Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Geospatial Assessment of Urban Development on

Land Surface Temperature in Abuja, Nigeria
O. O. Ajayi*, O. M. Adepoju1, G. K. James., Dr. I. Jega1, V. T. Salami1, O.M Aderoju1, O. I. Adedeji1
Strategic Space Applications Department,
National Space Research and Development Agency,
Abuja, Nigeria

Abstract:- This dissertation investigates assessment At present, most of the studies on regional LST are to
development in urban on land surface temperature using study a large scope of urban areas, but there are fewer
geospatial technique with land use land cover and studies on LST changes within the road areas. As an
variation between 1986 and 2016. The aim of this study important indicator of environmental change, most LST
is to examine the effect of urbanization on land surface researchers use it as an important characterization object for
temperature using GIS and Remote Sensing technique. surface energy involvement and environmental change
Satellite images used for this dissertation were Thematic (Marland et al, 2003; Islam and Islam 2013. However, in
Mapper ™ acquired on 1986, Enhance Thematic addition to the use of real-time monitoring and measurement
Mapper plus (ETM+) acquired on 2001 and Operational methods, researchers have made it more convenient to adopt
Land Imager (OLI) acquired on 2016. All satellite data satellite data to study regional LST wherefore, numerous
have 30-meter resolutions, Thematic Mapper and researchers use satellite data for regional LST analysis, not
Enhance Thematic Mapper plus images have spectral only from the inversion algorithms to explore and make
range of 0.45 to 2.35 micro meters (µm) with 8 bands, improvement but also in the selection of source data in
while the Operational Land Imager extends to band 12. different aspects of screening (Srivastava et al, 2010). For
The images were used to produce land use/land cover example, some researchers have summarized and analyzed
map of Abuja Municipal Area Council (AMAC) for several major LST inversion methods and compared their
effective analysis of land surface temperature for three accuracy. However, different parameter selection on the
epochs to know the feature contributes most to surface inversion algorithms will also cause differences in accuracy.
temperature and changes over time. Results of land
use/land cover shown that there is significant increment However, as an important part of the development of
of Built-up from 36.74 per square kilometer to 283.7 per social foundation, attention should be paid to the sustainable
square kilometer between 1986 and 2016, water body development in the continuous development process of road,
from 1.21 to 1.32 per square kilometer and bare surface (Reducing urban heat islands 2008), and in-depth research
from 571.5 to 607.5 per square kilometer. There is also and analysis should be conducted due to its own changes. Its
sharp decrement in vegetation from 714.4 to 452.34 per retrieval from remotely sensed thermal-infrared (TIR) data
square kilometer and rock outcrop from 132.52 to 111.48 provides spatially continuous LST measurements with
between 1986 and 2016. There is little rise in surface global coverage to examine the thermal heterogeneity of the
temperature from 1986 to 2016. Temperature rise from Earth's surface, and the impact on surface temperatures
15 to 26 degree Celsius (0C), built up, contributed most resulting from natural and human-induced changes (Oke,
to surface temperature. 1978Daytime land surface temperature is more tightly
coupled with the radiative and thermodynamic
Keyword:- Remote sensing, land use land cover, surface characteristics of the Earth's surface than standard air
temperature and GIS. temperature measurements. LST is also more sensitive to
changes in vegetation density and captures additional
I. INTRODUCTION information on the biophysical controls on surface
temperature, such as surface roughness and transpiration
Temperature influences human activity in many cooling. However, this study used remote sensing and GIS
waysinclude the survival of man and his social-economic approach to address effect of land surface temperature in
activities such as business, income, agricultural practices, AMAC.
civil servant jobs and reproduction among others activity.
Land surface temperature (LST) is a favorable indicator for  The objectives are
the study of environmental conditions, and the use of LST  Examination of spatial distribution of land surface
indicators as a useful research object in regional energy temperature in Abuja.
change discussion is also increasing. As the infrastructure  Identification of land surface temperature change from
construction of the strip land, the roads will have an impact 1986 to 2016.
on the LST of the area along the roads due to the  Determination of land-use and land cover and their
construction of it and the subsequent occupation due to the contribution to surface temperature in the Abuja.
land occupation and itself. In addition, with the continuous  Determinationof relationship between surface
development of social needs, the mileage of road networks temperatures acquired from satellite imageries and
is increasing, and the LST within the road’s domain will land-use/land-cover impact.
also change due to the influence of the road itself, thus
affecting the surrounding environment (Hegerl et al, 2007).

