A Geoinformation Approach for Assessing Urban Growth Impact on Farmlands Using Earth Observation Data: A Case Study of Owerri in Nigeria
Abstract
This study investigates the impact of rapid urban expansion on agricultural land in the southeastern part of Nigeria, leveraging Cloud computing, machine learning, remote sensing, GIS, and community engagement techniques. The unprecedented pace of urbanization poses significant challenges to sustainable development, particularly threatening food security by converting fertile agricultural lands into urban areas. Utilizing earth observation data, the study maps urban growth patterns and assesses their effects on the contraction of farmlands. Through a combination of supervised machine learning models—Random Forest, Support Vector Machine, and Classification and Regression Trees—applied on Landsat images from 1986, 2000, and 2022, the study identifies significant land cover changes, quantifying the transition from farmland to built-up areas. The findings reveal a notable increase in built-up land, from 8.2% in 1986 to 26.6% in 2022, alongside a variable trend in farmland coverage, initially decreasing from 37.3% in 1986 to 27.7% in 2000, then partially recovering to33.2% in 2022, yet not to its original extent. This transformation underscores the tension between urban development and agricultural sustainability, with implications for local food production and ecosystem services. Community engagement through surveys with local farmers further enriches the analysis, providing insights into the socio-economic impacts of urban sprawl, including altered farming practices and adaptive strategies. The study underscores the need for integrated urban planning and policy interventions to balance urban growth with agricultural preservation, contributing to the broader discourse on sustainable development and food security. Recommendations include fostering sustainable land-use practices, enhancing community adaptive capacities, and advocating for policy reforms to support affected farmers and ensure long-term sustainability of both urban and rural landscapes.
Full text
Guia pa a a o ma ação de eses Ve são 4.0 Janei o 2006
A Geoin o ma ion App oach o Assessing U ban G ow h Impac on
Fa mlands Using Ea h Obse a ion Da a: A Case S udy o Owe i in
Nige ia
Tochukwu Emmanuel Amaeze
ii
A Geoin o ma ion App oach o Assessing U ban G ow h Impac on
Fa mlands Using Ea h Obse a ion Da a: A Case S udy o Owe i in
Nige ia
Disse a ion supe ised by
P o . Ped o Cab al
(NOVA IMS, Po ugal)
Co-supe ised by
D . Bakh ia Feizizadeh
(Uni e si y o Müns e , Ge many)
&
P o . Filibe o Pla
(UJI, Spain)
Feb ua y 2024
iii
DECLARATION OF ORIGINALITY
I decla e ha he wo k desc ibed in his documen is my own and no om someone else. All
he assis ance I ha e ecei ed om o he people is duly acknowledged and all he sou ces
(published o no published) a e e e enced.
This wo k has no been p e iously e alua ed o submi ed o NOVA In o ma ion
Managemen School o elsewhe e.
Tochukwu Emmanuel Amaeze
[digi al signa u e]
OR
[ he signed o iginal has been a chi ed by he NOVA IMS se ices]
i
ACKNOWLEDGMENTS
I would like o exp ess my deepes g a i ude o P o . Ped o Cab al, my main hesis
supe iso , o his unwa e ing suppo , in aluable guidance, and schola ly insigh
h oughou he en i e esea ch p ocess.
I am also g a e ul o my co-supe iso s, P o . Filibe o Pla and D . Bakh ia
Feizizadeh o hei cons uc i e eedback and hough ul sugges ions, which
g ea ly con ibu ed o he success o his wo k.
I also ex end my app ecia ion o God and o my amily o hei encou agemen ,
unde s anding, and pa ience du ing his demanding pe iod. Thei lo e and suppo
ha e been a cons an sou ce o mo i a ion.
Finally, I would like o acknowledge my iends and colleagues who ha e sha ed
hei expe ise and p o ided encou agemen h oughou his academic jou ney.
This hesis would no ha e been possible wi hou he collec i e suppo and
inspi a ion om all hese indi iduals.
