Measuring regional inequality using nightlight satellite data

and population density for Nigeria

Zubairul Islam*

University of Abuja, Nigeria

Abstract: Measuring the spatial differences in regional development was the main
objective of this study. To meet this objective, spatial patterns & clusters of variables, viz.
nightlight & population density, were identified at the LGA level. Secondly, regression
analysis between the same variables was performed to find the spatial differences in the
night light. VIIRS Day/Night Band (DNB) data was chosen as the dependent variable, and
UN-Adjusted Population Density data was selected as an explanatory variable. Spatial
patterns & clusters were identified using spatial statistics. Global Ordinary Least Squares
(OLS) linear regression was chosen to model nightlight in terms of its relationships to
population density. Geographically Weighted Regression (GWR) regression was used to
model spatially varying relationships between the same variables. The results show
nightlight (z >97, p <0000) & population density (z >108, p <0000) are highly clustered.
The R2 obtained from OLS & GWR are 0.75 & 0.85, respectively. Moreover, model variables
& diagnostics results confirm the validity of both models.

Key Words: nightlight data, population density, regional inequality.

Article Info: Received: 23 July 2023; Revised: 24 May 2024; Accepted: 27 May 2024;
Online: 31 May 2024.

Introduction

Despite Nigeria's rich endowment in natural and human resources, the

country's development pace remains sluggish (Ene, 2020) and not equally
distributed (Dieteren & Bonfrer, 2021). Variations in resource distribution across
different regions contribute to disparities in regional economic development (Li,
2021). But, equal development of regions is vital for eco and sustainable
development (Gechbaia et al., 2021). Unequal development in Nigeria is a pressing
issue that has persisted for decades, driven by a complex interplay of factors,
including geography, history, governance, and resource distribution (Hodgson,

* Correspondence address
Address: Leadership Development Center, University of Abuja, Nigeria.
Phone: +91-9821405543| Email: [email protected]

©2024 Human Geographies; The authors

This work is licensed under a
Creative Commons Attribution 4.0 International License. DOI:10.5719/hgeo.2024.181.3
2 Zubairul Islam

2018). Nigeria's diverse geographic landscape, which spans from arid northern
regions to lush southern areas, has contributed significantly to regional disparities
in development (Adepoju, 2006). Historical legacies play a pivotal role in these
disparities. Nigeria's colonial past, which included British rule in the south and less
direct rule in the north, led to differential access to education, infrastructure, and
economic opportunities. The effects of these historical imbalances continue to
reverberate today, perpetuating unequal development (Suberu, 2001).
Resource allocation and management have been another significant driver of
regional inequality. The southern regions, particularly the Niger Delta, are rich in
oil reserves, which have historically provided a substantial portion of the country's
revenue. However, the benefits of this resource have not been equitably
distributed, leading to socio-economic and infrastructural deficits in other regions
(Eboh, 2014). Additionally, governance issues, including corruption,
mismanagement of resources, and weak public institutions, have exacerbated the
problem of unequal development. These issues hinder effective policy
implementation and public service delivery, disproportionately affecting regions
already struggling with development challenges (Omotola, 2011).
The consequences of unequal development are profound. Disparities in income,
education, healthcare, and living standards persist, fostering social unrest and
political tensions (Oladipo, 2006). Addressing these disparities is a critical
challenge for Nigeria's government, requiring strategic and targeted policy
interventions (Egwu, 2017).
Satellite-based nighttime illumination data can be used to map human
activities (Levin et al., 2020), economic activities (Anand & Kim, 2021), local
economic growth (Bertinelli & Strobl, 2013), settlement activities (Bluhm &
Krause, 2022) additionally, this methodology offers an alternative for considering
economic factors in less-developed countries, where data accessibility can often be
challenging. (Weidmann & Schutte, 2017). Nightlight and Machine learning help
estimate development where data are not precise (Otchia & Asongu, 2020).
Population and nightlight (NLT) data have been used to visualize the development
index (Zhao et al., 2021). Nightlight data (NLT) opens up new possibilities for real-
time measurement of regional disparities by facilitating the development of the
Nightlight Development Index (NLDI) (Ivan et al., 2020). It is easy now to monitor
and evaluate energy consumption with night light data (Xiao et al., 2018). NLT has
many applications, such as estimating population and GDP, urban mapping, and
monitoring fire and war. Remotely sensed man-made lighting radiances can
provide a level of human activity during the night (Ma, 2018). A high positive
association exists between night lights and GDP estimates (Bhandari &
Roychowdhury, 2011). Regular observations of night lighting and its profound
study have made it easy to study socio-economic dynamics (Krikigianni et al.,
2019). The use of NLT is growing daily in economic activity estimation (Prakash et
al., 2019) and in GIS and remote sensing studies. Human-created light provides a
unique footprint of human activities and settlements (Zheng et al., 2018). The
satellite sensor detecting nighttime light data is supposed to be a reliable source
for mapping urban areas across large regions (Alahmadi & Atkinson, 2019).
Night light data offers a more straightforward approach to quantifying
urbanization levels. Its advantage lies in accurately reflecting the spatial extent of
urban land and effectively depicting the spatial pattern of socioeconomic activities
within urban areas (Wang et al., 2020).
In developing countries, incorporating energy planning into overall
development strategies is crucial due to the rapidly growing rates of economic
expansion and increasing energy demand (Lee et al., 2020). Currently, Nighttime
 Measuring regional inequality using nightlight satellite data 3

