| GISdevelopment.net ---> Application ---> Natural Resource Management |
Multicriteria evaluation in GIS environment for groundwater resource mapping in guwahati city areas, Assam
P. Phukon and S. Phukan
Department of Geological Sciences, Gauhati University, Guwahati-781014
Email:p_phukon@rediffmail.com
P. Das
The Geointel Group, 108B, B. K. Kakati Road, Ulubari, Guwahati-781007
B. Sarma
Department of Civil Engineering, Jorhat Engineering College, Jorhat- 785 007
Abstract
The city of Guwahati with a total municipal area of more than 313 sq. km, has witnessed a rapid growth in population particularly during the last one and half decades. The population jumped from 1,23,783 in 1971 to 5,77,791 in 1991 and as per census 2001, the figure stands at 8,14,575. As a result, there is tremendous pressure on the natural resources like groundwater. Although the mighty river Brahmaputra flows through the northern periphery of the city, water supply depends heavily on the dugwells and deepwells. However, it is found that due to the geological factors, aquifers with adequate yield are not well developed and are mostly laterally discontinuous. The problem of depleted groundwater table is acutely felt in the new growth centres like Beltola, Panjabari, Motoria etc. The high growth of built up areas and filling up of natural channel ways have adversely affected recharge of the ground water regime. In this backdrop, it is of vital importance to identify the groundwater potential zones of the city area taking into account the geological and anthropogenic factors. Geologically the city is characterised by mostly Precambrian granite gneisses, quartzites forming residual hills and occupying a major part of the landscape. Small elongated intermontane valleys with various thickness of sediment fill and alluvium form the rest of the areas. Study of satellite data supplemented by field check shows that a characteristic feature of the terrain is the presence of a number of paleochannels which are perceived to be old channel ways linked to the river Brahmaputra towards north. Using Analytical Hierarchy Principle (Saaty, 1980), a paired comparison matrix was prepared for the five criteria (geomorphology, slope, soil, geology and landuse) for Guwahati city. The computed values shows acceptable level of consistencies. The consistency ratio (CR) for all the five thematic layers is found to be 0.013, while for the individual criterion the CR values are : geomorphology (0.0039), soil (0.00528), slope (0.0176), geology (0.) and landuse (0.0053), which are within the acceptable limit of 0.10. The weights were fed into the SPANS 7.2 for multicriteria analysis to determine the groundwater potential zones for the city.
Introduction
Multi-criteria evaluation (MCE) techniques are numerical algorithms that define the suitability of a particular solution on the basis of the input criteria and weights together with some mathematical or logical means of determining trade-offs when conflict arise (Heydon et. al., 2003). In this technique, ‘weight’ is assigned to the data layers to reflect their relative importance. Ground water exploration in a geologically complex terrain requires consideration of a number of factors – both natural and anthropogenic. It is important to understand the control of these factors on the groundwater regime of any area for optimal exploitation and aquifer management particularly in and around urban growth centers. As such, to arrive at a clear picture of the situation, the controlling factors have to be treated and integrated giving weight that is specific to a particular area. In this backdrop, the present study is aimed at testing the efficacy of MCE technique used in GIS environment as framework for decision making problems pertaining to ground water exploration.
Study area
The present study area, measuring about 229.94 sq. km., encompasses southern part of the Greater Guwahati Municipality in Kamrup (Urban) district of Assam (Fig. 1). It is bounded by graticule lines 91°34¢-91°52¢N and 26°5¢-26°13¢E. The city is extended more in an E-W trend occupying the position between the Brahmaputra river towards north and Precambrian hills of Shillong Plateau towards south. It is home to a population of over 8,14,575 (Census 2001) occupying mostly the narrow tracts of alluvium and sediment filled low lands interspersed with Precambrian residual hills and inselbergs. Intensified anthropogenic activities particularly in and around the hills has led to high rate of aggradation in the low lying areas clogging the city drainage system. The city enjoys a subtropical humid climate with an average rainfall of 1600mm. April to September are the months that receive abundant rainfall occasionally causing flashflood in many areas of the city. Southwest monsoon reaches this area by mid of May and recedes back by mid of October. The temperature fluctuate between 17°C to 38°C with maximum summer temperature between 30°C -38°C and minimum winter temperature between 17°C to 10°C.