A Geoin o ma ion App oach o Assessing U ban G ow h Impac on
Fa mlands Using Ea h Obse a ion Da a: A Case S udy o Owe i in
Nige ia
ABSTRACT
This s udy in es iga es he impac o apid u ban expansion on ag icul u al land in he
sou heas e n pa o Nige ia, le e aging Cloud compu ing, machine lea ning, emo e sensing,
GIS, and communi y engagemen echniques. The unp eceden ed pace o u baniza ion poses
signi ican challenges o sus ainable de elopmen , pa icula ly h ea ening ood secu i y by
con e ing e ile ag icul u al lands in o u ban a eas. U ilizing ea h obse a ion da a, he
s udy maps u ban g ow h pa e ns and assesses hei e ec s on he con ac ion o a mlands.
Th ough a combina ion o supe ised machine lea ning models—Random Fo es , Suppo
Vec o Machine, and Classi ica ion and Reg ession T ees—applied on Landsa images om
1986, 2000, and 2022, he s udy iden i ies signi ican land co e changes, quan i ying he
ansi ion om a mland o buil -up a eas. The indings e eal a no able inc ease in buil -up
land, om 8.2% in 1986 o 26.6% in 2022, alongside a a iable end in a mland co e age,
ini ially dec easing om 37.3% in 1986 o 27.7% in 2000, hen pa ially eco e ing o33.2%
in 2022, ye no o i s o iginal ex en . This ans o ma ion unde sco es he ension be ween
u ban de elopmen and ag icul u al sus ainabili y, wi h implica ions o local ood p oduc ion
and ecosys em se ices. Communi y engagemen h ough su eys wi h local a me s u he
en iches he analysis, p o iding insigh s in o he socio-economic impac s o u ban sp awl,
including al e ed a ming p ac ices and adap i e s a egies. The s udy unde sco es he need
o in eg a ed u ban planning and policy in e en ions o balance u ban g ow h wi h
ag icul u al p ese a ion, con ibu ing o he b oade discou se on sus ainable de elopmen
and ood secu i y. Recommenda ions include os e ing sus ainable land-use p ac ices,
enhancing communi y adap i e capaci ies, and ad oca ing o policy e o ms o suppo
a ec ed a me s and ensu e long- e m sus ainabili y o bo h u ban and u al landscapes.
This s udy con ibu es o he achie emen o he ollowing sus ainabili y de elopmen goals 2
and 11 which aim a achie ing ze o hunge , and sus ainable ci ies and communi ies
espec i ely.
i
KEYWORDS
Ea h Obse a ion
Geog aphical In o ma ion Sys ems
Machine Lea ning
Remo e Sensing
Spa ial Analysis
Sus ainable De elopmen Goals
ii
ACRONYMS
ANN – A i icial Neu al Ne wo k
CART – Classi ica ion and Reg ession T ee
EO – Ea h Obse a ion
FGD – Focus G oup Discussion
GEE – Google Ea h Engine
GIS – Geog aphic In o ma ion Sys em
LST – Land Su ace Tempe a u e
LULC – Landuse and Landco e
ML – Machine Lea ning
MLH – Maximum Likelihood
NDBI – No malized Di e ence Buil -up Index
NDVI – No malized Di e ence Vege a ion Index
NIR – Nea In a ed
RF – Random Fo es
ROI – Region o In e es
SDGs – Sus ainable De elopmen Goals
SVM – Suppo Vec o Machine
SWIR – Sho Wa e In a ed
iii
INDEX OF THE TEXT
Page
TITLE PAGE ..............................................................................................................ii
DECLARATION OF ORIGINALITY …………..................................................... iii
ACKNOWLEDGMENTS …………......................................................................... i
ABSTRACT ………...................................................................................................
