Estimating Surface Urban Heat Island Effects of Abeokuta Within the Context of Its Economic Development Cluster in Ogun State Nigeria: A Baseline Study Utilising Remote Sensing and Cloud-Based Computing Technologies

Abstract

The demands for growth and prosperity in developing countries have prompted Ogun State to initiate six economic development clusters oriented around its urban areas. However, little attention has been given to the environmental impact of these clusters in relation to temperature change and thermal consequences. Serving as a baseline study for the Abeokuta Cluster, whose implementation is still ongoing, this study analysed the surface urban heat island (SUHI) effects for 2003, 2013, and 2023 to determine whether variations in these effects exist over time. The study utilised satellite imagery from Landsat sensors and the cloud computing power of Google Earth Engine for data collection and analysis. Findings revealed that Abeokuta City experienced varying degrees of high SUHI effects, while the surrounding areas proposed for residential and industrial development in the Abeokuta Cluster showed low SUHI effects. The differences in SUHI effects within Abeokuta City across the years were found to be statistically significant (F_within = 3.158, p = 0.044; F_between = 5.065, p = 0.025), though this was not the case for the Abeokuta cluster as a whole. This study recommends urban planning strategies and policy interventions to combat SUHI effects in Abeokuta City, along with precautionary measures for the Abeokuta Cluster.

1. Introduction

Worldwide, anthropogenic activities driven by urbanisation and industrialisation have negatively impacted urban climates. One significant effect is the persistent high temperature in urban areas, characterised by increased thermal energy and heat accumulations. For some decades now, urban heat island (UHI) effects have been of interest among many urban climate researchers and scientists. Studies [1,2,3,4] have shown that UHI results from the growth and development of cities, driven by increasing demands for housing, transportation, and industry. These demands have led to alterations in natural land cover whereby built-up areas, impervious surfaces, and heat absorbent materials dominate urban landscapes. These changes, combined with emissions from buildings, factories, and automobiles, have made cities hotspots for temperature rise. All these factors contribute to the increasing threats of global warming and climate change, which manifest as more frequent and severe extreme weather events evident in many urban areas [5,6,7,8], compromising their climate resilience and sustainability.

In the Global South, rising temperatures and associated thermal effects, which are particularly pronounced in cities, have been well-documented in the literature and are known to vary across cities based on local climate differences. Deng et al. [9] noted that UHIs are prevalent in developing countries during the summer, particularly in Asian countries like Pakistan and Iraq, whereas north-eastern China’s snow cities experience the most significant changes during autumn. In Africa, seasonal UHI effects are observed, with higher temperatures occurring from late summer to winter, especially in countries within the Greater Horn of Africa, such as Uganda, Ethiopia, South Sudan, and Rwanda [10]. Narrowing down to Nigeria, which has distinct seasonal variations, significant temporal changes in UHI intensity have been documented, showing stronger effects in the dry season and weaker effects in the wet season [11,12]. Studies [12,13,14,15] have also reported that UHI effects in major Nigerian cities and others across the Global South are characterised by poor air quality, thermal discomfort, heat-related illnesses, and deaths. These impacts were further broken down by Tong et al. [16] to include decreased labour productivity, reduced outdoor working capacity, heat strokes, kidney disease, mental illness, and cardiorespiratory mortality and morbidity.

Despite the UHI effects in Nigerian cities, their drivers continue to escalate, particularly due to the rapid pace of urbanisation and industrialisation along with increasing anthropogenic activities [17,18]. For instance, Lagos State was estimated to have a population of over 12 million in 2019 [19], hosting the highest concentration of industries in the country [20,21]. This concentration has influenced neighbouring Ogun State [22], with a population of almost 6 million in 2019 [19]. Recently, the Ogun State Government flagged off six economic development clusters oriented around its urban areas to further boost industrialisation. These clusters include Abeokuta, Ijebu, Remo, Magboro, Atan-Agbara-Ota, and Imeko-Afon-Aworo. According to the official document [23], the clusters are created with the ambition to position Ogun State as the fastest-growing economy in Nigeria, targeting an average annual growth rate of 25 per cent by diversifying sectors such as agriculture, transport, logistics, manufacturing, and construction. However, little attention has been paid to the environmental impacts of these clusters, particularly in terms of temperature change, UHI formation, and its consequences in urban areas that serve as pivots for these clusters. This lack of focus contradicts the global effort to align with Sustainable Development Goals (SDGs), particularly SDG 11 (sustainable cities and communities) and SDG 13 (climate action). In alignment with these goals, this study aims to inform local policies that promote climate adaptation measures, such as mitigating industrial emissions and promoting green infrastructure within Abeokuta Cluster.

Abeokuta Cluster was purposively selected out of the six clusters because it is centred around Abeokuta, the capital and most populous city in Ogun State, serving as an intermediary between industrialised cities like Lagos and Ibadan. This makes Abeokuta particularly vulnerable to urbanisation and industrialisation impacts that could exacerbate UHI effects. Furthermore, the lack of empirical data and tools for understanding UHI effects in Abeokuta and the entire cluster might hinder climate mitigation and adaptation efforts. Using Google Earth Engine and leveraging open remote sensing data on vegetation and land surface temperature (LST) for UHI analysis, this study aims to assess the Surface Urban Heat Island (SUHI) effects [4,24]. The research focuses on addressing the following research questions: (i) What are the changes in LSTs and SUHI effects in Abeokuta Cluster between 2003, 2013, and 2023? (ii) What are the SUHI effects in Abeokuta Cluster during this period? and (iii) What are the SUHI effects within Abeokuta City over the same timeframe? This research serves as a baseline study, providing empirical information on SUHI effects within the Abeokuta Cluster using remote sensing data and analysis. It also combines remote sensing with inferential statistics to establish comparative SUHI effects in Abeokuta City and the wider Abeokuta Cluster, considering urban and regional differences. This information is imperative for understanding and monitoring current thermal environmental conditions, which can aid the planning and management of Abeokuta City and Abeokuta Cluster to better mitigate local urban climate and heat-related consequences.

The literature [24,25,26] clearly defines UHI as a phenomenon whereby urban areas exhibit higher temperatures compared to their rural surroundings, characterised by the energy exchange dynamics between urban land surfaces and the atmosphere [24]. UHI can be categorised into three types: (i) atmospheric urban heat island (AUHI), (ii) surface urban heat island (SUHI), and (iii) subsurface urban heat island (SSUHI), each differentiated by their measurement methods and preferences [27,28,29,30,31]. Streutker [27] explained that AUHI is measured through air temperature data obtained from automobile transects and weather station networks. The in situ data generated have the benefit of a long data record and high temporal resolution, but suffers from poor spatial resolution. On the contrary, SUHI is measured through land surface layer or skin temperature using airborne or satellite remote sensing [27]. The remotely sensed data has the advantage of high spatial resolution and distribution but low temporal resolution with a shorter data record. The SSUHI is measured through underground soil temperature by in situ measurements and satellite sensors, grossing both their advantages and disadvantages [28,30]. Among these types, AUHI and SUHI are the most frequently analysed, often assessed using surface air temperature (SAT) and land surface temperature (LST) indices, respectively.

With a specific focus on SUHI, different patterns and effects have been documented using the LST index. Zhou et al. [24] conducted a global systematic review, finding an exponential rise in LST- and SUHI-based research since 2005, with distinct preferences for criteria such as region, time of day, season, research area, and sensor. Landsat and MODIS satellite sensors accounted for 70 per cent of studies, which indicated that they are mostly used for SUHI research. This might be because these sensors provide quality datasets that are open source, thus readily available for research use without any financial obligations. This insight guided the selection of satellite sensors for this study, which also employed remote sensing techniques.

