URBAN HEAT ISLAND ANALYSIS USING
LAND SURFACE TEMPERATURE: A
GIS-BASED STUDY OF KOCHI
Using Satellite Remote Sensing Data and ArcGIS
Pro: A Geospatial Remote Sensing Case
Study Submitted by:
Ashish Chaurasiya 2026
1. INTRODUCTION
1.1 Urban Development and Urbanization
Urbanization is one of the major forms of land-cover change associated with human
development. The expansion of cities results in the conversion of natural surfaces such as
vegetation, agricultural land and wetlands into built-up areas, roads, commercial structures and
other impervious surfaces. Such changes alter the physical characteristics of the Earth's surface
and influence the local thermal environment.
The increasing concentration of buildings, transportation infrastructure and human activities can
modify the energy balance of an urban area. Concrete, asphalt and other impervious materials
absorb and store solar energy during the day and release stored heat gradually. At the same time,
the reduction in vegetation decreases evapotranspiration and natural shading. These processes
contribute to increased surface temperatures within urban environments.
Kochi has experienced considerable urban development and infrastructure expansion. The
transformation of land cover associated with this growth makes the city an appropriate area for
investigating spatial variations in Land Surface Temperature (LST).
1.2 Urban Heat Island Effect
The Urban Heat Island (UHI) effect refers to the phenomenon in which urban areas experience
higher temperatures than surrounding less-developed or rural areas. The phenomenon can occur
because of changes in surface properties, reduced vegetation, increased heat storage by buildings
and roads, and anthropogenic heat emissions.
The UHI phenomenon can be studied through two related temperature concepts:
● Air temperature, which is measured using meteorological instruments.
● Land Surface Temperature (LST), which represents the temperature of the Earth's
surface and can be estimated using thermal remote sensing.
1.3 Detection through Remote Sensing
Remote sensing provides a method for observing the Earth's surface without direct physical
contact. Different materials have different spectral characteristics in terms of reflection and
absorption. Thermal infrared remote sensing is particularly useful for investigating surface
temperature.
Satellite thermal imagery can provide spatially continuous information over large urban regions
and therefore has advantages over measurements from individual weather stations.
Satellite-derived temperature data can subsequently be processed using GIS software to:
● visualize temperature distribution;
● classify temperature zones;
● calculate the area of different thermal classes;
● identify potential heat hotspots; and
● support urban environmental analysis.
In this practical, ArcGIS Pro was used for processing and visualization of the LST data and for
preparing the final thematic map.
PROBLEM STATEMENT
Rapid urban development causes significant changes in land-use and land-cover patterns. The
replacement of vegetation, wetlands and open land by buildings, roads and other impervious
surfaces can alter the thermal characteristics of an urban area.
The growth of urban areas without adequate consideration of green spaces and natural cooling
systems may result in localized heat accumulation. Such thermal hotspots can influence urban
environmental quality and increase vulnerability to heat-related conditions.
2. Scope of the Study
The scope of this study is to examine the spatial variation of Land Surface Temperature (LST)
within the selected study area of Kochi using Geographic Information System techniques. The
study focuses on processing and classifying the available temperature raster into different
thermal zones and representing their spatial distribution through a thematic map.
The study covers the following aspects:
● Spatial analysis of Land Surface Temperature in Kochi.
● Classification of the temperature raster into five temperature zones.
● Calculation of the number of raster cells and area covered by each temperature class.
● Preparation of a thematic LST map using ArcGIS Pro.
● Identification of comparatively cooler and warmer regions.
● Interpretation of the observed temperature pattern in relation to general urban surface
characteristics.
● Application of GIS techniques for urban environmental analysis.
The study is primarily focused on surface temperature mapping. It does not independently
quantify the relationship between LST and vegetation or built-up area using indices such as
NDVI or NDBI. Such analysis can be considered as an extension of the present study.
3. OBJECTIVES
To analyze and map the spatial distribution of Land Surface Temperature in
Kochi using GIS and ArcGIS Pro techniques for understanding the spatial
characteristics of the Urban Heat Island environment.
Specific Objectives
1. To obtain and prepare the required spatial temperature data for the study area.
2. To process the LST raster using ArcGIS Pro.
3. To classify the temperature values into five different thermal zones.
4. To calculate the pixel count and area corresponding to each temperature class.
5. To prepare a thematic Land Surface Temperature map of Kochi.
6. To identify areas exhibiting comparatively high and low surface temperatures.
7. To interpret the spatial distribution of the identified thermal zones.
8. To demonstrate the application of GIS in urban environmental and heat analysis.
4. STUDY AREA
4.1 Kochi, Kerala
The study area selected for this practical is Kochi, located in the state of Kerala on the
southwestern coast of India.
