3. INTRODUCTION
• Land use/cover are two separate terminologies which are often
used interchangeably.
• Land cover refers to the physical characteristics of earth’s surface
(vegetation, water, soil and other physical features of the land,
including those created by human activities e.g., settlements.
• land-use refers to the way in which land has been used by
humans.
4. • Land use/cover changes is a widespread and accelerating process,
mainly driven by natural phenomena and anthropogenic activities,
affecting natural ecosystem.
• Accurate and up-to-date land cover change information is
necessary to understanding and assessing the environmental
consequences of such changes.
• The basic premise in using remote sensing data for change
detection is that Information is generally required about the
‘‘from-to’’ analysis.
5. OBJECTIVES OF STUDY
• To utilize GIS and Remote Sensing applications to discern the
extent of changes occurred over particular time period.
• To identify and delineate different LULC categories and pattern of
land use change
• To examine the potential of integrating GIS with RS in studying
the spatial distribution of different LULC changes
• To determine the shift in LULC categories through spatial
comparison of the LULC maps produced.
6. Why remote sensing
• Application of remotely sensed data made possible to study the
changes in land cover in less time, at low cost and with better
accuracy in association with GIS that provides suitable platform
for data analysis, update and retrieval.
• The advent of high spatial resolution satellite imagery and more
advanced image processing and GIS technologies, has resulted in
updating land use/cover maps.
7. Success in methodology
Regardless of the technique used, the success of change
detection from imagery depend the
1. Nature of the change involved and
2. The success of the image preprocessing and classification
procedures.
Example - If the nature of change within a particular scene is
either abrupt or at a scale appropriate to the imagery
collected then change should be relatively easy to detect.
8. CLASSIFICATION TECHNIQUE
• Unsupervised classification or clustering
• Supervised classification
• PCA
• Hybrid classification
• Fuzzy classification
These are the most commonly applied techniques used in
classification
9. CASE STUDY - 1
Study area
The Hawalbagh block of District Almora of the
Uttarakhand state.
• Study analysed for a period of 20 years from 1990-2010 .
10.
11. SOURCE OF DATA
• Landsat-TM images represent valuable and continuous records
of the earth’s surface during the last 3 decades.
• The entire Landsat archive is now available free-of-charge to the
scientific public, for identifying and monitoring changes in
manmade and physical environments.
12. STEPS INVOLVED
• Landsat Thematic Mapper at a resolution of 30 m of 1990 and
2010 were used for land use/cover classification.
• The TM sensor primarily detect reflected radiation from the
Earth’s surface in the visible and near-infrared (IR) wavelengths.
• The TM sensor have seven spectral bands.
• The wavelength range for the TM sensor is from the visible,
through the mid-IR, into the thermal-IR portion of the EMS.
13. • The satellite data covering study area were obtained from
global land cover facility (GLCF) and earth explorer site.
• To work out the land use/cover classification, supervised
classification method with maximum likelihood algorithm was
applied in the ERDAS Imagine 9.3 Software.
• Maximum likelihood algorithm (MLC) is one of the most
popular supervised classification methods used with remote
sensing image data
• These data sets were imported in ERDAS Imagine version 9.3
satellite image processing software to create a false colour
composite (FCC).
14. • The layer stack option in image interpreter tool box was used to
generate FCCs for the study areas.
• The sub-setting of satellite images were performed for extracting
study area from both images by taking geo-referenced outline
boundary of Hawalbagh block map as AOI (Area of Interest).
• The spectral distance method is used for classifying those pixels
that were unclassified.
• Ground verification was done for doubtful areas.
• Based on the ground truthing, the misclassified areas were
corrected using recode option in ERDAS Imagine.
15. • For performing land use/cover change detection, a post-
classification detection method was employed.
• Post-classification comparison proved to be the most effective
technique.
• A pixel-based comparison was used to produce change
information on pixel basis and thus, interpret the changes
more efficiently taking the advantage of ‘‘-from, -to’’
information.
16. • Classified image pairs of two different decade data were
compared using cross-tabulation in order to determine
qualitative and quantitative aspects of the changes for the
periods from 1990 to 2010.
• A change matrix was produced with the help of ERDAS
Imagine software.
• Areal data of the overall land use/cover changes as well as
gains and losses in each category between 1990 and 2010
were then compiled.
17. LAND USE COVER CHANGE DETECTED
Five land use/cover types are identified in the study.
1. vegetation
2. agricultural land
3. barren land
4. built-up land
5. water body
20. Categories 1990 2010 Change 1990-2010
km2 % km2 % km2 %
Vegetation 146.5 54.75 155.88 58.26 9.39 3.51
Agriculture 84.73 31.69 80.67 30.17 -4.06 -1.52
Barren 31.17 11.65 16.58 6.19 -14.59 -5.46
Built-up 2.72 1.01 12.2 4.56 9.48 3.55
Water body 2.42 0.9 2.2 0.82 -0.22 -0.08
total 267.53 100 267.53 100 0.00 0.00
Area and amount of change in different land
use/cover
21.