Several empirical studies are available on SUHI at regional and local scales. Peng et al. [4], for example, addressed SUHI in 31 Chinese cities using remote sensing approaches based on LST data from MODIS, in conjunction with night-time light, meteorological, and socioeconomic data. The study discovered that the range and intensity of SUHI increased during the summer daytime in new urban expansion areas of two-thirds of cities. The study established high increase in SUHI range and intensity in low-latitude areas with high socioeconomic level and vice-versa. Considering the determinants of SUHI, Zhou et al. [1] studied how city size and urban form influence UHI in Europe using data from CORINE urban morphological zone and LST data from MODIS. The study established that the intensity of SUHI increases with size, form and density based on evidence from the largest 5000 cities. Similarly, Xu et al. [32] assessed how urban spatial form influenced SUHI in China. The study employed remote sensing and GIS approaches using LST data from Landsat sensor, OpenStreetMap data, digital elevation model datasets, amongst others. It found that building height and the fraction of permeable surfaces are primary determinants of SUHI. The SUHI is least affected by topography and the industrial area ratio shows a significant warming effect, but the local variable water area ratio shows a clear cooling effect. However, the locations for these studies are characteristically different from the one under investigation which might make their findings not absolutely adoptable.

In Nigeria, several studies have focused on UHI effects, mainly on atmospheric rather than surface UHI. For example, Balogun et al. [33] investigated UHI effects in Akure, finding that night-time UHI intensity is higher during the dry season, while daytime intensity is higher in the wet season. Similarly, using data from field measurements in both urban and rural locations within Akure, Balogun and Balogun [34] investigated the impact of urbanisation on human bioclimatic conditions. The study found increased frequency of high temperatures in the city centre, suggesting a serious risk of heat stress and health risk in urban areas of Akure. Ojeh et al. [35] addressed the temperature differences between rural and urban areas in their assessment of UHI effects. The study also found higher temperature for night-time UHI in dry season. Although these studies attested to the existence of UHI effects in Nigerian cities, they focused only on AUHI effects and not SUHI as intended in this study.

Regarding SUHI effects in Nigeria, Adeyeri et al. [36] analysed the relationship between LST and multiple spectral indices and land cover classes in Abuja, finding that LST was positively correlated with Normalised Difference Built-up Index (NDBI) and Ratio Vegetation Index (RVI), but negatively correlated with Normalised Difference Vegetation Index 705 (NDVI705) and Modified Soil Adjusted Vegetation Index 2 (MSAVI2). While significant cold spots were observed on vegetated surfaces, significant hot spots with high LST were observed in bare surfaces and built-up areas. The study also established that different land use land cover (LULC) classes have different influences on urban heat. Similarly, Ofordu et al. [37] investigated the effects of land LULC on LST, revealing a significant upward trend in temperature influenced by urban expansion. However, neither study addressed citywide and regional differences in SUHI effects nor incorporated location-dependent measures over multiple years, a gap this study seeks to fill. By incorporating location-dependent measures, this research aims to capture spatial and temporal variations across Abeokuta Cluster and Abeokuta City, thus providing insights into persistent heat-risk zones and how land-use changes influence temperature trends, enabling more targeted and effective climate mitigation strategies.

2. Materials and Methods

2.1. Study Area

The study focuses on the Abeokuta Economic Development Cluster (Abeokuta Cluster), located in Ogun State, Nigeria, within the African continent (Figure 1a,b). It cuts across 5 out of the 20 local government areas (LGAs) in the State. These five LGAs are Abeokuta North, Abeokuta South, Remo North, Obafemi-Owode, Odeda and Ewekoro (Figure 1c). The whole of Abeokuta South LGA is covered by the cluster, while some portions of the other LGAs are covered by it. Although the cluster has an imaginary boundary, we theoretically estimated the total land area to be 25,047.48 square Kilometres in ArcGIS Desktop 10.3. The major urban area within the cluster is Abeokuta and other small towns include Kobape, Kajola, Odeda and Wasinmi (Figure 1d). Abeokuta is the Capital of Ogun State and also a secondary city. The metropolitan Abeokuta is estimated to have 544,000 population in 2021, which implies a growth of 2.06% over 2020 [38]. Situated within Ogun State, Abeokuta Cluster possesses the climate and vegetation attributes of the South-West region of Nigeria. The cluster is therefore located in the tropical rainforest and tropical climate region. This region has two distinct seasons, each lasting around 130 days: the rainy season and the dry season. The region also experiences approximately 28 °C, 1270 mm and 1100 mm of annual mean temperature, precipitation and evaporation, respectively.

2.2. Research Design

This research employed a design that incorporated contemporary approaches and techniques in the geospatial field. This comprised satellite remote sensing, cloud-based computing and geographic information system (GIS). This design is important for analysing SUHI effects in order to leverage open source high-resolution remotely sensed data and cloud-based computing technologies. The design was selected based on its advantage of leveraging open remote sensing data to provide large-scale and timely data collection and spatial analysis in providing empirical-based evidence that aid decision-making on environmental issues. The data for this study were collected and analysed in Google Earth Engine (GEE). Quantitative data from the results were migrated to Statistical Package for Social Sciences (SPSS) for statistical analysis. The framework for the methods used is provided in Figure 2. It provides the graphical descriptions of the data acquisition and analytical methods utilised in this study.

2.3. Data Collection

Satellite images from Landsat sensors were used for this study. These images are readily available in the catalogue of Google Earth Engine (GEE). The service providers are the United States Geological Survey and the National Aeronautics and Space Administration (USGS-NASA). We acquired the Level 2, Collection 2 and Tier 1 surface reflectance products from the GEE catalogue. The products were used because they have been corrected for atmospheric, radiometric and geometrical errors and are currently made available and recommended by the service provider. The data were imported into GEE as image collections. The descriptions of the product for each year are provided in Table 1. The study dates are 2003, 2013 and 2023. These were chosen because they fall within the Millennium, marking Africa’s urbanisation surge since 2000, and the initial date of 2003 marks a politically significant year in Ogun State for settlement development [39]. Additionally, a ten-year interval was selected to align with the availability of decadal land use and cover change data. The shapefiles of Ogun State’s boundary were obtained from this website (https://www.geoboundaries.org/simplifiedDownloads.html (accessed on 15 January 2024)). This data is trusted and used because it is an open license resource from a database that is validated based on the ISO 3166-1 [40] alpha-3 encoding with each boundary having a globally unique ID. The data was used to prepare the boundary shapefiles for Abeokuta Cluster in ArcGIS Desktop 10.3 (ArcMap). The shapefiles were imported into GEE as asset files and used as the area of interest.

2.4. Data Analysis

2.4.1. Data Pre-Processing

The Landsat images acquired have been already pre-processed by the service providers. Hence, they are radiometrically, geometrically and atmospherically corrected surface reflectance images. However, due to typical artefacts in the image collections, they required further pre-processing. For 2003, the filtered dates ranged from 31 December 2002 to 31 December 2003. The images for nearest month of the preceding year (1–31 December 2002) were added to backfill for scan line corrector (SLC) errors identifiable with Landsat 7 from 31 October 2003. The image collection was masked for cloud cover and cloudshadow using the cloudsBitMask and cloudShadowBitMask functions, respectively. The image collection was mapped to minimise the cloud computing power for processing large datasets. The results were reduced to a median image for 2003. This image was later clipped to the area of interest which is the boundary of the study area. The same procedure was followed for the pre-processing of the image collections for 2013 and 2023. The exception is that the filtered dates for these image collections were from 1 January to 31 December. The reason being that there is no data gap or loss of data to SLC errors during these years.

2.4.2. Data Processing

The median images resulting from the pre-processed image collections for the years are further analysed to estimate the SUHI effects. There are different phases of the analysis. The first is to compute the normalised difference vegetation index, land surface emissivity and land surface temperature. The land surface temperature is further used to compute the surface urban heat island and the urban thermal field variance index. The Normalised Difference Vegetation Index (NDVI) is used for quantifying vegetation and has been applied in analysis of forest change, deforestation and urbanisation, amongst others [43,44]. The NDVI is computed as the difference between near-infrared band and red band as provided in Equation (1) and the NDVI values range from −1 to +1. Using the equation, we derived the NDVI for each year. For 2003, the band 4 (ST_B4) which is the near infrared channel was subtracted from band 3 (ST_B3) which is the red channel and the result divided by the addition of the two bands. For 2013 and 2023, the band 5 (ST_B5) which is the near infrared channel was subtracted from band 4 (ST_B4) which is the red channel and the result divided by the addition of the two bands.