Kochi is an important urban and commercial centre with a mixture of densely developed areas,
residential zones, transportation infrastructure, vegetation, wetlands and water bodies.
This diversity makes Kochi suitable for thermal analysis because different land-cover types have
different abilities to absorb, store and release heat.
4.2 Criteria for Selection
Kochi was selected because:
● It is a rapidly developing urban region.
● It contains both highly developed and relatively natural areas.
● It has extensive water bodies and wetland environments.
● Urban development has resulted in changes in land cover.
● Satellite-based thermal data can be used to study its spatial temperature pattern.
● GIS provides an effective method for visualizing these variations.
4.3 Importance of the Coastal Environment
Kochi's coastal location and presence of water bodies are important when interpreting its thermal
pattern. Water and vegetated areas can exhibit lower surface temperatures than many impervious
urban surfaces.
Therefore, the LST map provides an opportunity to observe the contrast between relatively
cooler and warmer portions of the study area.
5. LITERATURE REVIEW
5.1 Thermal Remote Sensing and Urban Climate
Voogt and Oke examined the application of thermal remote sensing in urban climate studies and
highlighted its importance in understanding the Urban Heat Island effect. Thermal remote
sensing allows researchers to examine spatial variations in urban surface temperatures and
provides information that cannot be obtained from a limited number of meteorological stations.
5.2 Satellite-Based Analysis of Urban Surface Temperature
Dousset and Gourmelon (2003) investigated summertime urban microclimates in Los Angeles
and Paris using satellite thermal infrared imagery. Their study combined thermal information
with land-cover classification and GIS analysis. This demonstrated the usefulness of integrating
thermal remote sensing with land-cover information to understand urban thermal patterns.
5.3 Land-Cover Change and Urban Heat Island
Chen et al. (2006) investigated the relationship between Urban Heat Island intensity and
land-use/land-cover changes in the Pearl River Delta region of China. Landsat TM and ETM+
imagery were used to obtain temperature information and land-use/land-cover characteristics.
The research demonstrated the importance of examining urban growth and land-cover change
when studying thermal patterns.
5.4 Vegetation, Impervious Surfaces and LST
Yuan and Bauer (2007) examined the relationship between Land Surface Temperature,
impervious surface area and NDVI using Landsat imagery. Their study compared vegetation and
impervious surface indicators as measures of surface Urban Heat Island effects. The research
demonstrated the importance of vegetation and built-up surfaces when interpreting spatial
temperature patterns.
5.5 Relevance to the Present Study
The literature demonstrates that satellite-based thermal information, combined with GIS, can be
used to investigate urban thermal environments. Previous research has commonly examined:
● Land Surface Temperature
● Land-use/land-cover change
● Vegetation
● Built-up surfaces
● Urban morphology
● Thermal hotspots
The present practical applies the same general concept to Kochi, but focuses specifically on the
generation and classification of an LST map using ArcGIS Pro.
6. METHODOLOGY
The methodology adopted for the present study involves processing and analyzing the Land
Surface Temperature (LST) raster using ArcGIS Pro. The analysis was carried out to classify the
temperature values, determine the spatial extent of each temperature class, and prepare a
thematic map of Kochi.
6.1 Data Preparation
The available Land Surface Temperature raster was imported into ArcGIS Pro as the primary
input dataset. The spatial reference and extent of the raster were checked to ensure that the data
were suitable for further analysis.
6.2 Study Area Extraction
The Kochi study-area boundary was used to extract the required portion of the temperature
raster. This restricted the analysis to the selected study area and removed areas outside the region
of interest.
6.3 LST Processing and Classification
The temperature raster was processed and classified into five categories based on the observed
temperature range. The classes were defined as:
● Extremely Low Temperature: 24.26–26.247°C
● Low Temperature: 26.248–27.175°C
● Moderate Temperature: 27.176–27.936°C
● High Temperature: 27.937–28.897°C
● Extremely High Temperature: 28.898–32.705°C
This classification makes the continuous temperature data easier to visualize and interpret.
6.4 Attribute Table Generation
After classification, a raster attribute table was generated in ArcGIS Pro. The table contains
information such as the class value, number of pixels, temperature range, temperature zone and
area represented by each class.