22. CASE STUDY - 2
Study area
In the northwestern coast of Egypt
For a period of 14 years from 1987-2001.
23. METHODOLOGY USED
• Landsat TM images acquired in 1987-2001.
• Ground information was collected for the purpose of supervised
classification and classification accuracy assessment.
24. IMAGE PRE-PROCESSING
Geometric correction
• Change detection analysis is performed on a pixel-by-pixel basis.
• Any mis-registration greater than one pixel will provide an anomalous
result of that pixel.
• To overcome this problem, the root mean-square error (RMSE)
between any two dates should not exceed 0.5 pixel.
• The RMSE between the two images was less that 0.4 pixel which is
acceptable.
25. Image inhancement and visual interpretation
• To improve the visual interpretability.
• To optimize the complementary abilities of the human mind
and the computer.
• Contrast stretching was applied on the two images and two
false color composites (FCC) were produced.
• Some classes were spectrally confused and could not be
separated well by supervised classification and hence visual
interpretation was required to separate them.
26. Image classification
• A supervised classification was carried for the two images individually
with the aid of ground truth data
• The overall objective of the image classification procedure is to
automatically categorize all pixels in an image into land cover classes
or themes.
• Using ancillary data, visual interpretation and expert knowledge of
the area through GIS further refined the classification results.
• Post-classification change detection technique was used to produce
change image through cross-tabulation
27. CROSSTAB
Categories of one image are compared with those of a second
image.
The result of this operation is a table listing the tabulation
totals as well as several measures of association between the
images.
28. • A legend is automatically produced showing these
combinations.
• In order to increase the accuracy of land cover mapping of the
two images, ancillary data and the result of visual
interpretation was integrated with the classification results
using GIS.
• The module used is the overlay module in IDRISI Kilimanjaro
software.
29. The area was classified into eight main classes:
• seawater
• salt marshes
• Sabkha – (Arabic name for low lying , high water table area
• cropland
• grassland
• bare land
• urban and
• quires - (Areas with active excavation and mining)
31. year categories 1987 (Area in ha)
Salt
marsh
Sabkh
a
Crop
land
Grass
land
Bare
land
Urban Quire
s
Total
2001
Salt marsh 1960 4116 386 145 50 0 0 6657
Sabkha 28 3890 289 2027 532 0 0 6766
Crop
Land
14 5079 21,189 35,459 15,220 0 0 77,266
Grass
Land
0 2715 185 60,951 12,857 0 98 78,806
Bare
Land
0 924 8 60,737 131,911 0 340 193,92
0
urban 67 2262 202 3100 4563 1543 89 11,826
Quires 0 17 0 345 1391 0 242 1995
total 2069 19,00
3
22,259 163,06
4
166,524 1543 769 375,23
6
Cross tabulation of land cover classes between 1987-2001
32. CASE STUDY -03
• Study area
Simly watershed, Islamabad, Pakistan.
for the years 1992 and 2012
33. METHODOLOGY
• satellite data obtained from Landsat 5 and SPOT 5 for the years
1992 and 2012
• Supervised classification
• Data were pre-processed in ERDAS imagine 12 for geo-
referencing, mosaicking and subsetting of the image on the
basis of Area of Interest (AOI).
34. The watershed was classified into five major land cover/use
classes viz-
• Agriculture
• Bare soil/rocks
• Settlements
• Vegetation and
• Water
Resultant land cover/land use and overlay maps generated in
ArcGIS 10 software.
35. SATELLITE DATA SPECIFICATIONS
Data Year of
acquisition
Band/colour Resolution
(m)
Source
Landsat 5 TM 1992 Multi-spectral 250 USGS glovis
SPOT imagery 2012 Multi-spectral 2.5 SUPARCO
36. The delineated classes were
• Agriculture
• Bare soil/rocks
• Settlements
• Vegetation and
• water class
38. LULC classes
1992 2012
Area (ha) % Area (ha) %
Agriculture 1775 11 4681 29
Bare soil 1648 10 2691 16
Settlement 1038 6 1870 11
Vegetation 11,342 69 7008 43
Water 603 4 155 1
Land cover/land use classes and areas in
hectares from year 1992-2012
39. CONCLUSION
• The objectives of this study were to detect land cover types
and land cover changes that have taken place.
• Integration of visual interpretation with supervised
classification using GIS and remote sensing was found to the
best combination of detecting changes.
• Integration leads to increase in the overall accuracy.
• Especially the area having spatial distribution of different land
cover changes.
• Assessment of land degradation, and future planning.