(1)$NDVI={\displaystyle \frac{NIR-RED}{NIR+RED}}$

where:

• NDVI = Normalised Difference Vegetation Index;

• NIR = near infrared channel;

• RED = red channel.

Emissivity is defined as the “ratio of the actual emitted radiance to the radiance emitted from a black body at the same thermodynamic temperature” [45]. The land surface emissivity (ε) therefore describes the thermal radiation that is produced by a thin layer of earth’s surface and its values range from 0 to 1. For satellite images, spectral emissivity is required to correct for at-satellite brightness temperature referred to as black body temperature in order to calculate the land surface temperature, while the emissivity correction is mostly determined by the land use type and being estimated using NDVI per pixels values [46]. The formulae for computing the land surface emissivity are provided in Equations (2) and (3). Based on the above equations, we first of all computed the minimum and maximum NDVI for 2003, 2013 and 2023. The values for each year were used to calculate the proportion of vegetation for that year which was then used to calculate the land surface emissivity for the same year.

(2)$\mathsf{\epsilon }=0.004Pv+0.986$

where:

• ε = Emissivity;

• Pv = Proportion of Vegetation.

(3)$Pv=\left({\displaystyle \frac{NDVI\ast {NDVI}_{min}}{{NDVI}_{max}-{NDVI}_{min}}}\right)$

where:

• NDVI = Normalised Difference Vegetation Index;

• NDVI_min = Minimum value of NDVI;

• NDVI_max = Maximum value of NDVI.

The Land Surface Temperature (LST) temperature of a thin layer that exists at the interface between the atmosphere and surface elements such as vegetation, soil and bare lands [47]. It is calculated using the formula provided in Equation (4). Based on this equation, we computed the LST for each year using the thermal bands available for 2003, 2013 and 2023. The thermal band for 2003 is the band 6 (ST_B6) while for 2013 and 2023 is band 10 (ST_B10).

(4)$T_S={\displaystyle \frac{T_B}{1+\left(\frac{\lambda \ast T_n}{\rho }\right)\ast \ln \mathsf{\epsilon }}}-273.15$

where:

• T_S = Land Surface Temperature [LST] (°C);

• T_B = at-satellite brightness temperature (K);

• λ = Wavelength of emitted radiance (11.5 μm);

• ρ = 1.438 × 10⁻² mK;

• ε = Emissivity (unitless).

The Surface Urban Heat Island (SUHI) is empirically determined by comparing the land temperatures in towns and cities with their surrounding areas. According to Waleed and Sajjad [46], a normalised method is used in computing the SUHI based on Equation (5). We computed the SUHI index by substituting the required parameters using GEE reducers.

(5)${UHI}_N={\displaystyle \frac{T_s-T_m}{T_{Std}}}$

where:

• UHI_N = normalised SUHI;

• T_S = Land Surface Temperature;

• T_m = Mean of Land Surface Temperature;

• T_Std = Standard Deviation of Land Surface Temperature.

The SUHI effects are widely estimated using the Urban Thermal Field Variance Index (UTFVI) [48,49,50]. The UTFVI quantitatively measured the thermal comfort level of any urban area. It is computed based on Equation (6). We estimated the UTFVI by substituting the required parameters using GEE reducers. These ranges and interpretations of UTFVI (<0.000: none; 0.000–0.005: weak; 0.005 ≤ UTFVI < 0.010: middle; 0.010 ≤ UTFVI < 0.015: strong; 0.015 ≤ UTFVI < 0.020: stronger; UTFVI ≥ 0.020: strongest) are provided by Moisa and Gemeda [51] in understanding the SUHI effects. In reference to these ranges, the urban thermal comfort level (UTCL) is measured as excellent, good, normal, bad, worse and worst, respectively.

(6)$UTFVI={\displaystyle \frac{T_s-T_m}{T_s}}$

where:

• UHI_N = normalised SUHI;

• T_S = Land Surface Temperature;

• T_m = Mean of Land Surface Temperature.

2.4.3. Hypothesis Testing

We put forward two hypotheses for this study which are restated below in their null (H₀) and alternate (H₁) formats. The SUHI effects are derived using the UTFVI, hence the null and alternate hypotheses are reformulated based on the latter.

• H₀1: There is no statistically significant difference in UTFVI in Abeokuta Cluster between 2003, 2013 and 2023.

• H₁1: There is a statistically significant difference in UTFVI in Abeokuta Cluster between 2003, 2013 and 2023.

• H₀2: There is no statistically significant difference in UTFVI in Abeokuta City between 2003, 2013 and 2023.

• H₁2: There is a statistically significant difference in UTFVI in Abeokuta City between 2003, 2013 and 2023.

The hypotheses are tested using the Repeated Measures Analysis of Variance (ANOVA). This statistic is used in analysing whether a statistically significant difference or variation exists between the means of three or more groups in which the same participants or subjects are involved in each group [52]. The assumptions for this type of test are contained in the literature [53]. Of utmost importance is the assumption of sphericity which is verified using the Mauchly’s W test. When this assumption is violated, the Huynh–Feldt correction is applied, if the estimate is greater than 7.5 (Epsilon > 7.5), otherwise the Greenhouse–Geisser is applied (Epsilon < 7.5).

The Repeated Measures ANOVA is calculated as provided in Equation (7). In performing the ANOVA analyses, we used constant random seeding (seed = 0) to extracted 1000 sampled points as location dependent measures of UTFVI for 2003, 2013 and 2023 in Abeokuta Cluster and Abeokuta City. The datasets were inputted into SPSS 20 for data wrangling and statistical analysis.

(7)$F={\displaystyle \frac{{MS}_{Model}}{{MS}_{Error}}}$

where:

• F = F-statistic;

• MS_Model = Mean sum of squares between groups;

• MS_Error = Mean sum of squares within groups.

3. Results

3.1. LSTs and SUHI Effects in Abeokuta Cluster

The minimum and maximum values of NDVI, emissivity and LST for 2003, 2013 and 2023 for Abeokuta Cluster are presented in Figure 3. In 2003, we found that the minimum NDVI is −0.310 which suggests areas with little or no vegetation and the maximum NDVI is 0.838 which indicates areas with dense vegetation. By 2013, this range had widened to −0.834 and 0.999 which points to areas with significantly low and high vegetation. This trend continued in 2023 as the NDVI values ranged from −0.583 to 0.976, thereby indicating some recovery in vegetation cover but still showing signs of decline in certain areas. We also found that the minimum emissivity is −0.986 and the maximum emissivity is 0.990 for 2003, 2013 and 2023. These values remained stable across the three years as they consistently fell between this range. This implies that the thermal properties of the land surface did not change significantly. This might suggest that although the vegetation cover changed, the underlying soil and built-up surfaces remained mostly unchanged, thereby causing slight changes in the land surface. On considering the LST for 2003, the minimum is 15.380 °C and the maximum is 58.590 °C. By 2013, this range shifted to 19.635 °C and 52.957 °C which indicates a decrease in maximum temperature but a slight increase in minimum temperature. In 2023, the range narrowed further to 21.150 °C and 52.822 °C which suggests an increase in baseline temperatures in Abeokuta Cluster.

For the estimation of the SUHI, the minimum and maximum values are obtained (Figure 4 and Figure 5). In 2003, SUHI values ranged from −7.045 to 4.740 signifying cooler and hotter locations within the cluster that correspond to vegetation and urban areas. By 2013, the range expanded from −15.685 to 5.020, therefore, showing more extreme temperature differences between urban areas and its surroundings that comprise vegetation. In 2023, the SUHI values ranging from −4.961 to 5.930 showed continued urban heat effects, although with some reduction in the range of temperature differences. The UTFVI values also show significant fluctuations over the years. We found that the range of UTFVI values from−1.339 to 0.386 in 2003 indicates moderate stress due to SUHI. By 2013, the range from −6.573 to 3.108 was much broader and suggests higher heat stress that is possibly linked to increased temperatures in the cluster. In 2023, the UTFVI values from −0.682 to 0.326 narrowed in range which suggests some reduction in heat stress in the cluster. In categorising the effects using a geometric interval which minimised the sum of squares of the values, the locations with low, medium and high classes of UTFVI values were established for all the years.

