Evaluation of causes and effects of environmental shifts in the Lower Digaru River Basin, Assam, India

– Environmental issues like deforestation, biodiversity loss, soil erosion, siltation, wetland shrinkage, river bank erosion, etc. in drainage basin perspectives today are in a rapid flux triggered by population growth and developmental activities. Focusing on the Lower Digaru River basin of North-East India, the present study investigates the changes in the Land Use / Land Cover (LULC) trend using geospatial techniques and the cause-effect relationship by identifying the various environmental issues operating within the study area. The study area has been classified into nine LULC classes: dense forest, open forest, scrub forest, fallow land, barren land, cropland, dense scrub, mixed built-up land, and water body. The results revealed that the area under dense forest has drastically declined from 60.25 % to 16.63 % during 1999-2020. In contrast, the other important categories like Mix built-up land (8.11 % to 14.05 %), scrub forest (2.1 % to 25.11 %), barren land (0.2 % to 6.63 %) and open forest (11.9 % to 18.34 %) show an increasing trend of change during the said period. Post-classification comparison of the classified images based on the transition matrix indicated that approximately 40 % of the total area under forest had been converted to scrub forest, followed by 25.09 % to open forest and, most significantly, 5.22 % to mix built-up land during 1999 - 2020. Increasing population pressure and growth of economic activities like the establishment of brick industries, coke industries, and sand mining were the major driving forces for such LULC changes. Environmental implications like wetland shrinkage, river bank erosion, alteration of human occupation and economy, etc., were the study area’s prime concern. The findings suggest that integrated watershed management and land use planning should be implemented in the Lower Digaru River Basin.


