Cities 50 (2016) 111–118

Contents lists available at ScienceDirect

Cities

journal homepage: www.elsevier.com/locate/jcit

Delhi's land cover change in post transit era

Sohail Ahmad a,b,1, Ram Avtar a,c,⁎, Mahendra Sethi a,d, Akhilesh Surjan e
a
United Nations University, Institute for the Advanced Study of Sustainability (UNU-IAS), Tokyo, Japan
b
Tokyo Institute of Technology, Tokyo, Japan
c
The University of Tokyo, Japan
d
National Institute of Urban Affairs, New Delhi, India
e
Faculty of Engineering, Health, Science and the Environment; Charles Darwin University, Darwin, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Growing urbanization and recent Mass Rapid Transit System (MRTS) play an important role in land cover change
Received 8 February 2015 in Indian cities. However, understanding about direction and magnitude of this change is limited, especially in
Received in revised form 8 September 2015 reference to MRTS introduction, which is required to assess sustainable urban futures. Thus, this study attempts
Accepted 9 September 2015
to assess pattern of land cover change, paying special attention to the development of MRTS (both metro lines
Available online 25 September 2015
and stations) in the National Capital Territory of Delhi. Land covers are classified using Landsat images from
Keywords:
year 2001 and 2011. In order to measure transformations in developed areas, this study employs maximum
Urban growth likelihood supervised classification and performs buffer analyses along the metro lines and stations. The results
Land cover change reveal that growth of built-up area is higher in peripheral districts, whereas relatively low along the MRTS.
Built-up area This study indicates that ongoing development process needs corrective measures, such as increasing built-up
Urban transit areas across the metro stations and lines, and planned provisioning of physical and social infrastructure in
Remote sensing peripheral areas to induce sustainable urban development. To implement these spatial interventions, robust
Delhi implementation strategies are needed.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Urbanization provides opportunities for employment, better hous-
ing, education, knowledge and technology transfer, and ready markets
for the agricultural products, but enormous stress on natural resources,
existing social services and infrastructure (Rees, 1992). India, one of the
rapidly urbanizing and economically prospering nations of Asia, is full of
promises and challenges. Urbanization patterns of developing and de-
veloped economies differ in many aspects. Rapid growth and prolifera-
tion of informal settlements, in particular, are unique only to developing
countries. Urban India accommodated 377 million people (31.2% of total
population) as per Census 2011, the second largest urban population in
the world (after China), spreads across 7935 urban centers, including 53
Urban Agglomerations (UAs), which are defined as cities with over a
million people (Census of India, 2011). The annual urban population
growth rate during the last decade (2001–11) was 2.76%, relatively
high in comparison to developed economies (UNDESA, 2011). The
growth rate of built-up area is higher than the growth rate of
⁎ Corresponding author at: United Nations University, Institute for the Advanced Study
of Sustainability (UNU-IAS), 5-53-70 Jingumae, Shibuya-ku, Tokyo 150-8925, Japan.
Tel.: +81 3 5467 1241.
E-mail addresses: architectsohail@gmail.com (S. Ahmad), avtar@unu.edu,
ram.envjnu@gmail.com (R. Avtar).
1
Current Address: Mercator Research Institute on Global Commons and Climate
Change (MCC), Torgauer Straße 12-15, 10829 Berlin Germany.
population, for instance, global estimate shows that by 2030 built-up
area will be tripled, while urban population will be doubled (Angel
et al., 2005). This would create excessive pressure on land for develop-
ment projects, particularly in the developing economies. It was estimat-
ed that the overall demand of land for various Government projects/
schemes during the XII Five Year Plan (2007–12) would be over
0.3 mHa (Sethi, 2011). This high population growth rate, pressure on
land, along with indiscreet urban policy and institutional environment
has led to proliferation of unplanned and imbalanced development in
urban India.
In 1985, the National Capital Region (NCR) of Delhi came into exis-
tence. The motto of the NCR was to diffuse the population in the region
and alter the changes in the land use (NCR Board, 2005). In this context,
this study maps urban growth patterns of Delhi, the capital city of India,
particularly in reference to Mass Rapid Transit System (MRTS) and indi-
cates potential directions for sustainable urban futures. Open areas like
agriculture, forests, and scrubs have drawn greater attention among city
managers and policy-makers because of their growing role in providing
ecosystem services, as well as mitigating greenhouse gas emissions and
adaptation to climate change (Avtar, Yunus, Kraines, & Yamamuro,
2015; Emmanuel & Baker, 2012). Strong linkage has been established
between built-up density and energy consumption in cities, at least in
developed economies. Thus, it may also be construed that densification
is by and large a pre-requisite to urban sustainability. For instance, in-
creasing built-up density reduces gasoline consumption (or

