer tothe spatial relations among the floor polygons:such as, adjacency, around, above, and below.Therefore, the 3-D model of the city sodeveloped displays both form and function aswell as their spatial relationships.

The addition of the 3-D display function toGIS allows height data to be effectivelydisplayed. This is an especially important usein urban morphology of DEM data and thebuilding height data obtained from digitalphotogrammetry or from LIDAR. Figure 5illustrates a 3-D model of the campus of theUniversity of Georgia, generated using thefollowing types of geospatial technology: (1)digital photogrammetry with stereocorrelationto extract the heights of the buildings from apair of stereophotographs (in digital form); (2)a DEM of the campus (also produced fromstereocorrelation) to provide the topography ofthe terrain; (3) the creation of a 3-D modelusing the 3-D Analyst extension in theArcView GIS program; and (4) image drapingof DEM over the aerial photographic image.The 3-D Analyst also allows the whole 3-Dscene to be rotated for viewing at differentperspective angles. This example illustratesthat different types of geospatial technologycan be usefully utilized in combination toproduce a realistic three-dimensional map ofthe city for research. When compared toFigure 3, this example illustrates the greattechnological advance there has been inproducing a 3-D map of a city. Much of thisadvance can be credited to the automationbrought about in digital photogrammetry interrain data (X, Y and Z) extraction as well asGIS software development that makesvisualization in 3-D possible.


Urban simulation

The availability of powerful computers anddigital geospatial data on cities facilitates theuse of simulation to study city growth andexplore theories on how cities work andchange over time. One of the most popularmodels used for this purpose is the cellularautomata (CA). A cellular automaton, asoriginally introduced by von Neumann andUlam in the 1960s to model biological self-reproduction, consists of a lattice of cells, afinite set of states for the cells, a finiteneighbourhood, and a local transition function(Wolfram, 1994, p. 5). The cells in the latticedetermine the state of a CA, which evolves indiscrete time steps. The value of the variablein each cell is affected by the values ofvariables in neighbouring cells on the previoustime step. The values of these variables arecontinuously updated according to a definiteset of local transition rules. This originalconceptualization of CA has been adapted andmodified for use in urban growth modelling bygeographers, notably in the works of Clarke etal. (1997) and Batty et al. (1999). The CAmodel needs to be calibrated using historicalurban maps to show changes, and a set oftransition rules is developed to explain theprocesses that have driven the change. TheCA model is capable of dealing with changesin the form and function of the city. As such,it is particularly useful in urban morphologicalresearch on how the city evolves with time.The transition rules developed will be basedon an understanding of settlement geography,notably site and situation and economicactivities. Once the calibration of the CAmodel is complete, the growth of the city canthen be simulated into the future. CA modelsoftware is freely available. One of the mostpopular software packages is called SLEUTH,developed by Clarke (Clarke et al., 1997) withthe funding support of the USGS. SLEUTHderives its name from the six types of datainputs: Slope, Land cover, Exclusion, Urbanextent, Transportation, and Hillshade, all ofwhich are geospatial data obtainable fromsatellite image data, with the aid of GIS. Theresults of the simulation can be stored in a GIS

database and displayed as a series of time-sequential maps, which can be shown as aform of movie (animation). CA simulation hasbeen applied to many cities in different partsof the world, including cities in the UnitedStates (Baltimore-Washington DC by Jantz etal. (2003), Atlanta by Yang and Lo (2003),Sioux Falls (Goldstein, 2004), and Houston(Oguz et al., 2004)), Portugal (Lisbon andPorto by Silva and Clarke (2002)), Australia(Sydney by Liu and Phinn (2004) and China(Dongguan by Yeh and Li (2001) and Wuhanby Cheng and Masser (2003)).

Conclusions

This brief survey has illustrated the types ofgeospatial technology that can be applied inurban morphological research. The transform-ation of maps into digital form has opened upnew areas of research on town-plan analysis,such as the use of GIS to carry out networkanalysis of the street pattern. It also facilitatesmetrological analysis. The three-dimensionalform of a town or city as well as the variousfunctions of buildings can be readily studiedby the use of GIS. The ability of GIS todisplay maps and perform spatial analysismakes it an indispensable tool for research inurban morphology. All these advances ingeospatial technology have been stimulated bythe advances in computer technology in thepast decade, notably increasingly powerfulPCs at lower costs in recent years. Geospatialtechnology now provides data that allow urbansimulations to be carried out, and the popularCA model is particularly appropriate for useby urban morphologists to simulate urbangrowth. Geospatial technology, in combin-ation with computers, is transforming thetechniques that are available for research inurban morphology.