INTRODUCTION
A drainage basin, sometimes called a catchment or watershed, is a fundamental and critical geomorphic unit that accumulates precipitation, producing runoff and river flow (Gregory, 1976, Gregory & Walling, 1973. It is a natural unit for physical, economic, industrial, and social planning and development. Drainage basin studies are crucial as they provide insight into the geology and subsurface lithological characteristics, driving forces, and processes that determine the basin's hydro-geological behavior (Oyedotun, 2022). Analysis of the drainage basin provides a broader understanding of the interplay of natural resources and communities depending on it, which will enhance assessing the basin's sustainability regarding river water, discharge, sediments, and biodiversity. In the present context, environmental issues in drainage basins over spatiotemporal dimensions are in flux. The rate of environmental change in any place varies due to natural and anthropogenic processes operating on it (Marsh, 1864). It is a matter of grave concern that the present mankind has forgotten the environmental and ecological significance of natural vegetation (Miller, 1953). The rise in population has expanded their activities towards the river basin in various ways resulting in the growth of various geo-environmental sues within the basin ecosystem, especially in developing nations. Moreover, the extensive ecological degradation and loss of biological diversity resulting from basin exploitation have led to multifaceted consequences, viz., lowering of groundwater, flash flood, wetland shrinkage, declining water quality, loss of flora and fauna, etc. (Murty, 1994). Several studies have been carried out to understand geoenvironmental issues and river basin management (Yurova & Shirokova, 2020;Rotaru, 2010;Bird et al., 2010;Kolesnikova, 2020;Ruttoh et al., 2022;Bufebu & Ilias, 2021). These studies revealed that the complex and dynamic land use and land cover change at various scales have environmental implications. The main driving forces of this change can be traced to the increasing population's consumption demand within the particular river basin. Many recent studies suggest that the world's river basins are affected by different kinds of human interventions, especially during the last few decades. Human-induced LU/LC change triggers deforestation, biodiversity loss, habitat destruction, soil degradation, soil erosion, siltation, flooding etc. The effects are found mainly in the areas where the settlements and associated developmental activities are more. Historical evidence reveals that human activities have altered the earth's environment by changing land use and land cover over the past several centuries. But the intensity of the pace of change and the resultant effect was limited due to the sparse population. However, the recent past witnessed very complex and challenging LU/LC change dynamics, and the effects are also very catastrophic and eyecatching (Kipkeeva et al., 2018;Koehnken et al., 2019;Rentier & Cammerat, 2022;Abdalla, 2015;Tian et al., 2014;Li et al., 2022;Maitama et al., 2010;Wu et al.,2010;Kilonzo, 2014;Malede et al., 2023;Xu et al., 2023). The studies further emphasized identifying the linkages between various issues and their consequences to develop specific frameworks and regulations for land management, including all sectors of LULC, like forest cover, cropping, grazing, urbanization, etc.
Many Indian studies have also assessed the relationship between geo-environmental issues and their cause-effect relationship. They found that demographic changes and the growth of economic activities contribute more than any other causative factors of LULC changes in the river basins. They have also investigated and discussed the sustainability issues of the river basin environment and its management (De, 2015;Das, 2020;Kumbhar et al., 2014;Islam et al., 2023;Jose et al., 2014;Iyer & Roy, 2005). Few studies on river basins of Assam have also been carried out to understand the cause-effect relationship of LULC change, river bank erosion, wetland conservation, soil loss estimation, lowering of groundwater table etc., at the individual level. But a detailed investigation of all these aspects prevailing in the river basin is yet to be studied (Jaiswal et al.,2012;Kotoky et al., 2013;Bora & Goswami, 2012;Bora & Goswami, 2016;Barman & Goswami, 2015;Bhattacharjee, 2016).
Under a unique physiographic set-up exhibiting scenic beauty, Assam is a land of perennial and ephemeral rivers and tributaries originating from the hills and gliding through valleys, finally debouches into the mighty rivers of Brahmaputra and Barak. Though the river basins of Assam were once composed of rich biodiversity, unchecked economic activities accelerated by population growth have left their scars on the basin environment vividly. Therefore, it is high time to call for river basin investigation and related studies for management purposes based on a scientific understanding of sustainable development (Coates, 1972). In this context, Geographical Information System (GIS) plays a vital role as it facilitates the acquisition of data in digital format and can be used for the integration of multiple correlated spatial databases of all the controlling factors (Liu & Buhe, 2000;Plummer, 2000;Lillesand et al., 2015;Longley, 2005). The application of geospatial technologies (Geographic Information System, Remote Sensing, and Differential Global Positioning System) can serve as a valuable tool for providing up-to-date information at a cheaper cost in assessing, monitoring, and managing natural resources Chowdhury et al., 2020). Remote sensing system offers a unique and highly flexible tool to survey and monitor biophysical resources and their characterization and to track changes in composition, extent, and distribution of communities and ecosystems (Philliepe, 1997;Briassoulis, 2000;Chakraborty et al., 2009). Similarly, GIS helps in the visualization of geospatial data conveniently and effectively to communicate complex information and increase the level of understanding (Lambin et al., 1999). Amongst many indicators, the dynamic interaction between man and the environment can be analyzed by land use and land cover (LULC) change (Gondwe et al., 2021). Comprehensive assessment of LULC through mapping and quantifying the extent of change are crucial to identify the critical regions (Abebe et al., 2022). Information gathered from LULC change and responsible factors involved would help monitor and manage the environment and living conditions (Choudhury et al., 2020). In this respect, the study also tried to evaluate the LULC status to gain insight into the intensity of change that ultimately determines the sustainability of resources within the basin. Therefore, the present research aims to identify and assess the LULC change dynamics with related issues like threatened wetlands, river bank erosion, shifting of human occupation etc. in the lower Digaru river basin stretching from the foothills of Meghalaya to floodplains of the mighty river Brahmaputra (Assam part). As discussed, the study area has already witnessed various environmental issues; therefore, an in-depth study with the help of modern tools and technologies has given a precise understanding of the impact of multiple processes operating within it. The database and understanding generated through this study will be helpful for further planning, management, and development of the basin ecosystem. The aim of the present study has been achieved through the following objectives 1) To investigate the changing land use and land cover pattern in the study area and the driving forces responsible for it 2) To identify the consequences resulting from the driving forces on the environment and livelihood of the inhabitants.