Light (NLT) data is gathered by two satellite sensors: DMSP-OLS and VIIRS, as
noted by Zheng et al. (2019). The VIIRS sensor's NLT data offers a higher spatial
resolution than the DMSP-OLS night light data, as highlighted by Small et al.
(2013). A significant issue with DMSP-OLS is the extensive blurring in its images,
often referred to as 'overglow' or 'blooming', a point raised by Abrahams et al.
(2018). Since April 2012, NTL data from VIIRS has been available, effectively
overcoming some DMSP-OLS limitations, as Sahoo et al. (2020) mentioned.
Therefore, this research employs VIIRS nighttime light data. The use of VIIRS-
based data is crucial for investigating the spatial pattern of the urban system using
night light data, as Zhong et al. (2018) have indicated.
The low power generation in Nigeria has delayed its economic growth (Emovon
et al., 2018). The goal of the power sector in Nigeria is to transmit electrical power
to all regions professionally (Samuel et al., 2020). Researchers have used night
light data to assess the regional differences in power supply (Fan et al., 2014).
Researchers have used night light data to determine the regional differences in
power supply (Elvidge et al., 2012).
The general objective of this study was to measure regional inequality in the
development process of Nige-ria. Four specific objectives were framed to meet this
objective: first & second, to find the pattern and clusters of night light data and
population density; and third and fourth, to model night light distribution about
population density using OLS & GWR, respectively.

Material and Methods

This research was carried out at the district level across Nigeria, located on
Africa's western coast. Nigeria showcases diverse geographical features and a
range of climates, varying from dry to humid equatorial conditions. The
southeastern part of Nigeria experiences hot and moist conditions for most of the
year, whereas the northwest and inland areas are predominantly dry. A savanna
climate prevails in the west and north, characterized by distinct dry and wet
seasons, while the far north exhibits a steppe climate with lower precipitation
levels (Ajayi et al., 2020). Unequal development in Nigeria is a pressing concern
with far-reaching implications. Several factors contribute to this disparity,
including variations in natural resources, historical legacies, infrastructure, and
economic opportunities. The northern regions, for instance, face challenges
associated with arid climates and limited access to arable land, while the southern
regions often benefit from more favourable conditions for agriculture and oil
wealth. These disparities have led to inequalities in income, education, healthcare,
and overall quality of life. In 2020, population density in Nigeria exhibited
significant variation at the Local Government Area (LGA) level, reflecting diverse
demographic patterns across the country. Urban areas, particularly Lagos,
recorded very high population densities, with some LGAs exceeding 20,000 people
per square kilometre due to intense urbanization and economic activities.
Conversely, rural regions, such as those in the northern states of Yobe and Taraba,
experienced very low population densities, often below 50 people per square
kilometre, attributed to expansive land areas and lower urban development. This
stark contrast highlights Nigeria's uneven distribution of population density,
driven by factors such as urbanisation, economic opportunities, and regional
development disparities.
Understanding these regional development differences is crucial for several
reasons. Firstly, it informs policymakers about the specific needs and challenges
faced by different regions, facilitating the allocation of resources and the design of
4 Zubairul Islam