3.2. Hypothesis Testing: SUHI Effects in Abeokuta Cluster

In this study, two hypotheses are tested using the Repeated Measures ANOVA test. The results of the first hypothesis are presented as descriptive and inferential summaries (Table 2, Table 3, Table 4 and Table 5). Based on the 1000 sampled points, the minimum UTFVI for 2003, 2013 and 2023 in Abeokuta Cluster are −0.2309, −0.7446 and −0.4160, respectively, while the maximum UTFVI are 0.2426, 0.2508 and 0.2639, respectively (Table 2). The means and standard deviations of the UTFVI for Abeokuta Cluster are −0.086 ± 0.776, −0.085 ± 0.970, and 0.064 ± 0.8173 for 2003, 2013 and 2023 (Table 2). To understand the mean difference for each year, the inferential results indicate whether SUHI effects in Abeokuta Cluster differ between the years being studied. The Mauchly’s test (Mauchly’s W = 0.985, p = 0.001) indicates that there is a statistically significant difference between the variances of the years being studied, which violates the condition for sphericity (Table 3). In correcting for this violation, we utilised the Huynh–Feldt correction because the estimate of sphericity is greater than 7.5 (Epsilon = 0.987) (Table 3). We found that there is no statistically significant difference in the within-subjects’ effects of UTFVI (F_wth = 0.297, p = 0.740) between 2003, 2013 and 2023 in Abeokuta Cluster (Table 4). However, there is a statistically significant difference in between-subjects’ effects of UTFVI (F_btw = 12.846, p = 0.000) across the years in Abeokuta Cluster (Table 5). These results indicate SUHI effects of the same locations in Abeokuta Cluster in the different years are not significantly different but those at different points across the years are significantly different.

3.3. Hypothesis Testing: SUHI Effects in Abeokuta City

The results of the second hypothesis are also presented as descriptive and inferential summaries (Table 6, Table 7, Table 8 and Table 9). Based on the 1000 sampled points, the minimum UTFVI for Abeokuta City in 2003, 2013 and 2023 are −0.3390, −0.5452 and −0.3443, respectively, while the maximum UTFVI are 0.2189, 0.2263 and 0.2708, respectively (Table 6). The means and standard deviations of the UTFVI for Abeokuta City are −0.0036±0.0860, −0.0091±0.9552, and 0.0035±0.8530 for 2003, 2013 and 2023 (Table 6). The mean difference also indicates whether SUHI effects differ in Abeokuta City between the years. The Mauchly’s test (Mauchly’s W = 0.973, p = 0.000) also violated the condition for sphericity, thus indicating that there is a statistically significant difference between the variances of the years being studied (Table 7). The Huynh–Feldt correction also showed the estimate of sphericity is greater than 7.5 (Epsilon = 0.976) (Table 7). For the within-subjects’ effects of UTFVI (F_wth = 3.158, p = 0.044) in Abeokuta City, a statistically significant difference in UTFVI was established between 2003, 2013 and 2023 (Table 8). Similarly, there is a statistically significant difference in between-subjects’ effects of UTFVI (F_btw = 5.065, p = 0.025) across the years in Abeokuta City (Table 9). These results imply that the SUHI effects of same locations in Abeokuta City in the different years and those at different locations across the years are significantly different.

4. Discussion

Developing countries like Nigeria are experiencing a rapid pace of urbanisation and an increasing number of urban areas. As these towns and cities evolve, evidence of changes in land use, land cover, and ecosystems has emerged. Obateru et al. [54] documented that a number of studies have assessed land use and land cover changes at subnational and national levels, highlighting the rapid growth of urban areas at the expense of natural landscapes. This transformation process in Nigerian cities is not limited to its current state but will continue into the future, as indicated by projections estimating a population of 295 million by 2050 [55] and other urban growth statistics [56]. With the current initiative of Abeokuta Cluster that primarily focuses on city and regional development based on social and economic functions in Ogun State, Nigeria, similar historical land transformation is expected within the delineated clusters with associated consequences if not properly implemented. Most especially, the temperature change and thermal effects within the new Abeokuta Cluster which have not been quantified and therefore remain unknown. Serving as a baseline study for a cluster still in its implementation phase, we have provided information on SUHI in Abeokuta Cluster and Abeokuta City over a period of 20 years at ten-year intervals.

First, our findings show evidence of SUHI in Abeokuta Cluster for the years 2003, 2013, and 2023. This is indicated by higher SUHI values in the major city, Abeokuta, compared to the surrounding hinterlands. These hinterlands include Kobape, Olorunda, Odeda, Kajola and Wasinmi. Looking at the variation over time, the formation of SUHI is higher in 2023 than 2013 as well as in 2013 than in 2003. This implies that there is increasing formation of SUHI over time. These findings show a rising trend as posited in a number of existing studies [1,4] in other regions that are experiencing extreme temperatures. With the plethora of information linking temperature to SUHI formation [3,24,57], the implication is that there is higher temperature in Abeokuta City than its rural areas which makes the former warmer than the latter. Considering these findings from regional planning perspective, it is pertinent to understand that the hinterlands such as Wasinmi and Kobape are the areas targeted by the Abeokuta Cluster initiative for residential and industrial developments. The intention is for the developments to initiate population and economic growth in these locations. This implies that human activities associated with the growth and development will likely make the hinterlands urbanised through deforestation and increased built-up areas. This might result to formation of SUHI in these locations with devastating effects. In addition, urbanisation in Abeokuta city is driven by its administrative status as the capital of Ogun State, and real estate development that is mainly residential for its increasing urban population [58,59]. As also stated for other parts of Ogun State [60], the regional development plan has been less effective to curtail the effects of urbanisation in Abeokuta City. As such, the city’s dense population is concentrated around the traditional city centre as well as other locations around the old and new administrative centres. These are the locations that might have reflected the high temperatures and effects of SUHI within Abeokuta City and Abeokuta Cluster.

We further considered the SUHI effects in Abeokuta City and Abeokuta Cluster differently. The differences in SUHI effects of same locations and at different locations within Abeokuta City across the years are established to be statistically significant. The finding with significant difference shows the thermal effects are more evident within the city limit. This implies a localised heat intensity within the cityscape of Abeokuta which is typical of urban areas. Such heat intensity might be due to high heat retention of materials used in building and road construction, large extent of bare grounds and less of vegetations and cooling areas such as greeneries and water bodies [36,61,62]. The non-significant difference at the cluster level suggests that when the city is expanded to include the hinterlands, the thermal effects are weakened by the characteristics of the sub-urban and rural areas. This confirms the findings of Yang et al. [63] that an extensive rural land with natural vegetation surrounding the cityscape have the capacity to mitigate urban heat island effects in cities. It is therefore evident that the major urban area which is Abeokuta City experienced varying high degree of temperature and SUHI effects while the surrounding areas proposed as residential and industrial development points for Abeokuta Cluster have low effects.

The need to manage climate impact like UHI effects in cities amongst other challenges has prompted global efforts geared towards achieving the SDGs. In light of the ongoing Abeokuta EDC and other clusters in Ogun State, there is a need to consider its industrialisation in line with the 9th SDG (industry, innovation and infrastructure). All stakeholders have to build resilient infrastructure, promote sustainable industrialisation and foster innovations. In line with environmental sustainability, the goals to build sustainable cities and communities and also ensure climate action is particularly emphasised by the 11th and 13th SDGs. In achieving these goals, the Ogun State Government must balance their social and economic plans with local environmental plans as well as aligning with global environment goals spelt out in different climate summits of the Conference of the Parties (COP), especially the recent COP28 [64]. Much emphasis has been laid on achieving zero carbon and net zero in an urgent need to significantly reduce GHGs emissions to align with the 1.5 °C climate target. Hence, it is expected that the Ogun State Government emphasise that developers, investors and other stakeholders adopt latest sustainable building materials, low-carbon industrial technologies, clean energy sources relating to renewables and more sustainable innovations of pollution abatement and removal. These are critical pathways to positive climate outcomes in achieving the 2030 agenda [65,66].