LOCATION OF THE STUDY AREA
The lower Digaru river basin lies in Assam with a geographical extension between 25° 30' 15" N to 26° 14' 18" N latitude and 91° 34' 15" E to 92° 0'15" E longitude (Fig 1), covering 217.5 km 2 . Numerous tributaries join the river, forming a remarkable southern sub-basin of the Brahmaputra river system. The present research study is centered on the lower part of the Digaru River that falls in the Kamrup (M) district of Assam. The river originates from the Khasi hills of Meghalaya bordering Assam near the south of the village Raitong at an elevation of 1067 m. The river is known as Umtrew in Meghalaya. At the same time, as it enters Assam, it is renamed Digaru River, glides from south to north direction and discharges into Kolong River in the Kamrup Metropolitan district of Assam. The basin comprises several small dissected hillocks with elevations ranging from 200 to 600 meters. The topography is undulating in the southern part. Outcrops of rock at various locations are visible during the non-monsoon season while the terrain remains flat or plain towards lower reach located in Assam. The basin is endowed with lush and dense subtropical vegetation and scrubland towards the upper reach. On average, the basin receives a maximum temperature of 35°C in the summer while winter remains mild at 15°C. The study area's population is unevenly distributed in its upper course due to hilly characteristics. In contrast, the plain area of the basin is heavily populated due to the presence of favorable conditions. Important urban centers of the study area are Sonapur, Byrnihat, and Jorabat, which have few newly set up industries.

DATABASE AND METHODOLOGY
The present study analyzes a database collected from the field, generated from the ancillary and satellite data supplemented by secondary data collected from various offices and public sources. The satellite data used in this study have been described in Table 1. The present study has adopted both primary and secondary data. The basin boundary has been identified from the Survey of India (SOI) topographical map (78 N/16, 78 O/9) at a 1:50000 scale. The secondary data regarding different physical, climatic, socio-economic, and cultural profiles have been collected from various central and state government offices and other published and unpublished sources. The satellite data viz. IRS LISS III and LISS IV have been acquired from NRSC, Hyderabad, and tentative thematic maps have been prepared and modified after the field visit. For field surveys, structured questionnaires have been prepared to fulfill various objectives regarding socioeconomic information, land use/land cover change, wetland change, shifting cultivation, brick industry, etc. The survey has been followed using a simple random sampling technique. For LU/LC change detection analysis, the visual interpretation technique (on-screen manual digitization) has been found more suitable than the digital interpretation technique, which is a pixel-based interpretation. Sometimes information needed for image analysis and understanding is not represented in pixels but in meaningful image objects and their mutual relations. Hence, the visual interpretation approach is favorable for micro-level study (Gorte, 1998;Baatz, 2000;Blaschke et al., 2000). The land use/ land cover polygons, as seen in the satellite data of the study area, have been delineated on screen using NRSA standard classification system, and a preliminary interpretation map has been prepared. Before initial interpretation, all the imageries of different years have been rectified, and various image enhancement techniques, viz. contrast enhancement and histogram equalization (Abdulaziz et al., 2009), have been adopted for a better view of the land use/land cover patches.
After preliminary interpretation, fieldwork has been conducted to establish the relationship between image elements and tentatively identified LULC categories. A minimum of four patches for the individual land use category with smaller area coverage and a maximum of ten patches for the individual land use category with larger coverage have been verified. Then the delineation of LULC categories prepared during the preliminary interpretation phase has been modified based on information collected from the fieldwork. During this phase, the classes with accuracies have been aggregated to the nearest LULC class.
Finally, the digitized layers of different years have been superimposed in a GIS environment to assess the changing pattern. Thus, the error matrix, as well as the accuracy of the work, has been obtained. Village-wise population data has been collected from the census office and integrated with the village layer to prepare the population density map. Data regarding industrial and economic set-ups have been collected from the field, and their locations have been recorded by GPS.

