Embed Size (px)
Citation preview
________________________________________________________________________________________________
__________________________________________________________________________________________
Page | 24 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
MONITORING LAND USE OF KHAGRACHHARI UPAZILA OF BANGLADESH USING REMOTE SENSING, GPS AND GIS TECHNOLOGIES
Mozammel Haque Sarker1 , Mahbubur Rahman1 and Humayun Kabir2
1 :National Land Zoning Project, Ministry of Land, Bhumi Bhaban, Nlkhet, Dhaka-1205, Bangladesh
email: [email protected], [email protected] *Corresponding Author
2:Computer Source Ltd, 9/A, Rangs KB Square, Satmasjit Road, Dhanmondi, Dhaka-1205; email: [email protected]
Key Words: Land use, Remote Sensing, GIS, GPS, AEZ, and Contour.
ABSTRACT: Land use and land cover of Khagrachhai upazila, Bangladesh has been investigated using multi-spectral satellite images of the year 2015. Unsupervised (ISODATA) clustering as well as supervised classification based on parallelepiped and maximum likelihood has been used to classify the dominant features such as green crops, forest, urban/built up areas and water bodies. The study has been complimented with GIS for inclusion of the transport network, interesting/tourism places and features which could not be captured by 5 m resolution RapidEye image using the Google imagery. This has been done on screen digitization in ERDAS imagine environment. Contour lines have been generated from ASTER-DEM image by 20 meters spacing. Height elevation was observed 240m in this upazila. Agro Ecological Zones (AEZ) has also been used as a secondary data for observation of zone-wise agricultural intensity. It is reveals that the entire Khagrachhari Sadar upazila had been divided into 20 land use classes. Afterward, the thematic layers were clustered into 11 classes. The analyzed shows that Herb-shrub with scattered trees plays dominate role around 67%. On the other hand forest/plantation around 6%, Agriculture land use around 15% including Jhum (shifting cultivation) and rests of the land use are less than 2%.
Objectives: The main objective of this study is to generate the present land use and land cover map of Khagrachhari upazla using RS, GPS and GIS technologies.
1. Introduction
Land use and land cover plays an important role in global environmental change and sustainability, including response to climate change, effect on ecosystem structure and function, species and genetic diversity, water and energy balance and
agro-ecological potential. Land use and land cover is one of the most important typical applications of remote sensing data (Chrysoulakis et al., 2004). Remotely sensed data are a useful tool and have scientific value for the study of human environment interactions, especially land use and land cover changes (Dale et al., 1993). Land use is characterized by the activities of the people that involve using of lands for habitation, agricultural, forestry and industrial development and all sorts of small and large scale livelihood activities using lands of different cover types (FAO/UNDP, 1999). Land use is a product of interactions between a society's cultural background, state, and its physical needs on the one hand, and the natural potential of land on the other hand. The expansion of population and need for increased food production and socio-economic development are the major drivers causing the land-use change in the country. Thus, it is of importance to map the land-use pattern and study the land-use and monitor the land cover changes. Such information is highly useful for planning the future land-use and industrial and urban development and management of natural resources. Various techniques have been used for providing information about natural resource management. But remote sensing satellite observation with its unique capability to provide time-saving and cost-effective support in compiling the latest information source extracting natural features in landscape repeatedly (Stach,N et al 2007). RS technique has ability to represent of land use/land cover analysis by means of digital classification process of multispectral satellite image (Campbell, J.B. 1996). Satellite remote sensing, in conjunction with GIS, has been widely applied and been recognized as a powerful and effective tool in detecting land use and land cover mapping (Kumar N et al., 2015, Ehlers et al., 1990; Meaille and Wald, 1990; Treitz et al., 1992; Westmoreland and Stow,
________________________________________________________________________________________________
______________________________________________________________________________________
Page | 25 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
1992; Harris and Ventura, 1995; Yeh and Li,1997). The integrated use of remotely sensed data, GPS and GIS will enable consultants and natural resource managers and researchers in government agencies, conservation organizations, and industry to develop management plans for a variety of natural resource management applications (Philipson & Lindell, 2003). Remotely sensed data with the help of GIS may give the appropriate solutions for these problems in relatively small amount of time (Agarwal, 2000; Bakr, et al., 2010). For land-use mapping, supervised and unsupervised approaches are normally adopted for spectrally discriminate the surface features, well described in Campbell (2002), Kerle et al. (2004), Lillesand, et al. (2004), Jensen et al. (1996), Liu et al. (2002), Lunetta et al., (2004) and Coppin et al., (2004). In this study attempt has been made to prepare a present land use map of Khagrachhari upazila using Remote Sensing, GIS and GNSS technologies.