http://dx.doi.org/10.1016/j.cities.2015.09.003
0264-2751/© 2015 Elsevier Ltd. All rights reserved.
112 S. Ahmad et al. / Cities 50 (2016) 111–118
expenditure in public and private vehicles) (Ahmad, Baiocchi, &
Creutzig, 2015; Creutzig, Baiocchi, Bierkandt, Pichler, & Seto, 2015;
Newman & Kenworthy, 1989). In addition, densification promotes the
use of public transport systems, which is predominantly used by disad-
vantaged communities that leads to inclusive mobility. Depending upon
the nature of spatial policy, MRTS follows densification or sprawl, as ob-
served in case of transit oriented development. In this backdrop, this
study investigates the influence of MRTS on land cover change in
Delhi. Findings reveal that built-up area has increased, whereas agricul-
ture area has decreased significantly. Notably built-up area has largely
been added in peripheral areas rather than along transit. Based on
these findings we suggest for transit oriented development, and provi-
sion of urban services in peripheral areas through robust implementa-
tion strategies.
Following introduction, Section 2 presents literature review on
understanding the impact of urban rail system on land cover change.
Subsequently, Section 3 specifies the methodological framework, and
Section 4 presents results from the remote sensing–geographical infor-
mation system (RS–GIS) based analysis. Finally, Section 5 synthesizes
the findings and presents policy implications.
2. Urban rail systems and land cover change
Urban rail systems influence land cover/use change, urban mobility,
property values, and greenhouse emissions, among others, which affect
urban sustainability in both ways positive as well as negative. For positive
outcomes, merely introduction of transits is not sufficient, but require
comprehensive interventions. Today one of the major challenges in land
use planning is how it deals with urbanization, particularly rapid land
cover change, over and above that induced by mass rapid transit systems.
For decades, inquiry related to influence of transportation infrastructure
investment on urbanization patterns has been front and center of urban
and regional planning discourse. Historically, changes in urban forms, re-
gardless of its shape and density follow the evolution of transportation
technology, especially the innovation of automobile (Muller, 2004). The
contribution of several transportation infrastructures, such as highways,
to suburbanization has been analyzed intensively, whereas discussion
concerning urban rail system is limited (Baum-Snow, 2007). The evolu-
tion of sprawl in American cities, in particular, can be traced back to
how car-oriented infrastructure created urban sprawl as development
reinforced the way people travel (Handy, 2005). Over the last decade,
research concluded that automobile-oriented urbanization has become
the main engine for metropolitan growth pattern, including directing its
population movement (Baum-Snow, 2007; Handy, 2005).
Despite several studies investigating the relations between urban rail
system and urbanization, yet many aspects remain unexplored, particu-
larly related to emerging economies. The existing literature primarily fo-
cuses on challenges and prospects of urban rail development to
residential and commercial property, including causal effect between
rail stations and property values (Anas & Armstrong, 1993; Bowes &
Ihlanfeldt, 2001; Cervero, 2004; Cervero & Duncan, 2002; Dewees,
1976), and rail transit and economic development (Bollinger &
Ihlanfeldt, 1997; Cervero, 1984, 2004). Studies also explore how urban
rail systems curb sporadic urbanization and direct metropolitan growth
(Guiliano & Agarwal, 2010).
Dulal, Brodnig, and Onoriose (2011) look at the role of urban design
forms – settlement density – housing and employment activities and
the effects they could have in reducing travel demands, motor vehicle
dependency and GHG emissions. They infer that urban planning can
be very effective in shifting private vehicle dependency to public and
other alternative environmentally friendly modes of transports (such
as walking and cycling), in the long term. A mixture of high residential
development and employment density could influence shorter com-
muting trips and a reduction in private vehicle use, if supported by an
efficient public transport system and appropriate fiscal and regulatory
instruments. Santos, Behrendt, and Teytelboym (2010) rightly suggest
that a sustainable model for transport policy requires integration with
land-use policies. These may be somewhat limited within the bounds
of existing cities, but as cities grow and new cities are built, urban
planners must emphasize on land use for sustainable transport in
order to reduce congestion and CO2 emissions.
In this context, Suzuki, Cervero, and Iuchi (2013) for a World Bank
study explore the complex process of transit and land-use integration
in rapidly growing cities in developing countries. They report that
well-integrated transit and land development create urban forms and
spaces that reduce the need to travel by private motorized vehicles.
Areas with good access to public transit and well-designed urban
spaces, that are walkable and bikeable, become highly attractive places
for people to live, work, learn, play and interact. The study draws lessons
from global best case examples of transit-oriented metropolises that
have direct relevance to cities in developing countries. It differentiates
articulated densities (that are strategically introduced across parts of
the metropolitan area) and average density, and seeks for highly
concentrated development rather than dispersed ones for transit and
land-use integration. It further mandates that in order to integrate
transit and land-use, there is a need for strategic vision and enabling
institutional and regulatory framework.
From a methodology perspective, a limited number of researchers
have relied on remote sensing data to assess direct and indirect effects
of urban rail system and urbanization (Seto, Güneralp, & Hutyra,
2012). They developed spatially explicit probabilistic forecasts of global
urban land-cover change and direct impacts on biodiversity hotspots
and carbon biomass. However, some researches utilize transportation
models to examining the relationship between the availability of trans-
portation infrastructure and services and the pattern of house prices in
an urban area (Martínez & Viegas, 2009). Moreover, empirical investiga-
tion from developing countries that may account for the effect of urban
rail system on land cover change is limited.
An urban rail system, in theory, is seen as a tool to reduce the use of