RESULTS AND DISCUSSION
The pattern of LULC and its changing scenario From the perspective of sustainable development in a river basin, a detailed and comprehensive micro-level review of the LU/LC change is most convenient. Based on the season and knowledge gathered during the field visit, the study area has been classified into nine major types of land use and land cover as prescribed by the National Remote Sensing Agency (NRSA) classification scheme. These are viz. dense forest, open forest, scrub forest, fallow land, barren land, cropland, dense scrub, mixed built-up land, and water body. A large part of the region is covered by notified reserve forests. Under this circumstance, the open scrub category has been renamed scrub forest as per the NRSA classification (NRC LULC Manual, 2006). Additionally, the cropland category has been minutely observed in the satellite imagery supplemented by ground truth.

Land use/Land cover of 1999
The LULC of 1999 illustrates the category of dense forest dominating the highest area, accounting for 60.25 % of the basin's total area. This forest cover mainly comprises various species of giant trees inside the reserve forest. There is also a high concentration of fallow land, having 16.4 % of the total area, reflecting the human population's pressure on land. The area shares by other LULC categories, viz. cropland, open forest, scrub forest, barren land, dense scrub, mixed built-up land, and water body, is 5.11 %, 5.48 %, 0.97 %, .09 %, 1.7 % and 1.9 % respectively (Fig 2). It has been observed from the LULC map that most categories, like mixed built-up land, cropland, fallow land, and barren land, have been found along the national highway and the river valleys (Fig 3).

Land use/Land cover of 2020:
The LULC map of 2020 represents most of the categories sharing proportionate areas of the study area where scrub forest occupies 54.5 km 2 (25%), followed by open forest (18%) and dense forest (17%). However, a negligible portion has been covered by waterbody (2%), cropland (2%), and barren land (1%), as illustrated in Fig. 4.

Changing pattern between 1999-2020
Change in LULC pattern in river ecosystem is dynamic and complex due to human interference that is determined by a host of factors (Lambin et al., 2003. A similar situation is witnessed in the study area showing drastic change in the land use land cover during the last decade. From the analysis, it is clear that there was a massive reduction of dense forest from 130.8 to 36.1 km 2 during 1999-2020. It is mainly attributed to the unauthorized cutting of large trees inside the reserve forest for timber and firewood consumption. A significant portion of these products are carried to Guwahati city for various economic purposes. Though the area is under a protected reserve forest, the government still needs to take the initiative. Compared to the dense forest, the area under scrub forest has tremendously increased from 2.1 to 54.5 km 2 from 1999-2020. The reason for positive change is due to the factors mentioned above. In the case of open forest, it also shows an increasing trend from 11.9 km 2 in 1999 to 39.8 km 2 in 2020. Meanwhile, the estimated overall accuracy value is 92 % ( Table 2 and Fig 6).  It can also be used as one of the indicators of accuracy assessment (Nagaraja & Navalgund, 2003). This matrix is produced by multiplying each column in the transition matrix by the number of cells of corresponding land use in the later image through the pivot table. It has been generated by intersecting two LULC maps of different years in GIS environment (Nagaraja et al., 2003). For the nine-by-nine matrix (Nine LULC categories), as shown in Table 3, the rows represent older LULC categories (1999), and the column represents the newer categories of LULC (2020). Moreover, the highlighted numbers placed in the diagonal matrix (red mark) in the table indicate unchanged areas under each category. This matrix can be used as a direct input for the specification of the prior probabilities in the maximum likelihood classification of the remotely sensed imagery in the future (Nagaraja & Navalgund, 2003). It has been estimated that dense forest, cropland, fallow land, water body, and dense scrub categories have shown a significant negative conversion to other categories in their area from 1999 to 2020. In contrast, the other categories like mixed built-up land, open forest, scrub forest, and barren land have gained in their area from other categories during this period. Over the temporal period, out of the total area under cropland (11.14 km 2 ) in 1999, only 2.91 km 2 remained unchanged, whereas 4.81 km 2 and 1.68 km 2 of cropland were converted to fallow land and barren land in 2020.  Table 4.

