P. Vijay Sagar, R. Sivaganesavelan
EGIS Solutions Private Limited, Chennai, India
G. B Sukumaran
Professor and Head, Department of Geology, Presidency College, Chennai, India
Sri Ramakrishna College of Arts and Science For Women, Coimbatore, India
Avinashi College of Arts and Science, Deemed University, Coimbatore, India
The land – use change significantly due to various physical and socio – economic factors. The land – use pattern of an area is directly related with the level of tecno – economic advancement and the nature and degree pf covo;osatopm pf ots omahabitants (Whyte, 1961). Land – use is a dynamic phenomenon and both its value and pattern changes from one particular point of time to another and also from one geographical unit to another with varying efficiencies, abilities, priorities and needs (Bisht and Tiwari, 1996). The changes in land – use or land – cover due to natural and human activities can be observed using current and archived remotely sensed data (Luong, 1993) as information on land – use / land – cover pattern, their spatial distribution and changes over a time scale are the prerequisites for making development plans. Remote sensing plays a key role in providing information regarding the spatial distribution of various landuse classes, and multi-temporal data aids in detecting the changes that have taken place in a given terrain. Quantification of such changes is possible through GIS techniques even if the resultant spatial datasets are on different scales / resolutions. Change detection studies recognize that the biotic and abiotic components of the biosphere are linked and that human impacts on the earth now approach the global scale of biosphere processes (Hobbs, ed. 1990). However, no software can provide a full fledged solution to many of the problems posed by the user, and similarly it is very difficult to analyse spatial data in a GIS environment as intended by the user, and as many GIS software is provided with an extension tool which enables the user to create his own application software modules using these tools. In this context, an attempt is made to detect the changes automatically that have taken place in a given terrain over a period of time as it is often essential for academicians and decision makers.
The area chosen is part of the Palar River basin, in the state of Tamilnadu, India. The following figures show the two time data of the part basin considered for the years 1980 and 2000.
Objective of the Study
The chief objective of the study is the Development of Application Software for automatic detection of changes in spatial features over space and time.
About the study and Methodology
The above study enables the user to detect changes which have taken place over space and time, as satellite Remote sensing data with their repetitive nature aids in preparing land use / land cover patterns that changes with time. The quantification of such changes are possible even if the resultant spatial datasets are on different scales/resolution. The present study has been taken up in order to detect the changes automatically that have taken place in land – use.
The study encompasses the following modules:
- Data Acquisition, selection of two time remotely sensed data
- Generation of Map, preparation of Land – use map using visual interpretation techniques
- Data Conversion, using MapInfo software by registering the data to an earth coordinate system
- Digitisation of data, and to generate attribute information for the polygon data generated
- Conversion to Shape files, using Universal Translator available in MapInfo
- Spatial Integration of Data, by intersecting the two time data (1980 and 2000) using the module developed which has the facility to ‘Union’, ‘Intersect’ etc. the data as well as to classify the data using ‘Thematic shading’ facility.
- Invoking MapObjects,
- Creation of Utility Tool/Menus
- Selection of unchanged Areas based on Spatial query from the integrated data
- Automated detection of changed areas
- Creating Charts based on Changed and Unchanged areas
- Saving the layers
Consists of Login screen, Main Form, Map Display Form, Path Form, Data Integration Form
Where as the output design is formulated for Combine Form, Thematic map or Classification of display purpose Form, Change detection Form.
Creation of Utility Tool/Menus
Some of the utility tools and Menus developed using MapObjects and VB are shown
As far as integration tools are concerned the following have been included in the module
The following figure shows the integrated data with attribute data of the input themes stored in the output integrated map. The info table shows details of attribute data.
A careful study of the two time data indicate that there is quite a few changes with the Double crop areas fast depleting as also the forest area. The integrated map would give the detail about both time data (that is stored as an attribute). To detect the changes over a period of time it is necessary to compare the attribute data of each of the polygons and know the changes
In practice, none of the GIS software provides full-fledged solution, as it is next to impossible to cater to the needs of different users. In many GIS software provision is made to generate customized solutions. An important module in GIS is its Analysis and Modelling, which normally provides tools to query, both spatial and non – spatial database. However, not all modules are present in all the GIS software, hence, it may be necessary to develop tools in this regard. Further, in some cases special requirements similar to the objectives of this study, may require additional OCX tools and Programming languages.
Keeping in view the requirements of the study, it is proposed to develop an application software, which will be capable of detecting changes that have taken at a given location. The satellite data is interpreted to generate a thematic map depicting the actual use of land, that is the land use of the terrain for the years, 1980 and 2000. Land Sat (LS), TM and Indian Remote Sensing (IRS) 1 – C, LISS III data have been used. The various classes such as etc. are all represented in the landuse map. The two time data is integrated using ‘Intersection’, in Arc View and the output is taken as an ‘Input’ and analysed to find the changes that have taken place. For this purpose, a OCX tool ‘Map Objects’ and a programming software such as ‘Visual Basic’ has been used to develop an ‘Application Software’.
The first instance let us select all the unchanged areas of a particular class say, ‘Double Crop’.
From the above menu we can select the double crop areas or any other class later this can be compared with that of the changed areas. The following shows all that area under ‘Double Crop’, which has not been changed and are highlighted in yellow colour.
The following figure shows the Double crop area changed into Evergreen forest area.
Similarly all the changes that have taken place with respect to each other can be detected automatically and mapped, further, the area can also be calculated to know the changes in a given area. The following figure shows the major changes that have taken place in the sub – basin area of the Palar river.
It is also possible to create charts like Bar, Pie etc using the module where in we can display the changes in a graphic form along with the changes that have taken place in space.
The following figure shows how the integrated data can be represented graphically with both time data shown as Bars. Some times these charts will be of immense use when we try to depict the data or to compare the data of two different times. A careful study reveals that almost all land use classes have changed and this study along with the water level data would throw light on the actual reason for the changes.
The process of developing application software involves many a steps. In this study, various GIS software were thoroughly studied and found that an automated tool to detect changes over space and time is the main requirement. The module has been developed as a cost effective and efficient tool. This system has been developed in such a way that an Interface in the form of a menu control is generated. Simultaneously the various classes in different themes which are being integrated are automatically populated into a common database which is available to the user for further analysis. The most striking feature of the system is the capability to generate spatial as well as non-spatial data into one single theme, then it is further processed to detect unchanged areas and changed areas. The automatic detection of changes in landuse and land cover classes form the base for the study.