J. Cent. South Univ.

(2018) 25: 570−585

DOI: https://doi.org/10.1007/s11771-018-3762-3

Potential landfill site selection for solid waste disposal using GIS and
multi-criteria decision analysis (MCDA)

S. Kapilan1, K. Elangovan2
1. Department of Civil Engineering, Akshaya College of Engineering and Technology, Kinathukadavu,
Coimbatore–642109, India;
2. Department of Civil Engineering, PSG College of Technology, Coimbatore–641004, India
© Central South University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract: Proper solid waste disposal is an important socioeconomic concern for all developing countries.
Municipalities have their own policies, individual approaches and methods to manage the solid wastes. They consider
wastelands outside the urban area as the best suitable for the solid waste disposal. Such improper site selection will
create morphological changes that lead to environmental hazards in the urban and its surrounding areas. In this research,
the site selection for urban solid waste disposal in the Coimbatore district used geographical information system (GIS)
and multi-criteria decision analysis (MCDA). Thematic layers of lineament density, landuse/landcover, population
density, groundwater depth, drainage density, slope, soil texture, geology and geomorphology were considered as
primary criteria and weights for criteria, and sub-criteria were assigned by MCDA analysis. The resultant weight score
was validated by consistency ratio so that the efficiency of the selected criteria was justified. The overlay analysis in
GIS environment provides 17 potential zones in Coimbatore district, among which, four suitable sites were screened
and refined with the help of field investigation and visual interpretation of satellite image. The result of landfill
suitability map shows the effectiveness of the proposed method.

Key words: municipal solid waste; landfill site; multi-criteria decision analysis; remote sensing; GIS; Coimbatore

Cite this article as: S. Kapilan, K. Elangovan. Potential landfill site selection for solid waste disposal using GIS and
multi-criteria decision analysis (MCDA) [J]. Journal of Central South University, 2018, 25(3): 570–585. DOI:
https://doi.org/10.1007/s11771-018-3762-3.

rapid growth of cities with huge population creates

1 Introduction a lack of basic infrastructure services for municipal
solid waste (MSW) disposal [2]. The main reason
The increasing population and rapid for urban waste accumulation is somehow related
urbanization are of great concern to the municipal with the living standard and well-being of the
authorities for the management of solid wastes. The population through industrialization [3]. The
rapid growth of population demands the use of management of solid waste disposal is still a
non-renewable resources and dumps its by-products complicated issue for the city planners, as it
such as toxic solid waste into the environment. requires financial, environmental and political
Such type of solid wastes has resulted in consideration [4]. Site selection for solid waste
deteriorated soil, air and water bodies, which causes disposal is the primary task for solid waste
serious health hazards for the public [1]. The management. Solid wastes in the urban areas
disposal of generated solid wastes is formed once in mostly include wastes of plastics, glass, fabrics, and
a day in urban and semi-urban areas. However, metals, which have complex composition and late
Received date: 2016−11−11; Accepted date: 2017−03−23
Corresponding author: S. Kapilan, Assistant Professor; Tel: +91−7448571066; E-mail: [email protected], [email protected];
ORCID: 0000-0002-2656-9700
J. Cent. South Univ. (2018) 25: 570–585 571

degradable characteristics, creating more harm to its administrative boundary. In Coimbatore city, the
the environment. The present waste management major waste generation points are mostly the
techniques rely on reuse, recycling, reduction and households, hotels, restaurants, industries, hospitals,
recovery of energy concept [5]. Nevertheless, market places, slums, bus-stands and community
elimination of all forms of waste is almost halls. The municipal corporation collects the solid
impossible and such waste can be handled by wastes by street sweeping, door to door collection,
adopting suitable environmental friendly methods. collection from bins and open dumping and
One such suitable method is the landfill, although it collection of drain silt. However, they execute the
is found at the bottom of waste management processing and the disposal of the solid wastes in a
hierarchy and requires huge attention to reduce its common way without consideration of socio-
environmental impact [1]. Due to its low cost of physiographic parameters such as settlements,
handling with respect to other forms of waste groundwater and drainage network, agriculture and
management techniques, the developing countries other environmental considerations. The socio-
adopt landfill as a suitable method for solid waste environmental factors considered in this work is
disposal. based on the guidelines of Municipal Solid Waste
Enormous literatures on landfills are (Management & handling) Rules, India (2000),
theoretically available for selection of solid waste Central Pollution Control Board, India (CPCB,
landfill sites [5–11]. The site suitability for landfills 2003) and Central Public Health and Environmental
as outlined in the literature requires complex and Engineering Organization (CPHEEO, 2000).
multidisciplinary approach, which includes
ecological, social, technical, economic and 2 Study area
environmental considerations. Moreover, settlement
of solid wastes is based on the waste composition, The study area is located in the north-western
density and moisture content [12], hence, part of the Tamil Nadu state with the elevation of
geomorphic condition of landfill sites should satisfy about 410 m above mean sea level (MSL). The
these considerations. The management of such huge coverage of Coimbatore district is 4732 km2, which
disaggregated data by traditional means is a tedious includes 10 taluks and 295 villages. The district is
and time-consuming task and often results in poor surrounded by Nilgiri biosphere reserve forest in
output. Geographical information system (GIS) the North, Valparai reserve forest in the south,
based decision analysis tools help in resolving this Western Ghats in the west and the eastern part is
problem and suggests suitable sites for landfilling, predominantly dry, bounded by Erode and Tiruppur
waste segregation and recycling process [6, 13–17]. districts. The location map of the study area is
The decision-making is a kind of data mining shown in Figure 1. The study area is the second
process helps to solve day-to-day problems using largest city in Tamil Nadu and fastest growing
the standard optimization techniques. There are tier-II cities in India. The district preserves
several techniques available in literature, however, salubrious climate with an average annual
few of them are feasible for site selection temperature of 26.3 °C. The average annual rainfall
applications. The multi criteria decision analysis is 693 mm amongst northeast and southwest
(MCDA) is an analytical hierarchy process (AHP) monsoon contributing to 47% and 28%,
which is developed by SAATY [18]. This technique respectively. Besides, Coimbatore district is a major
provides a mean of decomposing the criteria into hub for textiles, manufacturing industries, education
sub-criteria that can be hierarchically ranked on a and commerce. Owing to these salient features,
numerical scale [19, 20]. The MCDA technique people are likely to settle in Coimbatore other than
integrated with GIS provides qualitative and any districts of Tamil Nadu. The district has a
quantitative results compared to that of the simple population count of 3458045 among which
weighted average techniques [5, 21–23]. 2618940 are residing in the urban area and 839105
The present study aims to identify specific are in the rural area (Census of India, 2011). The
sites for solid waste disposal in the Coimbatore waste generated in the municipal corporation is 601
district. Coimbatore municipal corporation is Mt/d, out of which 76% of the solid wastes are
divided into five major zones covering 100 wards in biodegradable that are primarily originated from the
572 J. Cent. South Univ. (2018) 25: 570–585

Figure 1 Location map of proposed study area

residences and markets. methodological framework is presented in Figure 2.

