1 Introduction

Water availability for different purposes particularly for irrigation and domestic use is of great concern in the near future. Hence, it becomes necessary to tap the maximum possible water within the river basin. Water harvesting technique is one of the techniques to tap the water by constructing storage4 structures like check dams, storage tanks, stop dams etc. Water storage serves like an insurance mechanism acting as a barrier to the variability of rainfall (Payen et al. 2012). Water storage structures are one of the important components of watershed development which not only collects and stores water but can also be utilized to recharge the ground water (Ahmad and Verma 2016). Storage structures require considerable investment and hence it is important to identify the suitable location of these structures before execution. Decision making and planning about the required number and type of water storage structures to be constructed using remote sensing and GIS techniques is extremely important to avoid huge investment (Singh et al. 2009). These techniques enable us to perform watershed analysis in shorter time and in a cost effective manner.

Runoff is one of the important decision making parameter for water storage site. It can be stored by constructing suitable structures (Ramakrishnan et al. 2009). There are number of approaches for estimating potential runoff at the water storage site. The Soil Conservation Service-Curve Number (SCS-CN) method developed by National Resources Conservation Service (NRCS), United States Department of Agriculture (USDA) in 1969 (NRCS 2004), is a simple, predictable and stable conceptual method for estimation of direct runoff depth based on storm rainfall depth (Nagarajan and Poongothai 2012). This method is a quantitative description of land use / land cover / soil complex characteristics of a basin.

A number of studies have been reported for site suitability using multi-criteria evaluation system and analytic hierarchy process (Ahmad and Verma 2017; Ahmad and Verma 2016; Ahmad et al. 2013; Bamne et al. 2014; Banai-Kashani 1989; Bodin and Gass 2004; Gavade et al. 2011; Haas and Meixner 2005; Saaty 1980; Salih and Al-Tarif 2012; Teknomo 2006; Triantaphyllou and Mann 1995). Analytic Hierarchy Process (AHP) is used as a decision-making tool to determine the percentage importance of various parameters used in the determination of suitable sites. The objective of the present paper is to identify suitable location for water storage site using multi-criteria evaluation technique under the guidelines of Integrated Mission for Sustainable Development (IMSD) (IMSD 1995). This study provides the suggested sites for water storage in a planned manner for better conservation.

2 Material and Methodology

2.1 Study Area

A portion of upper Sheonath river basin was considered for this study. The study area extends between latitudes 20°25′00” N and 21°00′00” N, and longitudes 80°26′00″ E and 80°55′00″ E. The climate of Chhattisgarh state is tropical and most of the parts are non-arid. The monsoon season is from late June to October and maximum rainfall occurs during this period. The study area comprises of Durg (93.52 sq. km), Rajnandgaon (991.83 sq. km) districts of Chhattisgarh state and Gondia (13.15 sq. km), Garchirolli (297.39 sq. km) districts of Maharashtra state. Total geographic area of the upper Sheonath basin considered is 1395.89 sq. km. Study area was influenced by four raingauge stations namely Amabagarh Chowki, Dongargaon, Doundi Lohara and Mohala. Daily rainfall data were collected from Chhattisgarh Water Resources Department, for a period of 4 years, i.e. from 2009 to 2013. The Indian Remote Sensing satellite Resourcesat 2: AWiFS data for land use land cover was downloaded from website (bhuvan.nrsc.gov.in/) available at a scale of 1:250 K for the year 2011–12. The location map of the study area is presented in Fig. 1.

Fig. 1
figure 1

Location map of study area upper Sheonath basin

Full size image

2.2 Methodology

The data obtained were digitized and converted into digital format using in GIS for creating various criteria based layers. GIS based multi-criteria evaluation is used to identify the most suitable sites for water storage. The input parametric criteria are potential runoff, hydrologic soil group, land use, lineament, slope, stream order and settlement. AHP is applied to these parameters to determine the percentage importance. The methodology adopted to achieve the objective is shown with the help of flowchart in Fig. 2.