targeted development programs. Secondly, it can help address social and political
tensions arising from unequal development, contributing to social cohesion and
stability.
This study, conducted at the district level in Nigeria, is instrumental in
shedding light on the spatial variations in development within the country. Using
spatial analysis and regression techniques provides a data-driven understanding
of factors contributing to regional disparities, such as the relationship between
nightlight and population density. With a more granular understanding of these
dynamics, Nigeria's leaders can make more informed decisions to promote
balanced development and improve the livelihoods of citizens across the nation,
ultimately fostering economic growth and social well-being. In essence, this study's
findings have the potential to influence policy reforms that can address long-
standing disparities, promoting a more equitable and prosperous future for
Nigeria as a whole.
Two primary datasets were used to measure regional inequality in Nigeria for
2020. The first dataset, the Night Light data, captures information about nightlight
emissions. This data was stored in TIFF format with a band named "avg_rad." It
measured night light in Watts per square centimetre per steradian and represented
the mean monthly values for 2020. The second dataset was Population Density
data, which estimates the number of individuals per 30 arc-second grid cells and
is also in TIFF format. This data helps us understand how people are distributed
across different areas in Nigeria in the same year. By combining these datasets,
researchers will gain valuable insights into the regional disparities and inequalities
in economic development and population distribution within Nigeria during 2020.

Table 1. Data Source and Resolution of Satellite Images

Dataset Source Resolution Date
Night Light Data VIIRS Day/Night Band 2.8 km 2020
Population Gridded Population of the World 30 arc-second grid 2020
Density (GPW), v4 cells
Source: Earth Engine Catalog, 2024

To measure regional inequality in Nigeria using nightlight satellite data and

population density, we utilized Google Earth Engine (GEE) and ArcGIS. Nightlight
data from the Visible Infrared Imaging Radiometer Suite (VIIRS) was collected
using GEE, which has cloud-based capabilities for accessing and processing
extensive satellite datasets. The pre-processed nightlight data was then imported
into ArcGIS, where various spatial analyses were performed. ArcGIS facilitated
tasks such as reprojecting data, conducting zonal and hot spot analyses, and
visualising the relationship between nightlight intensity and population density.
This integration of GEE for data collection and ArcGIS for spatial analysis enabled
a comprehensive examination of regional inequality across Nigeria.
The "Global Moran's I" statistics were used to predict the pattern of nightlight
and population density at the local government administrative level in Nigeria.
This method assesses spatial autocorrelation by simultaneously considering the
location and attribute values. The resulting pattern could be dispersed, clustered,
or random, contingent on the features' specific location and attribute values. The
Moran’s I statistic, which quantifies spatial autocorrelation, is delineated in
equations 1 to 5.
 Measuring regional inequality using nightlight satellite data 5

𝑛 ∑"#$% ∑"&$% 𝑤#,& 𝑧# 𝑧& (1)

𝐼=
𝑆! ∑"#$% 𝑧#(
In this context, zi represents the deviation of a feature's attribute from its mean,
while the spatial weight between feature j and i is denoted. The term n stands for
the total number of features. S0 refers to the aggregate of all spatial weights.
" "
(2)
𝑆! = ) ) 𝑤#,&
#$% &$%

The zi-score for the statistic is computed as:

𝐼 − 𝐸[𝐼] (3)
𝑧) =
.𝑉[𝐼]
Where:
𝐸[𝐼] = −1/(𝑛 − 1) (4)
𝑉[𝐼] = 𝐸[𝐼( ] − 𝐸[𝐼]( (5)
The inverse distance & Euclidian distance method was chosen to conceptualise
spatial relationship parameters. Distance Threshold was given as 71,668.8 Meters,
representing the maximum distance over which spatial interactions were
considered significant for the analysis.
The statistic “Getis-Ord Gi*" was used to identify the hot spots for the nightlight
and population density dataset. The resultant p-values and z-scores indicate low
or high values cluster spatially, eq. 06 to 08.
∑"#$% 𝑤#,& 𝑥& − 𝑋7 ∑"&$% 𝑤#,& (6)
𝐺#∗ =
(
9𝑛 ∑"#$% 𝑤#&( − :∑"&$% 𝑤#,& ; <
𝑆8 𝑛−1
In this formula, xj indicates the attribute value of feature j, and wij represents
the spatial weight between feature i and j. The variable n signifies the total count
of features, and the following details are elaborated upon further in the text:
∑"&$% 𝑥& (7)
𝑋7 =
𝑛
(8)
∑"&$% 𝑥&(
𝐺#∗ = 8 − (𝑋7)(
𝑛

This statistic is the z score, so no further calculation is required.

Ordinary least squire (OLA) regression was used to find the nightlight global
variability in relation to population density within Nigeria (eq. 9):

𝛾 = 𝛽! + 𝛽% 𝑋% + 𝛽( 𝑋( + 𝛽+ 𝑋+ +. . . . . . . 𝛽" 𝑋" + ɛ (9)

In this equation, γ stands for the dependent variable, β denotes the coefficients,
X represents the explanatory variables, and ɛ indicates the random error.
Geographically weighted regression (GWR) was used to find the local variability of
nightlights in relation to population density at the LGA level in Nigeria. OLS
constructs a single equation, but GWR constructs separate by taking variables of
features under the bandwidth of each target feature. The shape and size of the
bandwidth chosen are given as follows. Kernel type: For the spatial analysis, an
adaptive kernel type, specifically the Gaussian kernel, was chosen. The number of
neighbouring features determines this kernel's functioning. In areas where
6 Zubairul Islam

features are densely distributed, the spatial context defined by this kernel is
smaller. Conversely, the spatial context becomes larger in regions with sparse
feature distribution. Bandwidth method: CV (cross-validation) was selected, and
the tool found the optimal distance or number of neighbours for the analysis.

Results and Discussion

Spatial Pattern of Nightlight and Population Density
The result indicates that night light in Nigeria's LGA level is too clustered (z =
97.01). Given the z-score of 97.01, there is a <1% likelihood that this clustered
pattern may be due to random chance. This means that the distribution of night
lights in Nigeria's LGAs is not random. Instead, night light appears to be
concentrated in certain areas of LGAs. These clusters may represent regions with
higher economic activity, urbanisation, or infrastructure development, resulting in
more illumination during the night. A high z-score indicates an extreme deviation
from the mean. In this case, a z-score of 97.01 is exceptionally high, suggesting that
the clustering of night light is far from what would be expected by random chance.
These findings are essential for understanding the spatial distribution of
development and can affect policy decisions to address regional disparities and
promote balanced growth across the country.
Population density at the LGA level in Nigeria is highly clustered (z = 108.8).
Given the z-score of 108.18, there is a <1% probability that this highly clustered
pattern may be due to random chance. This means that the distribution of
population density in the LGAs of Nigeria is not random. Instead, it indicates that
people are concentrated in certain areas or LGAs, and there are notable spatial
patterns. These clusters may represent regions with higher population density due
to urbanisation, economic opportunities, or historical settlement patterns. These
findings are essential for policymakers as they highlight areas where the
population is concentrated, and targeted policies may be needed to address issues
related to infrastructure, services, and resource allocation in densely populated
regions.

Hot Spots of Night Light and Population Density

The result of the night light data hotspots is given in Figure 1, which indicates
that there are 04 well-marked hot spots for the night light. Lagos has the highest
mean nightlight intensity of 12.88 Watts/cm²/sr, signifying its status as a
prominent urban and economic centre. Delta, Kano, Rivers, Edo, Ogun, Bayelsa,
and Jigawa follow in descending order. These rankings provide insights into the
varying degrees of urbanisation, economic activity, and development across these
states, which have implications for understanding regional disparities and policy
considerations.
The result of hotspots of population density data is presented in Figure 2,
indicating two well-marked hotspots of population density. These hotspots are
organised based on different confidence levels, with each section detailing the
number of Local Government Areas (LGAs) and the mean population density. At a
99% confidence level, Lagos has 20 LGAs with an average population density of
13,635 people per 30 arc-second grid cells, signifying a high population density
across its LGAs. Similarly, Kano has 38 LGAs with an average population density
of 3,241, indicating substantial population concentrations. Ogun has 14 LGAs with
a mean population density of 932, which is also relatively high, whereas Jigawa,
with only 2 LGAs, has a mean population density of 370, indicating lower
 Measuring regional inequality using nightlight satellite data 7

population density areas. The total number of LGAs considered at the 99%
confidence level is 74, with a combined average population density of 4,544.