2. Materials and Methods
2.1 Study Area:
Khagrachhari is an Upazila (sub district) of Khagrachari District in the Division of Chittagong,
Bangladesh. It is located at 91 53 10.06 E to 92
04 14.82 E and 23 14 43.85 N to 23 00
03.85N.
The Chengi is one of the main river. Populations of Khagrachhari Sadar upazila are 48000, 61000, 91000 and 112000 in the census year of 1981, 1991, 2001 and 2011 respectively (Population census 2011). Tribal population (main ethnic group) are Chakma (31431), Tripura (24300), Marma (12939) and other (282). Figure 1 shows the study area (Khagrachari Sadar Upazila).
2.2. Data used
2.2.1 Remote sensing data
RapidEye satellite images serves as a major information source extracting natural features in land use of Khagrachhari Sadar Upazila. The RapidEye constellation of five satellites stands apart from other providers of satellite-based geospatial information in its unique ability to acquire high-resolution, large-area image data on a daily basis. All five satellites are equipped with identical sensors and are located in the same orbital plane. It has five spectral bands and having multi-spectral push broom imager sensor. Dry season of RapidEye images were used to identify present land use and land cover of the area. A number of moderate resolution satellite images (30m) such as Landsat8-OLI (Operational Land Imager) were used to delineate the Boro (winter rice) and Rabi crops (winter crop), Sentinel-MSI (Multi Spectral Imager) images were used to understand broad cropping pattern and seasonal land cover like; Aman rice (monsoon crop) and Google images were used to identify the micro-level land use features and Terra-ASTER (Advanced Space borne Thermal Emission and Reflection Radiometer) data were used for generation the contour from DEM of the Chittagong Hill Tracts. Table 1 shows the satellite images and its characteristics.
2.2.2 Field survey data
Global Positioning System (GPS) is a space-based navigation system that provides reliable positioning, navigation, and timely services to worldwide users on a continuous basis in all weather, day and night. It also provides three-dimensional locations plus precise time (Richharia, M 2003). In this study a handheld GPS (Germin) was used to collect the Ground Control Point (GCP) for geo-referencing the remotely sensed data as well as field verification of land use data interpretation. Easily identifiable points such as, road crossing, end of a bridge, corner of a building, etc. were used as reference points. Geographical location of the Important/Tourist places, etc. of the study area were also been extracted using GPS.
________________________________________________________________________________________________
______________________________________________________________________________________
Page | 26 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
Rectified images and the point data collected from GPS were then integrated into a GIS database. Figure 2 shows the integration of GPS, RS and GIS.
2.2.3 Ancillary data
Soil data collected from Soil Research Development Institute (SRDI) and Local Government Engineering Department (LGED) use map as a reference guide for land use mapping.
The GIS layer of Upazila boundary of Bangladesh collected from Bangladesh Space Research and Remote Sensing Organization (SPARRSO) has been used for sub-setting the area of interest (Khagrachhari upazila) from the satellite imageries. Statistical data collected from the Bangladesh Bureau of Statistics (BBS) have been used for validation of remote sensing results. The field investigation has also been performed to validate the interpretation of the imagery.