IJISRT22FEB089 www.ijisrt.com 287

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
II. THE STUDY AREA s/n Data Type Date Spatial Resolution
1 Landsat 4-5 (TM) 1986 30meters
Abuja, the Federal Capital of Nigeria is located at 2 Landsat 7 (ETM) 2001 30meters
8.8230 N and 9.1870 Nof the Equator and 7.240 and 7.560 E 3 Landsat 8 (OLI) 2016 30 meters
in (figure 1). According to the 2006 population census the Table 1: Shows characteristics of Landsat Imagery
Federal Capital Territory (FCT) has total population of
1,402,201 (2006 population census)and land area of 1,769 IV. METHODS
square kilometres (km2).
A. Data Pre-Processing
The satellite imageries were pre-processed in order to
correct the error during scanning, transmission and
recording of the data. The pre-processing steps used were:
 Radiometric correction to remove effects of atmosphere
error ;
 Geometric correction i.e. to correct image distortion by
establishing a projection matches a specific projection
surface or shape ; and
 Noise removal to remove any unwanted noise during
acquisition or transmission process.

B. Image Classification3.2.2
The false Colour Composite was the first and foremost
layer stack; the three bands (4, 3 and 2) were combined for
thematic mapper and Enhance Thematic mapper, while
bands (5, 4 and 3) were combined for Operational Land
Imager. These false colours were classified using the
maximum likelihood classification algorism (pixel
classification process). This was carried out by selecting
sample representative sites of known cover type called
Training Sites or Area. So also, Google Earth map was used
to validate classification where necessary. The computer
Fig. 1: Abuja Study Area Map algorithm then uses the spectral signatures from these
training areas and classifies images into; residential, bare-
 Climate and Weather surface, water-body, vegetation, road-network and rock-
The Abuja municipal Area Council (AMAC), outcrop. All the images for these classifications were
experiences two climate seasons, wet and dry season with a acquired in 1986, 2001 and 2016 respectively.
brief of Harmattan, occasioned by the movement of the
northeast trade wind, with the main features of dust haze, C. Derivation of Land Surface Temperature3.2.3
intensified coldness, and dryness. Rainfall in the AMAC Single-Channel Algorithm (SC)
starts by March and end in October. Humidity in raining The SC algorithm developed by Jiménez-Muñoz et al.
season is high and also temperature around this time is (2014) for the estimation of LST. The mathematical
moderate. The dry season stars by November and end in structure of SC algorithm is:
April, humidity in this period is very low and the Ts= У [1/ԑ (Ѱ1Lsen+Ѱ2) + Ѱ3] +Ϭ………….1
temperature is high due to free cloud cover Balogun Where ԑ is the surface emissivity and (y, Ϭ) two parameters
(2001).The soil in Abuja Municipal Area Council is good given as
for agriculture as it good for cash crop. Y= T2sen/by Lsen; Ϭ = Tsen– T2sen/by………..2
Where Tsen is the at-sensor brightness, temperature, Ѱ1, Ѱ2
III. METHODOLOGY and Ѱ3 are atmospheric function, given as
Ѱ1= 1/t; Ѱ2 = -Ld-Lu/t; Ѱ3 = Ld……………..3
 Data Acquisition and Source
The Landsat data was acquired from the global land- V. RESULTS AND DISCUSSION
cover website at the University of Maryland, USA. The
images are thematic mapper (TM) image acquired on 18th A. Results
December 1986, Enhance Thematic Mapper plus (ETM+) The analyses and results of this work was carried
image acquired on 6th February 2001 and the Operational outfrom satellite images, for three epochs of different years
land Imager (OLI) acquired on 5th of February 2016 as in mapping and monitoring of surface temperature in Abuja
shown in Table 1. The satellite data have 30m spatial Municipal Area Council. Table 2 shown reports of land use
resolutions and the TM and ETM Plus images have spectral and land cover for different decades for which surface
range of 0.45-2.35 micro meter with bands 1,2,3,4,5,6,7 and temperature was measured. Figures 2, 3 and 4 present land
use and land cover maps of the study area in 1986, 2001 and
8 while the Operational Land Imager (OLI) extends to band
12. 2016 of three decade and Figures 5, 6 and 7 present land

IJISRT22FEB089 www.ijisrt.com 288

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
surface temperatures map for 1986, 2001 and 2016
respectively.