Driving forces
Unprecedented rates of human-induced activities have converted the natural landscape haphazardly during the last few decades, particularly in developing nations. (Twisa and Buchroithner, 2019). A comprehensive assessment of LULC change dynamics over a spatiotemporal dimension is crucial to identify the driving forces for land management strategies. Updated and precise LULC maps would aid in sound planning, monitoring, managing natural resources, and implementing strategies (Seyam et al., 2023). As the world faces acute transformation of LULC change, the factors responsible have also been observed as the standard drivers in the study area. The present study examines the underlying drivers affecting the terrestrial surface of the lower Digaru basin. From field surveys, it has been witnessed that human-induced factors such as population growth, deforestation, agricultural expansion, industrial development as well as different developmental activities are continuously operating throughout the basin. Based on the knowledge gathered from time to time, the following factors have been recognized for changing the nature of LULC in the study area.

Increasing population pressure
Rapid population growth and its consequences, particularly in the LULC change, have greatly concerned social scientists. The tremendous population growth in the recent past has directly influenced the LULC pattern of the lower Digaru basin. A total of 66 villages within the study area have been selected. The basin's population increased from 34100 to 52568 during 2001-2011. Pieces of evidence state that the reserve forest of the study area is now under continuous threat of human encroachment. The average population density recorded has increased from 156 persons/km 2 in 2001 to 241 persons/ km 2 in 2011. The villages have been classified into different population density categories viz. very low (less than 50 person/ km 2 ), low (50-200 person/ km 2 ), moderate (200-600 person/ km 2 ), high (600-1000 person/ km 2 ) and very high (above 1000 person/km 2 ). Among the 66 villages of the study area, 56 % are located in rugged terrain, and the rest in floodplain areas. Therefore, shifting of settlement towards plain regions is a common phenomenon observed. It has been estimated that 24 % of the total villages fell under the very low category in 2001, which decreased to 10 % in 2011. Again 21 % of the total village belonged to the low category in 2001, which significantly changed to 22 % in 2011. The most remarkable changes have been observed in the medium density category, where 33 % of villages have increased to 48 % from 2001 to 2011, shown in Fig. 7. Slight change has been witnessed in the remaining two categories of high and very high categories. Sonapur gaon has the highest population (6801) and density (4301 persons/ km 2 ), followed by Baruabari gaon (2250 person/ km 2 ), Barkhat gaon (1021 person//km 2 ), and Kasutali Pathar (692 person/ km 2 ) depicted in fig: 8a and 8b. The high concentration of population in these villages is accompanied by easy access to transport and communication and fertile agricultural land. Villages in hilly and rugged topography bear the lowest population density, viz. Dhemai N.C (1.8 person/ km 2 ), Ghagua No. 2 (10 person/ km 2 ), and Tepesia N.C (22 person/ km 2 ) are shown in Table 4.

