Urban green cover assessment and site analysis in Chennai, Tamil nadu - a remote sensing and gis approach

Abstract

Green space distribution plays a imperative role in urban planning since they contribute significantly in enhancing ecological quality of metropolitan areas. It improves air quality, urban health, conserving biodiversity, reducing noise, etc. Removal of vegetation cover can be identified as one of the poorest effects of urbanization. Proper distribution of green spaces in urban environments is consequently more inevitable for the sustainable development and healthy living. Hence, it is necessary to identify the green space requirement quantitatively and spatially. To achieve the goal, high resolution Cartosat-1 satellite data, were used to analyse the spatial pattern. Spatial features like Point feature and polygon features were demarcated from imagery. Individual trees, group of trees, bushes, building area (covering both residential/industrial area), water bodies (lakes, ponds, reservoir, streams, rivers etc.), parks and temples has been considered. The tree cover area covers 72.82Sqkm, Buildings covers 241Sqkm, Parks covers 9.28Sqkm, Water bodies covers 35.73 Sq.km and other area 104.40Sqkm out of 464sq.km area coverage of Chennai municipality. Subsequently, green spaces required to be created are calculated with respect to WHO standards of green spaces per capita for healthy living (9.5 m2/ person) and a methodology is developed to spatially define appropriate areas to establish them.

Full text

VOL. 10, NO. 5, MARCH 2015 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2015 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
2239
URBAN GREEN COVER ASSESSMENT AND SITE ANALYSIS
IN CHENNAI, TAMIL NADU - A REMOTE SENSING
AND GIS APPROACH
Meera Gandhi. G, Nagaraj Thummala and Christy. A
Faculty of Computing, Centre for Remote Sensing and Geo informatics, Sathyabama University, Rajiv Gandhi Road, Jeppiaar Nagar,
Chennai, India
E-Mail: professorgandhi29@gmail.com
ABSTRACT
Green space distribution plays a imperative role in urban planning since they contribute significantly in enhancing
ecological quality of metropolitan areas. It improves air quality, urban health, conserving biodiversity, reducing noise, etc.
Removal of vegetation cover can be identified as one of the poorest effects of urbanization. Proper distribution of green
spaces in urban environments is consequently more inevitable for the sustainable development and healthy living. Hence, it
is necessary to identify the green space requirement quantitatively and spatially. To achieve the goal, high resolution
Cartosat-1 satellite data, were used to analyse the spatial pattern. Spatial features like Point feature and polygon features
were demarcated from imagery. Individual trees, group of trees, bushes, building area (covering both residential/industrial
area), water bodies (lakes, ponds, reservoir, streams, rivers etc.), parks and temples has been considered. The tree cover
area covers 72.82Sqkm, Buildings covers 241Sqkm, Parks covers 9.28Sqkm, Water bodies covers 35.73 Sq.km and other
area 104.40Sqkm out of 464sq.km area coverage of Chennai municipality. Subsequently, green spaces required to be
created are calculated with respect to WHO standards of green spaces per capita for healthy living (9.5 m2/ person) and a
methodology is developed to spatially define appropriate areas to establish them.
Keywords: green pace, Chennai Corporation, tree area, crown area, remote sensing, GIS.
1. INTRODUCTION
Tree cover in urban areas around the world, is
declining and inflexible cover is increasing due to the
demand of the land for development Forest Survey of
India (FSI) has been assessing country’s forest cover since
the 1980’s using data from remote sensing satellites on a
two-year cycle. Due to a substantial number of trees tree
cover is not captured by the Satellite data and reported as
tree cover for the first time in 2001 assessment. The
planned development of Chennai city present a clean and
green with trees, plants, lakes and parks and towns in
Chennai. It is growing at fast pace in terms of
urbanization, technology, infrastructure, and environment.
The pace of urbanization is harmfully affecting the green
cover in the urban areas. Trees provide numerous
Environmental, Social and Economic benefits to people
and their services in maintaining environment are been
universally accepted. The tree canopies shows moderate
temperature, provide shade to building, area of sidewalk,
streets and reduce pollution. Urban areas Kuhelmeister,
G., 1998 can comprises large variety of green spaces, such
as Parks/ gardens green space near institution, Industrial
area green spaces ( Heinze, J., 2011), and private green
spaces (Boone et al, 2010). It includes woodlands, farm
lands, public gardens and play areas. Green spaces play a
major role in urban areas through their environmental,
aesthetic, social and economic contributions to residents’
health and wellbeing (Cavanagh et al. 2009). (e.g. Faryadi
and Taheri, 2009).
In order to design an appropriate urban green
cover assessment I.P. Senanayake 2014, spatial features
must be evaluated. An attempt is carried out in this study
to map the status of green coverage land use and land
cover of the Chennai city area using high resolution
Cartosat-1 satellite data. In order to achieve the goal, high
resolution Cartosat-1 satellite data, were used to analyse
the spatial pattern of land cover change in the area and the
future growth was modelled by applying CA-Markov
model described in Shikhar deep 2014. Spatial features
like Point feature and polygon features were demarcated
from imagery. Individual trees, group of trees, bushes,
building area (covering both residential/industrial area),
water bodies (lakes, ponds, reservoir, streams, rivers etc.
parks and temples has been considered detect the land
consumption rate and the changes that have taken place
particularly in their built-up area.
2. STUDY AREA AND DATASET
Chennai District is bounded by Northern
Latitudes of 12° 59’ 10” and 13° 08’ 50” and eastern
Longitudes of 80° 12’ 10” and 80° 18’ 20” according to
Survey of India Topographical Maps Nos. 66 C / 4 and
and 66 D 1 and 5. (Santhiya et al 2010). The North East
monsoon during the months of October, November and
December essentially contributes the rainfall for the
district. The average annual rainfall of the district is in
the range of 1285.6 to 1232.7mm. Location map of the
Chennai Corporation Study Area is shown in Figure-1.
GIS datasets are common data sources used for
geo processing and are useful for automated data
processing and GIS analysis. Datasets are used as inputs,
and new datasets are derived as results for various geo
processing tools. Geo processing helps you to automate
many tasks as a series of operations so they can be run as a

