BangaloreT. V. RAMACHANDRAIndian Institute of Science, India

BHARATH H. AITHALIndian Institute of Technology Kharagpur, India

INTRODUCTION

Urbanization or urban growth is a dynamicprocess involving changes to vast expansesof land cover with the progressive concen-tration of human population. The processentails a switch from a spread-out patternof human settlements to compact growth inurban centers. The process of urbanizationgained impetus with the Industrial Revo-lution 200 years ago and accelerated in the1990s with globalization and the consequentrelaxation in the market economy. Rapidlyurbanizing landscapes reach a huge pop-ulation size leading to gradual collapse inurban services; the result is problems withhousing, slums, lack of a treated water supply,inadequate infrastructure, higher pollutionlevels, poor quality of life, and so forth.Urbanization is a product of demographicexplosion and poverty induced rural–urbanmigration. Globalization, liberalization, andprivatization are the agents fueling urban-ization in most parts of India. However,unplanned urbanization coupled with thelack of a holistic approach is leading to lack ofinfrastructure and basic amenities. Therefore,proper urban planning with operational,developmental, and restorative strategies isrequired to ensure the sustainable manage-ment of natural resources. The process ofurbanization involves migration from ruralto urban areas, increased urban populationdensity, increased levels of consumption, cor-responding lifestyle changes, and increased

The Wiley Blackwell Encyclopedia of Urban and Regional Studies. Edited by Anthony Orum.© 2019 John Wiley & Sons Ltd. Published 2019 by John Wiley & Sons Ltd.DOI: 10.1002/9781118568446.eurs0014

energy consumption; these shifts promote anincrease in carbon emissions.

Urban growth is the spatial pattern of landdevelopment to accommodate anthropogenicdemand that influences other land uses (e.g.,open spaces, water bodies, etc.). The surfaceof the earth has been altered considerablyby humans over the past 50 years throughurbanization. On a global scale, 2.5 billionpeople were living in urban areas in 1950; thenumber is expected to be 6 billion by 2030(Ramachandra, Aithal, and Durgappa 2012).Continuing urban growth raises concernsabout the degradation of the environment andits ecological health. Understanding urbangrowth and change is critical to city plan-ners and resource managers in these rapidlychanging environments. Dynamic urbanchange processes, through which the produc-tive agricultural lands, vegetation, and waterbodies are irretrievably lost and transformedat an alarming rate is often referred to as rapidurbanization. Unplanned rapid urbanizationchanges the structure of the landscape,influencing its functioning quite apart fromthe lack of basic infrastructure, amenities,enhanced levels of pollution, and changesin local climate and ecology (Aithal andRamachandra 2016a). This phenomenon isvery rapid in India with its urban populationgrowing at around 2.3 percent per annum,and certain cities like Bangalore have a higherpopulation growth at 4.6 percent per annum(Ramachandra, Aithal, and Sowmyashree2013). Cities in developing countries havegrown more compact and more clus-tered, spreading beyond the boundariesof central cities (Ramachandra, Aithal, andSowmyashree 2015). This dispersed growthclose to the large urban forms having a mixedland use (Aithal and Ramachandra 2016a,2016b) is known as “urban sprawl.” This

2 BANGALORE

leads to the inefficient use of land resourcesand energy and large-scale encroachmentonto agricultural lands, traffic congestion,shortages in urban services and facilities, andmajor problems of urban poverty. Cities areexpanding in all directions, which is resultingin large-scale urban sprawl and changes inurban land use (Ramachandra, Aithal, andSowmyashree 2015). Large-scale land-useand land-cover changes combined withunplanned urbanization and sprawl has animpact on the environment and the sustain-ability of natural resources (Ramachandra,Aithal, and Barik 2014; Ramachandra, Aithal,and Sowmyashreeb 2014).

Understanding land-use and land-coverchanges is essential to evolving appropri-ate strategies for sustainable managementof natural resources and monitoring envi-ronmental changes such as greenhouse gasemissions and urban heat island effects(Ramachandra, Aithal, and Bharath 2013).Spatial data acquired through space-bornesensors since the early 1970s and advance-ments in geo-informatics have helped inunderstanding and visualizing landscapedynamics. The collection of remotely senseddata at regular intervals facilitates synopticanalyses of the earth: system function, pat-terning, and change at local, regional, andglobal scales over time. Further evaluatingthe impact of urban growth in this formon the environment and understanding thedynamics of complex urban systems involvesmodeling and simulation, which requireinnovative analytical methods and robusttechniques. A number of analytical andstatic urban models have been developedthat are based on diverse theories such asurban geometry, size relationship betweencities, economic functions, and social pat-terns with respect to the city. However, thesemodels explain urban expansion and evolv-ing patterns instead of predicting futureurban development. In urban modeling,

dynamic agent-based modeling as an urbansimulation tool has rapidly gained popular-ity in recent years among urban plannersand geographers. Considerable researchefforts have developed different dynamicagent-based models (stochastic based on cel-lular automata, Markov model, etc.) for urbanand environmental applications (Aithal,Vinay, and Ramachandra 2014). All thesemodels have some common features, such asthe use of transition probabilities in a classtransition matrix (Aithal and Ramachandra2016a). Cellular automata algorithms definethe state of the cell based on the previous stateof the cells within a neighborhood, using aset of transition rules. Coupled Markov chain(MC) and cellular automata (CA) eliminatesthe shortcomings of CA and MC respectively.MC quantifies future changes based on pastchanges, thereby serving as a constraintfor CA, which addresses spatial alloca-tion and the location of change. AlthoughCA–Markov gives promising results, it failsto achieve accurate results since the drivingforces are not accounted for in this model.

Fuzzy analytical hierarchical process(AHP)-based CA–Markov modeling usesanalytical hierarchical process-based model-ing to simulate the land-use dynamics basedon the agents of change including drivers orconstraints, socioeconomic and infrastruc-tural activities, and human actions (Aithal,Vishwanatha, and Ramachandra 2015).Agent-based modeling (ABM) weights/ranksthe growth factors and constraints as reflectedby the real-world scenarios to develop sitesuitability maps in order to model the landuse; it has emerged as a promising approachin understanding the complex urban pro-cesses. This provides ample opportunitiesand challenges which complement or extendto other approaches. The site suitabilitymaps provide the transitional areas, describ-ing where the particular land use has theprobability to change or retain its state. The

BANGALORE 3

site suitability maps are combined with theCA–Markov in order to simulate and predictthe land-use dynamics.