The detailed description of data collection and
3 Materials and methods processing methods is described in the subsequent
sections.
Methodology in this strategy includes thematic
map generation, assigning weight factor, weighted 3.1 Thematic data
overlay analysis by multi-criteria decision analysis In this work, nine thematic layers namely,
and site prioritization. The schematic structure of lineament density, landuse/landcover, population
J. Cent. South Univ. (2018) 25: 570–585 573

Figure 2 Schematic structure of methodological framework used in present study

density, drainage density, groundwater table, slope, by support vector machine (SVM) algorithm with
soil texture, geology and geomorphology have been adjustable learning parameters [25]. The data of
used for site suitability analysis. Lineament feature groundwater table for Coimbatore district are
in the present study area is extracted from the retrieved from annual report of Central Ground
panchromatic image of Landsat 7 ETM+ data. Water Board (2014–2015). The data were imported
Landsat data (path: 144 and row: 52, 53) acquired as a point feature in ArcGIS and the interpolation
on 21 February, 2002 (cloud coverage less than map was generated by using IDW spatial analysis
3.64) is used for lineament extraction. The images tool.
were geometrically corrected by radiometric Village wise population data for 2011 is
correction algorithm [24]. Sobel filter with different obtained from census of India website
orientation (N-S, NE-SW, E-W and NW-SE) is (http://censusindia.gov.in/). The data were used to
applied to the panchromatic images for extraction create population density map in ArcGIS
of linear features. These linear lines were assigned environment. The SRTM DEM with 30 m
to binary values (1 for line, 0 for others) and resolution data is used to extract drainage density
imported into ArcGIS 10.3 platform. The line present in the region. The DEM data are imported
features were converted to vectors by raster to into River Tool software (version 4.0,
vector spatial analysis toolbox. The length of http://rivix.com/) and extract drainage density
polyline features measured by geometry features using drainage network module. Further,
calculations and actual lineaments are identified density features were imported into ArcGIS
through onscreen visual interpretation using Google software and transformed to point features by
earth. Further, line density tool is used to generate spatial analyst toolbox. The points are converted
lineament density map of the present study region. into a raster map by IDW spatial interpolation
Likewise, Landsat 8 (OLI/TIRS) multispectral data technique. On the other hand, the DEM is used to
acquired on 20 March 2016 is used for landuse/ extract slope features by ArcGIS spatial analysis
landcover (LULC) classification. Prominent tool. Soil features in the present study region is
features like agriculture, fallow land, scrub land, extracted from existing soil map provided by the
settlement, forest and water bodies were classified agricultural engineering department (AED), Tamil
574 J. Cent. South Univ. (2018) 25: 570–585
Nadu, in the year 2012. Similarly, geology and parameters are converted into thematic layers for
geomorphology features are extracted from the decision-making using analytic hierarchy process
district resource map published in 1995 by the (AHP) [18, 20]. The score for each criterion and its
Geological Survey of India, Hyderabad, India. sub-criteria is fixed based on relative importance to
each other as well as guidelines of existing research
3.2 GIS-based multi criteria decision analysis publications (Table 1) [26–33]. The rating of the
(GIS–MCDA) scale is assigned on a 9 point continuous scale (9, 8,
The GIS–multi criteria decision analysis is a 7, 6, 5, 4, 3, 2, 1, 1/2, 1/3, 1/4, 1/5, 1/6, 1/7, 1/8,
decision making process that transforms the 1/9). If the factor value is on the left side of 1,
geographical data into the hierarchical network comparison matrix constituted with actual value
based on the importance of criteria and its otherwise constituted with reciprocal value. The
sub-criteria. The methodology of the GIS–MCDA score 9 for the most important and 1 for equal
contains the following processes. At first, the most importance whereas 1/9 represents the least
important parameters (criteria) are assigned for the important [34].
problem. In the present case, population, The normalization matrix method is used to
groundwater table, landuse/landcover, slope of the calculate priority vector (weight) of each criterion
area, drainage density, soil type, lineament density, and its sub-criteria. The value of each cell in the
geomorphology and geology are the crucial matrix is constituted by score value of the
parameters for landfill site selection. The comparison matrix divided by the column total.

Table 1 Grading values for selected criteria

Range/classification with Grading value (1—less preferred;
Site No. Name of criteria Criteria guideline
respect to present study area 9—extremely preferred)
>3000 1
2001–3000 2
1001–2000 4
1 Population/km–2 Ref. [44]
501–1000 6
101–500 8
0–100 9
0–2 1
2.01–5 2
5.01–10 4
2 Ground to water table/m Refs. [28, 30, 33]
10.01–20 6
20.01–30 8
>30 9
Water bodies 1
Forest 2
Settlement 3
3 Landuse/landcover Refs. [29, 44]
Agriculture 6
Scrub land 8
Fallow land 9
>30 1
20.01–30 2
10.01–20 3
4 Slope/(°) Refs. [26, 32]
5.01–10 5
2.01–5 7
0–2 9
to be continued
J. Cent. South Univ. (2018) 25: 570–585 575
Continued
Range/classification with Grading value (1—less preferred;
Site No. Name of criteria Criteria guideline
respect to present study area 9—extremely preferred)
>5 1
4–5 2
Drainage density/
5 Ref. [31] 3–4 5
(km·km–2)
2–3 7
<2 8
Clay loam 1
Silt clay loam 2
Coarse granule loam 4
6 Soil Refs. [27, 28]
Coarse sandy loam 5
Loamy sand 7
Sandy clay loam 9
>0.8 1
Lineament density/
7 Refs. [23, 28, 30] 0.41–0.8 5
(km·km–2)
0–0.4 7
Structural hills 1
Flood plain 2
Colluvial fan 4
8 Geomorphology Ref. [31]
Buried pediment 5
Shallow weathered buried pediplain 6
Moderately weathered buried pediplain 9
Granites 1
Basic anorthosite 2
Granitoid and gneiss 3
9 Geology Refs. [23, 28]
Garnet silliminate graphite gneiss 4
Magmatites 6
Charnockite gneiss and pyroxene
9
granulites/fluvial