The study has its limitations. One, it is based on decadal measurements because of non-availability of Landsat images which are open-source remote sensing datasets for certain dates in the region. Nevertheless, the ten-year time gap is one of the standard intervals for assessing environmental impact. Two, we were unable to consider seasonality by considering the regions’ dry season (December-March), because the Landsat image collection for the season were not available for some of the years considered. Therefore, the synthetic median images used were the best available options for this study. Lastly, we did not account for spatial dependencies in the hypothesis testing because the interest was not in spatial clustering analysis of sampled locations within each year. These limitations provide some implications for further studies. Future research could therefore focus on understanding the annual long term temperature change and SUHI effects in the region through a time series analysis. Satellite images with high temporal resolution from MODIS or those from commercial data providers could be accessed for the dates. Mixed methods research on SUHI and self-reported health could also be conducted for urban dwellers to understand their perception of SUHI effects in the cluster and region at large.

5. Conclusions

The demands for growth and prosperity in developing countries have prompted Ogun State to initiate six economic development clusters focused around its urban areas. However, limited attention has been given to the environmental impact of these clusters, particularly regarding temperature changes and their thermal consequences. The formation of SUHI and its effects in cities is a multifaceted issue that has received global attention. Amid numerous SUHI studies, our local context-based research provides empirical evidence of SUHI effects in Abeokuta Cluster and Abeokuta City for the years 2003, 2013, and 2023. Addressing the first research question, we observed a rising trend in minimum land surface temperatures and increasing variability in urban thermal conditions between 2003 and 2023. In response to the second research question, we found that the urban thermal effects at the same locations within Abeokuta Cluster over different years are not significantly different; however, these effects do vary significantly across different points in time. Lastly, for the third research question, we found significant differences in urban thermal effects both for the same locations in Abeokuta City across the different years and for different locations across these years.

Our study highlights that temperatures in Abeokuta city are higher than its surrounding hinterlands, with increasing SUHI effects observed over the years. From these findings, we infer that efforts to mitigate SUHI effects should adopt a city-wide approach, with a specific focus on Abeokuta, rather than employing a broader regional approach. However, care should still be taken in the hinterlands, given the ongoing regional development in the cluster. We therefore recommend urban planning strategies and policy interventions to combat SUHI effects in Abeokuta City and to maintain the current conditions of Abeokuta Cluster. First, we suggest that climate adaptation and mitigation techniques such as the development of green-blue infrastructure should be implemented in Abeokuta City. As the core of the economic cluster, the city’s high temperature could be minimised through parks and green spaces, green roofs and walls, and artificial lakes. Second, in areas earmarked for development in the surrounding region, forest conservation should be prioritised. The existing forest areas should be integrated with new developments. Sustainable buildings and structures incorporating green areas, reflective pavements, roofing, and energy-efficient materials and amenities should also be provided. Finally, road infrastructure and other hard landscapes should be constructed using light-coloured reflective pavements and heat-resistant materials to minimize heat absorption by buildings and paved surfaces, helping to keep the overall environment cooler.

Author Contributions

Conceptualization, O.M.O. and A.R.; methodology, O.M.O. and A.R.; software, O.M.O.; validation, O.M.O.; formal analysis, O.M.O.; investigation, O.M.O.; resources, A.R.; data curation, O.M.O.; writing—original draft preparation, O.M.O.; writing—review and editing, A.R.; visualization, O.M.O.; supervision, A.R.; project administration, A.R.; funding acquisition, O.M.O. and A.R. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Data are from secondary sources and are freely available in the public repositories through this link: https://developers.google.com/earth-engine/datasets/catalog/landsat (accessed on 22 May 2024).

Acknowledgments

The authors extend their gratitude to the Alexander von Humboldt Foundation for the 2023 Postdoctoral Fellowship awarded to Oluwafemi Michael Odunsi (Grant Number: 3.4-1230757-NGA-HFST-P). We appreciate Ruhr University Bochum, Germany for hosting the researcher, and the Institute of Geography for providing the enabling environment for the study to be conducted.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Map of the Study Area. (a) Nigeria within the context of Africa (b) Ogun State within the context of Nigeria (c) Abeokuta cluster within the context of Ogun State (d) Abeokuta and its surrounding areas within the context of Abeokuta Cluster and its Local Government Areas (LGAs).

Figure 2. Framework for research methods (Keys: NDVI—Normalised Difference Vegetation Index, EM—Emissivity, LST—Land Surface Temperature, SUHI—Urban Heat Island, UTFVI—Urban Thermal Field Variance Index).

Figure 3. Annual Composite of (a) Normalised Difference Vegetation Index (NDVI), (b) Land Surface Emissivity (Emissivity), and (c) Land Surface Temperature (LST) for Abeokuta Cluster in 2003, 2013 and 2023.

Figure 4. Annual Composite of (a) Surface Urban Heat Island (SUHI) Index, (b) quantitative Urban Thermal Field Variance Index (UTFVI), and (c) categorised Urban Thermal Field Variance Index (UTFVI) for Abeokuta Cluster in 2003, 2013 and 2023.

Figure 5. Urban Thermal Field Variance Index (UTFVI) Profiles for Abeokuta City in 2003, 2013 and 2023.

Table 1

Description of remote sensing data.

Serial Number	Year	Satellite Sensor	Product Identifier	Number of Bands
1.	2003	Landsat 7 Enhanced Thematic Mapper Plus (ETM+)	LANDSAT/LE07/C02/T1_L2	8
2.	2013	Landsat 8 Operational Land Imager/Thermal Infrared Sensor (OLI/TIRS)	LANDSAT/LC08/C02/T1_L2	11
3.	2023	Landsat 9 Operational Land Imager/Thermal Infrared Sensor 2 (OLI-2/TIRS-2)	LANDSAT/LC09/C02/T1_L2	11

Source: USGS [41,42]

Table 2

Descriptive statistics of UTFVI for Abeokuta Cluster.

	N	Minimum	Maximum	Mean	Std. Deviation
2003_UTFVI	1000	−0.2309	0.2426	−0.008643	0.0776285
2013_UTFV	1000	−0.7446	0.2508	−0.008455	0.0970450
2023_UTFV	1000	−0.4160	0.2639	−0.006743	0.0817256
Valid N (listwise)	1000

Table 3

Mauchly’s Test of Sphericity ^a of UTFVI Measure for Abeokuta Cluster.

Within Subjects Effect	Mauchly’s W	Approx. Chi-Square	df	Sig.	Epsilon b
					Greenhouse-Geisser	Huynh-Feldt	Lower-Bound
Year	0.985	14.963	2	0.001	0.985	0.987	0.500

Tests the null hypothesis that the error covariance matrix of the orthonormalised transformed dependent variables is proportional to an identity matrix. ^a Design: Intercept; Within Subjects Design: Year; ^b May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table.

Table 4

Tests of Within-Subjects Effects of UTFVI Measure for Abeokuta Cluster.

Source		Type III Sum of Squares	df	Mean Square	F	Sig.
Year	Sphericity Assumed	0.002	2	0.001	0.297	0.743
	Greenhouse-Geisser	0.002	1.971	0.001	0.297	0.740
	Huynh-Feldt	0.002	1.975	0.001	0.297	0.740
	Lower-bound	0.002	1.000	0.002	0.297	0.586
Error (Year)	Sphericity Assumed	7.366	1998	0.004
	Greenhouse-Geisser	7.366	1968.703	0.004
	Huynh-Feldt	7.366	1972.565	0.004
	Lower-bound	7.366	999.000	0.007

Table 5

Tests of Between-Subjects Effects for Abeokuta Cluster.

Source	Type III Sum of Squares	df	Mean Square	F	Sig.
Intercept	0.189	1	0.189	12.846	0.000
Error	14.735	999	0.015

Table 6

Descriptive statistics of UTFVI for Abeokuta City.