2.2.4 Data Generation
Generation of land use and land cover data sets involves data collection, pre- processing (geo-referencing, sub-setting, etc.), thematic layer generation (land use classes), data analysis, field validation and finally produce output of study. Many geo-referencing methods have been used to geo-reference multi-spectral satellite imagery (ERDAS imagine-user guide). One of the methods is collection of Ground Control Point (GCP) from study area using Global Positioning System (GPS). In this method GCP collected from study area and transformed this geographical coordinate to multi-spectral satellite image (same area images) by using geometric model of ERDAS imagine software. In geo-referencing/re-sampling method Bangladesh Transverse Mercator (BTM) projection system was used. RapidEye images of the upazila were the key data of the land use and land cover mapping. The multi-temporal Landsat imagery has been used which helps identification of perennial and seasonally varying features of land-use. The
RGB false colour composite image has been prepared for visualizing the land features. Then the multi-spectral classification using unsupervised (ISODATA) and supervised classification (Parallelepiped and Maximum Likelihood techniques) and results have been compared. In some cases the response of forest/plantation canopies and that of herb-shrub with scattered trees was similar in RapidEye image and cannot be
separated using digital classification technique. Another limitation was arisen from the geometric dimension of the plants/gardens (pineapple, mango, horticulture, Betel vine, litchi, etc.) and it was difficult to get clusters of homogeneous RapidEye 5m pixels for the features of interest of this study. The solutions of above two limitations are described in section 3.1. Based on the field investigation and secondary data, it is found that the supervised classification using maximum likelihood provide best results. Procedure of data generation is shown in figure 3. The entire Khagrachhari Sadar upazila had been divided in details into several thematic layers/features. Afterward, the thematic layers were clustered into following classes shown in Table-2
3. Analysis of Results
Land-use and land cover such as settlement, forest, agriculture, water bodies, urban area have been analyzed based on the multispectral digital classifications as well as on screen digitization of satellite/Google imageries. The results of the analysis are discussed below.
3.1 GPS based collected data analysis As described in chapter 2.2.4 that in some cases the response of forest/plantation canopies and that of herb-shrub with scattered trees was similar in RapidEye 5m image and cannot be separated using digital classification techniques. This limitation was overcome by generating herb-shrub as a master
Figure 2: Integration of GPS, RS and GIS
________________________________________________________________________________________________
______________________________________________________________________________________
Page | 27 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
layer (almost 67%) in ArcGIS environment and forest/plantation (around 6%) generating by on screen digitization from Google images based on field information using GPS. Pineapple, mango, rubber garden has also been
extracted from by on screen digitization from
Google images. In this case extensive field information has been collected from the field using GPS. Figure 4a
and 4b shows the RapidEye images as well as photographs of herb-shrub and pineapple garden collected during field observation. Roads, streams, etc. have also been extracted from Google images using on screen digitization technique in ERDAS/ArcGIS environment.
3.2 Analysis of AE sub-zones
Bangladesh has been divided into 30 agro ecological zones and 88 sub- zones of which 3 sub-zones belong to Khagrachhari Sadar Upazila. These are 29a (Mainly high hill range), 29b (Low hills and Piedmont Plains) and 29c (Low hills). Figure 5 shows the AE sub-zones area.
From table 3 is clearly seen that Jhum (shifting) cultivation activities concentrated more in high hill areas 29a (around 1485 ha) compare to medium and low land area. It has also planted in slope of hill. Contour map of figure 6 shows the slope of hill between 100 to 240 m. On the other hand Rice and Robi crops generally planted in low land areas 29c (around 910 ha). Table 3 and contour map of figure 7 shows such areas.
3.3 Contour Analysis
It is clearly seen from the figure 6 that contour lines are divided into five classes. First two classes are belongs to gradual
slope (<30) because pattern of contour lines shows wider between each other (far apart). Contour lines are between 161 to 240 shows closer due to steep slope of
Figure 5: Agro-Ecological Sub-zones
Figure 4b: Pineapple garden identified during field
Table 3: Statistics of Agricultural wrt AEZ
Descripction of Agricultural AEZ (29a) AEZ (29b) AEZ (29c)
land use Area (ha) Area (ha) Area (ha)
Rice and Robi 449.85 389.90 910.66
Horticulture 158.20 72.10
Batel Vine 0.03 0.14 0.04
Jhum (shifting crop) 1485.58 83.06 35.77
Other crop 319.53 308.10 592.07
Figure 4a: Herb-Shrub area collected during
field
observation
False Color Composite of RapidEye
image 5m
________________________________________________________________________________________________
______________________________________________________________________________________
Page | 28 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
hills (>50). Lines are form “V” pattern shows valleys.