Further, principal eigenvector (weight) of each reference of SAATY [18, 20], the RI value for
factor is calculated from the average value of row factors is assigned as 1.41. In individual, the factors
matrix. The consistency of the comparison matrix is such as population density, groundwater table,
evaluated by the consistency ratio (CR, Rc). The landuse/landcover, slope, soil type, geomorphology
CR value for each matrix can be calculated as and geology use 6 sub-criteria and therefore the RI
follows: value is 1.24. Conversely, RI values for drainage
CI density (with 5 sub-criteria) and lineament density
Rc  (1) (with 3 sub-criteria) are 1.12 and 0.58, respectively.
RI
The derived weight values are assigned to each
(max  n)
CI= (2) thematic layer and reclassified it by ArcGIS raster
n 1
reclassification tool. Further, site suitability index
where λmax is the product between each element of (SSI, Iss) is calculated by:
the weight and the column total of the comparison
matrix, n is the number of criteria or sub-criteria. I ss  ( Pw  Psw )  (GWt w  GWtsw )  ( LULC w 
The random consistency index (RI) is a constant LULCsw )  ( Sp w  Spsw )  ( Dw  Dsw ) 
parameter which is assigned on the basis of criteria ( Sl w  Slsw )  ( Lw  Lsw )  (Gmpw  Gmpsw ) 
and its sub-criteria we considered. Based on the (Geol w  Geolsw ) (3)
576 J. Cent. South Univ. (2018) 25: 570–585
where subscript ‘w’ is the weight of the factor and Conversely, areas in south of Mettupalayam, east of
‘sw’ is the weight of the sub-factor. The letters of Thondamuthur, east of Vellalur and southeast of
the decision criteria are P–population density, GWt– Kinathukadavu have water depth of more than 30 m.
groundwater table, LULC–landuse/landcover, Sp– The groundwater depth map is divided by five
slope, D–drainage density, Sl–soil type, L– grades based on the guidelines of published
lineament density, Gmp–geomorphology and Geol– research works [23, 30]. The shallow groundwater
geology. The result of SSI raster image has different depth areas are the least considered zones for
pixel values based on the weights of the criteria and landfilling while deep groundwater depth areas are
sub-criteria. The raster data were reclassified into highly preferable for landfilling. The weight and
three classes: highly suitable, less suitable and not score value for groundwater depth criteria is shown
suitable using standard classification scheme in Tables 1 and 2. It is revealed that least weight of
(quantile classification method) [35]. 0.04 falls on shallow depth areas and maximum
weight of 0.35 falls on high groundwater depth
4 Weight score for criteria and sub- areas.
criteria
4.3 Landuse / landcover
4.1 Population density The landuse/landcover (LULC) was prepared
The area covered by Coimbatore Municipal with five major classes namely agriculture,
Corporation is almost occupied by households, settlements, fallow land, scrubland and forest. Of
hotels, restaurants, industries, hospitals, market the total area, forest covered 38.62%, followed by
places, slums, bus-stands and community halls. settlements 18.15%, agriculture 13.14%, fallow
Landfilling sites within these areas might be land 12.66%, scrubland 10.27% and water bodies
harmful for the human beings and therefore less 7.17%. The areas within forest, settlement and
population density areas are mostly preferred. The water bodies are considered inappropriate for
population density data are categorized into landfilling sites and are therefore given less
minimum of 100 km–2 and maximum of above importance. The same is applied to agriculture land,
5000 km–2. The score value for least population as they are sensitive to ecological system. The
density is given most importance and high fallow lands are uncultivated lands that are mostly
population density is given least importance. The not used to raise crop. The northeastern part of the
weight and score value for each sub-criterion are study area is mostly of fallow lands due to
shown in comparison matrix (Table 2). The table considerable urban expansion. The scrublands
reveals that least weight of 0.04 falls on urban areas composed of shrubs or grass dominated lands are
and the maximum weight of 0.41 falls on rural and unsuitable for human habitation due to the presence
forest areas. However, least population sites far of natural disturbance such as dense plants and
away from the urban areas are considered as poisonous insects. The fallow and scrub lands are
unsuitable due to the expensive waste away from the human activities hence, they are
transportation. considered as most suitable areas for landfilling
sites. The weights for each sub-criterion in Table 2
4.2 Groundwater table shows least weight of 0.04 for water bodies and
The groundwater table is said to be the maximum weight of 0.31 for scrub land.
distance between the ground surface and the water
table [33]. The depth of groundwater table provides 4.4 Slope
an insight of contamination risk of groundwater in The stability of land surface is measured by
order to limit anthropogenic contamination. the slope value present in the region. In the landfill
Groundwater data of Coimbatore district showing suitability analysis, slope parameter plays a vital
shallow water depth range from 1 to 5 m are role to identify suitable sites with respect to the
observed at Karamadai, Sulur, Thnodamuthur and waste load into the surface. The steep slope with
Pollachi. The area from Kinathukadavadu to overload leads to landslide in the region. Moreover,
Pollachi has water depth range no more than 10 m vehicles consume more fuel for transportation.
which delineates enrichment of groundwater source. Therefore, flat or gentle sloping areas are more
J. Cent. South Univ. (2018) 25: 570–585 577