KEYWORDS ……………………………………………………………………… i
ACRONYMS............................................................................................................ ii
INDEX OF TEXT ….………………………………………………………….…. iii
INDEX OF TABLES………………………………………………………….……. x
INDEX OF FIGURES ……………………………………………………………... xi
INDEX OF EQUATONS ………………………………………………………….. xi
1 INTRODUCTION
1.1 Gene al P oblem …….…...…………………………………………………..1
1.2 Rele ance o S udy …………………………………………………………. 2
1.3 Resea ch Objec i es ………….………………………………………….….. 3
2 LITERATURE REVIEW
2.1 Concep o U ban G ow h …….…………………………………………..... 4
2.1.1 Fac o s In luencing U ban G ow h …………………………………………. 5
2.1.2 Challenges o U ban G ow h ………………………………………………...5
2.2 Machine Lea ning & Remo e Sensing ……………………………………… 7
2.3 Rela ed S udies ……………………………………………………………… 8
3 METHODOLOGY
3.1 Me hodology F amewo k.………………………………………………...... 11
3.2 S udy A ea …………………………………………………………………..12
3.3 Da a ………………………………………………………………………… 13
3.3.1 Remo e Sensing Da a ………………………………………………………. 13
3.3.2 Roads and Admin Bounda y ……………………………………………….. 13
3.3.3 Popula ion Da a …………………………………………………………….. 13
3.3.4 Su ey Da a ………………………………………………………………… 13
3.4 Me hods …………………………………………………………………….. 14
ix
3.4.1 Da a Access & P epa a ion in GEE …………………………………………. 14
3.4.2 Spec al Index Calcula ion …………………………………………………... 14
3.4.3 Landuse and Landco e Classi ica ion ……………………………………… 14
3.4.4 LULC Classi ica ion Accu acy Assessmen ………………………………… 17
3.4.5 LULC Change De ec ion Analysis ……………………………...…………... 17
3.4.6 Densi y Map …………………………………………………………………. 18
3.4.7 Bu e Analysis …………………………………………………………….... 18
3.4.8 Su ey ……………………………………………………………………….. 18
4 RESULTS AND DISCUSSION
4.1 Region o In e es Ex ac ion ……………………………………………….. 20
4.2 Classi ica ion and Accu acy Assessmen Resul s ……………………………21
4.3 LULC Change De ec ion …………………………………………………… 21
4.4 No malized Di e ence Vege a ion Index (NDVI) Analysis ………………. 28
4.5 U ban - Fa mland In e ace A eas ………………………………………… 30
4.6 Buil up Expansion Pa e n ………………………………………………… 32
4.7 Su ey ……………………………………………………………………… 31
5 CONCLUSION & RECOMMENDATION …………………………………… 39
REFERENCE ………………………………………………………………………. 41
4
2. LITERATURE REVIEW
This chap e gi es a gene al o e iew o he concep o u ban g ow h, he ac o s ha in luence
u ban g ow h, and he challenge i poses o he en i onmen , and pa icula ly o ag icul u al
lands. Rela ed s udies in is a ea a e men ioned, as well as he concep o machine lea ning
applica ion h ough google ea h engine o ea h obse a ion s udies.
2.1 Concep o U ban G ow h
U ban g ow h e e s o he inc ease in size and popula ion o u ban a eas o e ime. This
phenomenon is d i en by a ious ac o s, including popula ion g ow h, economic
de elopmen , and in as uc u e ini ia i es (Mah uza e al., 2021). Th ough he p ocess o
u baniza ion, a u al a ea ans o ms in o u ban a eas whe e popula ion densi y ela i ely g ows
highe in compa ison o su ounding a eas (Debna h & Amin, 2015). U ban g ow h is a global
phenomenon ha is occu ing in esponse o he inc easing popula ion and he need o
economic and in as uc u al de elopmen (El-Magd e al., 2015). The concep o u ban g ow h
encompasses bo h planned de elopmen and unplanned u ban expansion.
Planned de elopmen e e s o he in en ional physical expansion o u ban a eas in esponse o
popula ion g ow h and in as uc u e needs. This ype o u ban g ow h is ca e ully planned and
o ganized, wi h conside a ions o ac o s such as land use, anspo a ion sys ems, and
ameni ies (Didichenko e al., 2019). Unplanned u ban expansion, on he o he hand, occu s
when ci ies g ow haphaza dly wi hou p ope planning o egula ion. This ype o u ban g ow h
o en esul s in u ban sp awl, which is cha ac e ized by poo ly planned and uncon olled
de elopmen in p e iously un ouched a eas o land (Unplanned U ban De elopmen : A
Neglec ed Global Th ea , n.d). U ban sp awl is a signi ican conce n as i can lead o a ious
nega i e impac s, including inc eased a ic conges ion, loss o g een spaces, and ine icien
land use pa e ns (Fabolude & Aighewi, 2022). I is impo an o dis inguish be ween u ban
g ow h and u ban sp awl as hey ha e di e en implica ions o he sus ainable de elopmen
o ci ies. U ban g ow h, when p ope ly planned and managed, can con ibu e o he
de elopmen and p ospe i y o ci ies. Howe e , i u ban g ow h is cha ac e ized by unchecked
5
sp awl, i can nega i ely impac he en i onmen and quali y o li e o esiden s (T ao e &
Wa anabe, 2017).