	N	Minimum	Maximum	Mean	Std. Deviation
2003 UTFVI	1000	−0.33896	0.21892	−0.0036184	0.08604748
2013 UTFVI	1000	−0.54522	0.22625	−0.0090921	0.09521672
2023 UTFVI	1000	−0.34434	0.27077	−0.0034868	0.08529961
Valid N (listwise)	1000

Table 7

Mauchly’s Test of Sphericity ^a of UTFVI Measure for Abeokuta City.

Within Subjects Effect	Mauchly’s W	Approx. Chi-Square	df	Sig.	Epsilon b
					Greenhouse-Geisser	Huynh-Feldt	Lower-Bound
Year	0.973	27.222	2	0.000	0.974	0.976	0.500

Tests the null hypothesis that the error covariance matrix of the orthonormalised transformed dependent variables is proportional to an identity matrix. ^a. Design: Intercept; Within Subjects Design: Year; ^b. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table.

Table 8

Tests of Within-Subjects Effects of UTFVI Measure for Abeokuta City.

Source		Type III Sum of Squares	df	Mean Square	F	Sig.
Year	Sphericity Assumed	0.020	2	0.010	3.158	0.043
	Greenhouse–Geisser	0.020	1.948	0.011	3.158	0.044
	Huynh–Feldt	0.020	1.951	0.010	3.158	0.044
	Lower-bound	0.020	1.000	0.020	3.158	0.076
Error (Year)	Sphericity Assumed	6.474	1998	0.003
	Greenhouse–Geisser	6.474	1945.647	0.003
	Huynh–Feldt	6.474	1949.395	0.003
	Lower-bound	6.474	999.000	0.006

Table 9

Tests of Between-Subjects Effects for Abeokuta City.

Source	Type III Sum of Squares	df	Mean Square	F	Sig.
Intercept	0.087	1	0.087	5.065	0.025
Error	17.249	999	0.017

References

1. Zhou, B.; Rybski, D.; Kropp, J.P. The role of city size and urban form in the surface urban heat island. Sci. Rep.; 2017; 7, 4791. [DOI: https://dx.doi.org/10.1038/s41598-017-04242-2] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28684850]

2. Ng, E.; Ren, C. China’s adaptation to climate & urban climatic changes: A critical review. Urban Clim.; 2017; 23, pp. 352-372. [DOI: https://dx.doi.org/10.1016/j.uclim.2017.07.006] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32363139]

3. John-Nwagwu, H.O.; Samuel, S.K.; Johnson, N.G. Analysis of the Effect of Population Growth on Land Surface Temperature in Lokoja Metropolis. Coou Afr. J. Environ. Res.; 2024; 5, pp. 240-258.

4. Peng, J.; Qiao, R.; Wang, Q.; Yu, S.; Dong, J.; Yang, Z. Diversified evolutionary patterns of surface urban heat island in new expansion areas of 31 Chinese cities. npj Urban Sustain.; 2024; 4, pp. 1-11. [DOI: https://dx.doi.org/10.1038/s42949-024-00152-1]

5. Depietri, Y.; Welle, T.; Renaud, F.G. Social vulnerability assessment of the Cologne urban area (Germany) to heat waves: Links to ecosystem services. Int. J. Disaster Risk Reduct.; 2013; 6, pp. 98-117. [DOI: https://dx.doi.org/10.1016/j.ijdrr.2013.10.001]

6. Fares, A.; Habibi, H.; Awal, R. Chapter 1-Extreme events and climate change: A multidisciplinary approach. Climate Change and Extreme Events; Elsevier: Amsterdam, The Netherlands, 2021; pp. 1-7. Available online: https://www.sciencedirect.com/science/article/abs/pii/B9780128227008000196?via%3Dihub (accessed on 14 July 2024).

7. Kashi, S.M.H.; Farrokhzadeh, S.; Baharvandi, S.; Zolfani, S.H. Effects of extreme weather events and climate change on cities’ livability. Cities; 2024; 151, 105114. [DOI: https://dx.doi.org/10.1016/j.cities.2024.105114]

8. Yang, X.; Wang, Z.-H.; Wang, C.; Lai, Y.-C. Megacities are causal pacemakers of extreme heatwaves. npj Urban Sustain.; 2024; 4, pp. 1-10. [DOI: https://dx.doi.org/10.1038/s42949-024-00148-x]

9. Deng, X.; Yu, W.; Shi, J.; Huang, Y.; Li, D.; He, X.; Zhou, W.; Xie, Z. Characteristics of surface urban heat islands in global cities of different scales: Trends and drivers. Sustain. Cities Soc.; 2024; 107, 105483. [DOI: https://dx.doi.org/10.1016/j.scs.2024.105483]

10. Garuma, G.F. Tropical surface urban heat islands in east Africa. Sci. Rep.; 2023; 13, pp. 1-15. [DOI: https://dx.doi.org/10.1038/s41598-023-31678-6]

11. Oluwamimo, S. The Temporal Structure of the Urban Heat Island in Lagos State, Nigeria; Geophysical Research Abstract European Geosciences Union: Munich, Germany, 2006; Volume 8, 00045.Available online: https://meetings.copernicus.org/www.cosis.net/abstracts/EGU06/00045/EGU06-J-00045.pdf (accessed on 4 August 2024).

12. Bassett, R.; Young, P.J.; Blair, G.S.; Samreen, F.; Simm, W. The Megacity Lagos and Three Decades of Urban Heat Island Growth. J. Appl. Meteorol. Clim.; 2020; 59, pp. 2041-2055. [DOI: https://dx.doi.org/10.1175/JAMC-D-20-0059.1]

13. Heaviside, C.; Macintyre, H.; Vardoulakis, S. The Urban Heat Island: Implications for Health in a Changing Environment. Curr. Environ. Health Rep.; 2017; 4, pp. 296-305. [DOI: https://dx.doi.org/10.1007/s40572-017-0150-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28695487]

14. Obe, O.B.; Morakinyo, T.E.; Mills, G. An assessment of WRF-urban schemes in simulating local meteorology for heat stress analysis in a tropical sub-Saharan African city, Lagos, Nigeria. Int. J. Biometeorol.; 2024; 68, pp. 811-828. [DOI: https://dx.doi.org/10.1007/s00484-024-02627-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38360928]

15. Zhuo, S.; Zhou, W.; Fang, P.; Ye, J.; Luo, H.; Li, H.; Wu, C.; Chen, W.; Liu, Y. Cost-effective pearlescent pigments with high near-infrared reflectance and outstanding energy-saving ability for mitigating urban heat island effect. Appl. Energy; 2023; 353, 122051. [DOI: https://dx.doi.org/10.1016/j.apenergy.2023.122051]

16. Tong, S.; Prior, J.; McGregor, G.; Shi, X.; Kinney, P. Urban heat: An increasing threat to global health. BMJ; 2021; 375, n2467. [DOI: https://dx.doi.org/10.1136/bmj.n2467] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34697023]

17. Anthony, A.F.; Abidemi, K.R. Rapid urbanization of Ikorodu town and the implications on the metropolitan form and space. J. Soc. Rev. Dev.; 2024; 3, pp. 1-7. Available online: https://www.dzarc.com/social/article/view/433 (accessed on 22 July 2024).

18. Lamond, J.; Falade, J.B.; Monroy, J.; Hansen, J.; Walker, J.; Awuah, K.B.; Ojo, A. Urbanisation in Nigeria: A Scoping Review; Report University of Lincoln: Lincoln, UK, 2024; Available online: https://repository.lincoln.ac.uk/articles/report/Urbanisation_in_Nigeria_a_scoping_review/24365815 (accessed on 22 August 2024).

19. NBS. Demographic Statistics Bulletin; National Bureau of Statistics: Abuja, Nigeria, 2020; Available online: https://nigerianstat.gov.ng/pdfuploads/DEMOGRAPHIC%20BULLETIN%202020.pdf (accessed on 15 July 2024).