Fig 1. Location map of the study area
Database
SoI topomaps (78N/12 and 78N/16) in the scale of 1:50,000 were used to prepare the reference map for the study area with supplementary information from satellite images (IRS IA, Geocoded FCC, B2 3 4; IRS ID, LISS III PAN sharpened). Two vector layers were generated for spot height and elevation contours at 20m interval to work out the DEM and DEM derivatives. The database for geology and geomorphology were prepared from satellite images in combination with field work and input from available literature on the subject. Small scale published maps of NBSS/LUP and a few reference borehole data were used to generate the thematic layer on soil.
Methodology
Five thematic layers were selected for determination of groundwater potential zones for the City. They are: Geology, Geomorphology, Soil Thickness, Slope and Landuse. Different classes were identified for each criterion and these were arranged in decreasing order of weight. Using Analytical Hierarchy Principle (Saaty, 1980), a paired comparison matrix was prepared for the five criteria (geomorphology, slope, soil, geology and landuse) for Guwahati city and individual class weights and map scores were worked out. The paired comparison matrix was prepared for each criterion using Saaty’s nine point scale. The rating used for paired comparison is as follows:
Table 1: Rating scale used in saaty’s AHP model
We thus get a matrix A of order n, where n is the number of classes in the criterion. The matrix A is reciprocal and should be consistent. For each element aij of the matrix, the following condition is satisfied:
By solving the matrix, we can find the weight of each class. In order to solve for weights, the following equation was used:
where I is an identity matrix of order n and X is the n×1 weight matrix and is the eigenvalue. For a matrix of order n, we get n number of eigenvalues. To calculate the weights, we select the largest value of
.
By using the Geometric mean method, the weight (X) of each class is obtained which are then normalized so that their sum is 1. These are then used to calculate the largest positive eigenvalue
. We select the maximum to recalculate the weights.
The values in the paired comparison matrix A must be checked for consistency. It is important to check for consistency because there may be inconsistency in judgement while comparing the various classes. The following formula is used for calculating the consistency index:
Here RI is the randomized consistency index for a matrix of order n. We have used the following values for the RI:
Table 2: Average consistency index
or an acceptable level of consistency, we should have CR < 0.1. If the value is found greater, we revise our pairwise comparison matrix until we get the required CR.
A programme was developed to calculate the eigenvalues and the weights of the different criteria. The algorithm is as follows:
Input the Matrix A of order n
For each row i of the matrix
Weight[i] = geometric mean of the elements of a row
Weightnormalized [i] = Sum of the weights / n
For each row i
For each column j
lambda[i] =
For each row i
lambda[i] = lambda[i] / weightnormalised
lambdamax = Average of all the lambdas
This value of calculated using this algorithm gives a very good approximation of the largest value of eigenvector. After calculation of the , we can calculate the consistency ratios by using equations (3) & (4) and Table 2.
The five thematic layers selected are as follows:
Geology
The study area represents a Precambrian terrane which is an extension of the Shillong plateau. It is characterised by the Gneissic Complex composed of granite gneiss, biotite schists and gneiss and quartzite. They are affected by two dominant sets of joints and intruded by quartz veins, aplite, and pegmatites (Maswood and Goswami, 1974, Maswood, 1981, Shukla et. al., 1989). These Archaean gneisses and schists form the basement over which a variable thickness of quaternary alluvium composed of unconsolidated sand, silt and clay are deposited. Along many tracts occupied by the palaeochannels, the typical Brahmaputra sand with abundant biotite and mostly silty, are encountered (Deka, 2001). Thin layers of residual clays, which are the weathered product of feldspar, are found intertwined with the alluvium at places. In general, the alluvium occupies 31.5% of the study area, while the Precambrian Basement occupies 68.49 % (Table 3).