2.1.1 Fac o s In luencing U ban G ow h
U ban g ow h is in luenced by a a ie y o ac o s ha con ibu e o he expansion and
de elopmen o u ban a eas. These ac o s include popula ion g ow h, economic de elopmen ,
and in as uc u e ini ia i es (San os e al., 2021). Popula ion g ow h plays a signi ican ole in
u ban g ow h as i leads o an inc eased demand o housing, se ices, and in as uc u e
(Mah uza e al., 2021). As he popula ion expands, ci ies need o accommoda e he g owing
numbe o people and p o ide hem wi h adequa e housing and ameni ies. Economic
de elopmen and in as uc u e ini ia i es also con ibu e o u ban g ow h. As ci ies g ow
economically, hey a ac in es men and de elopmen , leading o he c ea ion o job
oppo uni ies and inc eased economic ac i i y (Tao, 2019).
This economic g ow h, in u n, a ac s mo e people o u ban a eas, ueling u he popula ion
g ow h and u ban expansion. Addi ionally, in as uc u e ini ia i es such as he cons uc ion o
anspo a ion ne wo ks, u ili ies, and public acili ies can also d i e u ban g ow h by
imp o ing accessibili y and enhancing he quali y o li e in ci ies (Al-Alola e al., 2021).
Ano he impo an ac o ha in luences u ban g ow h is land use planning and policies. The
way land is used and he pa e ns o land co e in u ban a eas can signi ican ly impac u ban
g ow h. Fo example, i he e is a lack o e icien land use planning and managemen , i can
lead o haphaza d de elopmen and sp awl (G ekousis & Moun akis, 2015). Unplanned u ban
g ow h can esul in ine icien use o land, inc eased a ic conges ion, and loss o g een
spaces, especially a mlands. Fu he mo e, na u al ac o s such as opog aphy and soil
cha ac e is ics can also con ibu e o spa ial he e ogenei y in u ban g ow h (Al-Bilbisi, 2019).
2.1.2 Challenges o U ban G ow h
While u ban g ow h can b ing economic oppo uni ies and de elopmen , i also p esen s
signi ican challenges o sus ainabili y (Leh ne , 2022). Some o he challenges associa ed
wi h u ban g ow h include:
6
1. Inc eased a ic conges ion: As u ban a eas expand, he numbe o ehicles on he oad also
inc eases. This leads o inc eased a ic conges ion, longe commu e imes, and dec eased ai
quali y due o ehicle emissions.
2. S ain on in as uc u e: Rapid u ban g ow h pu s a s ain on exis ing in as uc u e such as
oads, b idges, wa e supply sys ems, and sewage ea men plan s. The exis ing in as uc u e
may no be able o handle he inc eased demands, leading o inadequa e se ice deli e y and
in as uc u e ailu es.
3. En i onmen al deg ada ion: U ban g ow h can esul in he des uc ion o na u al habi a s,
de o es a ion, and loss o g een spaces. These en i onmen al impac s can ha m biodi e si y,
con ibu e o clima e change, and educe he a ailabili y o na u al esou ces.
4. Social inequali y and socio-economic dispa i y: U ban g ow h o en leads o he
concen a ion o weal h and esou ces in ce ain a eas, while ma ginalized communi ies a e le
behind. They may ace limi ed access o basic se ices and in as uc u e, leading o social
inequali y and socio-economic dispa i y.
5. Inadequa e housing and u ban po e y: Rapid u ban g ow h can lead o a sho age o
a o dable housing, esul ing in he eme gence o slums and in o mal se lemen s in he
pe iphe y o ci ies. These a eas o en lack basic se ices such as wa e , sani a ion, and p ope
housing condi ions, leading o u ban po e y and a lowe quali y o li e o esiden s.
6. Inc eased demand o esou ces: The g owing popula ion and u baniza ion pu inc eased
p essu e on esou ces such as ene gy, wa e , and ood. Mee ing hese demands can lead o
issues such as inc eased ene gy consump ion, pollu ion, ine icien use o esou ces, and s ain
on he ecosys em.