Cities across the world with increasedurbanization and urban populations alsoaccount for about two-thirds of the world’sprimary energy consumption and aboutthree-quarters of the greenhouse gas (GHG)emissions. Concentration of greenhousegases in the atmosphere has increased rapidlyover the last century due to ever-increasinganthropogenic activity; this has resulted insignificant increases in the temperature ofthe earth causing global warming. Majorsources of GHG are transportation (burningof fossil fuel), forests (due to human-inducedland-cover changes leading to deforestation),power generation (burning of fossil fuels),agriculture (livestock, farming, rice cultiva-tion, and burning of crop residues), waterbodies (wetlands), industry, and urban activ-ities (building, construction, transport, solidand liquid waste). Aggregation of GHG (CO2and non-CO2 gases), in terms of carbondioxide equivalent (CO2e), indicate the GHGfootprint.

GHG footprint is a measure of the impactof human activities on the environment interms of the amount of greenhouse gasesproduced. The study (Ramachandra, Aithal,and Shreejith 2015) calculated the amount ofthree important greenhouses gases, namelycarbon dioxide (CO2), methane (CH4), andnitrous oxide (N2O), thereby developingthe GHG footprint of the major cities inIndia. Energy consumption and related GHGemissions in cities largely depend on how thecities grow and operate and on the sourcesof energy that are used to support these pro-cesses. Increased emissions of CO2 and othergreenhouse gases are the main contributors toglobal warming (Ramachandra, Aithal, andShreejith 2015) with one-third of CO2 emis-sions resulting from land-use changes. Rapidchanges in land use leading to large-scale

degradation of natural resources (water bod-ies, tree cover) alter the carbon sink abilityand affect the environment. Land-use changein the form of urbanization is a main fac-tor that determines carbon dynamics andleads to global warming. Studies based onemissions and land use have established acorrelation between global warming andenhanced emissions of greenhouse gases andurban heat island effect across the urbanizingregions in the world. Urban heat island (UHI)has been garnering substantial attention inrecent decades. Extensive studies have beencarried out to explore the impact of pasturbanization on the UHI effect in cities withunique urban landscapes and histories – forexample, Tokyo, Paris, New York City, Bei-jing, Houston, Singapore – using numericalmodels. This emphasizes the need for under-standing spatial and temporal patterns ofurbanization, land-use changes, greenhousegas emissions, and urban heat island. Thisentry analyzes spatiotemporal patterns ofurbanization in Bangalore, one of the mostrapidly urbanizing cities with environmentalsustainability issues.

STUDY AREA

Greater Bangalore, at 741 km2, is the prin-cipal administrative, cultural, commercial,industrial, and educational capital of the stateof Karnataka (Figure 1).

Bangalore is located at 12∘59′ north lat-itude and 77∘57′ east longitude, almostequidistant from the eastern and westerncoasts of the South Indian peninsula, and issituated at an altitude of 920 m above meansea level. The mean annual total rainfall isabout 900 mm. The summer temperatureranges from 18 to 38 ∘C, while the wintertemperature ranges from 12 to 25 ∘C. Thus,Bangalore enjoys a salubrious climate aroundthe year. With its numerous parks andabundant greenery, it was called the “garden

4 BANGALORE

N

Bangalore

INDIA

0Kilometers

95190 380 570 0Kilometers

1.5 3 6 9760

37.1474° North

13.1429° North

7.9131° North12.8315° North

97.4

343° E

ast

77.7

841 ° E

ast

67.3

994°

East

77.4

568°

East

Figure 1 Study area: Bangalore (with 198 local administration wards)

city” of India. The Bangalore landscape islocated over ridges forming three watersheds.The undulating terrain in the region hasfacilitated the creation of a large number ofinterconnected lakes.

Bangalore has grown by more than 10times since 1951 (Figure 2). The presentname of the city, Bangalore (GreaterBangalore), is an anglicized form of Ben-galooru which according to popular beliefis derived from benda kaalu or boiled beansand ooru meaning a town. In 1862 twoindependent municipal boards were estab-lished: Bangalore City Municipality, andBangalore Civil and Military Station Munic-ipality. In 1949, the two municipalitieswere merged and the Bangalore City Cor-poration was formed. Subsequently, therehave been reorganizations with respect tothe zones and wards within the corpora-tion, rising from 50 divisions in 1949 to 95wards in the 1980s, 100 wards in 1995, andnow about 150 wards. In 2006 the Banga-lore City Corporation was reorganized asGreater Bangalore City Corporation with 198wards.

The urban agglomeration had an overallpopulation in 2011 of 8.4 million (Figure 2),including a workforce of 6.2 million, anda literacy rate of 87.6 percent according tothe latest census. The information and tech-nology (IT) sector is a major employer with45 percent of the workforce and the highestper capita income in the state at 202,176rupees (State and District Domestic Productof Karnataka 2016). This has significantlyincreased people’s purchasing power and hasencouraged the growth of real estate withconsequent land market dynamics, as well ascreating a lot of secondary employment inthe service industries.

Bangalore is home to numerous high-techknowledge hubs. This is evident from theestablishment of premier centers like theIndian Institute of Science (IISc), IndianSpace Research Organisation (ISRO),National Aerospace Laboratories (NAL),Defence Research and Development Organi-sation (DRDO), and Indian Institute of Man-agement (IIM). The city also houses a hugenumber of professional engineering and med-ical colleges at undergraduate and graduate

BANGALORE 5

01951 1961 1971 1981 1991 2001 2011

Year

20

40

60

80

100

120

140

0

100

200

300

400

500

600

700

800

Popula

tion (

lakh),

Popula

tion d

ensity

(pers

ons p

er

hecta

re)

Are

a (

km

2)

Area (km2)Population (lakh)Population density (persons/hectare)

Figure 2 Spatial and population dynamics in Bangalore, 1951 to 2011. Note: Population is in lakhs; onelakh = 0.1 million

levels. An important feature of the economicactivities of Bangalore is the huge concentra-tion of small and medium enterprises (SMEs)in diversified sectors across the city. Thusthe city has a thriving economy with a grossdomestic product of US$83 billion. The cityis also branded the country’s “Silicon Valley”because of its remarkable contributions intechnological innovation and its high HDI(human development index) of 0.73 (Kar-nataka Human Development Report 2005).