Geomorphology
The geological disposition is reflected in the landscape of the study area with low lying Precambrian residual hills dotting all around interspersed with elongated low lying plains. Broadly, three geomorphic units viz. the denudo structural hills (residual hills), the alluvial plains and the marshy lands including the static water bodies (Water bodies with paleochannels) can be identified from satellite images and topomaps. The major hills of the area are Sarania hill (193m), Nabagraha hill (217m), Nilachal hill (193m), Chunchali hill (293m), Tetelia hill (221m), Khanapara hill (303m), Narengi hill (168m), Chandrapur hill (205m) and Buragohain Pahar (500). A unique feature of the landscape of Guwahati is the presence of numerous partially silted waterbodies locally known as beel, the largest of which is the Deepar Beel presently covering about 5.7 Sq. km. in the western fringe of the city. Most of the earlier waterbodies however, have been converted into built up land during last couple of decades. Apart from the Brahmaputra towards northern extremity, two other rivers viz., the Basistha and Bharalu form the main drainage within the city. Water bodies along with paleochannels occupy only 7.82% of the total area, while the alluvial plain occupies 31.51% and residual hills occupy 68.49% of the total study area (Table 3; Fig. 2 and 3).
Fig 2. Geological map of guwahati
Fig 3. Geomorphological map of Guwahati
Slope
A DEM was generated using the two input vector layers viz., spot height and elevation contours and the DEM derivatives viz., shaded relief, slope and aspect were worked out using Geomatica V 9. Although the city encompasses many residual hills, the slope is generally gentle (<6°) occupying about 86.41% of total area. Near and over the hills the slope is found to be high but steep slope (>45°) is very rare and occupies only 0.25% of the total study area.(Table 3). In general, DEM profiles show an increase in slope towards the eastern part (Fig. 4).
Fig 4. Slope map of Guwahati
Soil thickness
Soil depth map (Fig. 5) was prepared on the basis of NBSS/LUP small scale maps, collateral geological maps and some borehole data. Four classes of thickness of soil were delineated viz., 0-20 m, 20-50m, 50-100m and >100m. The area occupied by the 0-20m class consists 30.75% of the total study area and is present over the residual hills. While soil thickness range from 20-50m surrounding the flank of the residual hills and covers 15.71% of the area, it is better developed in the western part of the study area (50-100m). The soil class of >100m thickness usually occupies the paleochannels that covers about 23.48% of the area.
Fig 5. Soil thickness map of Guwahati
Landuse
The landuse map (Fig.6) was prepared from PAN sharpened LISS-III image using supervised classification, for which 5 classes were identified. Water bodies cover 2.38% of the area and mainly represented by the Dipar Beel and some minor waterbodies in the western part. Thick vegetations were found towards the eastern part of the area, mainly over the residual hills and occupies 9.68% of the area. Sparse vegetal cover also mainly occupies the residual hills all over and consist 21.98% of the total area. Built up land is the dominant landuse class and occupies nearly half (49.94%) of the study area. Barren lands used for mostly paddy cultivation, are identified in the western and southern part and occupies 15.86% of the study area.
Fig 6. Landuse map of Guwahati
Table 3. Area occupied by the classes of various thematic layers
Results
Five thematic layers– geomorphology, slope, soil depth, geology and landuse, mentioned above were selected for the purpose of the work. Class weights and theme scores were given as follows:
Class weights and theme scores
All Thematic Layers
Consistency ratio: 0.013
Geomorphology
Consistency ratio: 0.0039
Geology
Consistency Ratio: 0.0
Slope
Consistency ratio: 0.0176
Soil
Consistency ratio: 0.00528
Landuse
CR=0.0053
These weights were fed into the multicriteria analysis of SPANS 7.2 software and a ground water potential map was prepared identifying 4 classes (Table 4, Fig. 7 & 8).