7. Loss o Ag icul u al Lands: U ban g ow h o en esul s in he con e sion o ag icul u al
lands in o u ban a eas, leading o a loss o highly p oduc i e land o ood p oduc ion. This
becomes a h ea o ood secu i y and can con ibu e o ood p ice in la ion and he eliance on
impo ed ood.
This s udy will ocus on he concep o u ban g ow h and i s implica ions on ood secu i y
because o he loss o ag icul u al land.
7
2.2 Machine Lea ning & Remo e Sensing
In ecen yea s, he applica ion o machine lea ning o emo e sensing has gained signi ican
momen um. Machine lea ning algo i hms, such as A i icial Neu al Ne wo ks (ANN), Random
Fo es (RF) and Suppo Vec o Machines (SVM) ha e been widely u ilized o emo e sensing
image classi ica ion (Bai e al., 2016). These algo i hms and o he s ha e p o en o be highly
e ec i e in accu a ely classi ying di e en land co e ypes and ex ac ing aluable
in o ma ion om emo e sensing da a. They ha e become inc easingly popula in he ield o
emo e sensing due o hei abili y o accu a ely classi y land uses, de ec signi ican ea u es,
and minimize classi ica ion e o s (Xue e al., 2021).
The in eg a ion o machine lea ning algo i hms in emo e sensing has e olu ionized he ield,
allowing o mo e p ecise and e icien classi ica ion o land co e ypes and ex ac ion o
aluable in o ma ion om emo e sensing da a (Lee e al., 2020). This in eg a ion has g ea ly
con ibu ed o ad ancemen s in a ious applica ions, such as land use mapping, u ban g ow h
de ec ion, and en i onmen al moni o ing (Ad ances in emo e sensing applica ions o u ban
sus ainabili y, n.d). Fu he mo e, he eme gence o deep lea ning echniques, pa icula ly deep
con olu ional neu al ne wo ks, has u he enhanced he capabili ies o emo e sensing image
classi ica ion (Liu & Shi, 2020).
Deep con olu ional neu al ne wo ks ha e shown ema kable p og ess in emo e sensing image
classi ica ion, including hype spec al image classi ica ion, and ha e achie ed s a e-o - he-a
esul s. These deep lea ning models can au oma ically lea n and ex ac ea u es om emo e
sensing images, allowing o end- o-end classi ica ion lea ning (Zhao & Feng, 2022).
The u u e o machine lea ning in emo e sensing holds g ea p omise. Wi h ad ancemen s in
echnology and he a ailabili y o mo e sophis ica ed algo i hms, i is expec ed ha machine
lea ning will con inue o play a c ucial ole in imp o ing he accu acy and e iciency o emo e
sensing analysis. Fu he mo e, he combina ion o machine lea ning algo i hms wi h o he
echnologies such as big da a analy ics and cloud compu ing will u he enhance he
capabili ies o emo e sensing analysis.
8
2.3 Rela ed S udies
O e he yea s, he e ha e been se e al s udies ha employed Ea h Obse a ion (EO) Remo e
Sensing (RS) and GIS me hods o accessing u ban g ow h, i s pa e ns, ex en , and how i
a ec s he en i onmen . Recen s udies ha e also employed ML me hods he eby p o iding
good esul s especially in he s age o sa elli e image classi ica ion which o ms he basis o
u ban g ow h analysis using he geospa ial app oach.
As LULC o ms he basis o u ban expansion s udies, a s udy by (A e e al., 2023) which
models he LULC changes o El-Fayoum Go e no a e in Egyp ocused on he use o RS, ML,
GIS, and Cloud (GEE) echnologies. Maximum Likelihood (MLH), Random Fo es , and
Suppo Vec o Machine me hods we e employed o classi ica ion a e which he accu acy
esul s we e compa ed, and he SVM me hod p o ided he bes accu acy. Google Ea h Engine
(GEE) was u ilized o p ocess he sa elli e images and hei classi ica ion. The s udy also
showed ha u ban sp awl has occu ed in he s udy a ea, and mos o he enc oachmen s we e
on ag icul u al land. The s udy sugges s ha accu acy o he echniques used o image
classi ica ion can be in luenced by a ious ac o s, which he e o e calls o mo e esea ch and
de elopmen in his ield.