Fig. 2
figure 2

Methodology adopted for determining the suitable site for water storage

Full size image

As per the Integrated Mission for Sustainable Development (IMSD) guidelines following criteria have been followed for selecting the suitable sites for water storage structures (IMSD 1995; Padmavathy et al. 1993; Singh et al. 2009):

  1. 1.

    The slope should be less than 15%.

  2. 2.

    The land use may be wasteland, scrubland, forest, agriculture river bed or water bodies.

  3. 3.

    The infiltration rate of the soil should be low.

  4. 4.

    The type of soil should be sandy clay loam.

The suitability of water storage structure is described with the help of Suitability Level Index. Gosschalk (2002) has described the suitability levels for regional dams scaled from 1 to 9. In this study suitability levels were considered in the range from 1 to 3. ‘1’ indicates the storage tanks, ‘2’ indicates the stop dams and ‘3’ indicates the check dams. The Suitability Level Index adopted in the study is presented in Table 1. Suitability level factors include runoff, land use, slope, soil type, stream order, settlement and lineament. The spatial data were used to prepare various thematic map’s viz. drainage map, stream order map, land use map, slope map, settlement map, soil map, lineament map and runoff potential map.

Table 1 Suitability level index for different parameter for identifying potential water storage structures
Full size table

2.3 SCS-CN Method

In this study loss from the basin was computed by the SCS-CN method. This model was developed by United States Department of Agriculture (USDA) Soil Conservation Services (SCS) in the year 1954 (Subramanya 2013). Curve number is essentially a coefficient that reduces total precipitation to runoff potential after losses i.e. evaporation, absorption, transpiration and surface storage. Basically, it is a quantitative descriptor of land use, soil characteristic and antecedent moisture condition (AMC). AMC is expressed in three levels (AMC I, AMC II and AMC III), as per the rainfall limits in growing and dormant seasons of the study area (Ramakrishnan et al. 2009). This method is based on three equations:

  1. 1.

    Water Balance Equation

$$ P={I}_a+F+R $$
(1)

Where.

P :

precipitation in mm (P ≥ R);

R :

runoff in mm;

S :

potential maximum retention in mm;

I a :

initial abstraction in mm;

F :

infiltration in mm;

Potential maximum retention accounts for potential maximum soil moisture retention after runoff begins and initial abstraction (Ia) accounts for depression (surface) storage, interception, and infiltration, occurring before runoff begins (Mishra et al. 2007).

  1. 2.

    Proportionality Hypothesis

The ratio of actual amount of runoff to maximum potential runoff is equal to the ratio of actual infiltration to the potential maximum retention (Subramanya 2013) and is expressed as

$$ \left(P-{I}_a-R\right)/S=R/\left(P-{I}_a\right) $$
(2)
  1. 3.

    The third hypothesis is that the amount of initial abstraction is some fraction of the potential maximum retention (Subramanya 2013) and thus expressed as

$$ {I}_a=\lambda S $$
(3)

for Indian conditions λ = 0.3S (Subramanya 2013)

$$ S=25400/ CN-254 $$
(4)

Solving eq. (1), (2) and (3) we get

$$ R={\left(P-{I}_a\right)}^2/\left(P-{I}_a+S\right)\; for\kern0.24em P\ge 0.3S $$
(5)

Equation (4) is the required equation for computing potential runoff of a basin.

2.4 Study Area Boundary Demarcation

The foremost task is to delineate the hydrological boundary of the upper Sheonath basin. This involves the application raster data i.e. 30 m resolution Aster Global Digital Elevation Model (GDEM) (USGS 2016) and vector data i.e. the drainage lines digitized with the help of topographic sheets in the scale 1:50000 procured from Survey of India (SOI). Geo-HMS toolbar in GIS is used for demarcation. Following procedure is adopted to demarcate the basin boundary (Ahmad and Verma 2015; Fleming and Doan 2009; Merwade 2012):

  1. 1.