20. Osayomi, T.; Adeniyi, O. Spatial econometric analysis of inbound foreign direct investments in Nigeria: A geographically weighted regression approach. Niger. J. Econ. Soc. Stud.; 2017; 59, pp. 91-120. Available online: https://www.econbiz.de/Record/spatial-econometric-analysis-of-inbound-foreign-direct-investments-in-nigeria-a-geographically-weighted-regression-approach-osayomi-tolulope/10012153040/Description#tabnav (accessed on 4 July 2024).

21. Adeleke, R.; Osayomi, T.; Adeoti, T. Does sub-national government revenue have an effect on socio-economic and infrastructural development in Nigeria? A geographical analysis. Reg. Sci. Policy Pract.; 2021; 13, pp. 1603-1614. [DOI: https://dx.doi.org/10.1111/rsp3.12438]

22. World Bank. Ogun State Economic Transformation Project. Project Proposal Document; Report No: PAD2803 International Development Association: Washington, DC, USA, 2020; Available online: https://documents1.worldbank.org/curated/en/598231582340468282/pdf/Nigeria-Ogun-State-Economic-Transformation-Project.pdf (accessed on 4 July 2024).

23. Ogun State Government. Gateway to Infrastructure and Industrial Investment Opportunities: Ogun State Economic Development Clusters; Ministry of Finance and Office of the Economic Adviser: Abeokuta, Nigeria, 2021.

24. Zhou, D.; Xiao, J.; Bonafoni, S.; Berger, C.; Deilami, K.; Zhou, Y.; Frolking, S.; Yao, R.; Qiao, Z.; Sobrino, J.A. Satellite Remote Sensing of Surface Urban Heat Islands: Progress, Challenges, and Perspectives. Remote Sens.; 2019; 11, 48. [DOI: https://dx.doi.org/10.3390/rs11010048]

25. Martin, P.; Baudouin, Y.; Gachon, P. An alternative method to characterize the surface urban heat island. Int. J. Biometeorol.; 2014; 59, pp. 849-861. [DOI: https://dx.doi.org/10.1007/s00484-014-0902-9]

26. Kaplan, G.; Avdan, U.; Avdan, Z.Y. Urban Heat Island Analysis Using the Landsat 8 Satellite Data: A Case Study in Skopje, Macedonia. Proceedings; 2018; 2, 358. [DOI: https://dx.doi.org/10.3390/ecrs-2-05171]

27. Streutker, D.R. Satellite-measured growth of the urban heat island of Houston, Texas. Remote Sens. Environ.; 2003; 85, pp. 282-289. [DOI: https://dx.doi.org/10.1016/S0034-4257(03)00007-5]

28. Benz, S.A.; Bayer, P.; Goettsche, F.M.; Olesen, F.S.; Blum, P. Linking Surface Urban Heat Islands with Groundwater Temperatures. Environ. Sci. Technol.; 2015; 50, pp. 70-78. [DOI: https://dx.doi.org/10.1021/acs.est.5b03672] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26595444]

29. Sheng, L.; Tang, X.; You, H.; Gu, Q.; Hu, H. Comparison of the urban heat island intensity quantified by using air temperature and Landsat land surface temperature in Hangzhou, China. Ecol. Indic.; 2016; 72, pp. 738-746. [DOI: https://dx.doi.org/10.1016/j.ecolind.2016.09.009]

30. Huang, F.; Zhan, W.; Wang, Z.-H.; Voogt, J.; Hu, L.; Quan, J.; Liu, C.; Zhang, N.; Lai, J. Satellite identification of atmospheric-surface-subsurface urban heat islands under clear sky. Remote Sens. Environ.; 2020; 250, 112039. [DOI: https://dx.doi.org/10.1016/j.rse.2020.112039]

31. Jabbar, H.K.; Hamoodi, M.N.; Al-Hameedawi, A.N. Urban heat islands: A review of contributing factors, effects and data. IOP Conf. Ser. Earth Environ. Sci.; 2023; 1129, 012038. [DOI: https://dx.doi.org/10.1088/1755-1315/1129/1/012038]

32. Xu, D.; Wang, Y.; Zhou, D.; Wang, Y.; Zhang, Q.; Yang, Y. Influences of urban spatial factors on surface urban heat island effect and its spatial heterogeneity: A case study of Xi’an. Build. Environ.; 2024; 248, 111072. [DOI: https://dx.doi.org/10.1016/j.buildenv.2023.111072]

33. Balogun, I.A.; Balogun, A.A.; Adeyewa, Z.D. Observed urban heat island characteristics in Akure, Nigeria. Afr. J. Environ. Sci. Technol.; 2012; 6, pp. 1-8. [DOI: https://dx.doi.org/10.5897/AJEST11.084]

34. Balogun, I.A.; Balogun, A.A. Urban heat island and bioclimatological conditions in a hot-humid tropical city: The example of Akure, Nigeria. DIE ERDE–J. Geogr. Soc. Berl.; 2014; 145, pp. 3-15. [DOI: https://dx.doi.org/10.12854/erde-145-2]

35. Ojeh, V.N.; Balogun, A.A.; Okhimamhe, A.A. Urban-Rural Temperature Differences in Lagos. Climate; 2016; 4, 29. [DOI: https://dx.doi.org/10.3390/cli4020029]

36. Adeyeri, O.; Akinsanola, A.; Ishola, K. Investigating surface urban heat island characteristics over Abuja, Nigeria: Relationship between land surface temperature and multiple vegetation indices. Remote Sens. Appl. Soc. Environ.; 2017; 7, pp. 57-68. [DOI: https://dx.doi.org/10.1016/j.rsase.2017.06.005]

37. Ofordu, C.S.; Agbor, C.F.; Aigbokhan, O.J.; Audu, M.A.; Adedoyin, E.D.; Ogoliegbune, O.M. Urban Heat Island and Land Use/Cover Dynamics Evaluation in Enugu Urban, Nigeria. J. Geosci. Environ. Prot.; 2022; 10, pp. 354-372. [DOI: https://dx.doi.org/10.4236/gep.2022.1012020]

38. Odunsi, O.M.; Ayedun, H.O.; Odunsi, M.Y.; Momodu, S.O.; Badiora, A.I.; Taiwo, A.O.; Adebara, T.M. Holding on to life: COVID-19 pandemic impact, adaptation and resilience in Abeokuta, Nigeria. Discov. Soc. Sci. Heal.; 2024; 4, pp. 1-20. [DOI: https://dx.doi.org/10.1007/s44155-024-00059-3]

39. Korah, A.; Koch, J.A.; Wimberly, M.C. Understanding urban growth modeling in Africa: Dynamics, drivers, and challenges. Cities; 2023; 146, 104734. [DOI: https://dx.doi.org/10.1016/j.cities.2023.104734]

40. Runfola, D.; Anderson, A.; Baier, H.; Crittenden, M.; Dowker, E.; Fuhrig, S.; Goodman, S.; Grimsley, G.; Layko, R.; Melville, G. et al. geoBoundaries: A global database of political administrative boundaries. PLoS ONE; 2020; 15, e0231866. [DOI: https://dx.doi.org/10.1371/journal.pone.0231866]

41. USGS. Landsat 7 (L7) Enhanced Thematic Mapper Plus (ETM+) Collection 2 (C2) Level 1 (L1) Data Format Control Book (DFCB). Earth Resources Observation and Science (EROS) Center. 2020; Available online: https://www.usgs.gov/centers/eros/science/usgs-eros-archive-landsat-archives-landsat-7-enhanced-thematic-mapper-plus (accessed on 22 July 2024).

42. USGS. Landsat 8-9 Operational Land Imager (OLI)—Thermal Infrared Sensor (TIRS) Collection 2 Level 1 (L1) Data Format Control Book (DFCB). Earth Resources Observation and Science (EROS) Center. 2020; Available online: https://www.usgs.gov/centers/eros/science/usgs-eros-archive-landsat-archives-landsat-8-9-operational-land-imager-and (accessed on 22 July 2024).