A s udy conduc ed by De ibew (2020) ocused on u ban expansion and i s impac s on pe i-
u ban ag icul u al and o es landscape in he sa elli e own o E hiopia while u ilizing he
objec -based image classi ica ion echnique which was in he pape , a gued o educe spec al
signa u e con usions in LULC classi ica ion. To assess he u ban expansion ex en and pa e n,
he Shannon’s En opy Index was employed. O he s udies like Shenbag aj e al. (2019) used
he index o s udy u ban g ow h in Chennai, India. Singh e al. (2019) also used he index o
analyze u ban sp awl dynamics in RooRkee planning a ea. I measu es he deg ee o spa ial
concen a ion and dispe sal o buil -up a eas o e ime, whe e highe alues o en opy index
sugges mo e dispe sed u ban g ow h. The s udy u he analyzed how u ban a eas ha e
expanded o e he yea s wi h g ea e spa ial gains om ag icul u al and o es land.
Ge u & Bha (2021) in hei s udy, ca ied ou an analysis o he u ban sp awl in Bahi Da ,
E hiopia. They u ilized a combina ion o RS and GIS, including Shannon’s en opy index o
e eal he deg ee o u ban sp awl and pa e n o u ban expansion in he s udy a ea. The
Maximum Likelihood supe ised classi ica ion echnique was employed o sa elli e image
classi ica ion. The s udy also in es iga ed he majo d i ing o ces ha in ensi y u ban
9
expansion and i s sp awl h ough in e iews and ocus g oup discussions (FGD), whe e i was
ealized ha ac o s such as apid popula ion g ow h, land use policy, weak land use plan and
poo law en o cemen , and low p ice o land in he su ounding communi y we e iden i ied as
con ibu ing o u ban sp awl.
(Dolui & Sa ka , 2023) in he assessmen o LULC changes and i s impac on ag icul u al
landscape in pe i-u ban Space o Bolpu own o Wes Bengal, India ca ied ou an in-dep h
analysis o LULC changes o he s udy a ea. The s udy employs GIS ools and suppo ec o
machine (SVM) lea ning algo i hms o assess he ends o u ban g ow h and i s impac on
ag icul u al land o e a 30-yea pe iod om yea 1990 o 2020. This s udy u ilized a mul i-
empo al Landsa da ase and alida ed he classi ied images wi h ac ual GPS da a, achie ing
a sa is ac o y esul wi h mo e han 86% accu acy. The esul s e eal a signi ican con e sion
o ag icul u al land and o es ed a eas in o esiden ial a eas o ou ism and ownship p ojec s.
In a s udy conduc ed by Ranagalage e al. (2021) whe e he u ban pa e ns o ou (4) apidly
g owing Sou h Asian ci ies we e analyzed using Sen inel-2 da a, spa ial me ics we e employed
o e alua e he dominance, composi ion, complexi y, agmen a ion, and connec i i y o u ban
a eas in he ci ies. The esul s e ealed a clea dominance o u ban a eas in all ou ci ies,
p o iding e idence o apidly u banizing landscapes. The s udy also applied a g adien analysis
o cap u e he spa ial dis ibu ion o land use and land co e (LULC) dynamics om he ci y
cen e o u al a eas, while ocusing on he buil -up LULC class, which e lec s he s udy's
ocus on u baniza ion. The pape also discusses he impac s o landscape changes on annual
mean land su ace empe a u e in opical moun ain ci ies. In summa y, he s udy uses a
combina ion o Sen inel-2 da a, machine lea ning models (neu al ne wo ks, k-nea es neighbo ,
andom o es , and suppo ec o machine) g adien analysis, and spa ial me ics o examine
he LULC dynamics and de ec he u baniza ion pa e ns o he ou ci ies unde s udy.