6.5 Area Calculation
The number of raster cells belonging to each temperature class was determined using the Count
field. The corresponding spatial area was calculated and recorded in the Total_Area field.
These values were used to compare the spatial distribution of the different thermal zones.
6.6 Thematic Map Preparation
The classified LST raster was symbolized using different colours to distinguish the five
temperature categories. A final map layout was prepared by adding the map title, legend, north
arrow, scale bar and study-area boundary.
6.7 Interpretation
Finally, the classified temperature distribution and area statistics were analyzed to identify
comparatively cooler and warmer regions of Kochi. The spatial pattern was interpreted in the
context of Urban Heat Island analysis and general characteristics of urban and natural surfaces.
7. Data Generation
Data generation refers to the preparation of the usable temperature dataset from the available
spatial data before performing classification and analysis.
For the present study, the main analytical dataset is the Land Surface Temperature raster of
the Kochi study area. The temperature raster contains continuous temperature values
representing the surface thermal condition of different locations.
The data-generation process involved the following stages:
7.1 Preparation of Temperature Data
The available LST raster was imported into ArcGIS Pro and examined for its spatial extent and
temperature values. The raster provides a continuous representation of surface temperature
across the study region.
7.2 Extraction of the Study Area
The Kochi study-area boundary was used to restrict the temperature data to the required
geographical region. This ensures that subsequent analysis is performed only on the selected
study area.
7.3 Temperature Range Identification
The minimum and maximum temperature values represented in the analyzed raster were
identified.
The temperature range obtained in the present analysis is approximately:
24.26°C – 32.705°C
7.4 Classification of Temperature
The continuous temperature values were divided into five classes to make the data easier to
visualize and interpret.
Class Temperature Range Temperature Zone
1 24.26–26.247°C Extremely Low
2 26.248–27.175°C Low
3 27.176–27.936°C Moderate
4 27.937–28.897°C High
5 28.898–32.705°C Extremely High
7.5 Generation of Attribute Information
After classification, the raster attribute table was generated. The table provides the class value,
pixel count, temperature range, temperature zone and calculated area.
This attribute information was subsequently used for quantitative analysis and interpretation.
8. GIS Data Integration
GIS data integration refers to combining spatial datasets and their associated attribute
information within a common GIS environment so that they can be analyzed together.
In this study, ArcGIS Pro was used as the central platform for integrating the temperature raster
with the geographical extent of the study area.
8.1 Integration of Spatial Data
The LST raster represents the temperature information spatially, while the Kochi boundary
defines the geographical extent of the analysis. Combining these datasets allows the temperature
analysis to be restricted to the required study region.
8.2 Raster and Attribute Integration
After classification, each temperature class is associated with attribute information. The raster
cells belonging to the same class are grouped together and their count can be determined.
This allows the spatial information contained in the raster to be converted into quantitative
statistics.
8.3 Map Integration
The classified raster and its attribute information were integrated into the final ArcGIS Pro map
layout. Cartographic elements such as:
● Legend
● North arrow
● Scale bar
● Map title
● Study-area boundary
were added to make the output easier to understand.
8.4 Importance of GIS Integration
GIS integration makes it possible to combine spatial and numerical information in a single
analytical environment. In the present study, this allows the spatial distribution of temperature to
be examined together with the area statistics of individual thermal classes.
9. Data Analysis
The classified Land Surface Temperature raster was analyzed in ArcGIS Pro to determine the
spatial distribution and area occupied by each temperature class. The raster attribute table
provides the pixel count and corresponding area for each class.
Temperature Zone Temperature Range (°C) Pixel Count Area
Extremely Low 24.26–26.247 49,605 44.6445
Low 26.248–27.175 129,582 116.6238
Moderate 27.176–27.936 163,904 147.5136
High 27.937–28.897 104,649 94.1841
Extremely High 28.898–32.705 28,377 25.5393
Table 9.1: Land Surface Temperature Analysis
The analysis indicates that the Moderate Temperature class has the highest pixel count, with
163,904 pixels, and covers approximately 147.51 area units as reported by the ArcGIS attribute
table. The High Temperature class contains 104,649 pixels, while the Low Temperature class
contains 129,582 pixels.
The Extremely Low and Extremely High categories have comparatively fewer pixels. The
highest recorded temperature range extends up to approximately 32.705°C, while the lowest
begins at approximately 24.26°C.
The quantitative analysis therefore shows that the study area contains a combination of cooler,
intermediate and warmer thermal conditions rather than a uniform temperature distribution.