Bangalore’s urban vegetation includestrees such as Alstonia scholaris, Artocarpusheterophyllus (jackfruit), Azadirachta indica(neem), Bombax cieba (red silk cotton), Buteamonosperma, Ficus bengalensis (alada mara),F. religiosa (ashwatha), Gmelina arborea,Kigelia pinnata (sausage tree), Lagerstroemiaspeciosa (pride of India), Mangifera indica(mango), Madhuca longifolia (mehwa orippe), Melia composite (Malabar neem),Michelia champaca, Neolamarkia kadamba(kadamba), Pongamia pinnata (honge), Ptero-carpus marsupium (honne), Syzigium cumini(jaamun), Saraca indica (seeta asok), Sweteniasp. (mahogani), Terminalia arjuna (arjuna),T. bellerica (tare), Tabebuia spectabilis, and

many more. Some of the gracious exotictrees found as avenue trees are Delonix regia(gulmhur), Enterolobium saman (rain tree),Parkia biglandulosa (badminton ball tree),Peltophorum pterocarpum (copper pod),Spathodea companulata (African tulip tree),Tabebuia spectabilis, and Polythia longifolia.

DATA

Urban dynamics were assessed using tem-poral remote sensing data for the period1973 to 2010. The time series spatial dataacquired from Landsat Series Multispec-tral Sensor (57.5 m) and Thematic Mapper(28.5 m), OLI (28.5 m) sensors for the period1973–2016 were downloaded from the publicdomain. The Survey of India (SOI) topo-graphic maps of 1:50,000 and 1:250,000scales were used to generate base layers ofcity boundary and other details. City mapswith ward boundaries were digitized fromthe BBMP (Bruhat Bangalore MahanagaraPalike) map. Ground control points to reg-ister and geo-correct remote sensing datawere collected using handheld pre-calibrated

6 BANGALORE

GPS (global positioning system), the SOItopographic maps, and online spatial portals.

Wetland delineation was performed usingfield data collected with a handheld GPS.Remote sensing data used for the study are:Landsat MSS data of 1973, Landsat TM of1992, Landsat ETM+ of 2000 MODIS 7 bandsproduct of 2002 and 2007 downloaded fromthe online portal, IRS LISS-3 data of 2006procured from NRSA, Hyderabad. IndiaGoogle Earth data served in the pre- andpost-classification process and validation ofthe results.

Vegetation cover analysis was doneusing Indian remote sensing (IRS) satel-lite Resourcesat 1 (multispectral with spatialresolution of 5.8 m) and Cartosat 1 (panchro-matic with spatial resolution 2.7 m). Thesedata were fused using hyperspectral colorspace resolution merge for better spatial andspectral resolution based on the universalimage quality index (UIQI).

METHOD

This includes understanding spatial pat-terns of urban growth, geovisualization ofurban dynamics, vegetation cover analysis,delineation of water bodies, sector-wise anal-ysis of greenhouse gas footprint, changes inland surface temperature, and urban heatisland.

Spatial pattern of urban growth in Bangalore

The remote sensing data obtained weregeo-referenced, rectified, and cropped per-taining to the city boundary (study area).Geo-registration of remote sensing data(Landsat data) was done using ground con-trol points collected from the field usingprecalibrated GPS and also from knownpoints (such as road intersections, etc.)collected from geo-referenced topographicmaps published by the Survey of India. The

Landsat satellite 1973 images with a spatialresolution of 57.5 m × 57.5 m (nominal reso-lution) were resampled to 28.5 m comparableto the 1989–2016 data which are 28.5 m ×28.5 m (nominal resolution). Landsat ETM+bands of 2010 were corrected for the SLC-offby using image enhancement techniques,followed by nearest-neighbor interpolation.

Land-use classification using remote sens-ing data involved (a) generation of false colorcomposite (FCC), which helped in identi-fying and locating heterogeneous patchesof landscapes; (b) selection of training datafrom the FCC (the training sites should coverat least 15 percent of the total area and bedistributed over the study area); (c) loading ofthese training data sites as polygon features toa GPS for collection of attribute informationfor the training data sites from the field;(d) supplementation of the information usingGoogle Earth. Seventy percent of the trainingdata were used for classification and the restfor validation of the classified outputs andaccuracy assessment. The classification ofthe data was completed using GRASS (Geo-graphic Resource Analysis Support System)open source GIS software by considering fourprime land-use classes.

Land-use analysis was carried out usingsupervised pattern classifier Gaussian max-imum likelihood algorithm as in Figure 3.This has been proved a superior classifier asit uses various classification decisions usingprobability and cost functions (Ramachan-dra, Aithal, and Durgappa 2012). Meanand covariance matrix are computed usingestimate of maximum likelihood estimator.Accuracy assessment to evaluate the per-formance of classifiers was done with thehelp of field data by testing the statisticalsignificance of a difference, computationof kappa coefficients, and proportion ofcorrectly allocated cases. Recent remotesensing data were classified using the col-lected training samples. Statistical assessment

BANGALORE 7

of classifier performance, based on theperformance of spectral classification con-sidering reference pixels, was done whichincluded computation of kappa (𝜅) statisticsand overall (producer’s and user’s) accura-cies. For earlier time data, training polygonalong with attribute details were compiledfrom the historical published topographicmaps, vegetation maps, revenue maps, andso on.

Land-use categories include (a) area undervegetation (parks, botanical gardens, grass-lands such as golf courses), (b) built-up(buildings, roads, or any paved surface,(c) water bodies (lakes/tanks, sewage treat-ment tanks), and (d) others (open areas suchas playgrounds, quarry regions, etc.) (seeTable 1).

Modeling and visualization of urban growth

Agent-based modeling was used in the formof fuzzy (to understand the behavior of theagent) as an input, analytical hieratical pro-cess (AHP) was used to create the agents’influence in the region, and CA–Markov tomodel the land-use transition. Influencingagents were prioritized which include slope,proximity to roads, industry, educationalinstitutions, and bus, railway, and metrostations. Factors such as hospitals and socio-cultural buildings have a minimal role inurbanization. All these factors were gener-alized using a fuzzy analytical process andprioritized using AHP in order to weight the

factors based on land use type. The weigh-tages from the AHP and the constraints asBoolean were used to generate transitionsuitability maps for different land use usingmulti-criteria evaluation (MCE) technique(Figure 3). The transition suitability (ofMCE) along with transition probabilities(from Markov) (were used to calibrate theparameters and simulate the sprawl for themost recent data based in cellular automatausing CA–Markov transitions. Validation ofthe simulated output was carried out withrespect to the classified data for recent data.Calibrated parameters were further used toderive the land use for the year 2020.

CA–Markov: Markov chains were usedto establish the transition probabilities toestimate the area under change based on thetransition rules incorporated with the landuse. Cellular automata were used to obtaina spatial location and distribution of prob-able land-use changes based on associatedand surrounding pixels. The probability/sitesuitability map from cellular automata andthe transition probabilities were used inconvergence in order to simulate the futureland-use scenarios.