The s udy conduc ed by (Samiullah e al., 2019) ocused on he e alua ion o u ban
enc oachmen on a mland and i s implica ions o u ban ag icul u e. I aimed a unde s anding
he cha ac e is ics o u ban ag icul u e, examine he dynamics o spa ial change in a ming
land, and in es iga e he ac o s esponsible o u ban enc oachmen on ag icul u al land. Da a
sou ces, including land capabili y maps, e enue eco ds, opog aphic maps, sa elli e images,
ocus g oup discussions, c op maps, image classi ica ion, and land use maps allowed hem o
analyze he ans o ma ion o a mland o e ime. The s udy e eals ha o e hal o he ci y
10
dis ic 's a ea is occupied by cul i a ed land, wi h a mland sp eading a ound he co e buil -up
a ea in all di ec ions which highligh s he dominance o ag icul u e in he egion and he
po en ial impac o u ban enc oachmen on ag icul u al land and explo ed he policy and
planning implica ions o u ban enc oachmen on ag icul u al land.
In a s udy conduc ed by (Mos a a e al., 2021) highligh he consequences o u ban g ow h ,
pa icula ly in de eloping coun ies whe e u baniza ion is o en andom and uncon olled. The
s udy ocuses on he decision-making p ocess o u ban planning, speci ically in he con ex o
eigh dis ic s in Egyp . The decision-making p ocess is based on i e c i e ia: popula ion
g ow h, employmen , local de elopmen , a ea, and socio-economic condi ions. I also
in oduces a machine lea ning based and emo e sensing-based me hodology o con inuous
moni o ing o land use and land co e (LULC). The s udy also highligh s he impo ance o
he Fuzzy Technique o O de o P e e ence by Simila i y o Ideal Solu ion (TOPSIS) me hod,
which elimina es unce ain ies in he da a and analysis o p oduce he u baniza ion isk map.
This map helps de e mine he suscep ibili y o he dis ic s o u baniza ion isk, especially in
cases whe e sa elli e images a e una ailable.
The s udy conduc ed by (Rahman e al., 2023) p o ided an in-dep h analysis o he impac s o
u ban a ea expansion on ag icul u al land and he he mal en i onmen in La kana, Pakis an.
The au ho s used he Random Fo es (RF) algo i hm and Google Ea h Engine (GEE) o assess
changes in land use and land co e (LULC) and land su ace empe a u e (LST) om 1990 o
2020. The esea ch ound ha he buil -up a ea in La kana inc eased signi ican ly o e he 30-
yea pe iod, which was la gely due o na u al popula ion g ow h and uncon olled mig a ion
om su ounding u al a eas, which esul ed in a haphaza d inc ease in he buil -up a ea o e
he ag icul u al land. The au ho s also no ed ha his unplanned u ban expansion led o a ious
en i onmen al and social issues, including conges ion and he enc oachmen o open a eas, and
land su ace empe a u e.
The e iewed pape s/s udies p o ide g ea insigh s on he exis ing body o li e a u e on u ban
expansion and i s impac s on ag icul u al land.
11
3. METHODOLODY
3.1 Me hodology F amewo k
This esea ch is based mainly on he in eg a ion o Cloud compu ing, Machine Lea ning,
Remo e Sensing, GIS, and Communi y engagemen echniques o assess he impac o u ban
expansion on a mlands wi hin a gi en s udy a ea. A e iden i ying he iable esea ch
p oblem, he me hodology o employ, equi ed da a, and a ailable sou ce we e iden i ied.
These will be discussed in his chap e while also elabo a ing on he p ocess o analysis and
esul s p esen a ion.
Figu e 1: Me hodology Wo k low Cha
12
3.2 S udy A ea
The s udy a ea is made up o h ee local go e nmen a eas, Owe i Municipal, Owe i Wes ,
and Owe i No h in Imo S a e, which lies in he sou h-eas e n geo-poli ical zone o Nige ia.
These local go e nmen a eas make up he capi al ci y o Imo s a e and a e gene ally e e ed
o as Owe i.
Figu e 2: Map Showing S udy A ea
I is loca ed a 5°29'06.0"N, 7°02'06.0"E, wi h a p ojec ed popula ion o 524,000 people as a
he yea 2022. This popula ion g ew om 289,721 in he yea 1991 o 403,425 in he yea 2006
popula ion census. The mean annual empe a u e eco ded in Owe i is 25.9 °C | 78.6 °F, while
he p ecipi a ion is abou 2412 mm | 95.0 inch pe annum (Clima e Da a, n.d.).