10. Output
The primary output of the analysis is the Land Surface Temperature map of Kochi 2026
generated using ArcGIS Pro.
Figure 10.1: Land Surface Temperature Kochi 2026
The map represents the classified temperature distribution using five thermal zones. A graduated
colour scheme is used to distinguish the temperature classes, with green representing relatively
cooler conditions and red representing relatively warmer conditions.
The final map contains the major cartographic elements required for interpretation, including:
● Map title
● Temperature legend
● North arrow
● Scale bar
● Study-area boundary
● Classified temperature surface
The map provides a visual representation of the spatial variation in Land Surface Temperature
and allows potential warmer and cooler areas to be identified.
OBJECTID Value Count Temp_Range Temp_Zone Total_Area
(°C)
1 1 49,605 24.26 – 26.247 Extremely Low 44.6445
Temperature
2 2 129,582 26.248 – 27.175 Low Temperature 116.6238
3 3 163,904 27.176 – 27.936 Moderate Temperature 147.5136
4 4 104,649 27.937 – 28.897 High Temperature 94.1841
5 5 28,377 28.898 – 32.705 Extremely High 25.5393
Temperature
Figure 10.2: LST Attribute Table
The attribute table provides the numerical information corresponding to the mapped classes,
including the number of raster cells and area associated with each temperature range. Therefore,
the map provides the spatial output, while the attribute table provides the quantitative output
of the analysis.
11. Results and Discussions
11.1 Results
The analysis shows a considerable variation in Land Surface Temperature across the study area,
with values ranging from approximately 24.26°C to 32.705°C.
The Moderate Temperature category represents the largest class, containing 163,904 pixels and
covering approximately 147.51 area units according to the ArcGIS output. This indicates that
intermediate thermal conditions occupy a substantial portion of the analyzed area.
The Low Temperature category is the second-largest class, with 129,582 pixels, while the High
Temperature category contains 104,649 pixels.
The Extremely Low and Extremely High classes occupy comparatively smaller areas. However,
the Extremely High class is particularly important for Urban Heat Island analysis because it
represents the locations with the highest surface-temperature values in the dataset.
11.2 Spatial Interpretation
The map demonstrates that surface temperature is not uniformly distributed across Kochi.
Different portions of the study area exhibit different thermal conditions.
The relatively cooler zones may be associated with surfaces influenced by vegetation, water or
greater moisture availability. Vegetation can reduce surface heating through shading and
evapotranspiration, while water bodies can influence local thermal conditions.
The warmer zones may occur over surfaces with greater heat absorption and storage, including
built-up and impervious surfaces. Concrete, asphalt and other urban materials can store solar
energy and contribute to elevated surface temperatures.
However, the present analysis does not include a separate land-use/land-cover, NDVI or NDBI
analysis. Therefore, these relationships should be treated as interpretations based on
established urban thermal principles, rather than statistically proven relationships from this
dataset.
11.3 Urban Heat Island Interpretation
The areas belonging to the High and Extremely High temperature categories can be considered
potential thermal hotspots. These locations are important for further Urban Heat Island
investigation because persistent high surface temperatures may indicate areas experiencing
greater thermal stress.
The presence of cooler areas alongside warmer zones demonstrates the heterogeneous nature of
the urban thermal environment. This spatial information can be useful for identifying locations
where additional vegetation, green infrastructure or other heat-mitigation measures could
potentially be considered.
11.4 Comparison with Literature
The observed spatial variation is consistent with the general findings of thermal remote-sensing
research. Voogt and Oke (2003) established the importance of thermal remote sensing for
examining urban surface-temperature patterns. Weng (2004) demonstrated the relevance of
vegetation when interpreting LST variation, while Yuan and Bauer (2007) highlighted the role of
impervious surfaces in surface Urban Heat Island studies.
The present study focuses on mapping and classifying Land Surface Temperature in Kochi using
ArcGIS Pro.
11.5 Overall Discussion
The results demonstrate that GIS provides an effective framework for converting temperature
raster data into an understandable thematic representation. Classification makes it easier to
identify the spatial extent of different thermal conditions, while the attribute table provides
quantitative information about each class.
The generated map can therefore serve as a preliminary thermal assessment of Kochi. For a more
detailed UHI investigation, the analysis could be extended by integrating NDVI, NDBI,
land-use/land-cover data and multi-temporal LST datasets. This would allow the relationship
between urban development, vegetation and surface temperature to be investigated quantitatively
rather than only through visual interpretation.