Acknowledgements

The author wishes to acknowledge the help givenby Mr Larry Lund of Optimal Geomatics ofHuntsville, Alabama, USA in the production of theLIDAR image of the city in Figure 4. The image


was produced from LIDAR data acquired byOptimal Geomatics, which kindly permits its use inthis paper.

References

Batty, M. (2004) ‘A new theory of space syntax’,Working Paper Series, Centre for AdvancedSpatial Analysis, University College London.

Batty, M., Xie, Y. and Sun, Z. (1999) ‘Modelingurban dynamics through GIS-based cellularautomata’, Computers, Environment and UrbanSystems 23, 205-33.

Cheng, J. and Masser, I. (2003) ‘Urban growthmodeling: a case study of Wuhan city, PRChina’, Landscape and Urban Planning 62,199-217.

Clarke, K. C., Hoppen, S. and Gaydos, L. (1997)‘A self-modifying cellular automaton model ofhistorical urbanization in the San Francisco Bayarea’, Environment and Planning B: Planningand Design 24, 247-61.

Conzen, M. R. G. (1960) Alnwick, Northumber-land: a study in town-plan analysis Institute ofBritish Geographers, Publication 27 (GeorgePhilip, London).

Gauthiez, B. (2004) ‘The history of urban morph-ology’, Urban Morphology 8, 71-89.

Goldstein, N. C. (2004) ‘Brains versus brawn –comparative strategies for the calibration of acellular automata-based urban growth model’,in Atkinson, P., Foody, G., Darby, S. and Wu, F.(eds) GeoDynamics (CRC Press, Boca Raton,FL) 249-72.

Hirano, A., Welch, R. and Lang, H. (2003)‘Mapping from ASTER stereo image data: DEMvalidation and accuracy assessment’, ISPRSJournal of Photogrammetry and Remote Sensing57, 356-70.

Holtier, S., Steadman, J. P. and Smith, M. G.(2000) ‘Three-dimensional representation ofurban built form in a GIS’, Environment andPlanning B: Planning and Design 27, 51-72.

Hudson, F. S. (1970) A geography of settlements(MacDonald and Evans, London).

Jantz. C. A., Goetz, S. J. and Shelley, M. K. (2003)‘Using the SLEUTH urban growth model tosimulate the impacts of future policy scenarioson urban land use in the Baltimore-Washingtonmetropolitan area’, Environment and PlanningB: Planning and Design 30, 251-71.

Jiang, B. and Claramunt, C. (2002) ‘Integration ofspace syntax into GIS: new perspectives forurban morphology’, Transactions in GIS 6,295-309.

Larkham, P. J. (2006) ‘The study of urban form inGreat Britain’, Urban Morphology 10, 117-41.

Lilley, K., Lloyd, C., Trick, S. and Graham, C.(2005) ‘Mapping and analysing medieval builtform using GPS and GIS’, Urban Morphology9, 5-15.

Liu, Y. and Phinn, S. R. (2004) ‘Mapping the urbandevelopment of Sydney (1971-1996) withcellular automata in a GIS environment’,Journal of Spatial Science 49, 57-74.

Lo, C. P. (1970) ‘Determining and presenting thethird dimension of a city center: a photo-grammetric approach’, Photogrammetric Record6, 625-39.

Lo, C. P. and Yeung, A. K. W. (2007) Conceptsand techniques of geographic informationsystems 2nd edn (Prentice Hall, Upper SaddleRiver, NJ).

Madhavan, B. B., Wang, C., Tanahashi, H., Hirayu,H., Niwa, Y., Kamamoto, K., Tachibana, K.,and Sasagawa, T. (2005) ‘A computer visionbased approach for 3D building modeling ofairborne laser scanner DSM data’, Computers,Environment and Urban Systems 30, 54-77.

Mesev, V. (2005) ‘Identification and character-ization of urban building patterns usingIKONOS imagery and point-based postal data’,Computers, Environment and Urban Systems29, 541-57.