Figure 1. Hot Spots of Night Light

Source: NOAA_VIIRS_DNB_MONTHLY_V1_VCMSLCFG

Figure 2. Hot Spots of Population Density

Source: CIESIN/GPWv411/GPW_Population_Density.
8 Zubairul Islam

At a 95% confidence level, Oyo has 6 LGAs with an average population density
of 4,166. Ogun is represented again at this confidence level with 3 LGAs and a mean
population density of 430. The total number of LGAs considered at the 95%
confidence level is 9, with a combined average population density of 2,298. At a
90% confidence level, Oyo has 9 LGAs with a mean population density of 4,378,
while Osun, at the same confidence level, has 6 LGAs with a mean population
density of 483. The total number of LGAs considered at the 90% confidence level
is 15, with a combined average population density of 2,431.
Figure 2 essentially provides insights into the distribution of population density
across Nigerian states at different confidence levels. Higher population densities
at specific confidence levels suggest regions with more significant population
concentrations, while lower densities indicate areas with relatively fewer
inhabitants. This information is vital for understanding the country's population
patterns and regional disparities.

Spatial relationships between Night Light & Population Density

Figure 3 presents a comprehensive analysis of standardised residuals of
nightlight (OLS) for various states and Local Government Areas (LGAs) in Nigeria.
These are a crucial indicator of how the actual nightlight levels differ from those
predicted by an Ordinary Least Squares (OLS) regression model.

Figure 3. Standard residuals at state level

Source: Ordinary Least Squares (OLS) model

In the context of this analysis, it's important to note that lower values of
standardised residuals indicate a situation where the actual nightlight levels are
significantly lower than the values predicted by the model, which may suggest
 Measuring regional inequality using nightlight satellite data 9

areas with less nighttime illumination relative to what would be expected based on
the population. On the other hand, higher positive values of standardised residuals
suggest areas where actual nightlight levels are significantly higher than the
model's predictions concerning population density, indicating regions with more
extensive or brighter nighttime illumination than what could be explained by
population alone.
For instance, Lagos stands out as an extreme case with a high positive
standardised residual (StdResid), indicating that it has exceptionally high
nightlight levels beyond what population density would predict, possibly due to its
significant economic and urban centre status. Conversely, areas with low
standardised residuals, such as Abia, Aba South, and Ugwunagbo, have nightlight
levels significantly below what population density alone would predict. This could
imply factors like limited urbanisation, economic activity, or infrastructure
development in these regions. This analysis is valuable for understanding
discrepancies between actual nightlight levels and what can be expected based on
population density, shedding light on the presence of "outliers" regarding
nighttime illumination across different states and LGAs in Nigeria.

