485 
In: Wagner W., Szekely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
the ontology classes allowed for more flexibility than some of 
the very specific NLUD classes. 
Land Use was a much more complicated subject. The link 
between the NLUD classification and the classes in the 
ontology did not present major problems. However, when it 
came to the actual classification using the data available there 
were a high number of polygons with unknown land use. Also, 
the diverse nature of land use meant that some polygons had 
several potential numbers of uses which could complicate the 
management and display of information for those using a 
detailed classification. Granularity is therefore an issue. 
There was a clear aid from aerial photography to distinguish 
residential areas from other that could be categorised as 
commercial or industrial buildings. It also helped to identify 
uses such as hospitals or schools and verify some land covers 
such as sand or scrub. However this assessment was done 
manually. More research is needed to investigate the possibility 
of an automatic way of identifying these areas. 
3.3 Discussion and further work 
The results from the trial in Bournemouth highlighted the 
following issues: 
- Need to asses other data sources. There are a few sources of 
ancillary data that could help complement the information that 
OS already hold. Some of them could come from directories 
such as Thomson, Yell, Experian and Valuation Office. One of 
the problems with these directories is that their information is 
generally not georeferenced. A test matching some of these 
directories to references on the ground achieved a 60% success. 
However it proved to be very time consuming. 
- Need to asses other methods of data capture, (i) Web 
harvesting. We performed a trial on web harvesting to 
complement and validate the information hold by Ordnance 
Survey. The first obstacle are legal constrains on use of the 
data which meant 80% of all sites could not be harvested. 
Furthermore, it is usually difficult to find out the currency of 
data provided in the web. The conclusion from this study was 
that automatic web harvesting is not suitable for land cover and 
land use information, (ii) Other sources of data capture to 
consider are surveyors and video surveying. Surveyors are used 
daily by Ordnance Survey to record changes in OS Mastermap 
layers. We have to explore the way of taking advantage of this 
resource so that they record land cover and land use information 
at the same time. A video capture trial is planned for this 
summer (iii) Automatic capture of land uses using aerial 
photography. The main objective of this assessment would be to 
determine what can be classified and in which way this 
information could complement that collected through other 
sources. 
- Need for user feedback. We need to know whether this 
approach answers any of the current of future land cover and 
land use user’s needs and could be a way of integrating 
different datasets. 
In conclusion, the design of an object-oriented ontology 
framework for land use and land cover that provides a high 
level classification seems like the right approach in order to 
provide a starting point and place of agreement for different 
users. However, there is still a lot of work to do in terms of data 
capture, data storage and maintenance, linkage with other 
widely used classifications and the understanding of user needs. 
References 
Fonseca, F., Egenhofer, M., Davis Jr. C.A., Borges, K.A.V. 
2000, Ontologies and Knowledge Sharing in Urban GIS, 
Computer, Environment and Urban Systems, 24, 251-271 
Di Gregorio, A. and Jansen, L.J.M, 2000, Land Cover 
Classification System (LCCS): Classification Concepts and user 
manual, FAO 
Eurostat (2000): Manual of Concepts on Land Cover and land 
Use Information Systems. Theme 5: Agriculture and Fisheries: 
Methods and Nomenclatures. Office for official Publications of 
the European Communities, Luxembourg, 
2000,110 pp 
Ldfvenhaft, K., Bjorn, C. and Ihse, M., 2002. Biotope patterns 
in urban areas: a conceptual model integrating biodiversity 
issues in spatial planning. Landscape and Urban Planning, 58 
(2/4), 223-240. 
MasterMap user manual v.2, 2006, Ordnance Survey 
Nunes, J., 1991, Geographic Space as a Set of Concrete 
Geographical Entities, in : Mark, D. and Frank, A., (eds), 
Cognitive and Linguistic Aspects of Geographic Space, pp. 9- 
33, Luwer, Dordrecht, The Netherlands 
Peterken, G.F. 1981 Woodland conservation and management. 
London: Chapman& Hall. 
ShimwelL, D. 1983. A conspectus of urban vegetation types. 
University of Manchester. Unpublished report to Nature 
Conservancy Council. 
References from websites: 
FAO www.fao.org, (accessed 5 lh May 2010) 
JNCC www.jncc.org (accessed 20 th April 2010) 
ACKNOWLEDGEMENTS 
I would like to thank Marion Seitz for her excellent work on 
web harvesting during her time at Ordnance Survey.