Fuzzy AHP/CA–Markov: Modeling, sim-ulation, and prediction of the land usedynamics were carried out using the combi-nation of fuzzy logic, analytical hierarchicalprocess (AHP), Markov chains and cellu-lar automata. Factors that have an impacton urbanization include roads, industries,educational institutions, bus stands, railway

Table 1 Land-use categories

Land use Categories included in land use

Urban Residential areas, industrial areas, paved surfaces, mixed pixels with built-uparea

Water Tanks, lakes, reservoirs, drainagesVegetation Forest, plantationsOther Rocks, quarry pits, open ground at building sites, unpaved roads, croplands,

nurseries, bare land

8 BANGALORE

LA

ND

-US

E A

NA

LY

SIS

MO

DE

LIN

G A

ND

PR

ED

ICT

ION

UR

BA

N S

PR

AW

LA

NA

LY

SIS

Satellite dataacquisition

Preprocessing

Land-useclassification

Accuracyassessment

Factors and constraints

Ground data,Toposheets,Bhuvan,Google Earth

Rasterization

Digitize

Factors1) Proximity to Industries, Major roads, Educational institutions, Bus and railway stations, Metro stations Cemeteries

Constraints

1) Water

2) Protected land based on City Development Plan 2015

Gradients andzonal analysis

LULC t0

LULC t1

2) Slope – DEM

Analytical hierarchicalprocess

Weights

Multi-criteriaanalysis

Site suitabilitymaps

Markov transitionprobability

LULC t2

Weights

CA-basedprediction

Validation

PredictedLULC t2

Shannon’s entropy Landscape metricanalysis

PredictedLULC tn

Fu

zzy

Figure 3 Flowchart illustrating the method of land-use analysis and visualization

stations, metro, population, city developmentplan (CDP), and water bodies were taken intoconsideration. Fuzzy logic was used to nor-malize factors such as slope, proximity maps(proximity to road, proximity to industries,proximity to educational institutions, prox-imity to bus stands, etc.) between 0 to 255, and

Boolean algebra was used on constrainingfactors of development such as water bodiesand CDP where the restricted locations forany new change would be zero vice versa.Factors of growth were prioritized/rankedand their effectiveness in change in landuse was estimated/mapped using the AHP.

BANGALORE 9

Prioritized weighted factors of growth alongwith the constraints were combined togetherthrough the multi-criteria evaluation processto derive site suitability maps. Site suitabilitymaps provide an insight into how the landuse can be managed. The site suitability mapsalong with the Markov transition probabili-ties are used in order to simulate the land usefor the year 2016 and predict across furtheryears. Both the above section on land useand visualization can be summed up as inFigure 3.

Vegetation cover analysis

Vegetation cover analysis was done usingIndian remote sensing (IRS) satelliteResourcesat 1 (multispectral with spatialresolution of 5.8 m) and Cartosat 1 (panchro-matic with spatial resolution 2.7 m). Thisinvolved (a) data fusion, (b) classification,and (c) analysis of tree distribution:

(a) Data fusion: Fusion of data from mul-tiple sensors aids in delineating objectswith comprehensive information dueto the integration of spatial informa-tion present in the high resolution (HR)panchromatic (PAN) image and spectralinformation present in the low resolu-tion (LR) multispectral (MS) images.Image fusion techniques integrate bothPAN and MS and can be performed atpixel levels. Data fusion (Resource satand Cartosat data) was performed usingalgorithms: Hyperspectral Color Spaceresolution (HCS) merge, High PassFilter (HPF) fusion, Modified IntensityHue Saturation (MIHS) fusion, WaveletFusion.

The output of different fusion meth-ods was analyzed for better accuracyand similarity in original data and fuseddata using the UIQI index. Universalimage quality index (UIQI) is used

to measure the similarity between twoimages. UIQI measures image distortionas a combination of three major factors:loss of correlation between data sets,radiometric imbalance, and contrastdistortion. UIQI index was calculatedusing equation 1:

UIQI = Q =𝜎AB

𝜎A𝜎B∗

2𝜇A𝜇B

𝜇2A + 𝜇2

B

∗2𝜎A𝜎B

𝜎2A + 𝜎2

B(equation 1)

HCS is more sensitive to image co-regis-tration accuracy. Inaccurate co-regis-tration may result in blurring edges inthe fused images. The MIHS transformfusion techniques are not sensitiveto the co-registration accuracy as thetextural features are replaced by theintensity replacement image. The visualevaluation and statistical analysis con-firm that HCS is a superior fusiontechnique for improving spatial detailof MS images with the preservation ofspectral properties.

(b) Classification: The land-use analysis wascarried out to classify the data into twoclasses, namely vegetation (trees) andnonvegetation, using the fused highspectral and spatial remote sensing datathrough the Gaussian Maximum Likeli-hood Classifier (MLC) algorithm withthe help of training data acquired fromthe field and supplementary data fromGoogle Earth. The MLC classifier usesthe probabilities to classify each pixelinto a particular land-use class. TheGaussian MLC classification technique(Ramachandra et al., 2011; Ramachan-dra, Aithal, and Sowmyashree 2013)has been used widely for analysis ofland use as this technique has provedto be superior to other classificationtechniques.

10 BANGALORE

(c) Analysis of tree distribution: The analysisof tree distribution was carried outbased on frequency distribution of thetree canopy area. The method involvedin assessing the distribution includes:data collection, frequency distribution,tree distribution in each ward:• Data collection: Five wards includ-

ing the Indian Institute of Sciencecampus (178 hectares, 22,000 trees)were sampled. The spatial extentof trees (with canopy size) coveredunder different species in selectedwards was delineated using pre-calibrated GPS and Google Earth.This aided in delineating canopies oftrees based on k-means algorithm inall wards (198 wards). The canopyarea was used to determine thefrequency distribution of trees withrespect to the crown area in eachward.

• Frequency distribution: Based on thefield data, wards were categorizedinto> 500 trees and< 500 trees. Thefrequency distribution of numberof trees versus average area of atree was analyzed to determine thestatistical distribution of tree speciesin the respective ward.

• Tree distribution in each ward: Thefrequency distribution under eachcategory was used to quantify theactual number of trees in eachward. The population for the year2013 was estimated for each ward(equation 2) based on the decadalgrowth to determine the tree distri-bution per person in each ward:

P2013 (i) = P2011 (i) ∗ (1 + n ∗ r (i))(equation 2)

where P2013(i) = population ofward i for the year 2013; P2011(i) =

population of ward i for the year2011; n = number of decades= 0.2;r(i) = incremental rate ofward i.