automobile, improve the environment while increasing urban density
and curb urban sprawl (American Planning Association, 2002). It is
normatively expected that urban rail investment will induce land
cover change yet research predicts ambiguous results according to the
status of existing transit system, as it may or may not change the relative
accessibility significant enough for residents and business to change
their property location in the proximity of rail system (Giuliano,
1995). Thus, this paper aims to build empirical evidence to fill the
research gaps in explaining whether and how urban rail system and
its subsequent complementary policies produce land cover change.
Using the newly built metro in the Delhi Metropolitan Area as the
case, this paper analyzes whether and to what extent the urban rail
transit initiates a change of land cover along the metro lines/stations
and the peripheral areas.
3. Methodology
To appreciate the relationship between urbanization and Land Use
and Land Cover (LULC), a quantitative and evidence based methodology
is adopted, at multiple scales—the city, the sub-city (district) and the
local station level. Decadal urban population data for the city is available
since 1931, enumerated by the Census of India. It includes significant
demographic indicators such as migration and natural increase, which
help in comprehensively understanding the pattern of urbanization.
LULC for Delhi have regularly been documented since the first Master
Plan of Delhi, 1962 (MPD), prepared by the Delhi Development Author-
ity (DDA). Thereafter, temporal data is available for successive plans i.e.
1981, 2001 and 2021. Since one of the key interests of this research is to
analyze local changes in land cover in the recent past, evidence available
from remote sensing data is used. Satellite imageries acquired from
Landsat satellite for the year 2001 and 2011 are used.
Remotely sensed data have been extensively used to monitor
and model urban change (Banerjee & Srivastava, 2013; Dewan &
 S. Ahmad et al. / Cities 50 (2016) 111–118
Yamaguchi, 2009; Xiao et al., 2006). Although historical maps and aerial
photographs have played an important role to characterize various
objects on the earth surface, satellite data from various sensors have
been widely used to monitor the spatio-temporal changes on the
earth surface with the recent advancement in the remote sensing
techniques (Avtar, Singh, Shashtri, & Mukherjee, 2011; Avtar,
Takeuchi, & Sawada, 2013a, 2013b). Taking cue from such studies, satel-
lite datasets of Landsat 7 Enhanced Thematic Mapper Plus (ETM+) and
Landsat 5 Thematic Mapper (TM) were radiometrically calibrated and
geometrically registered over the study area, which were acquired on
28 January 2001 and 16 January 2011 respectively. The images were
geometrically rectified to a common Universal Transverse Mercator
(UTM) WGS-84 coordinate system.
LULC classification is an important application of remote sensing,
which involves extraction of thematic information from satellite data.
It determines the spectral relationship between spectral signatures
and various classes that is of interest to the user (Wulder et al., 2008).
LULC information is also useful for effective management of land re-
sources and policy designing. Various classification approaches have
been developed to classify satellite data. This study uses maximum like-
lihood classification (MLC) technique employing ENVI 4.7 software. The
land cover map of Delhi for year 2001 and 2011 are classified into five
classes viz. built-up, forest, water, open/scrub land, and agriculture
(Fig. 1). Based on 120 randomly selected ground control points from
Google Earth with the help of very high resolution images, the classifica-
tion accuracy of the land cover map was assessed. The accuracy of the
land cover map 2001 and 2011 was about 79% and 83% respectively.
Both scenes of satellite data were selected for the same season (winter)
and same month (January) to minimize the influence of seasonality on
classification. We noticed some misclassification in the scrub land and ag-
riculture land because some farms were not covered by any vegetation. To
improve the quality of the classification, majority of the filter with win-
dow size 3 × 3 neighborhood pixels was applied during post classification.
Further, we have analyzed land cover change from 2001 to 2011, disag-
gregated at district level. Further changes in land cover classes namely
Fig. 1. Classified Landsat images of Delhi, 2001 and 2011.
113
built-up, forest, water, open/scrubland, agriculture and unclassified have
also been analyzed along the metro network by creating a buffer zone
of 750 m along the metro lines and metro stations respectively.
The satellite imagery used in this study is co-terminus with the Cen-
sus 2001 and 2011 that is being employed to study urbanization. In
order to assess the spatial distribution or variations of land cover chang-
es across the city, data has been tabulated at the district level. Delhi has
9 districts: South, South-West, North, North-East, North-West, East,
Central, New Delhi and West. For each district changes in urban popula-
tion, density, and land cover act as indicators for analysis. To investigate
the changes in land cover along the MRTS corridors, the two most
significant lines cardinally crisscrossing the urban fabric – blue line
(East–West) and yellow line (North–South) have been taken as
samples. Both lines were almost non-existent in 2001, (for which the
first satellite imagery is studied), but were under full operation by
2010 (for which 2011 satellite imagery is used). A buffer zone of
750 m has been considered along all the stations on these two lines, to
decipher the extent and nature of land cover change.
4. Results
4.1. Land covers change in Delhi, 2001–11
The result revealed that rapid increase in population, with decadal pop-
ulation growth rate of 21%, resulted in significant changes in land cover be-
tween 2001–2011 (Table 1). During this decade, built-up areas have
reached to 855 km2, an addition of 100 km2 (decadal growth rate 13.5%).
In the same period, the agriculture land cover reduced by 11% from
433 km2 to 384 km2. However, forest cover has increased by 1.64 km2,
which could be attributed to afforestation efforts of the government.
Results disaggregated by districts show that most of the built-up cover
has increased in outer districts of Delhi such as South-West and South,
while forest cover has increased in core districts. During 2001–2011,
each district shows diverse results for land cover in five different typolo-
gies, as explained below.
114 S. Ahmad et al. / Cities 50 (2016) 111–118