Classes 1986 Epoch 2001 Epoch 2016 Epoch

Area KM2 Area KM2 Area KM2
Water Body 1.210 1.140 1.320
Built Up 36.740 98.50 283.70
RockOutcrop 132.520 115.20 111.480
BareSurface 571.50 522.50 607.50
Vegetation 714.40 718.70 452.340
Total 1456.380 1456.320 1456.340
Table 2: Results of land use and land cover of study area for
1986, 2001 and 2016

Fig. 3: AMAC land use land cover 2001

Fig. 2: AMAC land use land cover 1986

Fig. 4: AMAC land use/land cover 2016

IJISRT22FEB089 www.ijisrt.com 289

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165

Fig. 5: AMAC Surface Temperatures 1986

Fig. 7: AMAC Surface Temperature 2016

B. DISCUSSION
The results of this analysis shows that land use land
cover changes were significant during the period of study
from 1986 to 2016. There is significant increase of built-up
area. On the other hand, the vegetation and rock out crop
decrease sharply. Water-body covers about 0.22 km2 of the
total land area of the study area, which is about 15.34%.
Whereas, built-up has about 60.99% spatial extent and
covers about 59.8 square kilometers (km2) in the total land
area of AMAC. Rock outcrop and bare surface contributed
minimally to both spatial extent and total land use of the
Abuja. Vegetation cover has a spatial extent of about 0.22
percent and covers about 1.59 square kilometer of the total
land use of the Abuja. In the years under review, 2001 to
2016, vegetation and water body contributes little to the land
use both in spatial extent and total land mass. The built up
experienced the increased of about 183.14 square kilometer
of the total land mass of Abuja, which is due to high
population growth and urbanization. Rock outcrop also
contributes little to both spatial extent and total land use of
the Abuja, and there is 82.46% increase in bare surface
andthe spatial extent of 13.59 percent.
Fig. 6: AMAC Surface Temperature 2001
Surface Temperature and-Land Use and Land-Cover
Impact 4.3.

The relationship between land-use land-cover and land

surface temperature for this period of investigation showed a
significant rise in temperature from 15.3 0C to 38.7 0C. In
1986 Figure 5, the temperature ranged between 15.3 to 30.4
0C (degrees Celsius). In 2001, temperatures ranged between

IJISRT22FEB089 www.ijisrt.com 290

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
16.3 0C to 38.7 0C Figure 6. The degree rise in temperature moderate temperature range, similar to that of 2001, but
in Figure 6 from 16.3 0C to 38.7 0C is because of the result with a mean temperature of 27.93 0C. The temperature of
of low vegetation cover in this period. The temperature vegetation and water bodies is the lowest, with mean
range in the year 2016 in Figure 7 is between 21.3 0C to temperatures of 27.440C and 22.720C, respectively.
37.4 0C. The features with the highest rise in temperature
are the built up and bare surface areas having an average Table 3 represents an error matrix for classified
temperature of 24.88 0C. This occurs as a result of very low Enhance thematic mapper image of 1986. This table
vegetation cover or no vegetation cover at all in these areas, compared the results of classified images of 1986 with
which can help in preventing higher emission of heat. The referenced high resolution imagery and was used to assess
water body fell within the features, having the least average the degree to which they correspond to individual classes. It
temperature of about 19.84 Celsius, followed by vegetation could be observed that the user’s accuracy for five classes
of its annual mean temperature of 22.46 Celsius and finally, ranged between 90 to 98% with an average accuracy of
rocky area having a mean temperature of 24.14Celsius. 93.8%. Similarly, the producer’s accuracy for four classes
also ranged between 83 to 100% with an average of 94%.
The temperature in 2001 had shown a significant Classification accuracy for 2001 was estimated 89.4%.
increase in degrees Celsius as the maximum temperature Table 4 provides results of classification accuracy for the
ranged from 30.9 0C to 38.7 0C, and then the minimum 2001 classified image of Abuja state using an error
temperature was between 16.2 0C and 25.8 0C in Figure 5. matrix/confusion matrix. This compares the result of actual
The built-up areas recorded the highest maximum classification alongside the referenced data to assess the
temperature of 30.22 0C followed by bare surface with a degree to which they correspond. The user accuracy for the
temperature of 29.20 0C. The least temperature was land use/land cover ranged between 78 to 100% with an
recorded for water body and vegetation cover with their average of 90.6% while the producer’s accuracy ranged
emitting temperature of 25.67 0C and 27.70 0C respectively. between 73 to 100% with an average of 89.4%. The overall
The rock outcrop falls within the region with a moderate accuracy assessment of this classification was estimated to
temperature range (27.6 0C to 28.7 0C) because of be 89.4%.
vegetation and mean temperature is 28.58 0C.
Table 5 shows the result of the Operational Land
The Land Surface temperature of year 2016 recorded a Imager image for 2016 of the study area using an error
maximum temperature of (30.200C and 37.40C), and also matrix: The matrix compared the result of the actual image
minimum temperature of (21.3 0C and 26 0C) as shown in classification with the ground truth reference information
Figure 7. Built-up is highest land use/land cover contributed and assessed the degree to which they correspond. In this
surface temperature of the environment, while vegetation assessment, the user’s accuracy which occurred due to
contributes least to the surface temperature of the study area. commission error was computed for individual land use
The results of the analysis revealed that there were few types in the classification image. The user’s accuracy ranges
changes in temperature between 2001 and 2016.The features from 78% to 100% with an average of 91% while the
having the maximum temperature are built up and bare producer’s accuracy ranges from 67% to 98% with an
surface areas with their mean temperatures of 29.23 0C and average of 89.4%. Overall classification accuracy for the
29.27 0C respectively. Rocky areas are characterized by a 2014 Landsat OLI image was 89.4%.