Spatially varying relationships between Night Light & Population Density

Negative standardised residuals indicate areas where the actual nightlight
levels are notably lower than predicted by the GWR model (Figure 4). These areas
may have lower nightlight levels relative to population density, suggesting
potential factors like lower economic activity or less urbanisation. Conversely,
positive standardised residuals point to areas with higher actual nightlight levels
than the model would predict based on population density alone. This may indicate
regions with more extensive or brighter nighttime illumination, potentially due to
economic activity, infrastructure development, or other contributing factors. For
instance, states like Lagos and Kano exhibit high positive standardised residuals,
suggesting they have significantly brighter nightlight than population density
alone would suggest. Lagos, a significant economic and urban centre, has
exceptionally high nighttime illumination. On the other hand, areas like Girei, Yola
South, or some LGAs in Akwa Ibom show notably negative standardised residuals,
indicating lower nightlight levels compared to what would be expected based on
population. Further investigation into these areas could help understand the
reasons behind these discrepancies.
The Geographical Weighted Regression (GWR) model offers advantages over
the Ordinary Least Squares (OLS) model, particularly when analysing spatial data.
GWR accounts for spatial heterogeneity, allowing for examining variations in the
relationship between variables at different locations. In analysing nightlight data,
it's essential to consider that the relationship between population density and
nightlight may vary across regions. GWR can capture this spatial variation, which
OLS cannot.
GWR provides localised parameter estimates. This means that you get a unique
set of parameter estimates for each specific location (e.g., state or LGA). This is
valuable for understanding the nuances of how population density and other
factors influence nightlight at a local scale. OLS provides a single set of parameters
for the entire dataset, which may not adequately capture the variations present in
spatial data. The results demonstrate that GWR can help identify spatial outliers
10 Zubairul Islam

with its standardised residuals. Areas with notably high or low standardised
residuals can be interpreted as locations with unexpected nightlight levels in
relation to population density. This information is critical for understanding
regional disparities and identifying areas where additional investigation or policy
intervention may be needed. GWR often results in a better model fit, as it accounts
for spatial autocorrelation. Spatial autocorrelation occurs when values in one
location are correlated with values in nearby locations. OLS assumes that
observations are independent, which can lead to issues in spatial data analysis.
GWR considers spatial autocorrelation, leading to more accurate and locally
relevant parameter estimates.

Figure 1. GWR - Standard Residual Map

Source: Geographical Weighted Regression (GWR) model.

So, GWR is a powerful tool for spatial decision-making and policy development.
Policymakers and researchers can make informed decisions and allocate resources
effectively by providing insights into how variables relate at the local level. This is
particularly important when addressing regional disparities and inequalities.

Conclusions

Based on the VIIRS/DNB nighttime lighting data, this paper studies the spatial
pattern, clusters, and spatial differences in the night light pattern in relation to
population density at the LGA level in Nigeria. The study addresses the critical
need to assess and understand the disparities in regional development. Regional
development is an essential concern for policymakers, as it can impact economic
growth, living standards, and social well-being. The study provides valuable
 Measuring regional inequality using nightlight satellite data 11

insights into development across different regions by measuring spatial

differences. The study identifies spatial patterns and clusters of variables,
specifically nightlight and population density, at the Local Government Area (LGA)
level. This is significant as it helps recognise areas with similar characteristics or
issues, which can inform targeted policy interventions and resource allocation. The
study reports high values of R2 for both the Ordinary Least Squares (OLS) and
GWR models, indicating that the chosen models provide a good fit for the data.
This adds credibility to the results, making them more robust and reliable. The
results of this research show that the following: (1) The pattern is highly clustered;
(2) There are four clusters of Nightlight but two clusters of population density; (3)
The variability of night light distribution was significantly explained by the density
of population. The use of nightlights is essential for identifying underdeveloped
areas and coordinating the country's development. Therefore, it is evident that the
study of the regional pattern of the night light at LGA level has high significance
and value. The VIIRS/DNB nighttime lighting satellite data has unparalleled
advantages for monitoring development and analysing spatial patterns of
nightlight data. This research paper utilised diverse data sources, including UN-
adjusted population density, Local Government Area (LGA) level administrative
boundaries of Nigeria, and night light data. It employs advanced statistical tools
and methodologies such as cluster analysis, pattern analysis, and regression
analysis. These techniques are instrumental in extracting nighttime light
thresholds. The processes involved in data calculation and the analysis of research
findings are characterised by a high degree of feasibility. Both OLS and GWR were
used to analyse spatial relations between night light and population density. GWR
was found to be a better model for this spatial data analysis, especially when
dealing with variations in relationships between variables across different
locations. Its ability to provide localised insights, identify spatial outliers, and offer
a better fit for spatial data makes it a valuable tool for understanding and
addressing regional diversity and inequalities, as demonstrated in the provided
tables.
Despite its utility, the practical application of this methodology faces
challenges, such as the low resolution of nighttime lighting satellite data and
population density data. This leads to reduced data accuracy in less urbanised
areas. Acquiring more detailed data through higher-resolution sources would be
beneficial. Quantitative research can potentially enhance this situation, suggesting
that further research in this area should be pursued.