Table 1
Land covers change in Delhi, by district, 2001–2011.

Year Land cover (km2) South South-West North North-East North-West East Central New Delhi West Total

167.81 43.76 51.17 190.62 52.29 13.52 17.39 77.45 754.09

Built-up 140.30 (54.87)
(40.85) (69.53) (71.31) (42.49) (79.40) (83.30) (49.44) (67.17) (50.88)
23.34 24.92 5.86 3.35 20.97 3.16 1.63 14.05 3.71 100.97
Forest
(9.13) (6.07) (9.31) (4.67) (4.67) (4.80) (10.03) (39.96) (3.22) (6.81)
2.08 1.37 0.49 2.45 0.94 1.81 0.04 0.04 0.84 10.06
2001 Water
(0.82) (0.33) (0.79) (3.41) (0.21) (2.75) (0.24) (0.10) (0.73) (0.68)
64.77 60.33 2.70 3.32 37.63 2.89 0.42 2.24 9.28 183.56
Open/Scrub
(25.33) (14.69) (4.29) (4.62) (8.39) (4.39) (2.59) (6.38) (8.05) (12.39)
25.20 156.36 10.12 11.47 198.48 5.70 0.62 1.45 24.02 433.39
Agriculture
(9.85) (38.06) (16.08) (15.99) (44.24) (8.66) (3.84) (4.13) (20.83) (29.24)
161.47 207.88 46.09 59.17 207.71 54.54 13.10 17.03 88.82 855.59
Built-up
(63.15) (50.60) (73.24) (82.46) (46.30) (82.82) (80.72) (48.42) (77.03) (57.73)
21.55 18.39 9.14 2.88 26.20 3.78 2.40 14.28 4.02 102.61
Forest
(8.43) (4.48) (14.52) (4.01) (5.84) (5.74) (14.81) (40.61) (3.48) (6.92)
2.87 1.44 1.00 4.05 2.98 2.76 0.09 0.07 1.54 16.80
2011 Water
(1.12) (0.35) (1.59) (5.65) (0.66) (4.20) (0.54) (0.20) (1.34) (1.13)
48.37 21.24 2.31 1.83 41.52 1.14 0.39 2.47 3.50 122.76
Open/Scrub
(18.92) (5.17) (3.67) (2.55) (9.26) (1.73) (2.39) (7.01) (3.04) (8.28)
21.42 161.84 4.39 3.83 170.22 3.63 0.25 1.32 17.43 384.31
Agriculture
(8.38) (39.40) (6.98) (5.33) (37.94) (5.51) (1.54) (3.76) (15.12) (25.93)
Built-up 15.09 23.87 5.34 15.64 8.97 4.31 −3.09 −2.05 14.67 13.46
Forest −7.65 −26.20 56.01 −14.17 24.97 19.47 47.68 1.63 8.27 1.63
Change %
Water 37.69 4.93 102.55 65.51 216.13 52.58 127.91 100.00 84.10 67.03
2001–2011
Open/Scrub −25.31 −64.79 −14.59 −44.72 10.34 −60.47 −7.92 9.91 −62.29 −33.12
Agriculture −14.99 3.51 −56.59 −66.66 −14.24 −36.36 −60.03 −8.99 −27.44 −11.33

Note: Parentheses values are the percentage share of the land cover.

Built-up: The built-up area has increased for seven districts, and
decreased for two districts, namely Central and New Delhi. The most sig-
nificant increase in numerical (40.07 km2) and percentage (23.87%) is ob-
served for South-West district. The area has witnessed major planned
developments like Dwarka sub-city by DDA and expansion of the Indira
Gandhi International Airport, in addition to several unplanned develop-
ments. On the contrary, there has been a marginal decline in built-up
areas of 2–3% in the central parts of the capital.
Forests: There is a slight increase of forest area, about 1.64 km2
(1.63%). Out of the nine districts of Delhi, there has been an increase in
forest cover in six, while three witnessed a decline. The decadal change
in land use varies across districts from −14.17% in North-East to 56.01%
in North. It is noteworthy that the destruction of forest cover is evident
in the same three districts — South, South-West, and North-East that ex-
hibits the maximum increase in land under built-up areas.
Water: During 2001–2011, Delhi has shown an increase in area under
water bodies. The increase from 10.06 km2 to 16.80 km2 (67.03%) comes as
a respite for the city that is located in hot-arid climatic region. The positive
change in this land use is witnessed in all the nine districts with results
ranging from 4.93% in South-West district to 216.13% in North West
district.
Open/Scrub land: Land cover under open scrub has decreased from
183.56 km2 in 2001 to 122.76 km2 in 2011. The decline of 33.12% is
significant in both percentage terms and distribution across the city.
Out of the nine districts, the decline is evident in seven districts ranging
from 7.92% to 64.79%. The only two districts showing a positive trend
are New Delhi (9.91%) and North-West (10.34%).
Agriculture: Similar to open/scrub, there is a decrease in agriculture
land cover from 433.39 km2 in 2001 to 384.31 km2 in 2011. The decline
may seem insignificant in percentage points (−11.33%), but in numerical
Airport Express), with a total length of 190 km, with 142 stations
(Table 2). The lines are combination of elevated, at-grade and under-
ground lines. While phase I and II have been completed, phases III and
IV are under construction and expected to be completed by 2016 and
2021 respectively, with the network spanning about 440 km by then
(The Indian Express, 2011). Normally, work on metro starts a few
years ahead of the operation, say three to four years, depending upon
the size and nature of construction works, for instance, part of the red
line 1 (Shahdara and Tis Hazari) started physical construction work in
October 1998 and inaugurated in December 2002, which took four
years. Therefore, this point should be kept in mind, while interpreting
land cover change along lines or stations.
Fig. 3 shows change in land use/land cover across MRTS. Across all
seven lines land cover of about 14.33 km2 has been changed to built-
up areas, which is 7% increase from 204 km2 to 218.3 km2. A large
share of change is attributed to urbanization and remaining construc-
tion of the metro (stations and lines). Delhi's decadal increase in built-
up areas owing to urbanization is about 13.5%, whereas, along MRTS
lines is about one-half. Fig. 2 clearly shows sparse built-up along MRTS
lines/stations. This difference indicates expansion of built-up areas in
peripheral areas, whereas large parts of metro lines were laid into
existing built-up areas.
Among all seven lines, Airport line experienced maximum change
both in terms of area and percentage (5.3 km2 and 25%) and only red
Table 2
History of Delhi's MRTS, 2002–2011.
First Last Length Character
Line operational extension Stations (km) #

terms of 49.08 km2, it is profound within the overall land use pattern of Red line-1 24-Dec-02 4-Jun-08 21 25.09 mostly E
the city, to an extent that all districts of Delhi (except South West) exhibit Yellow line-2 20-Dec-04 3-Sep-10 34 44.65 U and E
Blue line-3 31-Dec-05 30-Oct-10 44 49.93 A, E and U
a decline ranging from 8.99% (New Delhi) to 66.66% (North East).
Blue line-4 7-Jan-10 14-Jul-11 7 8.74 A, E and U
3-Apr-10 – 13 15.14 E
Green line-5
27-Aug-11 – 2 3.32 E
4.2. Land covers change along metro lines and stations, 2001–11 Violet line-6 3-Oct-10 14-Jan-11 15 20.04 E and U
Airport express 23-Feb-11 – 6 22.7 A, E and U
Delhi Metro, a rapid transit system, serves Delhi and adjoining areas Total 142 190