    DEM reconditioning: Since the vector drainage lines are digitized using topographic sheets and considered to be more reliable when compared to vector drainage lines processed directly using 30 m Aster GDEM. Therefore, it is required to adjust the DEM by superimposing the vector drainage lines over DEM.

  2. 2.

    Fill Sink: This operation is required to fill the cell surrounded by higher elevation cell and lower the cell surrounded by lower elevation cell, so that flow will occur smoothly overland. The resulting grid is known as hydrologically corrected DEM.

  3. 3.

    Flow direction: It defines the direction of flow in each cell. The eight-point pour algorithm specifies the eight possible directions of flow.

  4. 4.

    Flow accumulation: It determines the number of upstream cells draining to a given cell.

  5. 5.

    Stream definition: It classifies all the cells with flow accumulation greater than the user defined threshold belonging to stream network.

  6. 6.

    Stream segmentation: It divides the stream grid into number of segments up-to junction points of the stream that connects two successive junctions or outlets.

  7. 7.

    Catchment Grid Delineation: It delineates a sub-basin for every stream segments up-to junctions or outlets.

Finally, the catchment polygon processing creates a vector layer of sub-basin generated from the delineated grid and hence the study area boundary is demarcated.

2.5 Analytic Hierarchy Process

Analytic Hierarchy Process (AHP) is one of Multi Criteria decision-making method that was originally developed by Prof. Thomas L. Saaty (1987). Following are the important components in AHP (Saaty 1995; Schmoldt and Peterson 1997; Prato and Hajkowicz 1999; Mu and Rojas 2017):

  1. 1.

    Structured hierarchy of the parameters, arranged at different levels i.e. objective, criteria and alternatives.

  2. 2.

    Deriving priorities for the criteria, through pairwise comparison between the criteria with respect to desired objective. Derived priorities are then checked for consistency of judgments, to ensure reasonable level of consistency in terms of proportionality and transitivity.

  3. 3.

    Deriving local priorities for the alternatives i.e. deriving priorities of the alternatives with respect to each criterion separately and then check the consistency as required.

  4. 4.

    Deriving overall priorities i.e. all alternative priorities obtained are combined taking into account the weight of each criteria, to establish the overall priorities of the alternatives. The best decision will be the alternative with highest priority.

  5. 5.

    Performing sensitivity analysis to detect how the change in the weight of criteria affects the result obtained.

This method is used to determine the percentage importance of the parameters used in the identification of suitable sites for water storage in accordance with the guidelines laid by IMSD. The AHP procedure involves performing comparison of pairs of parameters within a set of reciprocal matrices. In comparing pairs of factors, the AHP scale of relative importance is used in the scale 1 to 9. Prato and Haikowitz (Prato and Hajkowicz 1999) presented that an open scale can be used. The AHP scale of paired comparison with its description is listed in Table 2 (Saaty 1987; Banai-Kashani 1989).

Table 2 AHP scale for pairwise comparison
Full size table

The number of comparison can be determined using

$$ no. of\ comparison=n\left(n-1\right)/2 $$
(6)

Where, n = number of parameter.

After formation of pairwise comparison matrix, priority vector is computed (the normalized Eigen vector of the matrix). The pair-wise comparison matrix is normalized by dividing the values by the sum of each column. A new matrix is formed and the normalized principal Eigen vector (or the priority vector) can be obtained by averaging across the rows.

The relative importance given to the parameters one over another is acceptable if the consistency ratio (CR) is less than 10%. If consistency ratio increases 10%, a new value is assigned in the pair-wise comparison matrix. CR is computed as:

$$ CR= CI/ RI $$
(7)

Where, CR = Consistency ratio.

CI = Consistency index

$$ =\left({\lambda}_{max}-n\right)/\left(n-1\right) $$

λmax = Principal Eigen value.

= value obtained from the summation of products between each element of Eigen vector and sum of columns of reciprocal matrix.

RI = Randomness Index.