Table 2 Pair wise comparison matrix, factor weights and consistency ratio of sub-criteria of factors
Factor A B C D E F Weight
A (>3000) 1 1.00 0.50 0.20 0.14 0.13 0.04
B (2001–3000) 1 1 0.50 0.20 0.14 0.13 0.04
C (1001–2000) 2 2 1 0.40 0.28 0.25 0.11
Population
D (501–1000) 5 4 2 1 0.71 0.63 0.14
density/km–2
E (101–500) 7 6 4 2 1 0.86 0.25
F (0–100) 8 7 5 3 1 1 0.41
Consistency ratio: 0.03
A (< 2) 1 1.00 0.33 0.20 0.14 0.13 0.04
B (2.1–5) 1 1 0.33 0.20 0.14 0.13 0.04
C (5.1–10) 3 2 1 0.60 0.43 0.38 0.10
Depth to water
D (10.1–20) 5 4 2 1 0.71 0.63 0.18
table/m
E (20.1–30) 7 6 4 2 1 0.86 0.29
F (>30) 8 7 5 3 1 1 0.35
Consistency ratio: 0.09
A (Water bodies) 1 1.00 0.50 0.20 0.14 0.13 0.04
B (Forest) 1 1 0.50 0.20 0.14 0.13 0.04
C (Settlement) 2 1 1 0.33 0.20 0.17 0.07
Landuse/
D (Agriculture) 5 4 3 1 0.50 1.00 0.20
landcover
E (Scrub land) 7 6 5 2 1 1.00 0.31
F (Fallow land) 8 7 6 1 1 1 0.31
Consistency ratio: 0.02
A (>30) 1 1.00 0.50 0.25 0.17 0.13 0.04
B (20.01–30) 1 1 0.50 0.25 0.17 0.13 0.04
C (10.01–20) 2 1 1 0.50 0.33 0.25 0.07
Slope/(°) D (5.01–10) 4 3 2 1 0.66 0.50 0.16
E (2.01–5) 6 5 4 2 1 0.75 0.26
F (0–2) 8 7 6 4 2 1 0.42
Consistency ratio: 0.03
A (>5) 1 0.50 0.25 0.17 0.14 0.04
B (4.01–5) 2 1 0.50 0.33 0.28 0.08
Drainage
C (3.01–4) 4 3 1 0.66 0.57 0.19
density/
D (2.01–3) 6 5 2 1 0.86 0.31
(km·km–2)
E (<2) 7 6 3 1 1 0.38
Consistency ratio: 0.09
A (Coarse granule loam) 1 1.00 0.33 0.25 0.17 0.13 0.04
B (Coarse sandy loam) 1 1 0.33 0.25 0.16 0.13 0.04
C (Loamy sand) 3 2 1 0.75 0.50 0.38 0.11
Soil type D (Sandy clay loam) 4 3 1 1 0.66 0.50 0.14
E (Silt clay loam) 6 5 3 2 1 0.75 0.25
F (Clay loam) 8 7 5 4 2 1 0.41
Consistency ratio: 0.06
to be continued
578 J. Cent. South Univ. (2018) 25: 570–585
Continued
Factor A B C D E F Weight
A (>0.8) 1 0.25 0.17 0.09
Lineament B (0.41–0.8) 4 1 0.66 0.34
density/
(km·km–2) C (0–0.4 ) 6 2 1 0.57
Consistency ratio: 0.07
A (Structural hills) 1 1.00 0.33 0.25 0.20 0.13 0.04
B (Flood plain) 1 1 0.33 0.25 0.20 0.14 0.04
C (Colluvial fan) 3 2 1 0.75 0.60 0.38 0.11
Geomorphology D (Buried pediment) 4 3 1 1 0.80 0.50 0.15
E (Shallow weathered buried pediplain) 5 4 2 1 1 0.63 0.19
F (Moderately weathered buried pediplain) 8 7 5 4 3 1 0.45
Consistency ratio: 0.02
A (Charnockite gneiss and pyroxene
1 1.000 0.50 0.33 0.20 0.13 0.05
granulites/fluvial)
B (Garnet silliminate graphite gneiss) 1 1 0.50 0.33 0.20 0.13 0.05
C (Granitoid and gneiss) 2 1 1 0.66 0.40 0.25 0.08
Geology D (Magmatites) 3 2 1 1 0.60 0.38 0.12
E (Basic anorthosite) 5 4 3 2 1 0.63 0.24
F (Granites) 8 7 6 5 3 1 0.47
Consistency ratio: 0.07