Fig 7. Ground water potential map of guwahati
Table 4. Area occupied by groundwater potential zones
Fig. 8. Pie diagram showing area occupied by various groundwater zones
Discussion and conclusion
Analysis of the results shows that the most potential zones for groundwater exploitation are the alluvial plains towards west and the areas occupied by the palaeochannels where sediment thickness is more than 100m. As evident from geomorphic analysis from satellite images, the palaeochannels, although obliterated due to largescale human activities, show clear link with the Brahmaputra river towards north. As such, these can be potential sites for natural recharging of the surrounding aquifers in the Greater Guwahati areas. The areas occupied by the residual hills with a thin veneer of soil and where the slope is high, are the least potential zones with certain exception of possible perched water condition developed in fractured and weathered crystalline rocks. However, the flanks of these hills and inselbargs with gentle slope and moderate development of soil profile provide a source for groundwater. As the ground situation reveals, dugwells are successfull in these areas but in many cases they are exhausted during dry season between February and March. A moderately potential zone characterized by sediment fill and occupying mainly the south central part of the city is identified that cover about 27% of the total area. This part of the city is also a built up area with high density of residential units that inhibit natural recharging of the aquifers through precipitation. Artificial seasonal recharge through injection wells might be an effective way of managing these aquifers.
References
P. Phukon and S. Phukan
Department of Geological Sciences, Gauhati University, Guwahati-781014
Email:p_phukon@rediffmail.com
P. Das
The Geointel Group, 108B, B. K. Kakati Road, Ulubari, Guwahati-781007
B. Sarma
Department of Civil Engineering, Jorhat Engineering College, Jorhat- 785 007
Abstract
The city of Guwahati with a total municipal area of more than 313 sq. km, has witnessed a rapid growth in population particularly during the last one and half decades. The population jumped from 1,23,783 in 1971 to 5,77,791 in 1991 and as per census 2001, the figure stands at 8,14,575. As a result, there is tremendous pressure on the natural resources like groundwater. Although the mighty river Brahmaputra flows through the northern periphery of the city, water supply depends heavily on the dugwells and deepwells. However, it is found that due to the geological factors, aquifers with adequate yield are not well developed and are mostly laterally discontinuous. The problem of depleted groundwater table is acutely felt in the new growth centres like Beltola, Panjabari, Motoria etc. The high growth of built up areas and filling up of natural channel ways have adversely affected recharge of the ground water regime. In this backdrop, it is of vital importance to identify the groundwater potential zones of the city area taking into account the geological and anthropogenic factors. Geologically the city is characterised by mostly Precambrian granite gneisses, quartzites forming residual hills and occupying a major part of the landscape. Small elongated intermontane valleys with various thickness of sediment fill and alluvium form the rest of the areas. Study of satellite data supplemented by field check shows that a characteristic feature of the terrain is the presence of a number of paleochannels which are perceived to be old channel ways linked to the river Brahmaputra towards north. Using Analytical Hierarchy Principle (Saaty, 1980), a paired comparison matrix was prepared for the five criteria (geomorphology, slope, soil, geology and landuse) for Guwahati city. The computed values shows acceptable level of consistencies. The consistency ratio (CR) for all the five thematic layers is found to be 0.013, while for the individual criterion the CR values are : geomorphology (0.0039), soil (0.00528), slope (0.0176), geology (0.) and landuse (0.0053), which are within the acceptable limit of 0.10. The weights were fed into the SPANS 7.2 for multicriteria analysis to determine the groundwater potential zones for the city.
Introduction
Multi-criteria evaluation (MCE) techniques are numerical algorithms that define the suitability of a particular solution on the basis of the input criteria and weights together with some mathematical or logical means of determining trade-offs when conflict arise (Heydon et. al., 2003). In this technique, ‘weight’ is assigned to the data layers to reflect their relative importance. Ground water exploration in a geologically complex terrain requires consideration of a number of factors – both natural and anthropogenic. It is important to understand the control of these factors on the groundwater regime of any area for optimal exploitation and aquifer management particularly in and around urban growth centers. As such, to arrive at a clear picture of the situation, the controlling factors have to be treated and integrated giving weight that is specific to a particular area. In this backdrop, the present study is aimed at testing the efficacy of MCE technique used in GIS environment as framework for decision making problems pertaining to ground water exploration.