Oguz, H., Klein, A. and Srinivasan, R. (2004)‘Modeling urban growth and landuse andlandcover change in the Houston metropolitanarea from 2002 to 2030’, Proceedings of theASPRS 2004 Fall Conference, Kansas City, MO.(American Society for Photogrammetry andRemote Sensing, Bethesda, MD).

Silva, E. A. and Clarke, K. C. (2002) ‘Calibrationof the SLEUTH urban growth model for Lisbonand Porto, Portugal’, Computers, Environmentand Urban System 26, 525-52.

Vogtle, T. and Steinle, E. (2005) ‘Fusion of 3Dbuilding models derived from first and last pulselaserscanning data’, Information Fusion 6, 275-81.

Welch, R., Jordan, T., Lang, H. and Murakami, H.(1998) ‘ASTER as a source for topographic datain the late 1900s’, IEEE Transactions onGeoscience and Remote Sensing 36, 1282-9.

Whitehand, J. W. R. (2001) ‘British urban morph-ology: the Conzenian tradition’, Urban Morph-ology 5, 103-9.

Wolfram, S. (1994) Cellular automata andcomplexity: collected papers (Addison-WesleyPublishing, Reading, MA).

Yang, X. and Lo, C. P. (2003) ‘Modelling urban


growth and landscape change in the Atlantametropolitan area’, International Journal ofGeographical Information Science 17, 463-88.

Yeh, A. G. O. and Li, X. (2001) ‘A constrained CA

model for the simulation and planning ofsustainable urban forms by using GIS’,Environment and Planning B: Planning andDesign 28, 733-53.

Urban morphological research at SCUT

The Urban Morphology Research Group (UMRG)at South China University of Technology inGuangzhou was established in September 2006.Professor Yinsheng Tian, three AssociateProfessors and four Lecturers in the Department ofArchitecture are the principal members of theGroup, which is located in the new Liwu ScienceBuilding. The principal funding sources aregovernment agencies. With the support of theChinese National Natural Science Foundation, theUMRG has completed two research projects: on‘The theory and practice of traditional development

in urban residential environments’ and ‘Thedevelopment of the spatial structure of the urbanrecreational system’. Researchers from both Chinaand overseas have contributed to the regularseminar series. The UMRG has established a closeresearch relationship with its namesake in theSchool of Geography, Earth and EnvironmentalSciences in the University of Birmingham, UK. Aresearch student will be spending 2 years in theUMRG at Birmingham, starting in November 2007.The UMRG at SCUT will be hosting ISUF 2009 inGuangzhou.

Society of Architectural Historians

The Society of Architectural Historians will hold itsSixty-First Annual Meeting from 23 to 27 April2008, in Cincinnati, Ohio, USA. Among the topicson which seminars will be held are:

• Identities in Greek and Roman architecture• Inter-disciplinary challenges in pre-modern

architectural history• Architecture and engineering: inter-disciplinary

contributions to architecture, 1946-2006• Entwined perspectives for the construction of

the colonized land: Asia before the Second

World War• The Muslim city: continuity and change• Spanish and Portuguese colonial architecture

and urbanism on the fringes• Modernist architecture in Africa• Bourgeois and middle-class Modernism in

German architecture, c. 1900 to 1940• Italian Modernism and the persistence of

tradition• Third World Modernism

Further information is available from www.sah.org

Journal of Urbanism

The new Journal of Urbanism: InternationalResearch on Placemaking and Urban Sustainabilityis being launched by Taylor & Francis in April2008. This multi-disciplinary journal will focus onresearch on the design of the built environmentwithin social, economic and environmentalcontexts. It is concerned with not only the form,pattern and design of human settlements but theirrelation to notions of sustainability, social justiceand cultural understanding. There will be three

issues per year, consisting of peer-reviewed articlesonly. The co-editors are Charles C. Bohl, School ofArchitecture, University of Miami; Matthew Hardy,Secretary of INTBAU; and Emily Talen, School ofPublic Affairs, Arizona State University. Papersand enquiries should be submitted [email protected]. The personal sub-scription is 74$ ( £38 or 59 Euros) per year. Theinstitutional subscription is 417$ (£214 or 334Euros) per year.

the application of geospatial technology to urban ... · the application of geospatial technology...

Documents