RI which is derived from a sample of size 500 of a randomly generated reciprocal matrix using the scale 1/9, 1/8. .. .. 1. .. .. 8, 9, is given by the size of the matrix (or the number of parameters, n, in the comparison matrix) (Saaty 1987; Saaty 1995; Banai-Kashani 1989). When n is 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 corresponding RI will be 0, 0, 0.58, 0.9, 1.12, 1.24, 1.24, 1.32, 1.41, 1.45, 1.49 respectively.

3 Results and Discussion

Spatial information on potential runoff, LULC, slope, soil, stream order, lineament and settlement plays a critical role in site selection for water storage structure. Based on the spatial information parametric layers were prepared as follows:

3.1 Potential Runoff

Curve number (CN) map is prepared to determine the potential runoff. As per United States Department of Agriculture (USDA) Natural Resources Conservation Services (NRCS) (NRCS 2004; Subramanya 2013), curve number is based on the composition of two parametric layers i.e. land use and Hydrological Soil Group (HSG). Land use map of the study area was obtained from Bhuvan portal of National Remote Sensing Centre (NRSC) Hyderabad (bhuvan.nrsc.gov.in/). Table 3 present the areal extent of land use classified as build up, plantation, forest, wasteland, agriculture (rabi, kharif, zaid, double/triple, current fallow) and water bodies. The same is presented in Fig. 3.

Table 3 Areal extent of land use in the study area
Full size table
Fig. 3
figure 3

Land use in the study area (NRSC (ISRO)

Full size image

HSG map is prepared from the soil series published by NBSS&LUP, Nagpur. Depending upon the infiltration and runoff potential the soil is categorized into four groups A, B, C and D known as HSG. The main characteristic of HSG (Weerasinghe et al. 2011) is presented in Table 4 .The areal extent of HSG is shown in Fig. 4.

Table 4 Main characteristic of hydrological soil group (HSG)
Full size table
Fig. 4
figure 4

Hydrologic soil group and stream order in the study area

Full size image

CN is basically a coefficient that reduces the rainfall to runoff. It depends upon two parameter land use and hydrologic soil group (HSG) (Ahmad et al. 2015). The CN ranges from 0 to 100, 0 implies no runoff condition and 100 imply rainfall is equal to runoff. Land use map was procured from National Remote Sensing Agency (NRSC) Hyderabad in the scale of 1:250,000. As per NRSC (ISRO) the study area has been classified in buildup, plantation, forest, wasteland, agriculture (rabi, kharif, zaid, double/triple, current fallow) and water bodies. Soil data has been procured from National Bureau of Soil Survey & Land Use Planning, Nagpur (Tamgadge et al. 2002). It is converted to digital format using ArcGIS. For the study purpose it is required to group the various soil. As per USDA Natural Resources Conservation Services (NRCS) (NRCS 2004; Subramanya 2013), depending upon the infiltration and runoff potential, soil is categorized into four groups A, B, C & D known as Hydrologic Soil Group (HSG). Group A soil very low runoff potential, Group B low runoff potential, Group C soil moderate runoff potential and Group D soil is having high runoff potential. These two parametric layers are merged together to get the CN grid map. The annual rainfall and corresponding runoff is presented in the Table 6. According to the weighted curve number (AMCII) the runoff potential was classified as low runoff potential if CNII ranges from 26 to 50, moderate runoff potential if CNII ranges from 50 to 75 and high runoff potential if CNII ranges from 75 to 100.