References

Abrahams, A., Oram, C., & Lozano-Gracia, N. (2018). Deblurring DMSP nighttime
lights: A new method using Gaussian filters and frequencies of illumination.
Remote Sensing of Environment, 210, 242–258.
Adepoju, A. (2006). Population Dynamics in Sub-Saharan Africa: Challenges and
Opportunities. In H. B. Tambo & M. Adepoju (Eds.), International Migration and
Development in Sub-Saharan Africa (pp. 49-74). Center for Migration Studies.
Alahmadi, M., & Atkinson, P. M. (2019). Three-fold urban expansion in Saudi
Arabia from 1992 to 2013 observed using calibrated DMSP-OLS night-time lights
imagery. Remote Sensing, 11(19).
Anand, A., & Kim, D. H. (2021). Pandemic induced changes in economic activity
around African protected areas captured through night-time light data. Remote
Sensing, 13(2), 1–15.
12 Zubairul Islam

Bertinelli, L., & Strobl, E. (2013). Quantifying the local economic growth impact of
hurricane strikes: An analysis from outer space for the caribbean. Journal of
Applied Meteorology and Climatology, 52(8), 1688–1697.
Bhandari, L., & Roychowdhury, K. (2011). Night Lights and Economic Activity in
India: A study using DMSP-OLS night time images. Proceedings of the Asia-
Pacific Advanced Network, 32(0), 218.
Bluhm, R., & Krause, M. (2022). Top lights: Bright cities and their contribution to
economic development. Journal of Development Economics, 157, 102880.
Dieteren, C., & Bonfrer, I. (2021). Socioeconomic inequalities in lifestyle risk
factors across low- and middle-income countries. BMC Public Health, 21(1).
https://doi.org/10.1186/s12889-021-11014-1
Elvidge, C. D., Baugh, K. E., Anderson, S. J., Sutton, P. C., & Ghosh, T. (2012). The
Night Light Development Index (NLDI): A spatially explicit measure of human
development from satellite data. Social Geography, 7(1), 23–35.
Emovon, I., Samuel, O. D., Mgbemena, C. O., & Adeyeri, M. K. (2018). Electric
Power generation crisis in Nigeria: A Review of causes and solutions.
International Journal of Integrated Engineering, 10(1), 47–56.
Ene Blessing Ita. (2020). Human Development in Nigeria. International Journal
of African and Asian Studies.
Eboh, E. (2014). Federalism and Nigeria's Development: A Paradigm Shift. African
Research Review, 8(2), 161-171.
Egwu, S. (2017). Political Leadership and Corruption in Nigeria Since 1960: A
Socio-economic Analysis. African Journal of Political Science and International
Relations, 11(8), 272-281.
Fan, J., Ma, T., Zhou, C., Zhou, Y., & Xu, T. (2014). Comparative Estimation of
Urban Development in China’s Cities Using Socioeconomic and DMSP/OLS
Night Light Data. Remote Sensing, 6(8), 7840–7856.
Gechbaia, B., Kharaishvili, E., Mushkudiani, Z., Goletiani, K., & Tsilosani, A.
(2021). Challenges of sustainable and equal development of regions in Georgia.
E3S Web of Conferences, 280, 11007.
Hodgson, S. (2018). Nigeria: Background to Unequal Development. In K. Kitchen,
S. Hodgson, & S. Ford, (Eds.), Global Inequality: A Sociological Perspective (pp.
269-288). Routledge.
Ivan, K., Holobâcă, I. H., Benedek, J., & Török, I. (2020). Potential of night-time
lights to measure regional inequality. Remote Sensing, 12(1), 1–15.
Krikigianni, E., Tsiakos, C., & Chalkias, C. (2019). Estimating the relationship
between touristic activities and night light emissions. European Journal of
Remote Sensing, 52(sup1), 233–246.
Lee, H. Y., Jang, K. M., & Kim, Y. (2020). Energy consumption prediction in
vietnam with an artificial neural network-based urban growth model. Energies,
13(17).
Levin, N., Kyba, C. C. M., Zhang, Q., Sánchez de Miguel, A., Román, M. O., Li, X.,
Portnov, B. A., Molthan, A. L., Jechow, A., Miller, S. D., Wang, Z., Shrestha, R.
M., & Elvidge, C. D. (2020). Remote sensing of night lights: A review and an
outlook for the future. Remote Sensing of Environment, 237(September 2019).
Li, Y. (2021). Analysis of Regional Economic Development Differences Based on
Intelligent Hybrid Algorithm. Complexity, 2021.
Ma, T. (2018). Quantitative responses of satellite-derived nighttime lighting
signals to anthropogenic land-use and land-cover changes across China. Remote
Sensing, 10(9).
 Measuring regional inequality using nightlight satellite data 13