Class Water body Built up Rock outcrop Bare surface Vegetation Row Total User Accuracy
Water body 37 0 0 0 0 37 93%
Built up 0 46 0 1 0 47 92%
Rock outcrop 3 0 43 1 0 47 90%
Bare surface 0 3 5 44 1 53 96%
Vegetation 0 1 0 0 49 50 98%
Total 40 50 48 46 50 235
Producer 100% 98% 91% 83% 98% Total
Accuracy Accuracy=93%
Table 3: Provides the result of classification accuracy assessment of thematic image of 1986

Class Water body Built up Rock outcrop Bare surface Vegetation Row Total User Accuracy
Water body 39 0 0 0 0 39 100%
Built up 1 35 1 0 0 37 95%
Rock outcrop 9 0 46 0 0 55 84%
Bare surface 0 12 1 49 0 60 78%
Vegetation 0 1 0 1 46 48 96%
Total 49 48 48 48 46 239
Producer Accuracy 80% 73% 96% 98% 100% Total Accuracy=94%
Table 4: Provides the result of classification accuracy assessment of thematic image of 2001

IJISRT22FEB089 www.ijisrt.com 291

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Class Water body Built up Rock outcrop Bare surface Vegetation Row Total User Accuracy
Water body 47 0 0 0 0 47 100%
Built up 3 48 2 1 0 54 89%
Rock outcrop 0 0 33 0 0 33 100%
Bare surface 0 1 4 45 1 51 88%
Vegetation 0 0 10 4 49 63 78%
Total 50 49 49 50 50 248
Producer 94% 98% 67% 90% 98% Total
Accuracy Accuracy=89.4%
Table 5: Provides the result of classification accuracy assessment of operational land imager of 2016

