The classificatıon accuracy were calculated by
comparing the results obtained from a digital
classification to the known identity of land
cover in test areas (75 randomly selected test
pixels were used) derived from an reference
area. Due to the effect of the high percentage
of mixed pixels in urban and suburban classes,
the overall classification accuracy's for each
date were obtained 84%, 8496 and 8596
respectively.
In Table 3, the areal context of the changes in
urban and green area obtained from the
classified 3 data set were given. In Figure 7,
the land cover changes as per the years are
shown.
Table 3. Land Use/Cover Change Assessment.
Land 1984 1990 1992
Use/Cover (hectares) | (hectares) (hectares)
Urban À 319.59 416.34 558.63 .
Green A. 4850.73 4561.92 4273.92
Land Cover / Use Change
5000 -— :
e 4000 + |
$ 30004 reine |
$ 2000 } | —— Green Areal |
I 1000 + {— O — Urban Areal |
om =m — 0 el —
1984 1990 1992 |
Year |
Figure 7. Land cover/use changes
as per the years.
4. Conclusion
Land use patterns change over time in
response to economic, social and
environmental forces. Type of any change in
the use of land resources is essential
information to proper planning, management
and regulation of the use of land resources.
As shown in this study, multi-temporal
remotely sensed images are capable for
identifying and delineating the changes
occurred in land use e.g. new logging areas or
new land developments such as settlements,
industrial complex and roads etc. by
comparing two (or more)
taken on different dates, pixel by pixel and
than updating the land use map of the study
area. It is possible to use many change-
detection techniques. The procedure that is
most appropriate to use in a given situation
Landsat scenes.
682
depends on the specific application (type of
environment, targets of interest), the amount
of detail required and an extensive knowledge
of the area to be studied and the logical and
spectral interrelationships between land use
classes.
As a result, it was concluded that since the
spatial structures of the Tuzla region became
more rigid and planning alternatives were
narrowed, the considerable progress must be
made in the creation of environmental
awareness and implementation of effective
legislation.
REFERENCES
Campbell, J.B., 1987, Introduction to Remote
Sensing , The Guilford Press, pp. 499-500.
Ingebritsen S.E. and R. J. P. Lyon, 1986,
Principal component analysis of multitemporal
image pairs, International Journal of Remote
Sensing, Vol.6, No. 5, pp. 687-696.
Lillesand, T.M. and R. W. Kiefer, 1987,
Remote Sensing And Image Interpretation,
John Wiley & Sons, pp.696-697.
Michalak, W.Z., 1993. GIS in land use change
analysis: integration of remotely sensed data
into GIS, Applied Geography, 13, pp.28-44.
Ormeci, C., and F. Sunar, 1994, Identification
and multispectral Analysis of changes of green
areas and coastal zones in the Kilyos-
Karaburun coastline of Istanbul, Turkey, 2nd
Thematic Conference on Remote Sensing for
Marine and Coastal Environments, New
Orleans, 31 January-2 February, pp.575-580.
Virag, LA. and J. E. Colwell, 1987, An
improved procedure for analysis of change
inthematic mapper image pairs, Twenty-First
International Symposium on Remote
Sensing of Environment, Ann Arbor,
Michigan, October, pp. 26-30.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996