suitable for landfilling sites. The slope map is density (>5 km/km2) secured lesser weights (0.04)
prepared with five categories (0–2°, 2.01°–5°, (Table 2).
5.01°–10°, 10.01°–20°, 20.01°–30° and >30°). The
central and northeastern part of the study area has 4.6 Soil type
most of gentle slope surface whereas northwestern Solid wastes applied to the soils that have less
and southern parts of the study area are steep slope permeability reduce surface erosion, runoff and
due to the presence of Western Ghats. The flat and improve the degradation rate [36]. The present
gentle slope areas have secured high score values study area has different soil types namely, clay
than others (Table 2). loam, silt clay loam, coarse granule loam, coarse
sandy loam, loamy sand and sandy clay loam. The
4.5 Drainage density soil group of coarse granule loam, coarse sandy
The drainage density data of present study loam and loamy sand have high permeability rate
reflect the close spacing of channels as well as than others. The soil group of loamy sand has
formation of surface and subsurface present in the moderated infiltration rate whereas sandy clay loam,
region. The drainage density is a key parameter to silt clay loam and clay loam have poor infiltration
evaluate surface runoff present in the region. The rate. It reveals that soil type with less infiltration
lesser drainage density reveals less possible of (permeability) rate is more suitable for solid waste
surface runoff whereas higher drainage density is disposal. Table 2 shows the weights of the soil type
prone to higher runoff prevailed in the region. The sub-criteria. As seen, soil texture of clay loam
obstacles present in the runoff path may lead to obtains higher weights (0.41) and soil texture of
serious flood prevailed in the region. Therefore, the coarse granule loam obtains lower weights (0.04).
areas that are away from the potential drainage
density are more suitable for landfilling sites. In the 4.7 Lineament density
present case, lesser drainage density (<2 km/km2) Lineament density represents the intensity of
secured higher weights (0.38) and higher drainage the surface fractures present in a unit area [37–39].
J. Cent. South Univ. (2018) 25: 570–585 579
The lineaments are linear features, which reflect and unconsolidated, porous formation. The rock
secondary porosity and permeability of the rock types such as charnockite gneiss and pyroxene
structure based on folding, faulting and fracturing granulites, fluvial and garnet silliminate graphite
[40]. The lineament trends in the present study area gneiss has high infiltration rate hence, it is not
are mostly of NW–SE, NE–SW and NEE–SWW suitable for landfilling process. On the other hand,
directions. The area with less lineament density is granitoid and gneiss and magmaties have less
more suitable for landfilling sites because rocks infiltration rate and therefore less suitable for waste
with infiltration rate act as a shield for ground disposal. The area with basic anorthosite and
water–solid waste contamination. The present study granites has poor infiltration rate that is highly
area is mostly of less lineament density (0– suitable for landfilling sites. Table 2 illustrates high
0.4 km/km2). The northwest and southern parts of weight for granites (0.47) and low weight for
the study area have moderate to high lineament charnockite gneiss and pyroxene granulites and
density due to the hilly terrain present in the region. fluvial (0.05).
The weights of lineament density is shown in
Table 2, which illustrates that the less lineament 5 Results and discussion
density attained weight factor of 0.57 and high
lineament density attained weight factor of 0.09. 5.1 Landfill suitability map
The suitable landfill site was delineated by
4.8 Geomorphology integrating the thematic layers with suitable weights
The surface geomorphology reflects the using GIS–MCDA techniques. The comparison
topography of the landforms, drainage pattern and matrix was developed for eight criteria and the
formation of surface sediments and their relation to weights were calculated for the factors by a
the geological features. The present study is made sequence of multiplication. The acceptable
up of structural and fluvial origin and has six major consistency ratio (0.0452) is obtained from the pair
classes namely, structural hills, flood plain, wise matrix. The calculated weight of each factor is
colluvial fan, buried pediment, shallow weathered presented in Table 3. The weight results show that
buried pediplain and moderately weathered buried the population is the most important factor (0.261)
pediplain. The structural hills present in the region and geology is the least important factor (0.0248).
are covered by hilly terrain with dense forest. The The weight factor of groundwater table (0.2022) is
bajada and colluvial plain are present in the comparable with the population factor. The reason
foothills with poor sorted sediments with bed. The for the significance given to the population and
flood plains are located parallel to the river flow groundwater table is that the disposal sites can
and the region has a possibility of flood overflow cause explosion, air and water pollution that are
due to runoff or heavy rainfall. Therefore, possible harmful to the human beings. Landuse/landcover
sites in these areas are not suitable for landfilling. factor is considered as another important factor
The buried pediments are deposited by stream (0.1578) because landfilling sites should be away
erosion and weathering and it has a gentle slope from the social criteria such as settlements,
nature so that likely to be suitable for landfilling cultivated lands and hilly terrains. Followed by
site. The shallow and moderate weathered buried slope, drainage, soil, lineament and geomorphology
pediplains are derived by continuous process of factors obtain the weight scores. The weights for
pedimentation and therefore it increases the waste criteria and sub-criteria are integrated to the
degradation. The weight score for thematic layer and landfill suitability map is
geomorphological parameters in Table 2 reveals prepared by site suitable index (SSI) analysis in
that buried pediplain attained high weights (0.45) GIS environment. The resultant map has been
and structural hills (0.04) attained less weight for categorized into highly suitable; less suitable and
selection of landfilling sites. not-suitable (Figure 3). The suitable sites resulted in
the site suitability index (SSI) map are mostly in the
4.9 Geology northeast and southern part of the Municipal
The geology of the study area is underlain by Corporation. This is due to the presence of drainage
consolidated hard rock formations of Archaean age and vegetation features in the west, southwest and
580 J. Cent. South Univ. (2018) 25: 570–585

Table 3 Pair wise comparison matrix, factor weights and consistency ratio of factors
Factor A B C D E F G H I Weight
Geology (A) 1 1 0.5 0.33 0.25 0.2 0.17 0.14 0.13 0.024809
Geomorphology (B) 1 1 0.5 0.33 0.25 0.2 0.17 0.14 0.13 0.024837
Lineament (C) 2 1 1 0.67 0.5 0.4 0.33 0.29 0.25 0.045789
Soil (D) 3 2 1 1 0.75 0.6 0.5 0.43 0.38 0.068183
Drainage (E) 4 3 2 1 1 0.8 0.67 0.57 0.5 0.093052
Slope (F) 5 4 3 2 1 1 0.83 0.71 0.63 0.122227
Landuse/landcover (G) 6 5 4 3 2 1 1 0.86 0.75 0.157843
Depth to water table (H) 7 6 5 4 3 2 1 1 0.86 0.202247
Population (I) 8 7 6 5 4 3 2 1 1 0.261013
Consistency ratio: 0.0452