VI. CONCLUSION The authors acknowledge the moral support of

Environmental and Climate change Division in the
According to analysis and findings, it can be concluded Department of Strategic Space Applications, National Space
that built-up and bare surfaces are part of land use and land Research and Development Agency (NASRDA) Abuja.
cover features that contributed more heat energy to the
environment of the study area than other features, most REFERENCES
likely as a result of absorbing little heat energy from the sun
and reflecting high thermal energy to the environment. The [1.] Artis, D.A., and Carnahan, W.H.,Survey of emissivity
built up and bare surface, which contributed maximally to variability in thermography of urban areas. United
the overall surface temperature, may be due to human States: N. p., 1982. Web.Doi: 10.1016/0034-
involvement in the use of fossil fuel in the area. Water 4257(82)90043-8.
bodies and vegetation are the least contributors to thermal [2.] Baleynaud (1999). Observing air quality over the city
energy in the study area, emitting little heat energy of the of Nantes by means ofLandsat thermal infrared data
surrounding is due to their ability to retain heat for longer International Journal of Remote Sensing, Taylor &
time before releasing to the surrounding. In contrast, despite Francis, 1999, 20 (5), pp.947-959
the global warming effect of rising temperature of surface, [3.] Balogun, (2001). The geology of the area is underlain
there are some little benefits of rising surface temperature. by basement complex rocks. The annual rainfall
For example commercial farmers may also use maps of land is highest within the FCC and its environs which is
surface temperature to assess water requirements for their about 1,631.7 mm. The annual mean
crops during the summer, when they are prone to heat stress. temperature ranges between 25.8 and 30.2°C (Balogun,
2001).
However, between the years 2001 to 2016, vegetation [4.] Hegerl G C, Zwiers F W, Braconnot P, Gillett N P,
and water body contributed little to the land-use both in Luo Y, Orsini J A M, Nicholls N, Penner J E and Stott
spatial extent and land area. The builtup experiences P A 2007 Climate Change 2007.
increased of 183.14 km2 of the land area of Abuja, which is [5.] Islam MS, Islam KS (2013) Application of thermal
due to a high rate of population growth and urbanization. infrared remote sensing to explore the relationship
The built-up experiences increased to about 183.14 km2 of between land use-land cover changes and urban heat
the total land mass of the study area, which is due to the Island effect: a case study of Khulna City. J
high rate of population growth, commercialization and Bangladesh Inst Plan 6:49–60
urbanization. [6.] Landsat Project Science Office, 2001 Landsat 7
Science Data User’s Handbook (Goddard Space Flight
Due to human inducement such as road construction, Center) Available online at:
building construction, mining activities, and deforestation, http://Landsathandbook.gsfc.nasa.gov/
vegetation and rock outcrop have experienced a high handbook/handbook_htmls/chapter11/chapter11.html
decrease in total land use and contribute little temperature (accessed 15 September 2010).
rise both in spatial extent and land use of the Abuja. Bare [7.] Ma, Y., Kuang, Y., and Ningsheng, H., 2010.Coupling
surface has increase to 82.46% with a spatial extent of about urbanization analysis for studying urban hermal
13.59% in Abuja because of provision of car park, environment and its interplay with biophysical
vegetation clearance and tarred road. The relationship parameters based on TM/ETM+ imagery.
between land use and land cover and land surface International Journal of Applied Earth Observation and
temperature for the period of investigation shows a Geoinformation, 12, 110-118.
significant rise in temperature from 15 0C to 39.7 0C [8.] Marland G, Pielke RAS, Apps M, Avissar R, Betts RA,
(degrees Celsius). The features with the highest rise in Davis KJ, Frumhoff PC, Jackson ST, Joyce LA,
temperature are the built up and bare surface areas having an Kauppi P, Katzenberger J, MacDicken KG, Neilson
average temperature of 24.9 0C. This occurs as a result of RP, Niles JO, NiyogiDDS, Norby RJ, Pena N,
very low vegetation cover or no vegetation cover at all in Sampson N, Xue Y (2003) The climatic impacts of
these areas, which can help in preventing higher emission of land surface change and carbon management, and the
thermal energy in the study area. implications for climate-change mitigation policy.
ACKNOWLEDGMENT Clim Policy 3:149–157

IJISRT22FEB089 www.ijisrt.com 292

Volume 7, Issue 2, February – 2022 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[9.] NASA, 2010. The Enhanced Thematic Mapper Plus.
Available online at: http://Landsat
.gsfc.nasa.gov/about/etm+.html (accessed 15
December 2010).
[10.] Oke, T. R. 1978: Boundary layer climates. London:
Methuen. xxi+372 pp. £5.75 paper. Published
by METHUEN YOUNG BOOKS (1978)
[11.] [10] Reducing Urban Heat Islands (2008):
Compendium of Strategies (October 2008) describes
the causes and impacts of summertime urban
heat islands and promotes strategies for lowering
temperatures in U.S. communities.
[12.] Srivastava PK, Majumdar TJ, Bhattacharya AK (2010)
Study of land surface temperature and spectral
emissivity using multi-sensor satellite data. J Earth
SystSci 11:67–74
[13.] Stathopoulou, M. and Cartalis, C. 2007. Daytime
urban heat islands from Landsat ETM+ and
Corine land cover data: An application to major
cities in Greece. Solar Energy. 2007, Vol. 81, pp. 358
-368
[14.] Wald, L., and J. M. Baleynaud, (1999). Observing air
quality over the city of Nantes by means of Landsat
thermal infrared data, International Journal of Remote
Sensing,20(5):947-959
[15.] Fatile J.O. (2013). Climate Change and Public Service
deliver in Nigeria. Journal of Business and Social
Science.7(1) 101 -102.
[16.] Ayoade. J. O. (2012). Introduction to Building and
Urban Climatology.Agho—AreoPublishers Ibadan.
[17.] Intergovernmental Panel on Climate Change (IPCC)
Fourth Assessment Report: Climate Change (2007).
Working Group II: Impacts. Adaption and
Vulnerahilil>htlp ://www.ipcc.chipublications_and
datalar4/wg2 ‘en’c h9s9-5- I .html.

IJISRT22FEB089 www.ijisrt.com 293