(Gurgaon, Noida and Ghaziabad). The network consists of six lines (plus Note: # A: at-grade; E: elevated; U; underground; Source: DMRC Website, 2011.
 S. Ahmad et al. / Cities 50 (2016) 111–118 115

Fig. 2. New (modern) development around a metro station on the selected line (upper panel) and low rise development along Anand Vihar Metro Station and Anand Vihar ISBT (lower panel).

line (number 1) experienced decrease in built-up area, which is extend-

Fig. 3. Change in built-up areas across metro lines, 2001–2011.
ed from Rithala to Dilshad Garden. Possible reason for decrease could be
removal of slum settlements along this line. After airport line, blue line
(number 3) and violet line (number 6) experienced similar change in
built-up areas to the tune of about 12%. Yellow line (number 2), blue
line (number 3) and green line (number 5) experienced increase in
built-up areas of about 6%, 5% and 2.3% respectively.
Two lines – yellow and blue – are further selected to assess built-up
area change across the stations. The yellow line was recently extended
from HUDA city center (Gurgaon, Haryana) to Jahangirpuri, while the
blue line from Dwarka Sector 21 to Noida city center (Noida, Uttar
Pradesh), thus both crossing the border of Delhi. But for our analysis,
we consider only metro stations located in Delhi, therefore we excluded
a few stations (see notes of Table 3).
Similar to metro lines, here we also draw a buffer zone of 750 m at
each station and measured change in built-up area during 2001 and
116 S. Ahmad et al. / Cities 50 (2016) 111–118

Table 3 Table 5
Change in built-up across stations of yellow line (2) and blue line (3) (750 m buffer). Change in length of roads, registered vehicles and modal split in Delhi.

Yellow line-2 (30 stations) Blue line-3 (39 stations) 2000–01 2010–11* % Change

2001 2011 % Change 2001–11 2001 2011 % Change 2001–11 Length of roads (in km) 28,508 31,183 9.38

Mean 1.24 1.29 5.01 1.51 1.53 3.08 Number of registered vehicles Cars and jeeps 920,723 2,173,323 136.05
Std. dev. 0.43 0.35 20.84 0.26 0.25 18.88 Two-wheelers 2,230,534 4,342,403 94.68
Min 0.08 0.16 −36.19 0.74 0.75 −31.16 Auto rickshaw 86,985 88,181 1.37
Max 1.77 1.77 47.44 1.77 1.77 65.57 Taxis 18,362 57,958 215.64
Busses** 41,483 61,471 48.18
Note: Stations located in other states, viz. Haryana and Uttar Predesh, are excluded (yellow
Good vehicles 158,492 209,370 32.10
line: Sikandarpur, MG Road, IFFCO Chowk, and Huda City Centre; blue line: Noida S15,
Total 3,456,579 6,932,706 100.57
Noida S16, Noida S18, Botanical garden, Golf course, and Noida city center).
Modal split – % of person trips Car 10.3 13.9 34.95
(excluding walk trips) Two-wheelers 17.2 21.5 25.00
Auto rickshaw 3.1 3.6 16.13
Bus 59.8 41.5 −30.60
Metro 0 4.1
2011. Descriptive statistics show that average built-up areas across
Trains (IR) 0.7 0.7 0.00
stations were 1.29 km2 for the yellow line and 1.53 km2 for the blue Bicycle 5.3 6.8 28.30
line in 2011. These areas correspondingly increased about 5% and 3% Cycle rickshaw 3.6 7.9 119.44
from 2001. It is noteworthy that built-up areas have decreased at sever- Note: *2007–08 data for modal split; **Ambulance & other passenger vehicles are included.
al stations of both lines (16 stations in blue line and 14 stations in yellow Sources: (Government of NCT of Delhi, 2012) and Tiwari (2011) for modal split adapted
line) with maximum decline in the range of 31–36% (Table 3). However, from Delhi Traffic and Forecast Study (2008) commissioned by DIMTS to RITES.
over time many stations have increased built-up areas in their vicinity.
We closely looked at which stations have maximum increased or
decreased and investigated possible causes for such change. Top three required to carry the motor vehicles plying in Delhi. During the same
stations with maximum decrease in built-up areas were Dwarka Sector period, the growth trend of registered vehicles figured highest for
21 (− 31.16%), Barahkhamba Road (− 24.70%) and Mandi Road taxis, followed by cars and jeeps, two wheelers, and busses. This ram-
(− 24.04%) in blue lines and Patel Chowk (− 26.54%), GTB Nagar pant increase in number of registered vehicles causes multitude of prob-
(−20.13%) and Civil Lines (−17.79%) in yellow line. Top three stations lems such as air pollution, noise pollution, and traffic congestion. The
with maximum increase in built-up areas were Yamuna Bank (65.6%), insufficient public transport system and the easy availability of bank fi-
Dwarka Sector 8 (49.3%) and Najafgarh Metro Depot (45.8%) in blue nancing for private vehicles have resulted increased vehicle ownership
line and Guru Dronacharya (90.11%), Arjan Garh (78.37%), and levels and their usage (Jalihal, Ravinder, & Reddy, 2005). To control the
Sultanpur (58.41%) in yellow line. In comparison to developed areas, trend of growth of increase in number of registered vehicles MRTS could
stations located in lesser developed or peripheral areas showed an in- play a significant role. However, connectivity of the public transporta-
crease in built-up areas. tion system to the MRTS station and dense metro network is a
The decadal growth rate of density is low (6.6%), about one-third of the challenge.
population and one-half of the built-up area in Delhi (Table 4). The growth
of built-up areas (r2 = 0.83, p b 0.01) and densification (r2=0.91, p b 5. Discussion and policy implications
0.01) are positively correlated with population growth. But growth of
built-up area along the transit (metro lines and stations) is lower than This study reveals that MRTS has not promoted positive land cover
the growth of built-up in non-transit areas (Table 3). The result indicates change (increasing built-up area), either along rail lines or stations, on
that intended densification have not materialized across the transit, as per the contrary it has influenced land cover change in peripheral areas.
the assumption. Therefore, enhanced built-up (and densification) should Similar influence of transits has also been noted elsewhere, such as
be pursued through spatial planning/policy interventions. Minneapolis, Minnesota (Hurst & West, 2014). While drawing lessons
The development of MRTS has to be seen in the context of existing from global best case examples of transit-oriented metropolises, that
transport infrastructure, including available modes. Table 5 shows have direct relevance to cities in developing countries, Suzuki et al.
change in roads, registered vehicles, and modal split over last decade (2013) show that transit systems reinforce and often accelerate
(2001–11) in Delhi. Over last decade, registered vehicles have been suburbanization to some degree, and therefore proactive planning is
doubled in comparison to 9.38% increase in the road network. The necessary for decentralized growth to take in the form of subcenters.
road network has grown at a much slower rate as compared to growth This study suggests that an urban transit may not by default lead to
in number of vehicles, that resulted a huge short fall in the capacity land cover changes or sustainable land use, as may appear normatively.