northwest region. In particular, northwest part is village, partial cultivated lands and they are near to
surrounded by dense forest with hilly terrain surface the river drainage. Sites 2, 3, 4, 8, 15 and 16 were
that create complex road accessibility. located too far from the municipal boundary and
highway road network. Therefore, they secured
5.2 Post-processing analysis rank 3 and were chosen as the less suitable sites for
The MCDA approach in conjunction with GIS solid waste disposal. In past decades, Coimbatore
based overlay analysis provides 17 suitable sites in Municipal Corporation has used existing landfilling
Coimbatore district (Figure 3). The result was based site, which is located at Vellalore (geocoordinate is
upon criteria and sub-criteria, which were selected 10°57'31.25"N and 77° 0'13.61"E). The local
from existing maps and remote sensing techniques. statistical report stated that around 5×105 t of solid
However, it does not provide accessibility of wastes were dumped at Vellalore over the span of a
potential zones in the present day condition. year. Several residents in and around the dump site
Therefore, subsequent screening and refinement seek to relocate the dump yard due to the increasing
was carried out to identify landfill suitable sites respiratory problems, skin rashes, irritation in the
based on field investigation. The potential sites eyes, groundwater pollution and mosquito menace
were reclassified and ranked based on road network which is reported in the Hindu newspaper [41].
availability and proximity to the Coimbatore Field photographs in Figure 4 show the improper
Municipal Corporation (Table 4) as the distance waste dumping as well as air pollution caused by
between site and major road network is important in burning of solid wastes at open air condition. The
the case of waste transportation and landfill sites present result helps to select an alternative landfill
should be located at a place where it can be reached sites and thereby to avoid socio-environmental
to urban areas. problems.
Sites 9, 11, 12, 13 and 14 were found to be in
close proximity to the Coimbatore Municipal 5.3 Implications to proper solid waste disposal
Corporation where it is located at Kurumbapalayam, Regions with developed resources and
Vellalore (12 & 13), Onappalayam and residence have more livelihood stability compared
Chettipalayam, respectively. They are very close to to that of the regions of less developed areas. The
the national highway (NH-47 and NH-209) and rapid increase in population density leads to
these areas are comprised of uncultivated land and encroachment of water bodies for their settlements
waste land. The coverage of landfilling area is high and creates a polluted infrastructure. Additionally,
in Kurumbapalayam (2.5052 km2) while other sites landfill sites in this region cause more harm to the
have less than 1 km2. These sites have secured rank environment. Furthermore, solid wastes may
1 and are chosen as the most suitable landfill sites contain broken glass, aerosol cans and potentially
in Coimbatore district. Although sites 1, 6 7 and 17 explosive chemicals or oils and therefore landfill
are closer to the highway, they are located away sites should be away from the urban, village and
from the municipal boundary and are comprised of industrial areas. Likewise, landfill sites near to the
J. Cent. South Univ. (2018) 25: 570–585 581

Figure 3 Landfill site suitability map of present study area

582 J. Cent. South Univ. (2018) 25: 570–585
Table 4 Selected landfill sites and its physiographic parameters (Rank # 1: most suitable, # 2: moderately suitable, # 3:
less suitable)
Geographical Distance from
Distance from
Site Area/ coordinates highway
Rank Site name available local Proximity coverage
No. km2 Distance/ Highway
Latitude Longitude roads/km
km type
Coimbatore Municipal
1 2 Kattupalayam 1.1939 11°21'18.78" 77° 6'38.81" 0.64 SH170 — north, Mettupalayam,
Puliampatti, Annur
Coimbatore Municipal
2 3 Solavampalayam 0.8920 11°18'38.90" 77° 4'51.53" 7.41 SH170 0.30 north, Mettupalayam,
Puliampatti
Coimbatore Municipal
3 3 Near Rasadi 0.5860 11°17'35.24" 77° 5'7.31" 5.62 SH170 0.16 north, Mettupalayam,
Puliampatti
Coimbatore Municipal
4 3 Near Alankuttai 0.7402 11°17'31.99" 77° 5'7.89" 10.04 SH170 1.09 north, Mettupalayam,
Puliampatti
Coimbatore Municipal
Chinna
5 2 0.7676 11°16'21.69" 76°57'57.44" 1.91 NH67 — north, Mettupalayam,
Thottipalayam
karamadai
Coimbatore Municipal
6 2 Periya Puthur 0.3243 11°14'7.07" 77° 1'56.38" 0.74 SH168 —
north, Annur, karamadai
Coimbatore Municipal
7 2 Nallichettipalayam 0.9239 11°14'4.29" 77° 3'51.66" 0.79 SH80 — northeast, Annur
karamadai
Coimbatore Municipal
northeast, Avinashi,
8 3 Anandhapuram 0.3306 11° 7'2.53" 77°13'19.39" 5.08 NH47 3.73
Karumathampatti,
Somanur
Coimbatore Municipal
9 1 Kurumbapalayam 2.5052 11° 5'19.32" 77° 2'47.11" 2.03 NH209 0.74
north
Coimbatore Municipal
10 2 Kaniyur 1.3613 11° 4'49.59" 77° 8'54.34" 1.02 NH47 —
northeast, Sulur
Coimbatore Municipal
11 1 Onappalayam 0.2276 11° 1'5.33" 76°51'55.94" 6.08 SH164 0.84
west
Coimbatore Municipal
12 1 Vellalore 0.9374 10°57'36.55" 77° 3'2.09" 1.59 NH47 0.00
southeast, Sulur
Coimbatore Municipal
13 1 Vellalore 0.6062 10°56'43.86" 77° 1'45.24" 1.63 NH47 0.00
southeast, Sulur
Coimbatore Municipal
14 1 Chettipalayam 0.5899 10°54'9.69" 77° 3'32.51" 2.02 SH163 0.00
southeast
Coimbatore Municipal
15 3 Pachapalayam 0.3142 10°53'25.49" 77° 4'58.64" 5.96 SH163 1.28
southeast
Coimbatore Municipal
16 3 Mettuvavi 0.3655 10°49'45.59" 77° 7'33.41" 4.36 SH19 1.06
southeast, Sultanpet
Coimbatore south,
17 2 Kinathukadavu 0.2708 10°48'48.27" 77° 1'50.16" 1.72 NH209 0.00
Pollachi

Figure 4 Field photos: (a) Improper waste dumping; (b) Air polluted by burning solid wastes at open air condition
(Vellalore dump yard)
J. Cent. South Univ. (2018) 25: 570–585 583

