Monitoring Land Use Land Cover Change for Dehradun District of Uttarakhand from 2009-2019

Land cover indicates the physical land type on the earths surface in the form of waterbodies, vegetation etc. whereas land use refers to the human adjustments with the land. Human has been modifying the land as a resource to fulfill their own needs since time immemorial but recently changes in land use land cover is unprecedented at local, regional as well as at the world level. These changes builds an enormous pressure on the surrounding environment and leading to climate change and loss of biodiversity. Thus, an attempt has been made to detect changes in land use land cover classes in Dehradun district of Uttarakhand state. The study has been carried out for 10 years (2009-2019) through remote sensing approach using satellite imageries of LANDSAT-5 "TM" for March 2009 and LANDSAT-8 "OLI" & "TIRS" sensor for April 2019. Methodology based on supervised classification has been applied using maximum likelihood in QGIS. The current analysis resulted that the district Dehradun has experienced land use land cover changes rapidly, as the area occupied by vegetation was about 46 percent during 2009 has decreased to 28 percent in 2019. About 27.54 percent area covered by vegetation gets turned into agriculture, 4.60 percent area into urban/ built-up and 6 percent into barren land. Agriculture and Urban/ Built-up area has increased immensely. Other land use land cover classes such as waterbody and barren land has also undergone changes. Monitoring and mediating the consequences of LULC classes has therefore become a major priority of researchers and policymakers around the world.


Introduction
Human relies on land as a resource for the fulfillment of there own needs thus land is an important natural resource from the developmental point of view. Basically land use and land cover are completely different terminologies but sometimes used interchangeably (Dimyati et al., 1996). Land cover indicates the physical land type on the earths surface in the form of waterbodies, vegetation etc. whereas land use refers to the human adjustments with the land. The land use land cover pattern of a region is an outcome of natural and socio-economic factors prevailing and their utilization by man over the time and space. Industrialization and other factors has recently encouraged the concentration of maximum populations in the urban areas (urbanization) and due to which giving way to depopulation of their rural counterparts, along with the intensification of agriculture and abandonment of the marginal lands.
Thus, land use land cover change detection is very essential for better understanding of the landscape dynamics during a given period of time (Kiefer Lillesand, 1987;Zhang, 2011) and to asses additions as well as the losses. Land use land cover change is an accelerating process worldwide which are mainly driven by the anthropogenic activities, which in turn changes the ecosystem (Ruiz-Luna and Berlanga-Robles, 2003;Turner and Ruscher, 2004).
The need of digital classification of land use land cover is necessary for the extraction of accurate results and thus lead to better understanding of relationships among human and various natural phenomenon and finally helps in decision making processes (Jokar Arsanjani et al., 2013;Pontius and Malanson, 2005).
Use of Geospatial technology has become important in the field of land use land cover mapping of a region as it gives a detailed information about the land features in very less time with better accuracy (Selcuk et al., 2003). Introduction of satellite imageries with very high resolution and advancement in GIS technology has given way for more consistent monitoring of changes in land use land cover classes over the earth.
Through this study an attempt has been made to map the status of land use land cover of Dehradun district of Uttarakhand for two time periods i.e. 2009 and 2019 and also to detect the changes that has been taken place during the last ten years using geospatial techniques.

Study Area
Dehradun district is one the 13 districts of the state Uttarakhand, also the capital of the state lies in the same district. As of 2011 Census, it is the second most populous District, encompassing an area of 3074.40 sq. km., extending in between 29 0 57' 56.44" North to 31 0 1' 127.13" North Latitudes and 77 0 38' 19.57" East to 78 0 14' 24.53" East Longitudes ( Figure 1). It has an area of 3088 sq. km. with the population size of 1696694. The district comprises of 7 tehsils, 6 developmental blocks and 767 villages. Few places of national importance are present in the district such as Forest Research Institute, Indian Millitary Academy, Lal Bahadur Shastri National Academy of Administration and Survey of India. Temperate climate is found in most of the places in the district with the elevation of 288 m to 3096 m.