Study area
The present study area, measuring about 229.94 sq. km., encompasses southern part of the Greater Guwahati Municipality in Kamrup (Urban) district of Assam (Fig. 1). It is bounded by graticule lines 91°34¢-91°52¢N and 26°5¢-26°13¢E. The city is extended more in an E-W trend occupying the position between the Brahmaputra river towards north and Precambrian hills of Shillong Plateau towards south. It is home to a population of over 8,14,575 (Census 2001) occupying mostly the narrow tracts of alluvium and sediment filled low lands interspersed with Precambrian residual hills and inselbergs. Intensified anthropogenic activities particularly in and around the hills has led to high rate of aggradation in the low lying areas clogging the city drainage system. The city enjoys a subtropical humid climate with an average rainfall of 1600mm. April to September are the months that receive abundant rainfall occasionally causing flashflood in many areas of the city. Southwest monsoon reaches this area by mid of May and recedes back by mid of October. The temperature fluctuate between 17°C to 38°C with maximum summer temperature between 30°C -38°C and minimum winter temperature between 17°C to 10°C.
Fig 1. Location map of the study area
Database
SoI topomaps (78N/12 and 78N/16) in the scale of 1:50,000 were used to prepare the reference map for the study area with supplementary information from satellite images (IRS IA, Geocoded FCC, B2 3 4; IRS ID, LISS III PAN sharpened). Two vector layers were generated for spot height and elevation contours at 20m interval to work out the DEM and DEM derivatives. The database for geology and geomorphology were prepared from satellite images in combination with field work and input from available literature on the subject. Small scale published maps of NBSS/LUP and a few reference borehole data were used to generate the thematic layer on soil.
Methodology
Five thematic layers were selected for determination of groundwater potential zones for the City. They are: Geology, Geomorphology, Soil Thickness, Slope and Landuse. Different classes were identified for each criterion and these were arranged in decreasing order of weight. Using Analytical Hierarchy Principle (Saaty, 1980), a paired comparison matrix was prepared for the five criteria (geomorphology, slope, soil, geology and landuse) for Guwahati city and individual class weights and map scores were worked out. The paired comparison matrix was prepared for each criterion using Saaty’s nine point scale. The rating used for paired comparison is as follows:
| Weight | Definition |
| 1 | Equally likely occurrence |
| 3 | Moderately likely occurrence |
| 5 | Strongly likely occurrence |
| 7 | Very strongly likely occurrence |
| 9 | Extremely strongly likely occurrence |
| The values 2,4,6 and 8 can be used to denote intermediary values | |
We thus get a matrix A of order n, where n is the number of classes in the criterion. The matrix A is reciprocal and should be consistent. For each element aij of the matrix, the following condition is satisfied:
 | (1) |
 | (2) |
where I is an identity matrix of order n and X is the n×1 weight matrix and is the eigenvalue. For a matrix of order n, we get n number of eigenvalues. To calculate the weights, we select the largest value of
.
By using the Geometric mean method, the weight (X) of each class is obtained which are then normalized so that their sum is 1. These are then used to calculate the largest positive eigenvalue
. We select the maximum to recalculate the weights.
The values in the paired comparison matrix A must be checked for consistency. It is important to check for consistency because there may be inconsistency in judgement while comparing the various classes. The following formula is used for calculating the consistency index:
Here RI is the randomized consistency index for a matrix of order n. We have used the following values for the RI:
| Order of matrix | Randomised Index |
| 3 | 0.90 |
| 4 | 1.12 |
| 5 | 1.24 |
| 6 | 1.32 |
| 7 | 1.41 |
| 8 | 1.45 |
| 9 | 1.51 |
or an acceptable level of consistency, we should have CR < 0.1. If the value is found greater, we revise our pairwise comparison matrix until we get the required CR.