VOL. 10, NO. 5, MARCH 2015 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2015 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
2240
single step. This helps to create a repeatable, well-
documented data processing workflow. Users also work
with Arc GIS datasets to perform spatial analysis. Visual
interpretation plays a major role in delineating spatial
features of the earth by a geospatial expert. It can be
concluded that, green space planning could be an essential
component of any urban development.
3. METHODOLOGY - URBAN GREEN COVER
MAPPING
Tree cover means the area covered by crown of
trees that is too small to be delineated by digital
interpretation of remote sensing data at 1:50,000 scales
used for forest cover assessment. India’s National Forest
Policy aims at maintaining 33 percent of country’s
geographical area under forest and tree cover.
(http://www.fsi.nic.in/sfr2003/treecover.pdf). The present
assessment of forest cover, carried out by digital
processing of satellite data at 1:50,000 scales, includes
forests and tree crops having 10 percent or more canopy
density and with an area of more than 1.0 ha. The tree
cover comprises of small patches of trees (< 1.0 ha) in
plantations and woodlots, or scattered trees on farms,
homesteads and urban areas, or trees along linear features,
such as roads, canals, bunds, etc. has been estimated by
mainly using field inventory methods.
Boundary of the grouped trees was demarcated
spatially and its spatial extension has been calculated.
From Ground truth information, the tree diameter is
identified. Number of trees within the spatial extent can be
attained by aggregating the trees diameter to total
demarcated area. Tree Cover in the urban areas should be
treated as important and essential constituent of urban
infrastructure. Estimation of tree population and tree cover
in urban area and publication of a report on the status of
tree cover in Municipal Corporations and municipalities
shall be of immense use for developing appropriate action
plan to improve tree cover for all urban areas.
Figure-1. Location map of the Chennai corporation study area.
3.1 Crown area mapping
Crown area is the area covered by the living
branches and foliage of trees. It is often expressed as a
percentage of total land area. Crown area is estimated
using Sky camera in the real field conditions. In our work,
Cartosat 1 data helps to find the crown diameter of
individual tree. It is well known fact the gaps of the trees
will not counted as crown area, which is eliminated in the
shape file. Average crown spread is one of the parameters
commonly measured as part of various champion tree
programs and documentation efforts. Other commonly
used parameters, outlined in Tree measurement, include
height, girth, and volume. Methodology of Tree height
measurement, Tree girth measurement, and Tree volume
measurement are presented in the links herein.
3.2 Water bodies and buildings coverage
Water bodies include tanks, ponds, lakes, rivers,
streams and artificial storage structures that are spatially
visible on the satellite images are identified. The size of
the water bodies may increase in width and length
depending on the seasonal inflow of water based on
monsoon rains and may decrease in size during non
monsoon seasons. Which are dynamic and temporal in
nature. Whereas, Cartosat 1 data acquired during the non
monsoon (with less per cent cloud cover) was adopted to
extract the required information of spatial extent. It is a
well known fact that all the major corporations in Tamil
Nadu possess large number of concrete structure to
accommodate all Government offices, residential complex
and business centres. Such features are delineated based
on reflectance properties of the satellite images that are
represented by light tonal variation due to highest
reflectance.
3.3 Delineating parks and temples
In Delineating spatial features such as parks and
temples, every pattern should be in the form of rectangle
or square. All these boundaries are delineated in shape file
format. Features have a similar pattern are seen through
Cartosat 1 Satellite imagery. Surface feature may be
represented by the open spaces at the centre of parks and
continuous plantation of trees at the boundaries as shown
in the Figure-5 and Figure-6. Minimum concrete structure
is available in the parks for recreation purpose.
4. RESULTS AND DISCUSSIONS
It is clear from the visual interpretation that urban
and built up has increased significantly based on the
assessment. The accuracy assessment will help in
validating green coverage area. The estimation of total
number of trees in Chennai Corporation is shown in Table-
1. The interpreted results of percentage of urban green
area assessment is shown in Table-2. Chennai
municipality boundary covers total area of 465 Sq.km in
which minimum area is occupied with temple area and
maximum area occupied with buildings. Spatial features
like Point feature and polygon features were distinguished