These two parametric layers are merged together to get the curve number (CN) grid map. Runoff curve number is taken from USDA, National Engineering Handbook (NEH) (2004). The CN grid shown in Fig. 5. The CN grid thus obtained contains CN for each grid ranging from 26 to 100. For potential runoff computation it is required to have one CN for entire basin. For this purpose area weighted composite curve number (CCN) for various conditions of land use and hydrologic soil conditions are computed as follows:

$$ \mathrm{CCN}=\left({\mathrm{CN}}_1\times {\mathrm{A}}_1\right)+\left({\mathrm{CN}}_2\times {\mathrm{A}}_2\right)+\left({\mathrm{CN}}_3\times {\mathrm{A}}_3\right)+\dots +\left({\mathrm{CN}}_{\mathrm{n}}\times {\mathrm{A}}_{\mathrm{n}}\right)/\mathrm{A} $$
(8)

Where A1, A2, A3, ….….,An represent areas of polygon having CN values CN1, CN2, CN3,……...,CNn respectively and A is the total area of the basin. The CCN for the basin is 75.59 and it represents the previous 5 days antecedent moisture condition (AMCII) of the ground surface. CN for AMCI is 57.58 and CN for AMCIII is 87.88, for previous 5 days moisture condition. A sample calculation of runoff is presented in Table 5. The runoff potential map is presented in Fig. 6.

Fig. 5
figure 5

Curve number grid in the study area rages from 26 to 100

Full size image
Table 5 Sample of daily rainfall-runoff computation
Full size table
Fig. 6
figure 6

Runoff potential in the study area

Full size image
Table 6 Estimated yearly runoff depth from SCS-CN method
Full size table
Table 7 AHP process in determining the relative weight of parameters
Full size table
Table 8 Results of the pairwise comparison of the alternatives with respect to each parameter
Full size table
Table 9 Summary of priority of the alternatives with respect to each parameter
Full size table
Table 10 Calculation of overall priorities of alternatives
Full size table
Table 11 The influence of a change in the parameter priority for the scenarios
Full size table

3.2 Stream Order, Slope, Lineament and Settlement

Stream order map was prepared with the help of DEM analysis in ArcGIS. Spatial analyst tool is being applied for extracting the stream lines and stream order tool is utilized to derive the order of the stream lines. As per the GIS analysis, stream order ranges from 1st order to 6th order. The stream order map is shown in Fig. 4. Slope is one of the important factor for deciding the location of storage structure. For deriving the slope layer ASTER GDEM (Property of METI and NASA) with 30 m resolution DEM is used to determine the slope. The slope of the study area ranges from 0 to 80.69%. The same is subdivided into 5 sub-groups; high slope (> 64%), moderate to high slope (48–64%), moderate slope (32–48%), gentle slope (16–32%) and level to gentle slope (0–16%). The slope map of the study area has been shown in Fig. 7. Lineament involves fractures, faults and folds as identified in the study area. Lineament data is collected form Chhattisgarh Council of Science & Technology and then analysed in GIS, creating a buffer of 100 m around the lineaments. As per the 2001 census, there are 269 villages and 2 towns in the study area. The locations of these settlements were identified in terms of latitude and longitude. A buffer zone of 500 m around the village location and 1000 m around the towns were created. The settlement and lineament is shown in Fig. 8.

Fig. 7
figure 7

Slope type in the study area (in percentage)

Full size image
Fig. 8
figure 8

Settlement and lineaments in the study area

Full size image

3.3 Application of AHP

The first step in AHP is to build a hierarchy for the decision. The first level of the hierarchy is the objective of the study i.e. best suitable site for water storage. The second level of the hierarchy is established by the criteria i.e. parameters viz. runoff potential, land use, slope, HSG, stream order, lineament and settlement. The third level consists of three alternatives i.e. site suitability index 1, site suitability index 2 and site suitability index 3. Figure 9 shows the proposed hierarchy.

Fig. 9
figure 9

Decision Hierarchy for suitable site for water storage

Full size image

The second step in the AHP process is to derive the relative priority for the parameters. From the objective of the study, it is very clear that all the parameters are not equally important. For example a hydrologist may give more importance to stream order rather than other parameters, while a planner may give more importance to land use rather than other parameters. Hence, the importance of each parameter will be different and due to this pairwise comparison of each parameter is required for determining the relative priority of each parameter with respect to one over another in the scale for comparison developed by Saaty (Salih and Al-Tarif 2012) shown in Table 2. For all the seven parameters the relative importance is derived (Table 7). Thus, a common scale (0% to 100%) is obtained from AHP procedure. The runoff parameter emerges as the most important (relative weight 58%) followed by HSG (11%), stream order (9%), slope (9%), land use (7%), lineament (4%) and settlement (4%).