A programme was developed to calculate the eigenvalues and the weights of the different criteria. The algorithm is as follows:
For each row i of the matrix
Weight[i] = geometric mean of the elements of a row
Weightnormalized [i] = Sum of the weights / n
For each row i
For each column j
lambda[i] =
For each row i
lambda[i] = lambda[i] / weightnormalised
lambdamax = Average of all the lambdas
This value of
calculated using this algorithm gives a very good approximation of the largest value of eigenvector. After calculation of the , we can calculate the consistency ratios by using equations (3) & (4) and Table 2.
The five thematic layers selected are as follows:
Geology
The study area represents a Precambrian terrane which is an extension of the Shillong plateau. It is characterised by the Gneissic Complex composed of granite gneiss, biotite schists and gneiss and quartzite. They are affected by two dominant sets of joints and intruded by quartz veins, aplite, and pegmatites (Maswood and Goswami, 1974, Maswood, 1981, Shukla et. al., 1989). These Archaean gneisses and schists form the basement over which a variable thickness of quaternary alluvium composed of unconsolidated sand, silt and clay are deposited. Along many tracts occupied by the palaeochannels, the typical Brahmaputra sand with abundant biotite and mostly silty, are encountered (Deka, 2001). Thin layers of residual clays, which are the weathered product of feldspar, are found intertwined with the alluvium at places. In general, the alluvium occupies 31.5% of the study area, while the Precambrian Basement occupies 68.49 % (Table 3).
Geomorphology
The geological disposition is reflected in the landscape of the study area with low lying Precambrian residual hills dotting all around interspersed with elongated low lying plains. Broadly, three geomorphic units viz. the denudo structural hills (residual hills), the alluvial plains and the marshy lands including the static water bodies (Water bodies with paleochannels) can be identified from satellite images and topomaps. The major hills of the area are Sarania hill (193m), Nabagraha hill (217m), Nilachal hill (193m), Chunchali hill (293m), Tetelia hill (221m), Khanapara hill (303m), Narengi hill (168m), Chandrapur hill (205m) and Buragohain Pahar (500). A unique feature of the landscape of Guwahati is the presence of numerous partially silted waterbodies locally known as beel, the largest of which is the Deepar Beel presently covering about 5.7 Sq. km. in the western fringe of the city. Most of the earlier waterbodies however, have been converted into built up land during last couple of decades. Apart from the Brahmaputra towards northern extremity, two other rivers viz., the Basistha and Bharalu form the main drainage within the city. Water bodies along with paleochannels occupy only 7.82% of the total area, while the alluvial plain occupies 31.51% and residual hills occupy 68.49% of the total study area (Table 3; Fig. 2 and 3).
Fig 2. Geological map of guwahati
Fig 3. Geomorphological map of Guwahati
Slope
A DEM was generated using the two input vector layers viz., spot height and elevation contours and the DEM derivatives viz., shaded relief, slope and aspect were worked out using Geomatica V 9. Although the city encompasses many residual hills, the slope is generally gentle (<6°) occupying about 86.41% of total area. Near and over the hills the slope is found to be high but steep slope (>45°) is very rare and occupies only 0.25% of the total study area.(Table 3). In general, DEM profiles show an increase in slope towards the eastern part (Fig. 4).
Fig 4. Slope map of Guwahati
Soil thickness
Soil depth map (Fig. 5) was prepared on the basis of NBSS/LUP small scale maps, collateral geological maps and some borehole data. Four classes of thickness of soil were delineated viz., 0-20 m, 20-50m, 50-100m and >100m. The area occupied by the 0-20m class consists 30.75% of the total study area and is present over the residual hills. While soil thickness range from 20-50m surrounding the flank of the residual hills and covers 15.71% of the area, it is better developed in the western part of the study area (50-100m). The soil class of >100m thickness usually occupies the paleochannels that covers about 23.48% of the area.
Fig 5. Soil thickness map of Guwahati
Landuse
The landuse map (Fig.6) was prepared from PAN sharpened LISS-III image using supervised classification, for which 5 classes were identified. Water bodies cover 2.38% of the area and mainly represented by the Dipar Beel and some minor waterbodies in the western part. Thick vegetations were found towards the eastern part of the area, mainly over the residual hills and occupies 9.68% of the area. Sparse vegetal cover also mainly occupies the residual hills all over and consist 21.98% of the total area. Built up land is the dominant landuse class and occupies nearly half (49.94%) of the study area. Barren lands used for mostly paddy cultivation, are identified in the western and southern part and occupies 15.86% of the study area.