3.4 Land Use Analysis.
Land-use of herb-shrub plays a domination role of
Khagrachhari upazila. It is clearly seen from the figure 7 and table 4 that herb-shrub occupied maximum around 17838 ha (67%), whereas dense forest area are depicted only 1716 ha (6%) of total area. Second highest land use agriculture area occupies about 2771 ha (10%) of total area and these area are belongs to low land and valley of hilly areas (figure 7). Shifting cultivation (Jhum) shows around 6% of total area and these areas are mainly concentrated in slope of high and medium hills (AE sub-zones 29b and 29c). Other land use contains brick-field, betel vine, and important/tourist places occupies only 28 ha (<1%). Number of brick-field was observed 11, betel vine garden 5 and important tourist places were 4. Rubber plantation land use occupies around 514 ha (2%) and settlement with homestead forest about 1415 ha (5%). Homestead forests are mostly concentrated in low land areas
(AE sub-zone 29c). Horticulture land use areas are about 300 ha (1%). It contains Banana, Pineapple, Mango, etc. garden. These are concentrated in medium (AEZ-29b) and high hill (AEZ-29c). Water
bodies are almost same as horticulture areas. It contains canal, creek, pond and river. River flows
over AE sub-zones 29b and 29c (figure 5 & 7). Only one build-up area in this upazila and occupied 339 ha area. It is belongs to agro-ecological sub-zone 29c and piedmont plains of 29a.
From bar diagram (figure 8), it is clearly seen that land use of herb-shrub with scattered trees shows height peak compare to others land use. Bar diagram also shows Agriculture, Forest, Settlement with homestead forest and Jhum (shifting cultivation) are around 2000 ha. Rest of the land use less than 1000 ha.
Figure 6: Contour map of study upazila
Table 4: Statistics of GIS based Land use of study
area
Figure 7: Present Land use (major) of study area
________________________________________________________________________________________________
______________________________________________________________________________________
Page | 29 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
5. Conclusions and Recommendations The study shows that the RapidEye 5m resolution and Landsat multi-date and multi-spectral imagery in association with Google images are capable of high quality land-use maps.
The land-use maps have been produced for Khagrachhari Upazila during the period 2015. The study reveals that the entire Khagrachhari Sadar upazila had been divided in details into 20 land use classes. Afterward, the thematic layers were clustered into 11 classes. The analyzed shows that the broad land-use feature identified is herb-shrub about 67%. On the other hand Forest/plantation around 6%, Agriculture land around 15% including Jhum (shifting cultivation) and rests of the land use are less than 2%. Field observation depicted that every year forest plantation is decreasing by deforestation and herb-shrub areas also deceasing due to shifting cultivation (Jhum). Settlement areas have also been increasing by increasing of population (Census 2011).
For future development planning and developing land-use strategies and policies land use change detection map preparation are encouraged. The satellite technology offers the best and quickest means of preparing such land-use maps and monitoring the land-use dynamics. This study demonstrates the usefulness of multi-temporal and multi-spectral satellite imageries for country wise land-use mapping
References
1. Agarwal, C.S., (2000), Textbook on Remote Sensing, in Natural Resourc Monitoring
and Management U.P., Ground Water Department, Lucknow, India.
2. Brammer H (2002). Land use and land use planning in Bangladesh, The University Press Limited, Dhaka.
3. Blaschke T, 2010. Object based image analysis for remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing 65 (2010) 2–16
4. Campbell, J. B. (2002) Introduction to Remote Sensing, London and New York, Taylor and Francis.
5. Coppin, P., Jonckheere, I., Nackaerts, K. & Muys, B. (2004) A review: Digital change detection methods in ecosystem monitoring: a review. International Journal of Remote Sensing, 25, 1565-1596.