Once the judgement is entered, they are checked for their consistency. Since the numeric values are derived from the subjective preferences of persons, it is very difficult to avoid some inconsistencies in the final matrix. The consistency results are as follows:

$$ {\displaystyle \begin{array}{l}\mathrm{Principal}\ \mathrm{Eigen}\ \mathrm{vector}=7.457\\ {}\mathrm{Consistency}\ \mathrm{Index}\ \left(\mathrm{CI}\right)=\left({\uplambda}_{\mathrm{max}}-\mathrm{n}\right)/\left(\mathrm{n}-1\right)\\ {}=0.076\end{array}} $$

For n = 7, RI = 1.32

$$ \mathrm{Consistency}\ \mathrm{Ratio}\ \left(\mathrm{CI}/\mathrm{RI}\right)=5.776\% $$

The consistency ratio is less than 10%, hence the judgement is considered reasonably consistence and we continue the process of decision making with AHP. This also implies that the comparison performed in step 1 of Table 7 is considered acceptable for further process.

The third step involves the derivation of local priorities i.e. what are the priorities of the alternatives with respect to the parameter. For this purpose pairwise comparison between all the alternatives with each parameter is performed. In the present study there are three alternatives compared with each parameter. The result of pairwise comparison of alternatives with respect to each parameter is presented in Table 8 and summarized in Table 9.

The local priorities have been obtained; in the fourth step, it is required to derive the overall priority for each alternative. An overall priority of the alternatives was calculated by multiplying the priority of each parameter by the local priorities of alternatives of each parameter and addition of these provides the overall priority of the alternative. Table 10 presents the calculation of overall priority of alternatives. This process is also called model synthesis. From the calculation it is very clear that alternative III (overall priority = 0.483) is more preferable compared to alternative II (overall priority = 0.343) and alternative I (overall priority = 0.174).

The overall priorities were influenced by the intensity of importance assigned to each parameters. At this stage, it is important to perform “what-if” analysis, to check whether there is any change in the decision, if we change the priority of the parameters. The process is called sensitivity analysis and constitutes the fifth step of AHP. To perform the sensitivity analysis, it is necessary to make changes in the priorities of each parameter and see how they change the overall priorities of the alternatives. To illustrate this following scenario were analysed: priority equally distributed to all parameter; 50% priority given to runoff potential, other parameter priority were equally distributed and similarly for other parameter one after another. The influence of a change in the parameter priority for the scenarios are shown in Table 11. The scenario 1 shows the case where all the parameters have the same priority, the best choice is alternative III. For scenario 2, 3, 7 and 8 i.e. giving 50% priority to runoff potential, land use, lineament and settlement respectively, alternative III is having the highest priority. For scenario 4 i.e. 50% priority given to slope, alternative II and III are the best choices. For scenario 5, i.e. 50% priority given to HSG, alternative I is suitable. Lastly, for scenario 6 giving 50% priority to stream order alternative II found to the best choice. Out of eight scenario, there five scenario in which alternative III is the best choice. This implies that the comparisons performed in Step 1 of Table 7 holds good for the percentage importance (or weightage) judged for each parameter and hence suitable for overlay.

3.4 Weighted Overlay Analysis

To implement the rules laid down in Table 1 and the information layers prepared with their weightage are overlaid in GIS to identify the suitable water storage structure sites. The generated layers are in vector format, for weighted overlay analysis the “Rasterization” of each layer is performed. The first step of data conversion is “Rasterization” for converting different lines and polygon coverage into raster data format. After this, reclassification of all the raster files is processed along with providing the scale value of each unit. A scale value in the range 1 to 3 is used in which ‘1’ is for storage tank, ‘2’ is for stop dam and ‘3’ is for check dam.