Fig 6. Landuse map of Guwahati
| Sl. No | Themes | Class | Area occupied | ||
| Area occupied(Sq. km) | % of total area | Cummulative Area(Sq. km) | |||
| 1 | Geomorphology | Water bodies withPaleo channels | 17.95 | 7.82 | 7.82 |
| Alluvial plain | 72.33 | 31.51 | 39.33 | ||
| Residual Hills | 139.26 | 60.67 | 100 | ||
| 2 | Geology | Alluvium | 72.32 | 31.51 | 31.51 |
| Precambrian Basement | 157.2 | 68.49 | 100 | ||
| 3 | Slope | <6° | 198.73 | 86.41 | 86.41 |
| 6°-25° | 30.239 | 13.15 | 99.56 | ||
| 6°-45° | 0.45 | 0.20 | 99.75 | ||
| >45° | 0.573 | 0.25 | 100 | ||
| 4 | Soil Thickness | 0-20m | 70.71 | 30.75 | 30.75 |
| 20-50m | 36.14 | 15.71 | 46.46 | ||
| 50-100m | 69.13 | 30.06 | 76.52 | ||
| >100m | 54.01 | 23.48 | 100 | ||
| 5 | Landuse | Water bodies | 5.81 | 2.53 | 2.53 |
| Thick vegetation | 22.25 | 9.68 | 12.21 | ||
| Sparse vegetation | 50.527 | 21.98 | 34.19 | ||
| Built-up land | 114.8 | 49.95 | 84.14 | ||
| Barren land | 36.47 | 15.87 | 100 | ||
Results
Five thematic layers– geomorphology, slope, soil depth, geology and landuse, mentioned above were selected for the purpose of the work. Class weights and theme scores were given as follows:
Class weights and theme scores
| Geomorphology | Soil thickness | Slope | Geology | Landuse | Weights | Normalised weight(´100) | |
| Geomorphology | 1 | 2 | 2 | 5 | 7 | 0.411 | 41.1 |
| Soil thickness | ½ | 1 | 1 | 3 | 5 | 0.227 | 22.7 |
| Slope | ½ | 1 | 1 | 3 | 5 | 0.227 | 22.7 |
| Geology | 1/5 | 1/3 | 1/3 | 1 | 3 | 00.09 | 09.0 |
| Landuse | 1/7 | 1/5 | 1/5 | 1/3 | 1 | 0.044 | 04.4 |
| Water Bodies and Paleochannels | Alluvial Plains | Residual Hills with piedmonts | Weights | Normalised weight(´100) | |
| Water Bodies and Paleochannels | 1 | 3 | 7 | 0.669 | 66.9 |
| Alluvial Plains | 1/3 | 1 | 3 | 0.243 | 24.3 |
| Residual Hills | 1/7 | ½ | 1 | 0.088 | 8.8 |
| Alluvium | Precambrian Basement | Weights | Normalised weight(´100) | |
| Alluvium | 1 | 9 | 0.90 | 90.0 |
| Precambrian Basement | 1/9 | 1 | 0.10 | 10.0 |
| (Degrees) | <6° | 6°-25° | 25° - 45° | >45° | Weights | Normalised weight(´100) |
| <6° | 1 | 2 | 3 | 5 | 0.469 | 46.9 |
| 6°-25° | ½ | 1 | 2 | 5 | 0.297 | 29.7 |
| 25° - 45° | 1/3 | ½ | 1 | 3 | 0.166 | 13.6 |
| >45° | 1/5 | 1/5 | 1/3 | 1 | 0.068 | 6.8 |
| (Thickness in metre) | >100 | 50-100 | 20-50 | 0-20 | Weights | Normalised weight(´100) |
| >100 | 1 | 2 | 3 | 7 | 0.495 | 49.5 |
| 50 – 100 | ½ | 1 | 2 | 5 | 0.291 | 29.1 |
| 20 – 50 | 1/3 | ½ | 1 | 2 | 0.147 | 14.7 |
| 0-20 | 1/7 | 1/5 | ½ | 1 | 0.0671 | 6.7 |
| Water bodies | Vegetation | Sparse vegetation | Barren land | Built up land | Weight | Normalised weight(´100) | |
| Water bodies | 1 | 3 | 4 | 4 | 7 | 0.496 | 49.6 |
| Vegetation | 1/3 | 1 | 2 | 2 | 4 | 0.215 | 21.5 |
| Sparse vegetation | ¼ | ½ | 1 | 1 | 2 | 0.116 | 11.6 |
| Barren land | ¼ | ½ | 1 | 1 | 2 | 0.116 | 11.6 |
| Built up land | 1/7 | ¼ | ½ | ½ | 1 | 0.060 | 6.0 |
These weights were fed into the multicriteria analysis of SPANS 7.2 software and a ground water potential map was prepared identifying 4 classes (Table 4, Fig. 7 & 8).
