Data - Historical Land Use/Land Cover Classification Using Remote Sensing - page 48

Chapter 4
Data
One decides what information and data is needed to achieve the purpose of a study.
The data collection is, however, often controlled by what is obtainable or what the
financial map will allow, rather than what is actually needed. During the searching
process, one may find other undecided data sources or types that are useful to the
achievement of the research. Data and information for the LULC-component is
available from administrative divisions, libraries, universities and private com-
panies can offer data. Processing time and resources require attention, depending
on the amount of data collected and organized. Enough data is essential to guar-
antee accuracy and answerability. At sensor type choosing stage, in relation to
rapid sensor development and various sensor configuration, one has to consider the
broad range of application sectors in an attempt to give potentials that meet actual
obligations, as it is impossible to find a specific sensor type to satisfy the all
specific needs of all cases. Here, to optimize the choice of the remotely sensed
data, we have to determine the purpose of the research and which dataset can
realize the two criterions of being cost effective and providing the relevant
information in relation to the research purpose. Finally, at these basic stages of
choosing and preparing the dataset, it is also significant to consider the relationship
between the used dataset and the required mapping scale (Liu and Mason 2009 ).
The purpose of the current study is to set maps of land uses and the natural
coverage of the ERB. The satellite images suggest basic inputs for a compre-
hensive study of this area, which is also reliant on other data and information to
achieve targets, such as topographic maps and statistical records. Lastly, the study
cannot be fulfilled without reliance on field observations.
4.1 Satellite Data
LANDSAT and ASTER general characteristics (Fig. 4.1 , Tables 4.1 , 4.2 ) are:
medium spatial resolution, medium area coverage, moderate revisit capability and
multispectral bands characteristic. The scale of the area coverage of the LANDSAT