6. Chrysoulakis, N. Abrams, M, Feidas, H and Velianistis, D (2004). Analysis of ASTER multispectral stereo imagery to produce DEM and land cover database for Greek Island; the REALDEMS project. Proceedings of e-Environment congress: Progressand Challenge, pp 411-424, Mexico
7. Dale, V. H.; O’Neill, R.V. 1993. Estimating the effect of land use change on global atmospheric CO2 concentrations, Canadian Journal of Forest Research 21; 87-90.
8. Ehlers, M., Jadkowski, M. A., Howard, R. R. and Brostuen, D. E. (1990). Application of SPOT data for regional growth analysis and local planning. Photogrammetric Engg. and Remote Sensing 56,175–180.
9. ERDAS imagine-user/tour guide.
10. FAO/UNEP, (1999) The Future of our land: facing the challenge. Guidelines for integrated planning for sustainable management of land resources. 71p.
11. Hadjimitsis, D. G., Papadavid, G., Agapiou, A., (2010) Atmospheric correction for satellite remotely sensed data intended for agricultural applications: impact on vegetation indices, In Natural Hazards And Earth System Science 10 (1), 89-95.
12. Harris, P. M. and Ventura, S. J. (1995). The integration of geographic data with remotely sensed imagery to improve classification in rural area. Photogram-metric Engineering and Remote Sensing 61, 993–998.
Figure 8: Bar diagram of major Land use of
Khagrachhari Upazila
________________________________________________________________________________________________
______________________________________________________________________________________
Page | 30 International Journal of Geosciences and Geomatics, Vol. 4, Issue 1, 2016
13. Jensen, J. R. (1996) Introductory Digital Image Processing: A Remote Sensing Perspective, US, Prentice Hall.
14. Kerle, N., Janssen, L. L. F. & Huurneman, G. C. (Eds.) (2004) Principles of Remote Sensing, Enschede, International Institute for Geoniformation Science and Earth Observation.
15. Kumar N., Yamaç S.S. and Velmurugan A. (2015) Applications of Remote Sensing and GIS in Natural Resource Management. Journal of the Andaman Science Association Vol. 20(1):1-6
16. Liu, X., Skidmore, A. K. & Van Oosten, H. (2002) Integrationof classification methods for improvement of land cover map accuracy. ISPRS Journal of Photogrammetry and Remote Sensing, 56, 257-268.
17. Lillesand, T. M., Kiefer, R. W. & Chipman, J. W. (2004) Remote Sensing and Image Interpretation, Fifth ed, New York, John Wiley and Sons.
18. Lunetta, R. S., Johnson, D. M., Lyon, J. G. & Crotwell, J. (2004) Impacts of imagery temporal frequency on land-cover change detection monitoring. Remote Sensing of Environment, 89, 444-454.
19. Meaille, R. and Wald, L. (1990). Using geographic information system and satellite imagery within a numerical simulation of regional urban growth. International Journal of Geographic Information Systems4, 445–456.
20. Philipson, P., & Lindell, T., 2003. Can coral reefs be monitored from space? Ambio, 32, 586−593
21. Richharia, M (2003), Mobile Satellite Communication.
22. Stach, N., Salvado, A, Petit, M., Ingrassia, F., Joubert, P. Deshayes, M. (2007) Monitoring land use and land cover changes in French Guiana by optical remote sensing. Forestsat 2007 Conference (Forest and remote Sensing: Methods and Operational Tools). Montpellier, France, Nov. 2007.
23. Treitz, P. M., Howard, P. J. and Gong, P. (1992) Application of satellite and GIS technologies for land-cover and land-use mapping at the rural-urban fringe: a case study. Photogrammetric Engineering and Remote Sensing58, 439–448.
24. Trimble eCognition Developer: http://www.ecognition.com
25. Weng, Q. (2001). A remote sensing-GIS evaluation of urban expansion and its impact on surface temperature in the Zhujiang Delta, China. International Journal of Remote Sensing, Forthcoming.
.