VOL. 10, NO. 5, MARCH 2015 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2015 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
2241
from imagery. Individual trees, group of trees, bushes,
building area (covering both residential/industrial area),
water bodies (lakes, ponds, reservoir, streams, rivers etc.
parks and temples has been considered. The tree area
covers 72.82Sq.km, Buildings covers 241Sq.km, Parks
covers 9.28Sq.km, Water bodies covers 35.73 Sq.km and
other area outlined as 104.40Sq.km out of 464 sq.km area
coverage of Chennai municipality Trees will be a distinct
feature that is visible in Cartosat satellite images. In
general, trees on the road side are mentioned as avenue
trees. These trees are seen as individual trees on the image
as shown in the Figure-3. In the case of large grouping of
trees, trees were grouped based on their shape and
distribution density categorised as either sparse or high
dense. In dense region, a composite map is shown with
dense grouping of trees.
4.1 Crown area calculation
Crown spread is taken independent of trunk
position. Spread should be measured to the tips of the
limbs, not to “notches” in the crown shape, and at
approximately right angles from each other according to
equation (1)
Average crown spread = (longest spread + longest cross-
spread)/2 (1)
With the increased availability of high resolution
air photos, crowns of individual trees can be distinguished
providing another option for measuring crown spread. The
latitude and longitude of the tree can be read directly from
Google Earth. Google Earth itself includes a ruler tool that
can be used to measure diameters or spokes across the
crown of the tree. Alternatively the crown area can be
measured and crown spread calculated from that value.
Easy Acreage V1.0 (demo version)
http://www.wildsoft.org / 2013 is a Google Earth area
measurement tool that calculates the area of any shape
outlined on the Google Earth display. Outline the edge of
the trees canopy, following the branches and hollows
around the canopy perimeter, including any enclosed
hollows within the canopy outline and read the area
provided by Easy Acreage. Average crown spread can be
determined with a simple formula in equation (2).
Crown spread = 2(area/π) ½ (2)
Here area is taken as the area of an equivalent circle.
2
RTT
TA NTCRA


 (3)
Where TA = Total tree area, TCRA = Total crown area
NT = Number of trees.
The total area of Chennai municipality comprises
of 464.65 Sq.km. Chennai Metropolitan Development
Authority map has been used for deriving boundary
considered as Greater Chennai (including Tambaram and
Avadi municipality. Layers were derived from Cartosat 1
imagery based on themes. Buildings are much congregated
at the central part of Chennai and therefore buildings were
digitized from imagery.
Figure-2. Natural colour composite (Cartosat image) of Chennai municipal corporation boundary.
During the path of the mapping techniques, area
of buildings as view from space a minimum size of
2m*2m size can be mapped. Larger buildings could be
spatially mapped in a convenient manner. From the
analysis, a strategy of an eight percent cover from the
obtained cover is employed to get the actual crown cover
from the total area. Each tree area is calculated, assuming
area is the area of an equivalent circle (circle area= ∏r2).
Approximately Circle Radius value was taken as 3m and
each tree area is (3.14*3*3) calculated by total number of
trees from total crown area divided by each tree area is
shown in equation (3).
Table-1. Estimation of number of trees in Chennai corporation crown area.
Corporation Total tree cover
area Sq.km Gaps between
tree area Sq.km Crown area
Sq.km Number of
trees
Chennai 72.82 5.83 66.99 23,70,504