Table 4
Change in population, built-up area and density in Delhi, by district, 2001–2011.

Population Built-up area (in km2) Density (person/km2)

District % Change
2001 2011 % Change 2001–11 2001 2011 % Change 2001–11 2001 2011
2001–11

South 2,267,023 2,733,752 20.59 140.30 161.47 15.09 16,158.59 16,930.70 4.78
South-West 1,755,041 2,292,363 30.62 167.81 207.88 23.87 10,458.20 11,027.47 5.44
North 781,525 883,418 13.04 43.76 46.09 5.34 17,859.43 19,165.41 7.31
North-East 1,768,061 2,240,749 26.73 51.17 59.17 15.64 34,553.02 37,869.36 9.60
North-West 2,860,869 3,651,261 27.63 190.62 207.71 8.97 15,008.59 17,578.64 17.12
East 1,463,583 1,707,725 16.68 52.29 54.54 4.31 27,991.18 31,311.42 11.86
Central 646,385 578,671 −10.48 13.52 13.10 −3.09 47,826.17 44,181.12 −7.62
New Delhi 179,112 133,713 −25.35 17.39 17.03 −2.05 10,301.96 7851.71 −23.78
West 2,128,908 2,531,583 18.91 77.45 88.82 14.67 27,486.73 28,503.52 3.70
Total 13,850,507 16,753,235 20.96 754.09 855.59 13.46 18,367.24 19,580.97 6.61
 S. Ahmad et al. / Cities 50 (2016) 111–118
Transit oriented development needs articulated densities, via increasing
floor area ratio and other means, across rail lines and stations. As noted
earlier, Delhi's transit system was envisaged to address urban mobility
issues rather than heaping rich ancillary benefits, including capturing
additional built-up areas and property values. Thus to harness co-
benefits in development, environmental and climate context (see
Puppim de Oliveira et al., 2013), several additional but inter-related
interventions are needed. These interventions could be in the form of
improvement of MRTS stations, both physical enhancement of the
stations such as a barrier-free system as well as improvement of
intermodal facilities and user-friendliness of transport facilities
(e.g., pedestrian walks, parking lots), enhancement of the convenience
level for transferring activities (e.g., shortening walking distance for
transfer), and providing safe and comfortable waiting space for transfer-
ring passengers. In addition, improvement of feeder bus services in
terms of service accessibility, frequency, and operational punctuality is
also required (Yagi & Nagayama, 2010). Doll and Balaban (2013)
argue that integration with other modes and sensible land use planning
is key for increasing co-benefits.
Although built-up area grew in peripheral districts and along the
peripheral section of MRTS (lines and stations), these areas have in-
efficient land management, leading to sprawl as well as unregulated
developments. Land under urban agriculture in peripheral areas has
been changed to other land uses, mostly in the form of informal set-
tlements. These informal settlements are often un-serviced by urban
amenities, social infrastructure, open and recreational areas, and
therefore, compromise on human well-being. Sridharan (2011) has
precisely highlighted spatial inequalities and the politics of urban
spaces, using the case of other metropolitan cities in India. Moreover,
integrating informal settlements with planned one is a difficult task
and poses one of the major challenges in the sustainable urban fu-
tures in Delhi (Ahmad, 2011; Ahmad & Choi, 2011; Ahmad,
Balaban, Doll, & Dreyfus, 2013; Ahmad, Sridharan, & Kono, 2012;
Pugh, 1991; Sivam, 2003). Inappropriate conversion has also led to
unbalancing the eco-system and increased urban heat island effect.
Nevertheless these sub-urban areas also bear potential to develop
as transit oriented development by articulating densities in selected
locations, as Suzuki et al. (2013) argue.
Appendix A
Fig. A1. Spatial location of understudied stations and lines (left panel) and change in land cover, 2001–2011 (right panel).
117
Depending upon the ground realities, our study suggests two
interventions – promoting transit oriented development particularly
in existing built areas and proactive spatial planning interventions in
peripheral areas – for sustainable urban development in Delhi or similar
other cities in developing economies. These interventions align with
most of the existing policies and programs in Delhi like MPD 2021, but
have grossly failed in execution. Thus, a major challenge in any future
policy prescription or research investigation would be to strategically
develop implementation strategies, duly considering the context of
defragmented governance. As Delhi's large share of population (over
one-fourth) is deprived, economically or socially, therefore, the sug-
gested interventions would also lead to inclusive growth.
While interpreting results and conclusions some of the limitations of
the study should be kept in mind. This study assumes that transits play
major role in land cover change. However, several additional factors,
such as, planning decisions and permits, land prices, and property owner-
ship patterns, influence land cover change (Hurst & West, 2014; Kim,
2015), which we did not control for this study. Second, our study con-
siders horizontal change in built-up areas, without considering vertical
change, which might have occurred in some parcels along MRTS. These