Growth of economic activities
The setting up of brick industries forms one of the significant developments that have taken place in the study area. It is believed that huge forests were cut to obtain fuel wood for brick-making. Brick-making takes place in rural industries belonging to small and informal sectors. The brick production process includes clay preparation, drying and firing operation, packaging, and transportation. The study area is close to Guwahati city, where there is a huge demand for bricks for excess developmental activities. It has been observed that 15 brick industries (table 6) have been established since 1990 near the bank of the Digaru River, causing many environmental problems. All the brick industries have been surveyed to understand their status, and a household survey has been conducted to know the impact of these industries on their surroundings. Fig. 9 depicts areawise locations of brick industries showing GPS points collected from field survey. It has been observed from the map that all the brick industries have been established along the Digaru River amidst the agricultural fields.
With the establishment of two brick industries viz. B.B.I and G.G.I in Kamalajhari and Damarapathar village in 1990, the area under brickfields has been expanding continuously. It has been estimated that brickfields occupy 77 hectares of land. The agricultural lands have been transformed into brickfields due to their growing demand. The brick kilns' survival depends on soil and silt availability. A survey has been conducted to understand the daily need and production of brick kilns, from which it has been estimated that one brick kilns need at least 40 trucks of earth and at least two trucks of silt per day for the production of 40 lakh pieces of brick in a month shown in table 5. Sand has been collected from different mining sites developed on the banks of Digaru River, viz. Kurkuria ghat, Luri ghat, Barni ghat, etc. A massive amount of coal and firewood is required to burn bricks. Hence, coal is collected from the coal mines in the Ri Bhoi district of Meghalaya, while firewood is gathered from the nearby reserve forests. These flourishing brick kilns along the Digaru River have drastically altered the existing land use pattern at an accelerated rate.   Another crucial industrial development confronted by the field study in the study area is the establishment of the coke industry in the border areas of Assam and Meghalaya. The coke or the coking coal is nothing but the black substances remaining after the coal gas and coal tar is removed from the coal. The formation of coke results from the heating of coal in the absence of oxygen. Thus, coke is used as a high-grade energy source in the manufacturing processes like iron and steel (Patowary, 2009). Altogether 18 coke industries have been developed in recent years near Byrnihat along the national highway No 40. The main reason for establishing this industrial belt is that the Meghalaya government relaxed some taxes for establishing industry within its boundary. These industries gradually enlarged their territory by encroaching into the forest lands. Table 7 shows the amount of coal burnt each day by the coke industries, where due to the rise in demand, the rate of burning has increased manifold in recent times leading to environmental issues.

Effect of LU/LC change
The consequences of land use change are complex and diverse. It has been found from the study that the changes in the pattern of land use/land cover led to several serious environmental problems. The problems, viz. shrinkage of wetland, soil erosion, extinction of flora and fauna from the reserve forest, lowering of the water table and their deterioration, siltation, river bank erosion, deterioration of air quality, and flooding in the downstream part, are remarkable. However, it is worth mentioning that though these effects are often considered natural, the augmentation of the same in most cases can be attributed to human interference in the study area. The following aspects have been identified as the prime concern at this stage for environmental equilibrium in the lower Digaru basin.

Shrinkage of Wetlands
The lower Digaru basin is the hub of various wetlands, swamps, and marshes locally known as "beels". Despite potentialities, the beels of the study area are subjected to multiple threats. Increasing population growth along with agricultural and industrial activities have reduced the size of the beels during the last few decades. Three significant wetlands viz Borbila beel, Domora beel, and Chong beel have been selected for the study using satellite imageries to observe the changing pattern of beels of the study area. Satellite imageries for two years have been taken to find out the changes, and a topographical sheet has been taken as a base map. It has been estimated from Table 8 and Figure 10 that the area under wetlands in 1970 was 206.05 hectares, which was reduced to 146.74 hectares in 1999 and decreased to 114.52 hectares in 2020. The decreasing rate from 1970 to 1999 was (-) 1.51 percent/ year, while from 1999 to 2020 was 2.00 percent/year, as indicated in Table 8. It has also been observed that the area under particular wetlands has also been significantly reduced over the years. Chong beel has been reduced by (-)  hectares in 1970, reduced to 15.9 hectares in 1999, with a reduced rate of (-) 2.5 percent/year. Furthermore, it reduced to 9.61 hectares in 2020 at (-) 3.6 percent/year. The Borbila beel has squeezed from 44. 4 hectares in 1970 to 20.78 hectares in 1999 and 16.44 hectares in 2020 at a decreasing rate of (-)1.51 percent/year and (-)2 percent/year, respectively, depicted in Fig. 10.  Agricultural expansion and industrial development are considered essential factors for the deterioration and spatial reduction of the wetlands of the lower Digaru basin. All the beels are located midway between the agricultural fields. The increasing population with growing demand for food directly exerts pressure on these wetlands as people of the nearby villages try to enhance the area under agriculture. Thus, a high concentration of paddy cultivation around the Damara beel and Chang beel have been observed. A similar condition prevails in Ghatua, Kukurakata, Sakuamara, and Bahtola beels (Fig 10). Due to the continuous expansion of paddy cultivation, these two beels, namely Ghatual and Bahtola, have recently been on the verge of extinction. Agricultural practices along the banks of these wetlands produce siltation to the wetlands making the water murky. This results in less sunlight penetration for microflora and fauna, thereby affecting the bio productivity of the wetlands subsequently.
Industrial developments such as the establishment of brick industries have adversely affected the ecosystem components of the wetlands. The problem arises mainly due to the cutting and piling of the earth for making raw bricks, contamination of water caused by the waste materials of the kiln, and the effect of thermal pollution. The Samaikuria beel near Kamalajari village is the worst affected, as three brick industries are located near this wetland. The width of the wetland has been diminishing owing to the piling of earth, raw bricks, finished bricks, burnt coal, and waste materials (Patowary, 2009). Moreover, industries near the Damara beel continuously provide water for making raw bricks, disrupting the ecosystem. Chong beel has also been encroached on by the newly established brick industries, and another factor responsible for the size reduction is siltation in the beel bottom during the summer season. Due to soil erosion in the upper catchment, the eroded materials are carried by the rainwater through the Digaru River and deposited in the downstream part. During summer, when the river is overloaded with excessive water, this sediment is transported to the beels. In addition to these, other developmental activities like the construction of roadways and railways have also affected the wetland environment.