drainage path may cause blockages that result in results will ultimately support the municipalities
surface water pollution, flooding and insanitary and government sectors to select the landfilling
conditions. Therefore, by no doubt these regions are sites. The awareness on proper waste disposal is
not suitable for solid waste disposal. In earthquake necessary for those who are generating the wastes
prone areas, it is necessary to examine the general in day-to-day life. Therefore, social welfare
trend of ground response prior to the selection of societies should take the initiative for awareness
landfill sites [42, 43]. Based on the above analysis, programme and provide training to people for
the landfill sites should be a wasteland away from proper solid waste management.
cultivated land, forest, settlements and hazardous
zones. It is also suggested that prior to landfilling, References
the solid wastes need to be processed by the
[1] RAHMAN M M, SULTANA K R, HOQUE M A. Suitable
window compost or aerobic composting and refuse sites for urban solid waste disposal using GIS approach in
derived fuel (RDF) methods. Khulna city, Bangladesh [J]. Proceedings of the Pakistan
Academy of Sciences (Pakistan), 2008, 45: 11–12.
6 Conclusions [2] HANNAN M A, ABDULLA AL MAMUN M, HUSSAIN A,
BASRI H, BEGUM R A. A review on technologies and their
usage in solid waste monitoring and management systems:
The abrupt increase in the accumulation of
Issues and challenges [J]. Waste Management, 2015, 43:
solid waste and its management is one of the 509–523.
greatest challenges faced by the municipal [3] OJHA C S P, GOYAL M K, KUMAR S. Applying fuzzy
authorities of Coimbatore. The present research logic and the point count system to select landfill sites [J].
provides optimal solution to characterize landfill Environmental Monitoring and Assessment, 2007, 135(1–3):
suitable sites in Coimbatore district using GIS 99–106.
[4] BABAN S M J, FLANNAGAN J. Developing and
based MCDA analysis. Unlike numerical models,
implementing GIS-assisted constraints criteria for planning
GIS has the capability to store, analyze and display landfill sites in the UK [J]. Planning Practice and Research,
huge spatial and aspatial data. The pair-wise 1998, 13(2):139–151.
comparison matrix performs decision-making and [5] KONTOS T D, KOMILIS D P, HALVADAKIS C P. Siting
assigns optimal weights to the criteria and sub- MSW landfills with a spatial multiple criteria analysis
criteria. The consistency ratio of 0.0452 which is methodology [J]. Waste Management, 2005, 25(8): 818–832.
[6] DELGADO O B, MENDOZA M, GRANADOS E L,
less than 0.10 endorsed the accuracy of the
GENELETTI D. Analysis of land suitability for the siting of
comparison matrix. The integrated overlay of inter-municipal landfills in the Cuitzeo Lake Basin, Mexico
thematic layers in GIS environment suggested 17 [J]. Waste Management, 2008, 28(7): 1137–1146.
potential sites around the Coimbatore Municipal [7] CHANG N B, PARVATHINATHAN G, BREEDEN J B.
Corporation. Later, the results were refined by the Combining GIS with fuzzy multicriteria decision-making for
post-processing techniques, in which four sites at landfill siting in a fast-growing urban region [J]. Journal of
Environmental Management, 2008, 87(1): 139–153.
Kurumbapalayam, Vellalore (2 sites), Onappalyam
[8] SHARIFI M, HADIDI M, VESSALI E, MOSSTAFAKHANI
and Chettipalayam were found to be more suitable P, TAHERI K, SHAHOIE S, KHODAMORADPOUR M.
for landfilling sites in Coimbatore district. These Integrating multi-criteria decision analysis for a GIS-based
areas are mostly wastelands that are near to the hazardous waste landfill sitting in Kurdistan Province,
municipal corporation and away from the human western Iran [J]. Waste Management, 2009, 29(10):
2740–2758.
dwelling places, thus eliminating the risk of social
[9] HIGGS G, LANGFORD M. GIScience, environmental
conflict. Geo-environmental parameters were justice, & estimating populations at risk: The case of landfills
considered as the key factors, which govern the site in Wales [J]. Applied Geography, 2009, 29(1): 63–76.
selection of potential landfill sites. The geo- [10] GENELETTI D. Combining stakeholder analysis and spatial
processing of entire methodological framework for multicriteria evaluation to select and rank inert landfill sites
delineating the landfill site in a developing country [J]. Waste Management, 2010, 30(2): 328–337.
[11] SIVAKUMAR B G L, ASCE M, REDDY K R, ASCE F,
like India by considering the key factors is a crucial
SRIVASTAVA A. Influence of spatially variable geotechnical
step for planning and execution. Future research properties of msw on stability of landfill slopes [J]. Journal
will include a public participatory campaign for of Hazardous, Toxic, and Radioactive Waste, 2014, 18(1):
assessing the performance and functioning of 27–37.
scientifically selected landfill sites. The present [12] SIVAKUMAR B G L, LAKSHMIKANTHAN P. Estimation
584 J. Cent. South Univ. (2018) 25: 570–585
of the components of municipal solid waste settlement [J]. [28] FRANTZIS I. Methodology for municipal landfill sites
Waste Management and Research, 2015, 33(1): 30–38. selection [J]. Waste Management Research, 1993, 11(5):
[13] JENSEN J, CHRISTENSEN E. Solid and hazardous waste 441–451.
disposal site selection using digital geographic information [29] MONTSERRAT Z, EMILIO M, ALVARO H, ALEJANDRO
system techniques [J]. Science of the Total Environment, G, ANGEL R. Evaluation of a municipal landfill site in
1986, 56: 265–276. southern spain with GIS-aided methodology [J]. Journal of
[14] SIDDIQUI M, EVERETT J, VIEUX B. Landfill siting using Hazardous Materials, 2008, 160(2, 3): 473–481.
geographic information systems: A demonstration [J]. [30] WANG Gui-qin, QIN Li, LI Guo-xue, CHEN Li-jun.
Journal of Environmental Engineering, 1996, 122(6): Landfill site selection using spatial information technologies