two considerations – controlling for additional factors in land cover
change, and vertical change – need much sophisticated method and
large data at small spatial scale, which we do not have for Delhi. Despite
these limitations, results indicate sustainable urban development path-
ways in Delhi, with special attention on the implementation strategies.
Acknowledgments
Earlier version of this article was presented in the IGU Kyoto Region-
al Conference (KRC), Kyoto, 4–9 August 2013. The authors wish to thank
the following organizations: The Japan Society for the Promotion of Sci-
ence (JSPS) for supporting completely or in part the work of S. Ahmad
through Grants-in-Aid for Scientific Research; UNU-IAS for supporting
M. Sethi through their PhD fellowship program and data analysis
works. Authors are grateful to Glovis-USGS and JAXA for providing sat-
ellite data. This work does not reflect the official position of any organi-
zation mentioned here.
118 S. Ahmad et al. / Cities 50 (2016) 111–118
References
Ahmad, S., Baiocchi, G., & Creutzig, F. (2015). CO2 Emissions from Direct Energy Use of
Urban Households in India. Environmental Science & Technology. http://dx.doi.org/
10.1021/es505814g.
Ahmad, S. (2011). Characteristics of uncontrolled urban settlements and housing
market in Delhi. (Ph.D) Seoul: Seoul National University (Retrieved from http://
library.snu.ac.kr/site/snu/viewer/SNUPDFViewer.jsp?cid=3685344&moi=
1781173&file=2479139).
Ahmad, S., & Choi, M. (2011). The context of uncontrolled urban settlements in Delhi.
Asien (Hamburg) (Januar 2011), 118, 75–90.
Ahmad, S., Balaban, O., Doll, C.N.H., & Dreyfus, M. (2013). Delhi revisited. Cities, 31,
641–653. http://dx.doi.org/10.1016/j.cities.2012.12.006.
Ahmad, S., Sridharan, N., & Kono, N. (2012). Housing supply system in unauthorized
settlements in Delhi: Process and outcomes. Paper presented at the 48th ISOCARP
congress, Perm, Russia.
Angel, S., Sheppard, S. C., Civco, D. L., et al. (2005). The Dynamics of Global Urban Expan-
sion. Transport and Urban Development Development, The World Bank.
American Planning Association (2002). Policy guide on smart growth. The Association.
Anas, A., & Armstrong, R. (1993). Land value and transit access: Modeling the relationship in
the New York metropolitan area.
Avtar, R., Singh, C.K., Shashtri, S., & Mukherjee, S. (2011). Identification of erosional and
inundation hazard zones in Ken–Betwa river linking area, India, using remote sensing
and GIS. Environmental Monitoring and Assessment, 182(1–4), 341–360.
Avtar, R., Takeuchi, W., & Sawada, H. (2013a). Full polarimetric PALSAR-based land cover
monitoring in Cambodia for implementation of REDD policies. International Journal of
Digital Earth, 6(3), 255–275.
Avtar, R., Takeuchi, W., & Sawada, H. (2013b). Monitoring of biophysical parameters of
cashew plants in Cambodia using ALOS/PALSAR data. Environmental Monitoring and
Assessment, 185(2), 2023–2037.
Avtar, R., Yunus, A.P., Kraines, S., & Yamamuro, M. (2015). Evaluation of DEM generation
based on interferometric SAR using TanDEM-X data in Tokyo. Physics and Chemistry of
the Earth. http://dx.doi.org/10.1016/j.pce.2015.07.007.
Banerjee, R., & Srivastava, P.K. (2013). Reconstruction of contested landscape: Detecting land
cover transformation hosting cultural heritage sites from Central India using remote
sensing. Land Use Policy, 34, 193–203. http://dx.doi.org/10.1016/j.landusepol.2013.03.
005.
Baum-Snow, N. (2007). Did highways cause suburbanization? The Quarterly Journal of
Economics, 775–805.
Bollinger, C.R., & Ihlanfeldt, K.R. (1997). The impact of rapid rail transit on economic
development: The case of Atlanta's MARTA. Journal of Urban Economics, 42(2),
179–204.
Bowes, D.R., & Ihlanfeldt, K.R. (2001). Identifying the impacts of rail transit stations on
residential property values. Journal of Urban Economics, 50(1), 1–25.
Census of India (2011). Provisional population totals paper 2 of 2011 India series 1. India:
Office of the Registrar General & Census Commissioner.
Cervero, R. (1984). Journal report: Light rail transit and urban development. Journal of the
American Planning Association, 50(2), 133–147.
Cervero, R. (2004). Transit-oriented development in the United States: Experiences,
challenges, and prospects, Vol. 102, Transportation Research Board.
Cervero, R., & Duncan, M. (2002). Transit's added value. At what point does locating near
transit raise real estate values? Urban Land, 61(2), 77–87.
Creutzig, F., Baiocchi, G., Bierkandt, R., Pichler, P. -P., & Seto, K.C. (2015). Global typology of
urban energy use and potentials for an urbanization mitigation wedge. Proceedings of
the National Academy of Sciences, 112(20), 6283–6288.
Dewan, A.M., & Yamaguchi, Y. (2009). Land use and land cover change in Greater Dhaka,
Bangladesh: Using remote sensing to promote sustainable urbanization. Applied
Geography, 29(3), 390–401. http://dx.doi.org/10.1016/j.apgeog.2008.12.005.
Delhi Traffic and Forecast Study (2008). Transport Demand Forecast Study & Develop-
ment of an Integrated Road Cum Multi-Modal Public Transport Network for the