Sand extraction and bank erosion
The combined impact of the lithological characteristics, climatogenetic forces, vegetation, slope, drainage density, and relief features determine the intensity of soil erosion. The leading cause of soil erosion is the large-scale deforestation in the upper part due to the dominant 'Jhoom' (shifting cultivation) practices. Due to the rise of population pressure, the jhoom cycle has been reduced to 1-2 years instead of 6-7 years. Such change in land use on the fragile hill slopes aggravated the topsoil erosion and caused deterioration of the soil quality. These eroded materials carried through the rills and gullies to the mainstream have been deposited in the downstream part of the Digaru River. Many sand extraction sites have been developed in the last two decades because of the availability of sand in the river bed and transportation access to the river bank. Some of them are leased out by the Assam government to the contractors locally called mahaldars through the sand extraction tendering process. However, some extraction sites remain illegal. The permitted sand extraction sites in the lower Digaru basin are Byrnihat, Tamulikuchi, Sonaiogaon, Alenga, Sonapur, and Kurkuria, as shown in Table 9.  Fig. 11 that extraction has rapidly increased since 2002, accompanied by rising production of bricks in the industries. Most of the sand is taken to the brick kilns for making raw bricks, while some amount is also supplied to the nearby areas and Guwahati city. From extraction to the process of disbursement of sand involves various steps wherein for most, sand is extracted from the river bed using a steel bucket and transported by boat. Then sand is unloaded at the river bank site, and after that, it is loaded in a truck and disbursed to different destinations. Earlier, the collection of sand was concentrated in some limited areas. However, because of the increasing demand for quality sand, the number of collection and depositing sites has increased faster. With accelerating demand, sand quarrying is now concentrated mainly in the sites near the bank to save time and labor costs. As a result, the depth of the river bed near the bank increases with time (Patowary, 2009). Therefore, during the winter season, the river water's depth deepens compared to the banks because of the lowering of the river bed. However, the process of unloading and loading sand exerts too much pressure on the bank of the river, causing bank erosion at the sites of sand quarries. It has been observed from the field study that from the Barnihat, located at the foothills of the National Highway, the Digaru River is subjected to bank erosion resulting in the widening of the river up to 80 meters at some places. This bank erosion significantly affects villages like Kamalajhari, Marakdola, Laflang, Kholabari, Moupur, and Luri. In Kamalajhari village, a long strip of around 200 meters collapsed on the river due to bank erosion indicated in plate 1. It happens mainly because of the rapid extraction of sand along the bank side, resulting in the incapability of preventing high-velocity water. Loading and unloading also aggravate erosion activities. Three sand depositing stations at Moupur village where bank erosion has taken place are significantly shown in Plate 2. It has been observed from the field study that the channel has been widened near Lurighat in comparison to the foothill zone and mouth due to bank erosion.

