On 7 December at 10:15 Kiira Mõisja will defend her doctoral thesis „Thematic accuracy and completeness of topographic maps“ at UT Senate Hall.
Senior Research Fellow Evelyn Uuemaa
Professor Tõnu Oja, TÜ ÖMI
Professor Menno-Jan Kraak, University of Twente, Netherlands
The availability of quality topographic databases and associated maps is critical for all users of spatial data. Various governmental agencies, first responders, utilities and GIS practitioners rely on the completeness and thematic accuracy of national topographic datasets. Spatial data quality has been the subject of discussions for almost 40 years. One of the biggest achievements of that period was agreement in international standards for spatial data quality. National mapping agencies involved in INSPIRE, are users of the ISO 19100 quality standards or the others spatial data quality standards. By contrast, the biggest problem is that data quality is analysed and presented at a generic global level rather than at more detailed levels of granularity. More detailed quality information is needed for data users and data producers as well.
An analysis of topographical Estonian Basic Map, empirical field inspections from 2003-2006 provides an excellent case study to investigate the effect of characteristics of field workers on spatial data quality. Mainly three quality elements are considered: classification correctness, omission, and commission. The error analyses published in II paper were performed on two levels: in general across all map sheets and in detail according to the field workers involved. The quality of topographical maps may vary spatially, also. The variation of the interpretation of orthophotos on the field may occur due to the differences in the complexity of the landscape, differences in the characteristics of individual field workers, and differences in their perception of the landscape. In the III publication the interaction between the characteristics of field workers, including their gender and years of experience (as a proxy for their mapping skills), and landscape heterogeneity were explored. In order to obtain landscape indicators describing landscape heterogeneity, the methodology of calculation of landscape indicators for vector data was developed (I publication).
Results showed the importance of error analyses on the level of a field worker and by landscapes as well. The outcomes reveal that the structure of errors on the general level and field workers’ level is different by geometry and error types. However, both systematic and individual errors were evident. Gender and years of experience of the field workers did not have a statistically significant impact on the mapping quality. By contrast, the results showed differences in the rates of misclassification, omission, and commission errors between field workers in different landscape types.
To improve the mapping quality concerning systematic errors, it is necessary to revise the definitions or methods of determination in a mapping specification or to consider whether the mapping of these features is absolutely necessary. In case of individual errors, monitoring field work to detect errors, so that workers can be trained to avoid such errors in the future, would also improve the mapping accuracy.