515–523. and AHP: A case study in Beijing , China [J]. Journal of
[15] FATTA D, SARAVANOS P, LOIZIDOU M. Industrial waste Environmental Management, 2009, 90(8): 2414–2421.
facility site selection using geographical information system [31] JAVAHERI H, NASRABADI T, JAFARIAN M H,
techniques [J]. International Journal of Environmental ROWSHAN G R, KHOSHNAM H. Site selection of
Studies, 1998, 56(1): 1–14. municipal solid waste landfills using analytical hierarchy
[16] KONTOS T D, KOMILIS D P, HALVADAKIS C P. Siting process method in a geographical information technology
MSW landfills on Lesvos island with a GIS-based environment in giroft [J]. Iran Journal of Health Science and
methodology [J]. Waste Management and Research, 2003, Engineering, 2006, 3(3): 177–184.
21(3): 262–277. [32] KUMAR S, HASSAN M I. Selection of a landfill site for
[17] SENER B, SUZEN M L, DOYURAN V. Landfill site solid waste management: An application of AHP and spatial
selection by using geographic information systems [J]. analyst tool [J]. Journal of Indian Society of Remote Sensing,
Environmental Geology, 2006, 49(3): 376–388. 2013, 41(1): 45–56.
[18] SAATY T L. The analytic hierarchy process, planning, [33] BABALOLA A, BUSU I. Selection of landfill sites for solid
piority setting, resource allocation [M]. New York: waste treatment in damaturu town-using GIS Techniques [J].
McGraw-Hill, 1980. Journal of Environmental Protection, 2011, 2(1): 1–10.
[19] ROY R. The analytic hierarchy process. strategic decision [34] AHMED B. Landslide susceptibility mapping using multi-
making: Applying the analytic hierarchy process [M]. criteria evaluation techniques in Chittagong Metropolitan
London: Springer London, 2004. Area, Bangladesh [J]. Landslides, 2015, 12(6): 1077–1095.
[20] SAATY T L. Decision making with the analytic hierarchy [35] SENER S, SENER E, KARAGUZEL R. Solid waste
process [J]. International Journal of Services Sciences, 2008, disposal site selection with GIS and AHP methodology: A
1(1): 83–98. case study in Senirkent–Uluborlu (Isparta) Basin, Turkey [J].
[21] LUKASHEH A F, DROSTE R L, WARITH M A. Review of Environmental Monitoring and Assessment, 2011, 173(1–4):
expert system (ES), geographic information system (GIS), 533–554.
decision support system (DSS), and their applications in [36] AWMFH. Chapter 5—Role of soils in waste management.
landfill design and management [J]. Waste Management and Part 651-Agricultural waste management field handbook
Research, 2001, 19(2): 177–185. [EB/OL]. [2009]. http://directives.sc.egov.usda.gov/6620.
[22] Al-JARRAH O, ABU-QDAIS H. Municipal solid waste wba.
landfill sitting using intelligent system [J]. Waste [37] RAVINDRAN K V, JEYARAM A. Groundwater prospects of
Management, 2006, 26(3): 299–306. shahbad tehsil, baran district, eastern rajasthan: A remote
[23] SUMATHI V R, NATESAN U, SARKAR C. GIS-based sensing approach [J]. Journal of the Indian Society of
approach for optimized siting of municipal solid waste Remote Sensing, 1997, 25(4): 239–246.
landfill [J]. Waste Management, 2008, 28(11): 2146–2160. [38] SAXENA R K, VERMA K S, CHARY G R, SRIVASTAVA
[24] CHANDER G, MARKHAM B L, HELDER D L. Summary R, BARTHWAL A K. IRS-1C data application in watershed
of current radiometric calibration coefficients for Landsat characterization and management [J]. International Journal
MSS, TM, ETM+, and EO-1 ALI sensors [J]. Remote of Remote Sensing, 2000, 21(17): 3197–3208.
Sensing of Environment, 2009, 113(5): 893–903. [39] KALIRAJ S, CHANDRASEKAR N, MAGESH N S.
[25] JOEVIVEK V, HEMALATHA T, SOMAN K P. Determining Evaluation of multiple environmental factors for site-specific
an efficient supervised classification method for groundwater recharge structures in the Vaigai River upper
hyperspectral image [C]// 2009 International Conference on basin, Tamil Nadu, India, using GIS-based weighted overlay
Advances in Recent Technologies in Communication and analysis [J]. Environmental Earth Sciences, 2015, 74(5):
Computing. IEEE, 2009: 384–386. 4355–4380.
[26] ALAVI N, GOUDARZI G, BABAEI A A, JAAFARZADEH [40] REDDY GPOBI, MOULI K C, SRIVASTAV S K,
N, HOSSEINZADEH M. Municipal solid waste landfill site SRINIVAS C V, MAJI A K. Evaluation of ground water
selection with geographic information systems and analytical potential zones using remote sensing data—A case study of
hierarchy process: A case study in Mahshahr County, Iran [J]. gaimukh watershed, bhandara district, maharashtra [J].
Waste Management and Research, 2013, 31(1): 98–105. Journal of the Indian Society of Remote Sensing, 2000, 28(1):
[27] ELAHI A, SAMADYAR H. Municipal solid waste landfill 19–32.
site selection using analytic hierarchy process method for [41] THE HINDU. Relocate Vellalore dump yard: AAP,
tafresh town [J]. Middle-East Journal of Scientific Research, Coimbatore edition [EB/OL]. [2016–08–10]. http://www.
2014, 22(9): 1294–1307. thehindu.com/news/cities/Coimbatore/relocate-vellalore-dum
J. Cent. South Univ. (2018) 25: 570–585 585
p-yard-aap/article4944417.ece, 2013-07-23/. Improving seismic monitoring system for small to
[42] JOEVIVEK V, CHANDRASEKAR N, intermediate earthquake detection [J]. International Journal
JAYANGONDAPERUMAL R. Evaluation of optimal of Computer Science and Security (IJCSS), 2010, 4(3):
wavelet filters for seismic wave analysis [J]. Himalayan 308–315.
Geology, 2016, 37(2): 176–189. (Edited by FANG Jing-hua)
[43] JOEVIVEK V, CHANDRASEKAR N, SRINIVAS Y.

中文导读

基于 GIS 和多准则决策分析的垃圾填埋场选址研究

摘要：妥善处置固体废物是所有发展中国家的一个重要的社会经济问题。通常市政有自己的策略和方
法来管理固体废物，他们认为城市以外的荒地是最适合固体废物处置的地方，而不当的场地选择会造
成形态变化，危害城市及其周围地区的环境。本文采用地理信息系统(GIS)和多准则决策分析(MCDA)
技术对 Coimbatore 地区城市固体废物进行选址。主要标准和权重包括线性密度、土地利用/土地覆盖、
人口密度、地下水埋深、河网密度、坡度、土壤质地、地质地貌，再采用 MCDA 分析方法对各子标
准进行划分。通过一致性比验证所选标准的权重，从而证明所选标准的有效性。通过 GIS 环境下的叠
加分析，选出 Coimbatore 地区的 17 个可能的区域，再通过野外调查和卫星图像的可视化筛选和细化
出 4 个适宜的地点。填埋场适宜性地图验证了该方法的有效性。

关键词：城市固体废物；填埋场；多指标决策分析；遥感；地理信息系统；哥印拜陀市（印度）