43. Gandhi, G.M.; Parthiban, S.; Thummalu, N.; Christy, A. Ndvi: Vegetation Change Detection Using Remote Sensing and Gis—A Case Study of Vellore District. Procedia Comput. Sci.; 2015; 57, pp. 1199-1210. [DOI: https://dx.doi.org/10.1016/j.procs.2015.07.415]

44. Huang, S.; Tang, L.; Hupy, J.P.; Wang, Y.; Shao, G.F. A commentary review on the use of normalized difference vegetation index (NDVI) in the era of popular remote sensing. J. For. Res.; 2021; 32, pp. 1-6. Erratum in J. For. Res. 2021, 32, 2719 [DOI: https://dx.doi.org/10.1007/s11676-020-01155-1]

45. Norman, J.M.; Becker, F. Terminology in thermal infrared remote sensing of natural surfaces. Agric. For. Meteorol.; 1995; 77, pp. 153-166. [DOI: https://dx.doi.org/10.1016/0168-1923(95)02259-Z]

46. Waleed, M.; Sajjad, M. Leveraging cloud-based computing and spatial modeling approaches for land surface temperature disparities in response to land cover change: Evidence from Pakistan. Remote Sens. Appl. Soc. Environ.; 2022; 25, 100665. [DOI: https://dx.doi.org/10.1016/j.rsase.2021.100665]

47. Li, Z.; Wu, H.; Duan, S.; Zhao, W.; Ren, H.; Liu, X.; Leng, P.; Tang, R.; Ye, X.; Zhu, J. et al. Satellite Remote Sensing of Global Land Surface Temperature: Definition, Methods, Products, and Applications. Rev. Geophys.; 2023; 61, e2022RG000777. [DOI: https://dx.doi.org/10.1029/2022RG000777]

48. Guha, S.; Govil, H.; Dey, A.; Gill, N. Analytical study of land surface temperature with NDVI and NDBI using Landsat 8 OLI and TIRS data in Florence and Naples city, Italy. Eur. J. Remote Sens.; 2018; 51, pp. 667-678. [DOI: https://dx.doi.org/10.1080/22797254.2018.1474494]

49. Kafy, A.-A.; Faisal, A.A.; Rahman, S.; Islam, M.; Al Rakib, A.; Islam, A.; Khan, H.H.; Sikdar, S.; Sarker, H.S.; Mawa, J. et al. Prediction of seasonal urban thermal field variance index using machine learning algorithms in Cumilla, Bangladesh. Sustain. Cities Soc.; 2021; 64, 102542. [DOI: https://dx.doi.org/10.1016/j.scs.2020.102542]

50. Waleed, M.; Sajjad, M.; Acheampong, A.O.; Alam, T. Towards Sustainable and Livable Cities: Leveraging Remote Sensing, Machine Learning, and Geo-Information Modelling to Explore and Predict Thermal Field Variance in Response to Urban Growth. Sustainability; 2023; 15, 1416. [DOI: https://dx.doi.org/10.3390/su15021416]

51. Moisa, M.B.; Gemeda, D.O. Assessment of urban thermal field variance index and thermal comfort level of Addis Ababa metropolitan city, Ethiopia. Heliyon; 2022; 8, e10185. [DOI: https://dx.doi.org/10.1016/j.heliyon.2022.e10185] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36033329]

52. Janczyk, M.; Pfister, R. Repeated-Measures Analysis of Variance (ANOVA). Understanding Inferential Statistics; Springer: Berlin/Heidelberg, Germany, 2023; Available online: https://link.springer.com/chapter/10.1007/978-3-662-66786-6_10 (accessed on 22 July 2024).

53. Field, A.P. Discovering Statistics Using SPSS: And Sex and Drugs and Rocks ‘n’ Roll; 4th ed. Sage: London, UK, 2013; Available online: https://books.google.com.hk/books/about/Discovering_Statistics_Using_SPSS.html?id=IY61Ddqnm6IC&redir_esc=y (accessed on 22 July 2024).

54. Obateru, R.O.; Okhimamhe, A.A.; Fashae, O.A.; Aweda, E.; Dragovich, D.; Conrad, C. Community-based assessment of the dynamics of urban landscape characteristics and ecosystem services in the rainforest and guinea savanna ecoregions of Nigeria. J. Environ. Manag.; 2024; 360, 121191. [DOI: https://dx.doi.org/10.1016/j.jenvman.2024.121191] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38759552]

55. World Bank. Urban Population—Nigeria. World Urbanization Prospects: 2018 Revision. Available online: https://data.worldbank.org/indicator/SP.URB.TOTL.IN.ZS?locations=NG (accessed on 16 July 2024).

56. Ojo, A.; Ojewale, O.; Ojo, A.; Ojewale, O. Nigeria’s urbanisation history, trends, drivers and implications. Urbanisation and Crime in Nigeria; Palgrave Macmillan: Cham, Switzerland, 2019; pp. 13-58. [DOI: https://dx.doi.org/10.1007/978-3-030-19765-0_2]

57. Sultana, S.; Satyanarayana, A.N.V. Assessment of urbanisation and urban heat island intensities using landsat imageries during 2000–2018 over a sub-tropical Indian City. Sustain. Cities Soc.; 2020; 52, 101846. [DOI: https://dx.doi.org/10.1016/j.scs.2019.101846]

58. UN-HABITAT. Abeokuta: City Context Report; Global Future Cities Programme: Nairobi, Kenya, 2018; Available online: https://www.globalfuturecities.org/sites/default/files/2020-07/Nigeria_Abeokuta_CCR.pdf (accessed on 14 July 2024).

59. Olayiwola, A.M.; Lawal, M.O. Urban Encroachment into Agricultural Land in Abeokuta. Ife Soc. Sci. Rev.; 2019; 27, pp. 13-27. Available online: https://www.researchgate.net/publication/377472893_Urban_Encroachment_into_Agricultural_Land_in_Abeokuta_Nigeria (accessed on 14 July 2024).

60. Odunsi, O.M.; Onanuga, M.Y.; Obaitor, O.S. The new millennium so far: Analysing land cover change in Ogun State Nigeria. J. Contemp. Afr. Stud.; 2024; 42, pp. 612-627. [DOI: https://dx.doi.org/10.1080/02589001.2024.2365633]

61. Zhang, B.; Xie, G.-D.; Gao, J.-X.; Yang, Y. The cooling effect of urban green spaces as a contribution to energy-saving and emission-reduction: A case study in Beijing, China. Build. Environ.; 2014; 76, pp. 37-43. [DOI: https://dx.doi.org/10.1016/j.buildenv.2014.03.003]

62. Amani-Beni, M.; Zhang, B.; Xie, G.-D.; Shi, Y. Impacts of Urban Green Landscape Patterns on Land Surface Temperature: Evidence from the Adjacent Area of Olympic Forest Park of Beijing, China. Sustainability; 2019; 11, 513. [DOI: https://dx.doi.org/10.3390/su11020513]

63. Yang, M.; Ren, C.; Wang, H.; Wang, J.; Feng, Z.; Kumar, P.; Haghighat, F.; Cao, S.-J. Mitigating urban heat island through neighboring rural land cover. Nat. Cities; 2024; 1, pp. 522-532. [DOI: https://dx.doi.org/10.1038/s44284-024-00091-z]

64. UNFCCC. Report of the Conference of the Parties Serving as the Meeting of the Parties to the Paris Agreement on Its Fifth Session. UN Climate Change Conference—United Arab Emirates. 2024 (30 November–6 December) FCCC/PA/CMA/2023/16/Add.1. Available online: https://unfccc.int/documents/637072 (accessed on 2 August 2024).

65. International Energy Agency. Empowering Cities for a Net Zero Future; IEA: Paris, France, 2021; Available online: https://www.iea.org/reports/empowering-cities-for-a-net-zero-future (accessed on 24 July 2024).

66. Sun, T.; Ocko, I.B.; Sturcken, E.; Hamburg, S.P. Path to net zero is critical to climate outcome. Sci. Rep.; 2021; 11, 22173. [DOI: https://dx.doi.org/10.1038/s41598-021-01639-y] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34772989]