Fig 7. Ground water potential map of guwahati
| Sl. No. | Class | Area (%) | Cumulative | Area (sq.km) |
| 1 | Very potential | 26.18 | 26.18 | 60.201 |
| 2 | Moderately potential | 26.80 | 52.98 | 61.640 |
| 3 | Less potential | 16.27 | 69.25 | 37.424 |
| 4 | Poor | 30.75 | 100.00 | 70.726 |
| Total | 100.00 | 229.991 | ||
Fig. 8. Pie diagram showing area occupied by various groundwater zones
Discussion and conclusion
Analysis of the results shows that the most potential zones for groundwater exploitation are the alluvial plains towards west and the areas occupied by the palaeochannels where sediment thickness is more than 100m. As evident from geomorphic analysis from satellite images, the palaeochannels, although obliterated due to largescale human activities, show clear link with the Brahmaputra river towards north. As such, these can be potential sites for natural recharging of the surrounding aquifers in the Greater Guwahati areas. The areas occupied by the residual hills with a thin veneer of soil and where the slope is high, are the least potential zones with certain exception of possible perched water condition developed in fractured and weathered crystalline rocks. However, the flanks of these hills and inselbargs with gentle slope and moderate development of soil profile provide a source for groundwater. As the ground situation reveals, dugwells are successfull in these areas but in many cases they are exhausted during dry season between February and March. A moderately potential zone characterized by sediment fill and occupying mainly the south central part of the city is identified that cover about 27% of the total area. This part of the city is also a built up area with high density of residential units that inhibit natural recharging of the aquifers through precipitation. Artificial seasonal recharge through injection wells might be an effective way of managing these aquifers.
References
- Deka, P. J., 2001: Reconstructing the Landscape of Guwahati, Jour. Geoscience, Vol., 5&6, pp101-104
- Heywood, I, Corneluis S, Carver, S, 2003: An Introduction to Geographic Information Systems, Pearson Education, 1st Indian Ed., 295p.
- Maswood Md. and Goswami D N D, 1974: Basic rocks from the Precambrian Terrain around Guwahati, Assam. Indian Mineralogist Vol. 15. Pg. 55-62
- Maswood, Md., 1981: Granite Gneisses around Guwahati, Assam. Jour. Geol. Mineral, Met.Soc. India. Vol. 53, No 3,4. Pg. 115-124
- Maswood, Md., 1982: Structural History of the Precambrian Rocks around Guwahati, Assam. Quart.Jour.Geol.Mim.Met.Soc, India, vol. 54, No 1,2, Pg. 33-38
- Saaty, Thomas L, (1980): The Analytic Hierarchy Process, McGraw Hill
- Shukla, R.C. Gopalkrishnan, R and Chakrabarty S, 1989: Study of Granitic rocks around Kamakhya Hill and adjoining area, Kamrup District Assam, GSI Record. -122 Pt IV P: 72-73