Plate 1: Bank erosion in Kamalajhari village
Plate 2: Bank erosion in Moupur village

Effect on human occupation and Economy
The growth of brick industries and sand quarries modified the traditional human occupation and living standards to a vast extent in the study area. They could not cultivate in their agricultural field as these lands became unfit for production due to removing topsoil for making bricks and the thermal pollution from brick kilns. A household (HH) survey has been conducted to understand the impact of brick industries on HH of the surrounding areas shown in Table  10. The field study was conducted in Kamalajhari, Sonai Gaon, Damarapathar, and Baruabari villages. Among 93 HH surveyed, 31 HH (33.3 %) sold their topsoil, and 12 HH (12.9 %) sold their agricultural land to owners of brick kilns. The inhabitants became the worst sufferers after selling their agricultural land and topsoil. Aftermath, they worked as daily laborers either in the brick industry or in sand extraction works under contractors at a meager wage.
Earlier, the laborers of the brick industries came from the western parts of Assam, especially from the districts of Dhubri and Goalpara, while few skilled laborers derived from Bihar and Jharkhand. However, nowadays, local laborers are readily available as they want to work for fewer wages than outsiders. Such transformation in their occupation made them poorer and drastically changed their social life. Moreover, in the terrains, the availability of hilltops inspired the local tribes to practice the age-old primitive jhoom cultivation from which income generation is poor, and people remained economically unsound as earlier.

CONCLUSION:
In a nutshell, the study has demonstrated that the use of geospatial techniques can assess and quantify the nature and trend of LULC changes and thereby contribute towards an improved understanding of the cause-effect relationship. IRS LISS III and LISS IV satellite imagery has been used to measure the pattern of LULC change between 1999 and 2020. The classified image and field survey has shown that the Lower Digaru River Basin has undergone significant LULC alteration during the mentioned period. The study area has experienced a decline in the dense forest from 60.27 % to 16.63 %, fallow land from 16.40 % to 14.05 %, and cropland from 5.11 % to 2.21 % during the 21-year-long interval. There is also a substantial increase in mixed builtup land from 8.11 % to 14.10 %, scrub forest from 0.96 % to 25.11 %, and barren land from 0.09 % to 6.63 %. The accuracy of the LULC map has been obtained through the calculation of the transition matrix. Most of the dense forest area has been converted to scrub forest, open forest, and mixed built-up land. Similarly, a significant portion of cropland and fallow land has been converted to barren land during the study period. The study reveals that the present LULC changes are primarily related to increasing population pressure and various economic activities. The Lower Digaru River basin's estimated population density increased from 156 persons/km 2 to 241 persons/km 2 between 2001 and 2011. The establishment of 15 brick industries along the Digaru River and 18 coke industries in the foothills of the Meghalaya plateau during the last few decades has worsened the environmental quality of the study area. The expansion tendency of these industries, along with the increase of mixed built-up land, have threatened the wetland and forest ecosystem to a large extent. Aggravation of sand mining from the Digaru River and soil from the nearby agricultural fields to meet the growing demand for brick supply led to river bank erosion and shifting of occupation and the economy of the rural villagers in the study area. Hence, it can be concluded that the lower Digaru River basin, once rich in various natural resources, is now facing severe threats due to human-induced determinants. Therefore, the LULC changes and cause-effect relationships as identified in this study require urgent attention and intervention from researchers, environmentalists, planners, and decision-makers to address the issues of environmental degradation in the study area. Based on the study's findings, it is recommended that an integrated watershed management program, in combination with scientific land use planning, would be helpful in sustainable resource management in the Lower Digaru River basin.