The ratio of number of trees ineach ward to population in eachward was determined to quantifytree distribution per person in eachward using equations 3 and 4:

TpP (i) =Tree (i)P2013 (i)

(equation 3)

TpP (B) =∑198

i=1 Tree (i)∑198

i=1 P2013 (i)(equation 4)

where TpP(i) = tree per person inward i; Tree(i) = number of treesin ward i; TpP(B) = trees per personin Bangalore.

• Validation: The validation was car-ried out with the field data usingGPS. Accuracy assessment wascarried out using equation 5:

Accuracy =

100 − (abs((ClassTree − GPSTree)∕

GPSTree) ∗ 100) (equation 5)

where ClassTree = tree count basedon classified data; GPSTree = treecount based on field census usingGPS.

Delineation of wetlands of Bangalore

K-means algorithm was used to partition thedata samples in the feature space into disjointsubsets or clusters. Accuracy assessment ofautomatically extracted wetlands was donewith field knowledge, visual interpretation,and Google Earth.

BANGALORE 11

Greenhouse gas (GHG) footprint assessment

The method involved (a) sector-wise quan-tification of GHG emissions, (b) computationof carbon dioxide equivalent (CO2e) ofthe non-CO2 gases using their respectiveglobal warming potential (GWP), and (c)aggregation of these CO2e to represent theGHG footprint. The major three greenhousegases quantified were carbon dioxide (CO2),methane (CH4), and nitrous oxide (N2O).The non-CO2 gases are converted to unitsof carbon dioxide equivalent (CO2e) usingtheir respective global warming potential(GWP), representing the GHG footprint.The total units of CO2e then represent asum total of the global warming poten-tial of all three major greenhouse gases,which represents the GHG footprint. Themajor categories considered for GHGemission inventory are energy (electric-ity consumption, fugitive emissions); thedomestic or household sector; transporta-tion; the industrial sector; agriculture-relatedactivities; livestock management; the wastesector.

National greenhouse gas inventories com-piled from various sources were used for cal-culation of GHG emissions. Country-specificemission factors were compiled from thepublished literature. In the absence ofcountry-specific emission factors, defaultemission factors of IPCC were used. Emissionof each GHG was estimated by multiplyingfuel consumption by the corresponding emis-sion factor. Total emissions of a gas from allits source categories, emissions are summedas given in equation 6:

EmissionsGas =∑

CategoryA × EF

(equation 6)

where EmissionsGas = emissions of givengas from all its source categories; A =amount of individual source category

utilized which generates emissions of thegas under consideration; EF = emissionfactor of a given gas type by type of sourcecategory.

Land surface temperature (LST) and urbanheat island

Digital number (DN) of Landsat ETM+was converted into spectral radiance LETMusing equation 7, and then converted toat-satellite brightness temperature (i.e., blackbody temperature, TETMSurface), under theassumption of uniform emissivity (𝜀 ≈ 1)using equation 8:

LETM = 0.0370588 × DN + 3.2(equation 7)

TETMSurface = K2∕ ln(

K1∕LETM + 1)

(equation 8)

where TETMSurface is the effective at-satellitetemperature in Kelvin, LETM is spectral radi-ance in watts/(meters squared × ster × μm);and K1 and K2 are pre-launch calibration con-stants. For Landsat-7 ETM+, K2 = 1282.71 Kand K1 = 666.09 mWcm−2sr−1μm−1 wereused (https://landsat.gsfc.nasa.gov/landsat-7-science-data-users-handbook/). The emissiv-ity corrected land surface temperatures Tswere finally computed by equation 9:

Ts =TB

1 +(𝜆 × TB∕𝜌

)ln 𝜀

(equation 9)

where λ is the wavelength of emitted radiancefor which the peak response and the averageof the limiting wavelengths (𝜆 = 11.5 μm)were used; 𝜌 = h × c/𝜎 (1.438 × 10−2 mK);𝜎 = Stefan–Boltzmann constant (5.67 × 10−8

Wm−2K−4 = 1.38 × 10−23 J/K); h = Planck’sconstant (6.626 × 10−34 Jsec); c = velocity oflight (2.998 × 108 m/sec); and 𝜀 is spectralemissivity.

12 BANGALORE

1973 1992 1999 2003

2008 2010 2016 2020

Built up Water Vegetation Others

Figure 4 Urban dynamics in Bangalore and visualization of growth by 2020

RESULTS AND DISCUSSION

Land-use analysis

Figure 4 depicts the unrealistic urban growthduring the last two decades. Overall accuracyof the classification was 72 percent (1973),77 percent (1992), 76 percent (1999), 80percent (2002), 75 percent (2006), and 78percent (2010) respectively. Urbanizationduring the period 1973 to 2016 (1005 percentconcretization or paved surface increase) hastelling influences on the natural resourcessuch as decline in green spaces (88 percentdecline in vegetation) and wetlands (79 per-cent decline). Analyses of the temporal datareveal an increase in urban built-up area of342.83 percent (1973 to 1992), 129.56 per-cent (1992 to 1999), 106.7 percent (1999 to2002), 136.92 percent (2002 to 2008), 105.19

percent (2008 to 2010), and 147.13 percent(2010 to 2016) and land use is as shown inTable 2.

Geo-visualization of urban growth

The transition suitability (of MCE) alongwith transition probabilities (from Markov)were used to calibrate the parameters andsimulate the sprawl for the year 2016 basedon cellular automata using CA–Markovtransitions. Validation of the simulatedoutput was carried out with respect to theclassified data for 2016. The simulated modelfor 2016 closely agrees with that of theclassified land use with kappa of 0.86. Thecalibrated parameters were further used toderive the land use for 2020. This predictionhas been done assuming water bodies toremain constant over all time frames, due

Table 2 Land-use statistics of urban pattern in Bangalore

Class 1973 1992 1999 2002 2008 2010 2016 2020

Built-up 7.97% 27.30% 35.37% 37.74% 49.54% 54.42% 76.91% 93.26%Vegetation 68.27% 46.22% 45.77% 38.72% 28.19% 23.41% 7.53% 2.96%Water 3.40% 2.62% 2.26% 1.85% 0.86% 0.90% 0.98% 2.09%Others 20.35% 23.86% 16.61% 21.70% 21.41% 21.27% 14.59% 1.69%

BANGALORE 13

to recent stringent norms. The simulationshows a predominant growth along the eastand south directions showing the effect ofupcoming industrial hubs in these directions.The process of urbanization in the simulatedresults is evident along the arterial roads andat the core of the city; the built-up densityincreases in lateral directions to a very largeextent almost to the verge of being saturated,followed by gradual sprawl along the bound-ary of the city. Geo-visualization of likelyland uses in 2020 through multi-criteriadecision-making techniques (fuzzy AHP)reveals a calamitous picture of 93 percent ofthe Bangalore landscape filled with pavedsurfaces (urban cover) and drastic reductionin open spaces and green cover (Table 2 andFigure 4).