Oladipo, S. E. (2006). Ethnic and Religious Conflicts: Challenges to Nigerian

Federalism and National Integration. Nordic Journal of African Studies, 15(3),
378-401.
Omotola, J. S. (2011). Politics and Leadership in Nigeria’s Fourth Republic. African
Spectrum, 46(2), 93-114.
Otchia, C., & Asongu, S. (2020). Industrial growth in sub-Saharan Africa: evidence
from machine learning with insights from nightlight satellite images. Journal of
Economic Studies.
Prakash, A., Shukla, A. K., Bhowmick, C., Carl, R., & Beyer, M. (2019). Night-time
Luminosity: Does it Brighten Understanding of Economic Activity in India?
Reserve Bank of India Occasional Papers, 40(1), 1–24.
Sahoo, S., Gupta, P. K., & Srivastav, S. K. (2020). Comparative analysis between
VIIRS-DNB and DMSP-OLS night-time light data to estimate electric power
consumption in Uttar Pradesh, India. International Journal of Remote Sensing,
41(7), 2565–2580.
Samuel, A., & TT, O. (2020). Power Generation In Nigeria: The Past, Present And
The Future. Journal of Earth and Environmental Sciences Research, 2(2), 1–8.
Small, C., Elvidge, C. D., & Baugh, K. (2013). Mapping urban structure and spatial
connectivity with VIIRS and OLS night light imagery. Joint Urban Remote
Sensing Event 2013, JURSE 2013, 856, 230–233.
Suberu, R. T. (2001). Federalism and Ethnic Conflict in Nigeria. US Institute of
Peace Press.
Wang, Y., Liu, Z., He, C., Xia, P., Liu, Z., & Liu, H. (2020). Quantifying urbanization
levels on the Tibetan Plateau with high-resolution nighttime light data.
Geography and Sustainability, 1(3), 233–244.
Weidmann, N. B., & Schutte, S. (2017). Using night light emissions for the
prediction of local wealth. Journal of Peace Research, 54(2), 125–140.
Xiao, H., Ma, Z., Mi, Z., Kelsey, J., Zheng, J., Yin, W., & Yan, M. (2018). Spatio-
temporal simulation of energy consumption in China’s provinces based on
satellite night-time light data. Applied Energy, 231(June 2018), 1070–1078.
Zhao, F., Song, L., Peng, Z., Yang, J., Luan, G., Chu, C., Ding, J., Feng, S., Jing, Y.,
& Xie, Z. (2021). Night-time light remote sensing mapping: Construction and
analysis of ethnic minority development index. Remote Sensing, 13(11), 1–26.
Zheng, Q., Weng, Q., Huang, L., Wang, K., Deng, J., Jiang, R., Ye, Z., & Gan, M.
(2018). A new source of multi-spectral high spatial resolution night-time light
imagery—JL1-3B. Remote Sensing of Environment, 215, 300–312.
Zheng, Q., Weng, Q., & Wang, K. (2019). Developing a new cross-sensor calibration
model for DMSP-OLS and Suomi-NPP VIIRS night-light imageries. ISPRS
Journal of Photogrammetry and Remote Sensing, 153(May), 36–47.
Zhong, Y., Lin, A., Zhou, Z., & Chen, F. (2018). Spatial pattern evolution and
optimization of urban system in the Yangtze River economic belt, China, based
on DMSP-OLS night light data. Sustainability (Switzerland), 10(10).