From the information given in Table 1 (Suitability level) and Table 2 (AHP scale) the combined suitability for water storage is given as (Weerasinghe et al. 2011):

$$ SI=\left\{\left({W}_{RO}\times {SL}_{RO_{it}}\right)+\left({W}_{LU}\times {SL}_{LU_{it}}\right)+\left({W}_{SL}\times {SL}_{SL_{it}}\right)+\left({W}_{HSG}\times {SL}_{HSG_{it}}\right)+\left({W}_{SO}\times {SL}_{SO_{it}}\right)+\left({W}_L\times {SL}_{L_{it}}\right)+\left({W}_S\times {SL}_{S_{it}}\right)\right\} $$
(9)

Where:

SI :

suitability index;

W RO :

weight index for runoff potential layer;

W LU :

weight index for land use layer;

W SL :

weight index for slope layer;

W HSG :

weight index for hydrologic soil group layer;

W SO :

weight index for stream order layer;

W L :

weight index for lineament layer;

W S :

weight index for settlement layer;

\( {SL}_{RO_{ij}} \) :

suitability level of ith row and jth column with respect to runoff potential;

\( {SL}_{LU_i} \) :

suitability level of ith row and jth column with respect to land use;

\( {SL}_{SL_{ij}} \) :

suitability level of ith row and jth column with respect to slope;

\( {SL}_{HSG_{ij}} \) :

suitability level of ith row and jth column with respect to hydrologic soil group;

\( {SL}_{SO_{ij}} \) :

suitability level of ith row and jth column with respect to stream order;

\( {SL}_{L_{ij}} \) :

suitability level of ith row and jth column with respect to lineament and

\( {SL}_{S_{ij}} \) :

suitability level of ith row and jth column with respect to settlement.

Based on this weighted overlay analysis a site suitability map is prepared and presented in Fig. 10. Applying the suitability index mentioned in Table 1, it was found that there are 31 numbers of sites suitable for creating water storage structures, out of which 11 sites suitable for check dam and 20 sites suitable for stop.

Fig. 10
figure 10

Site suitability map with existing Mahamara Anicut in the study area

Full size image

4 Conclusion

In the present paper AHP was proposed to evaluate the location of the most suitable water storage site. Seven criteria/ parameters were used to evaluate the suitable site using the suitability level index, laid by IMSD. Three suitability index has been selected representing storage tank, stop dam and check dam respectively. Suitability of the location for different types of water strage structures depends upon various factors such as runoff potential, land use, HSG, slope, stream order, lineament and settlement. As the number of parameters increases, it involves more complexity in decision making. AHP has been utilized to address the complexity in the decision making i.e. suitability of water storage between storage tank, stop dam and check dam. AHP seems to be a suitable method to rank the parameters. To ensure the decision to be stable and robust sensitivity analysis is additionally applied. The result of sensitivity analysis reveals that alternative III i.e. check dams is preferable for the study area.

The paper also estimates the daily runoff potential using SCS-CN method for preparation of runoff potential map. Though there are number of methods available for computing the runoff, SCS-CN method is simple, well acclaimed and produces better result (Ramakrishnan et al. 2009). A large number of methods are available for finding the suitable sites for water storages. However, this study used the GIS based method for computing the number of suitable sites based on specific criteria. The priority determined for each parameter using AHP is applied as percentage influence in overlay analysis using GIS. This integration of AHP and GIS could be useful in decision making and in handling large data for performing spatial as well as vector analysis for effective water resources management.

To appraise the suitability of the selected site, the suitability map was superimposed over the LULC map of the study area and it was found that one of site under the alternative III is exactly located over the existing storage structure and in local language it is known as “Mahamara Anicut” situated in the Rajnandgaon District of Chhattisgarh State, India. This study demonstrated the integrated approach of remote sensing, GIS and AHP.