Wetlands of Bangalore

The latest field survey (2015–2016) of lakesreveals that 98 percent of lakes have beenencroached on by illegal buildings (high riseapartments, commercial buildings, slums,etc.); 90 percent of lakes are sewage fed; 38percent are surrounded by slums; and 82 per-cent showed loss of catchment area. Also, lakecatchments were used as dumping groundsfor municipal solid waste or building debris.The surroundings of these lakes have illegallyconstructed buildings and slum-dwellersoccupy most of the adjoining areas. At manysites, water is used for washing and householdactivities, and fishing was even observed atone of these sites. Multistoried buildings havecome up on some lake beds, affecting thenatural catchment flow and thus leading to adeteriorating quality of water bodies. Unau-thorized construction in valley zones, lakebeds, and storm water drains (rajakaluves)highlight the apathy of decision-makers whilerevealing weak and fragmented governance.This is correlated with the increase in con-creted areas which severely affect open spaces

Table 3 Status of water body of Bangalore

Bangalore city(BMP, before 2004)

GreaterBangalore

1973 58 2072010 10 93

and, in particular, water bodies. Figure 5 andTable 3 show that Greater Bangalore had 207water bodies in 1973, which had declined to93 in 2010. The Bangalore landscape formsan important catchment for the lake systemand the present status of drains is reflectedin Figure 6. These interconnected lake sys-tems with their drainage networks help intransporting storm water beyond the citylimits. Bangalore, being a part of peninsularIndia, had the tradition of harvesting waterthrough surface water bodies to meet domes-tic water requirements in a decentralized way.These lake systems also helped in recharginggroundwater resources. The groundwatertable has declined to 300 m from 28 m (withthe removal of water bodies) and 400 to 500 min intensely urbanized areas such as White-field over a period of 20 years with the declinein wetlands and green spaces. Bangaloredepends on the groundwater resources to anextent of 40 percent. The disappearance oflocal water sources and consequent declinein the groundwater table have led to a severewater scarcity in the region.

Vegetation status

The vegetation cover in the city depicted inFigure 7 was derived from the fused remotesensing data (MSS IRS data of 5.8 m and Car-tosat data of 2.7 m) using the MLC classifier.This highlights that the vegetation coveris about 100.02 hectares (14.08 percent).Accuracy assessment of classified data showsan overall accuracy was 91.5 percent, thekappa of 0.86 indicating higher agreement ofclassified data with field reference data:

Missing lakesin 2011

Lakes in 2011

0 15 30

kilometers

NN

EE

SS

WW

Figure 5 Missing urban sinks

Hebbal NagavaraValley

KoramangalaChallaghatta Valley

SuvarnamukhiValley

VrishabhavathiValley

ArkavathiValley

0

Kilometers

1 2 4 6 8

N

Figure 6 Interlinked lake systems of Bangalore with catchment boundary

BANGALORE 15

N

0 1.5 3 6 9 12

Kilometers

Vegetation

Non-vegetation

Figure 7 Spatial distribution of vegetation in Bangalore

• Spatial patterns of vegetation cover: Banga-lore was divided into 1 km radius circles(with respect to the central businessdistrict) to assess the vegetation gradient.Figure 8 depicts vegetation density in eachconcentric region in 1973 and 2013. Ban-galore has an average vegetation densityof 0.14 (i.e., area of Bruhat Bangalore: 741km2, area under vegetation: 100.20 km2

and vegetation density: 0.14). Vegetationextent in each ward was computed byoverlaying the ward boundary, as shownin Figure 9. Wards such as Hudi, Aramanenagara, and Vasantha pura have highervegetation density of more than 0.4,while Chickpete, Laggere, Hegganahalli,Hongasandra, and Padarayanapura havelower density with less than 0.015.

Greenhouse gas emissions

The GHG footprint (aggregation of carbondioxide equivalent emissions of GHGs) was19796.5 Gg (19.796 million tonnes) and themajor contributing sectors are the trans-portation (43.48 percent), domestic (21.6percent), and industrial sectors (12.31 per-cent). Figure 10 shows the contributionsof various sectors to the city’s GHG foot-print. Emissions from transport are dueto the large-scale use of private vehicles,and mobility related to jobs accounts for 60percent of total emissions due to the lack ofan appropriate public transport system andhaphazard growth with unplanned urban-ization. There is a lack of integrated landuse and mobility planning, and therefore themajority of people commute long distances,

<0.05 0.05 - 0.1 0.1 - 0.15 0.15 - 0.2 >0.20

1973 2013

N N

0 1.5 3 6 9 12Kilometers

0 1.5 3 6 9 12Kilometers

Figure 8 Gradient-wise vegetation density

0 1.5 3 6 9 12Kilometers

N

<0.05 0.05 - 0.1 0.1 - 0.15 0.15 - 0.2 >0.20

Figure 9 Ward-wise vegetation density (2013)

BANGALORE 17

Domestic sector21.59%

Industry12.31%

Agriculture andlivestock1.31%

Waste and waste-water5.73%

Electricityconsumption

(other) ∗

15.46%Auxiliary

consumption andtransmission

loss0.13%

Road transport43.48%

Figure 10 GHG emissions sector-wise

thus contributing to emissions. Higher fuelconsumption, enhanced pollution levelsdue to the increase of private vehicles, andtraffic bottlenecks have contributed signifi-cantly to carbon emission. The consumptionof electricity has drastically increased incertain corporation wards with the adoptionof inappropriate building architecture forthe tropical climate region. This is evidentfrom higher per capita electricity consump-tion in the zones dominated by high risebuildings with glass facades, which require14,000–17,000 units (kWh) per year com-pared to the zones with eco-friendly buildings(1300–1500 units/person/year). In addition,mismanagement of solid and liquid wasteshas aggravated the situation. Dumping ofuntreated solid and liquid waste into thewater bodies has increased the anaerobiccondition, leading to emission of greenhousegases (methane, CO2, etc.).