VOL. 10, NO. 5, MARCH 2015 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2015 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
2242
Figure-3. Tree cover area of Chennai municipality.
Figure-4. Built up area of Chennai municipality.
The map scale has been fixed to 1:10000 scales in
order to obtain maximum features with precise accuracy.
Mapping of individual buildings have been shown in the
Figure-4.
Figure-5. Parks area coverage of Chennai municipal.
Figure-6. Tanks area coverage of Chennai municipal.
Table-2. Interpreted results of Chennai municipal
corporation.
Chennai municipal corporation
Name Area Sqkm % of area
coverage
Buildings 241.5794 19.23%
Parks 9.2899 0.50%
Tanks 35.7288 0.13%
Temples 0.8449 5.49%
Trees 72.8157 0.06%
Others 104.394 0.04%
Figure-7. Temple area Chennai city municipality.
Every individual shape file of buildings was merged as a
single polygon. Total area of the Chennai buildings
(241.57 Sq.km) is calculated using Arc GIS software.
Tanks possess uniform shape either as rectangle or square
or circle as viewed through Satellite data.
Chembarambakkam, puzhal, porur, Karanodai, are some
of the larger lake that supplies drinking water to Chennai
City. Boundary of tanks was digitized as area of the tank
as shown in the Figure-6. Water level of the tank may not
be considered as tank boundary, as water level of the tank
varies according to the monsoon rainfall. Table-2 depicts
the Chennai Municipal boundary percentage of urban area
assessment in which buildings occupy the highest
percentage of 19.23%. Park area of Chennai Municipal
Corporation that accounted for 9.29 Sq.km Cartosat 1
Satellite image outlines the area of feature that is greater
than 3m*3m with square or rectangular pattern as temple
boundary. Temples are widely distributed within the
Chennai Municipal Corporation. Old temples can be
derived from satellite imagery which acquire larger
boundary with more open spaces and small number of
trees. New temples look smaller in size having less open
spaces and greenery. Figure-7 depicts how every temple
feature is derived from satellite imagery in shape file
format in turn merged to a single layer.
CONCLUSIONS
A Green space assessment study has been carried
out to measure the existing green spaces in Chennai City
of Tamil Nadu quantitatively and to identify sites to create
new green spaces in order to upraise the green spaces for
the minimum required value recommended by WHO (i.e.
9.5 m2/ person). The methodology adopted in this study
can be utilized effectively in other urban centres to

VOL. 10, NO. 5, MARCH 2015 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
©2006-2015 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
2243
calculate the required amount of green spaces and to
identify the sites to create green spaces, in order to
enhance the environmental quality of the city based on
WHO standards.
REFERENCES
Boone, C.G., Cadenasso, M.L., Grove, J.M., Schwarz, K.,
Buckley, G.L., 2010. Land-scape, vegetation
characteristics, and group identity in an urban and
suburbanwatershed: why the 60s matter. Urban Ecosyst.
13, 255-271.
Cavanagh, J.E., Zawar-Reza, P., Wilson, J.G., 2009.
Spatial attenuation of ambientparticulate matter air
pollution within an urbanized native forest patch.
UrbanFor. Urban Green. 8(1), 21-30.
Faryadi, S., Taheri, S., 2009. Interconnections of urban
green spaces and environmental quality of Tehran.
International Journal of Environmental Research 3(2),
199-208.
NASA-Earth Observatory, 2012. Measuring Vegetation
(NDVI and EVI), Retrieved July 10, from
www.earthobservatory.nasa.gov.
Rusli, N. and Ludin, A.N.M., 2009. Integration of remote
sensing technique and geographic information system in
open space and recreation area evaluation, proceedings of
10th SENCAR and 1st CONVEESH 2009, Manado,
Indonesia.
United Nations. 2011. Population, distribution,
urbanization, internal migration and development: An
international perspective. Department of Economic and
Social Affairs, Population Division.
I.P. Senanayake, W.D.D.P. Welivitiya, P.M. Nadeeka.
2013. Assessment of Green Space Requirement and Site
Analysis in Colombo, Sri Lanka - A Remote Sensing and
GIS Approach, International Journal of Scientific and
Engineering Research, Vol. 4, Issue 12, December. ISSN
2229-5518.
Kuhelmeister, G. 1998. Urban Forestry: Present Situation
and Prospects in the Asia and Pacific region, FAO Asia-
Pacific Forestry Sector Outlook Study. Rome: Food and
Agriculture Organization of the United Nations, Forestry
Policy and Planning Division.