Database and Methodology
A research methodology is the theme of any research work; it defines the way through which conclusion for the problem can be taken out. Generally, it includes the very first step to the end result ( Figure 2). Collection of data for the present work is of secondary in nature. Toposheet from Survey of India, Remote Sensing Satellite Imageries acquired from US Geological Survey has been used.
The base map of the district has been prepared with the help of SOI Toposheet and freely available maps of the concerned area. The base map of the district has been digitized and proper attribute data has been inputted in the QGIS environment.
High resolution remote sensing satellite data of LANDSAT-5 with 7 spectral bands of March 31, 2009 and LANDSAT-8 with 9 spectral bands of April 28, 2019 has been used for identification, classification and change detection of land use land cover classes for the period from 2009 to 2019.
The main tools used for processing, analysis and interpretation are QGIS (Free Source Software) and MS-Excel.

Land Use Land Cover Classification
A number of methods has already been invented for land use land cover classifications, which are known as unsupervised and supervised classification. Land use classification can be carried out based on the relative spectral similarity among the pixels may be either by an unsupervised method in group cases, or by a supervised method based on similarity of cases of predefined classes that have been characterized spectrally. In the present study supervised method of classification with maximum likelihood algorithm has been applied in QGIS so as to get the higher accuracy in classification.
Majorly five land use land cover classes have been identified: Urban/ Built-up, Agriculture, Barren, Vegetation and Waterbody.

Change Detection of Land Use Land Cover and its Analysis
To find out changes in land use land cover classes, post-classification detection method has been applied. Change information has been extracted by comparing pixels of the same class and thus the interpretation of the change has been done.
Classified image pairs of two years i.e. 2009 and 2019 and compared using cross-tabulation in order to determine qualitative and quantitative aspects of the changes for the period of 10 years from 2009 to 2019. A change matrix or land conversion matrix has been produced using QGIS software. All the tabulations related to the gains and losses among the land use land cover classes between 2009 and 2019 has been done using MS-Excel.

Results and Discussion
The final results obtained through the analysis of multi-temporal satellite imageries are diagrammatically illustrated in Figure 3 (a) (b), 4 (a) (b) and data has been shown in Table 1 and 2.  Table 1 and 2 shows changes in different land use land cover classes over the span of 10 years. A brief account of these results is discussed in the successive paragraphs.

Land Use Land Cover Classification 2009-2019
The spatial distribution pattern of five land use land cover classes in Dehradun district for the year 2009 has been shown in Figure 3 (Table 1).  To understand how different land use land cover classes has changed into other uses over 10 years, a land conversion matrix (Table 2)

Conclusion
The work has been carried out for Dehradun district in the state of Uttarakhand advocates that multitemporal land use land cover classifications derived from high resolution satellite data provides suitable data to assess past as well as present changes in land use land cover classes.
The changes in land use land cover classes has been rapid. Urbanization has found to be among the major drivers of change, as the statistics of the change shows a drastic increase in the urban/ built-up area from 12.53 percent in 2009 to 18.39 percent in 2019. Also agriculture area has increased from 22.88 percent to 36.92 percent and barren land area from 11.54 percent to 13.43 percent. Vegetation cover in the study area has immensely reduced from 46.56 percent in 2009 to 28.49 percent in 2019.
Thus the results shows that there is a drastic increase of urban/ built-up area and agriculture area whereas reduction in green cover and waterbody area within the concerned district. As the rapid growth of the cities in Dehradun district as expected to progress, it can be expected that further urbanization will probably impact the overall vegetation cover in the .
The present work itself explains that how important is the applicability of Geospatial Technology, it helps to analyze the lengthy temporal as well as spatial datasets with faster results in a more accurate manner which is in another way not at all possible with the use of conventional mapping techniques.