Land surface temperature and urban heatisland

Temporal analysis of LST shows a gradualincrease from 19.03± 1.47 ∘C (in 1991),

26.57± 1.25 (2000), 31.24± 2.21 (2007),to 37.14± 2.68 ∘C (2016) with the increasein paved surfaces in the city. Relativelylower temperatures in the regions with veg-etation cover (25.79± 0.44), 24.20± 0.27(water bodies) highlights the role of naturalresources in moderating the local climate.Urban heat islands are formed with theincrease of surface and atmospheric temper-atures due to anthropogenic heat dischargedue to energy consumption, increased landsurface coverage by artificial materials hav-ing high heat capacities and conductivities,radiation by buildings due to inappropri-ate architecture, and the drastic declinein vegetation and water pervious surfaces(which aid in reducing the surface tem-perature through evapotranspiration, etc.).The implication for local climate is evident,with an increase of ∼2.5 to 3 ∘C tempera-ture during the last two decades) with thedecline of natural resources (decline in veg-etation cover and disappearance of waterbodies) that aided in moderating the localclimate.

18 BANGALORE

CONCLUSION

Cities’ origins can be traced back to the rivervalley civilizations of Mesopotamia, Egypt,the Indus Valley, and China. Initially thesesettlements were largely dependent uponagriculture; however, with the growth in pop-ulation the city size increased and economicactivity transformed to trading. The processof urbanization gained impetus with theIndustrial Revolution 200 years ago andaccelerated in the 1990s with globalizationand consequent relaxations in the marketeconomy. Migration to urban areas pushedthe growth of towns and cities with thehope of better living standards, consideringrelatively better infrastructural facilities (edu-cation, recreation, health centers, banking,transport, and communication) and higherper capita income. However, unplannedurbanization leads to large-scale land-usechanges affecting the sustenance of local nat-ural resources. Hence, understanding spatialpatterns of changes in the land and advancevisualization of growth is imperative forsustainable management of natural resourcesand mitigation of changes in climate. Thiswould help the city planners when planninghow to mitigate the problems associated withthe increased urban area and population, andultimately to build sustainable cities.

Bangalore has experienced unprecedentedrapid urbanization and sprawl, which hasled to large-scale land-cover changes andassociated serious environmental degra-dation, posing serious challenges to thedecision-makers in the city planning andmanagement process and involving manyserious challenges such as climate change,enhanced greenhouse gases (GHG) emis-sions, lack of appropriate infrastructure,traffic congestion, and lack of basic amenities(electricity, water, and sanitation) in manylocalities. Temporal analyses of urbaniza-tion using temporal remote sensing data

from 1973 to 2016 reveals 1005 percentconcretization (or paved surface increase)with a decline in green spaces (88 percentdecline in vegetation), wetlands (79 percentdecline), higher air pollutants, and a sharpdecline in the groundwater table. Quantifi-cation of the number of trees in the regionusing remote sensing data with field censusreveals that there are only 1.48 million treesto support Bangalore’s population of 9.5million, indicating one tree for every sevenpersons in the city. This is insufficient even tosequester respiratory carbon (540–900 g perperson per day). Geo-visualization of likelyland uses in 2020 through multi-criteriadecision-making techniques (fuzzy AHP)reveals a calamitous picture of 93 percent ofthe Bangalore landscape filled with pavedsurfaces (urban cover) and drastic reductionin open spaces and green cover. This wouldmake the region GHG rich, water scarce,nonresilient, and unliveable, deprivingcity-dwellers of clean air, water, and envi-ronment. Major implications of unplannedurbanization are (a) loss of natural resources,(b) groundwater decline, (c) altered regionalhydrology and recurring episodes of floods,(d) enhanced GHG footprint and significantcontributions to global warming, (e) reducedcarbon sequestration (with the removal oftree cover and water bodies), (f) alterationsin micro climate, (g) heat island forma-tion, (h) escalated energy consumption, (i)mismanagement of solid and liquid wastegenerated in the region, (j) health impacts(increased instances of kidney failure, respi-ratory ailments, cancer, etc.), (k) aggressivebehavior of humans (leading to frequentunrest with intra-family and societal con-flicts), and (l) domination of anti-social gangswith the fragmented governance of localadministration.

Conversion of wetlands to residential andcommercial areas has compounded the prob-lem by removing the interconnectivities in an

BANGALORE 19

undulating terrain. Encroachment of naturaldrains, alteration of topography involvingthe construction of high rise buildings,removal of vegetative cover, and reclama-tion of wetlands are the prime reasons forfrequent flooding even during normal rain-fall post-2000. Field investigations (during2015–2016) of 105 lakes revealed that 98percent of lakes have been encroached uponfor illegal buildings, 90 percent of lakes aresewage fed, 38 percent surrounded by slumsand 82 percent showed loss of catchmentarea. Also, lake catchments are being usedas dumping yards for either municipal solidwaste or building debris. Indiscriminatedisposal of solid and liquid waste (rich inorganic nutrient) has enriched nitrate levelsin the surrounding groundwater resources,threatening the residents’ health (such askidney failure, cancer, etc.). Washing, house-hold activities, vegetable cultivation, andeven fishing were observed in some con-taminated lakes. Multi-storied buildingshave come up on some lake beds, interferingin the natural catchment flow and leadingto a deteriorating quality of water bodies.Unauthorized construction in valley zones,lakebeds, and storm water drains highlightsthe apathy of decision-makers while mir-roring weak and fragmented governance.This correlates with the increase in unautho-rized constructions violating town-planningnorms (city development plan) which hasaffected severely open spaces and in partic-ular water bodies. Large-scale fish mortalityin recent months further highlights the levelof contamination and irresponsible manage-ment of water bodies. Sustained inflow ofuntreated sewage has increased the organiccontent beyond the threshold of remedia-tion capability of respective water bodies.Increasing temperature (of 34 to 35 ∘C) withthe onset of summer enhanced the biologicalactivities (evident from higher biochemical

oxygen demand and ammonia) that low-ered dissolved oxygen levels leading to fishdeath due to asphyxiation. Thus, unplannedurbanization not only contributes to globalclimate change by emitting the majority ofanthropogenic greenhouse gases but alsois particularly vulnerable to the effects ofclimate change and extreme weather. Thisemphasizes the need to improve urbansustainability through innovation whileaddressing technical, ecological, economic,behavioral, and political challenges to cre-ate cities that are low carbon, resilient, andliveable.