NCT of Delhi. RITES Ltd, MVA Asia Ltd, TERI June 9th 2008.
Dewees, D.N. (1976). The effect of a subway on residential property values in Toronto.
Journal of Urban Economics, 3(4), 357–369.
Doll, C.N., & Balaban, O. (2013). A methodology for evaluating environmental co-benefits
in the transport sector: Application to the Delhi metro. Journal of Cleaner Production,
58, 61–73.
Dulal, H.B., Brodnig, G., & Onoriose, C.G. (2011). Climate change mitigation in the
transport sector through urban planning: A review. Habitat International, 35(3),
494–500.
Emmanuel, M.R., & Baker, K. (2012). Carbon management in the built environment.
Routledge.
Giuliano, G. (1995). The weakening transportation-land use connection. ACCESS
Magazine, 1(6).
Government of NCT of Delhi (2012). Statistical abstract of Delhi. (Delhi).
Guiliano, G., & Agarwal, A. (2010). Public transit as a metropolitan growth strategy. In N.
Pindus, H. Wial, & H. Wolman (Eds.), Urban and regional policy and its effects, Vol. 3.
(pp. 205–252). Brookings Institution Press.
Handy, S. (2005). Smart growth and the transportation-land use connection: What does
the research tell us? International Regional Science Review, 28(2), 146–167.
Hurst, N.B., & West, S.E. (2014). Public transit and urban redevelopment: The effect of
light rail transit on land use in Minneapolis, Minnesota. Regional Science and Urban
Economics, 46, 57–72. http://dx.doi.org/10.1016/j.regsciurbeco.2014.02.002.
Jalihal, S.A., Ravinder, K., & Reddy, T. (2005). Traffic characteristics of India. Paper
presented at the proceedings of the Eastern Asia society for transportation studies.
Kim, J.H. (2015). Applied Geography, 60, 37–45. http://dx.doi.org/10.1016/j.apgeog.2015.
03.002.
Martínez, L., & Viegas, J. (2009). Effects of transportation accessibility on residential
property values: Hedonic price model in the Lisbon, Portugal, metropolitan area.
Transportation Research Record: Journal of the Transportation Research Board, 2115,
127–137.
Muller, P.O. (2004). Transportation and urban form-stages in the spatial evolution of the
American metropolis.
N.C.R. Board (2005). Master plan for NCR – 2021. National Capital Region Planning Board.
New Delhi: Ministry of Urban Development, Government of India.
Newman, P.W., & Kenworthy, J.R. (1989). Gasoline consumption and cities: A comparison
of US cities with a global survey. Journal of the American Planning Association, 55(1),
24–37.
Pugh, C. (1991). Housing and land policies in Delhi. Journal of Urban Affairs, 13(3),
367–382.
Puppim de Oliveira, J.A., Doll, C.N., Kurniawan, T.A., Geng, Y., Kapshe, M., & Huisingh, D.
(2013). Promoting win–win situations in climate change mitigation, local environ-
mental quality and development in Asian cities through co-benefits. Journal of
Cleaner Production, 58, 1–6.
Rees, W.E. (1992). Ecological footprints and appropriated carrying capacity: What urban
economics leaves out. Environment and Urbanization, 4(2), 121–130.
Santos, G., Behrendt, H., & Teytelboym, A. (2010). Part II: Policy instruments for
sustainable road transport. Research in Transportation Economics, 28(1), 46–91.
Sethi, M. (2011). Alternative perspective for land planning in today's context-minimizing
compensation issues. Indian Valuer, 43(4), 443–448.
Seto, K.C., Güneralp, B., & Hutyra, L.R. (2012). Global forecasts of urban expansion to 2030
and direct impacts on biodiversity and carbon pools. Proceedings of the National
Academy of Sciences, 109(40), 16083–16088.
Sivam, A. (2003). Housing supply in Delhi. Cities, 20(2), 135–141.
Sridharan, N. (2011). Spatial inequality and the politics of urban expansion. Environment
and Urbanization Asia, 2(2), 187–204.
Suzuki, H., Cervero, R., & Iuchi, K. (2013). Transforming cities with transit: Transit and
land-use integration for sustainable urban development. World Bank Publications.
The Indian Express (2011, 18 November). Centre green light for Metro Phase IV. The Indian
Express (Retrieved from http://www.indianexpress.com/news/centre-green-light-
for-metro-phase-iv/877497/).
Tiwari, G. (2011, 19–20 October). Metro systems in India: Case study DMRC. Paper
presented at the low carbon transport in India — stakeholders' consultation workshop.
New Delhi: IHC.
UNDESA (2011). World urbanization prospects, the 2011 revision. Retrieved from:
http://esa.un.org/unpd/wpp/Documentation/publications.htm
Wulder, M.A., White, J.C., Goward, S.N., Masek, J.G., Irons, J.R., Herold, M., ... Woodcock, C.E.
(2008). Landsat continuity: Issues and opportunities for land cover monitoring.
Remote Sensing of Environment, 112(3), 955–969.
Xiao, J., Shen, Y., Ge, J., Tateishi, R., Tang, C., Liang, Y., & Huang, Z. (2006). Evaluating urban
expansion and land use change in Shijiazhuang, China, by using GIS and remote
sensing. Landscape and Urban Planning, 75(1–2), 69–80. http://dx.doi.org/10.1016/j.
landurbplan.2004.12.005.
Yagi, S., & Nagayama, K. (2010). Opinions of intermodal transfer functions of urban
railway systems: A case study of Delhi Metro. Paper presented at the Transportation
Research Board 89th Annual Meeting.