The “smart cities” mission launched by thegovernment in India recently (June 2015)envisages developing physical, institutional,and social infrastructure in select cities withcentral assistance targeted at improving thequality of life as well as economic visibility ofthe respective urban centers. Four strategickey components are:

1. greenfield development through smarttownships by adopting holistic landmanagement;

2. pan-city development through adoptionof smart applications like transport, reuseand recycling of wastewater, smart meter-ing, recovering energy from solid waste,etc.;

3. retrofitting, to make existing areas moreefficient and liveable by reducing thegreenhouse gas footprint, improvingpower and treated water supply, improv-ing communication and infrastructureconnectivity, and security;

4. redevelopment of existing built-up areas,creation of new layouts through mixedland use, adoption of appropriate floorarea index (FAI) considering the level ofexisting and scope for improvement ofinfrastructure and basic amenities, whichhelps in keeping the city’s growth within

20 BANGALORE

the region’s carrying capacity so thaturban infrastructure becomes inclusive.

This all entails efficient decision-makingthrough integrated land-use planning as perthe city’s requirements considering mobility,etc., to minimize mobility related to jobs;enhancement of the functional capacitythrough user-friendly and economic publictransport support; development of massrapid-transport systems for easy mobilitywithin and between cities; effective use ofICTs as enabling technologies to improve thelevel of services. This is a welcome measure asmost cities are in civic and financial disarraybecause of unplanned rapid urbanization.Environmentally sound urban centers withessential basic amenities and advancedinfrastructures (such as sensors, electronicdevices, and networks) would stimulate sus-tainable economic growth and improvementsin services to citizens. The deployment ofinformation and communication technol-ogy infrastructures for effective governancesupport social and urban growth throughimproved economy and active participationof citizens. Indian cities, while exhibitingtechnological innovations and connect-edness, should also focus on increasedliving comfort through adequate infrastruc-ture and essential basic amenities to everycitizen.

ACKNOWLEDGMENT

We are grateful to the NRDMS division,the Ministry of Science and Technology(DST), Government of India; the Ministry ofEnvironment, Forests and Climate Change,Government of India; APN for ClimateChange; University of Southern California,USA; and ISRO-IISc Space Technology Cell,Indian Institute of Science for the financialand infrastructure support.

SEE ALSO: Cities in Developing Countries;Digital Cities; Global City; Global South/North;Globalization; Spatial Analysis; Sprawl

REFERENCES

Aithal, B. H., and T. V. Ramachandra. 2016a.“Modelling Urban Dynamics of Bhopal, India.”Journal of Settlements and Spatial Planning, 7(1):1–14.

Aithal, B. H., and T. V. Ramachandra. 2016b. “Vi-sualization of Urban Growth Pattern in ChennaiUsing Geoinformatics and Spatial Metrics.”Journal of the Indian Society of Remote Sensing,44(4): 617–633.

Aithal, B. H., S. Vinay, and T. V. Ramachandra.2014. “Landscape Dynamics Modelling throughIntegrated Markov, Fuzzy-AHP and CellularAutomata.” In Proceedings of International Geo-science and Remote Sensing Symposium (IEEEIGARSS 2014), Quebec City Convention Centre,Quebec, Canada, July 13–19, 2014.

Aithal, B. H., B. Vishwanatha, and T. V. Ramachan-dra. 2015. “Spatial Patterns of Urban Growthwith Globalization in India’s Silicon Valley.”In Proceedings of National Conference on OpenSource GIS: Opportunities and ChallengesDepartment of Civil Engineering, IIT (BHU),Varanasi, October 9–10, 2015.

Karnataka Human Development Report, 2005.Government of Karnataka. Accessed December20, 2017, at http://planningcommission.nic.in/plans/stateplan/sdr_pdf/shdr_kar05.pdf.

Ramachandra T. V., B. H. Aithal, and B. Barik.2014a. “Urbanisation Pattern of Incipient MegaRegion in India.” Tema: Journal of Land Use,Mobility and Environment, 7(1): 83–100.

Ramachandra, T. V., B. H. Aithal, and S. D.Durgappa. 2012. “Insights to Urban Dynam-ics through Landscape Spatial Pattern Analysis.”Journal of Applied Earth Observation and Geoin-formation, 18: 329–343.

Ramachandra, T. V., B. H. Aithal, and K. Shreejith.2015. “GHG Footprint of Major Cities in India.”Renewable and Sustainable Energy Reviews, 44:473–495.

Ramachandra, T. V., B. H. Aithal, and M. V.Sowmyashree. 2013. “Analysis of Spatial Pat-terns of Urbanisation Using Geoinformaticsand Spatial Metrics.” Theoretical and EmpiricalResearches in Urban Management, 8(4): 5–24.

BANGALORE 21

Ramachandra, T. V., B. H. Aithal, and M. V.Sowmyashree. 2014. “Urban Footprint of Mum-bai: The Commercial Capital of India.” Journalof Urban and Regional Analysis, 6(1): 71–94.

Ramachandra, T. V., B. H. Aithal, and M. V.Sowmyashree. 2015. “Monitoring Urbanizationand Its Implications in a Mega City fromSpace: Spatiotemporal Patterns and Its Indica-tors.” Journal of Environmental Management,148: 67–91.

Ramachandra, T. V., S. Bharath, and B. H. Aithal.2013. “Spatio-Temporal Dynamics along theTerrain Gradient of Diverse Landscape.” Jour-nal of Environmental Engineering and LandscapeManagement, 22(1): 50–63.

Ramachandra, T. V., B. Vishnu, B. H. Aithal,S. Bharath, and U. Kumar. 2011. “Expositionof Urban Structure and Dynamics throughGradient Landscape Metrics for SustainableManagement of Greater Bangalore.” FIIB Busi-ness Review, 1(1): 1–17.

State and District Domestic Product of Kar-nataka. 2016. Accessed December 20, 2017,at http://des.kar.nic.in/docs/sip/SDP%202016%20Report%20(1).pdf.

FURTHER READING

Ramachandra, T. V., and B. H. Aithal. 2016. “Ben-galuru’s Reality: Towards Unlivable Status withUnplanned Urban Trajectory.” Current Science,110(12): 2207–2208.

Ramachandra, T. V., V. Bajpai, G. Kulkarni, B. H.Aithal, and S. S. Han. 2017. “Economic Disparityand CO2 Emissions: The Domestic Energy Sec-tor in Greater Bangalore, India.” Renewable andSustainable Energy Reviews, 67: 1331–1344.

Sudhira, H. S., T. V. Ramachandra, and M. H. BalaSubrahmanya. 2007. “Bangalore.” Cities, 24(5):379–390.