13
3.3 Da a
3.3.1 Remo e Sensing Da a
A se o p e-p ocessed and spa ially e e enced mul i spec al and mul i- empo al image ies
we e used. The sa elli e images a e o he Landsa se ies; Landsa 5 TM (Thema ic Mappe )
image o he yea 1986; he Landsa 7 ETM+ (Enhanced Thema ic Mappe Plus) o he yea
2000; he Landsa 9 OLI (Ope a ional Land Image ) o yea 2022, all being su ace e lec ance
images and ha ing a spec al esolu ion o 30m. They we e ob ained h ough he Ea h Explo e
pla o m o he Uni ed S a es Geological Su ey.
3.3.2 Roads and Admin Bounda y
Le el 3 adminis a i e bounda y shape ile o Nige ia he s udy a ea was ob ained om he
O ice o he Su eyo Gene al o he Fede a ion, h ough he da a pla o m o he Uni ed
Na ions O ice o Coo dina ion o Humani a ian A ai s. Road Ne wo k da a was ob ained
om Open S ee Map h ough he GeoFab ik da a pla o m.
3.3.3 Popula ion Da a
Popula ion da a o he yea s 1991 and 2006 we e acqui ed om he Nige ian Na ional
Popula ion Commission h ough he Na ional Bu eau o S a is ics annual abs ac o s a is ics
publica ion o he yea 2010. P ojec ed popula ion o he yea 2022 was ob ained om he
Ci y Popula ion da abase.
3.3.4 Su ey Da a
To unde s and how a me s eel abou he u ban expansion phenomena in hei communi ies,
a su ey was conduc ed. This was achie ed by going di ec ly o he selec ed communi ies and
engaging wi h a ew a me s o hea om hem.
20
4. RESULTS AND DISCUSSION
4.1 Region o In e es Ex ac ion
Th ough he Google Ea h Engine (GEE) cloud pla o m, sa elli e images o he yea s 1986,
2000, and 2022 om Landsa 5 TM, Landsa 7 ETM+, and Landsa 9 OLI espec i ely. False
colou composi e images we e displayed by combining some bands o he images; bands 5,4,3
o Landsa 5 and 7, and 6,5,4 o Landsa 9. The geo e e enced bounda y shape ile o he s udy
a ea was impo ed o GEE and used o mask ou he s udy a ea om he image scenes as shown
in igu es 4, 5, and 6 below.
Figu e 5: Landsa 7 ETM+ (2000)
Figu e 6: Landsa 9 OLI (2022)
Figu e 4: Landsa 5 TM (1986)
21
4.2 Classi ica ion and Accu acy Assessmen Resul s
A e ca ying ou classi ica ion o he images o di e en yea s in GEE using di e en
machine lea ning models – Random Fo es (RF), Suppo Vec o Machine (SVM), and
Classi ica ion and Reg ession T ee (CART), he bes classi ica ion esul s we e chosen based
on he accu acy me ics. SVM had he be e esul s o he 1986 and 2000 classi ica ion while
CART pe o med be e o he yea 2022.
Yea 1986
Yea 2000
Yea 2022
RF
SVM
CART
RF
SVM
CART
RF
SVM
CART
O e all
Accu acy
0.975
0.979
0.969
0.973
0.985
0.984
0.983
0.989
0.994
Kappa
Coe icien
0.967
0.973
0.959
0.970
0.980
0.982
0.979
0.986
0.993
Table 4: Accu acy Resul o ML Classi ica ion Algo i hms
Using he e o ma ices de i ed om he es ing samples, o e all accu acy and he kappa
coe icien o he classi ica ion we e calcula ed and shown in ables 5, 6, and 7 below.
CLASS
1986
Wa e body
Fa mland
Vege a ion
Buil up
Land
Ba eland
Tes
To al
Wa e body
18
0
0
0
0
18
Fa mland
0
48
0
0
0
48
Vege a ion
0
0
58
0
0
58
Buil up
Land
0
0
0
48
1
49
Ba eland
0
1
0
2
18
21
To al
18
49
58
50
19
194
O e all
Accu acy
0.979
Kappa
Coe icien
0.973
Table 5: E o ma ix and accu acy esul s o 1986 image classi ica ion.
22
CLASS
2000
Wa e body
Fa mland
Vege a ion
Buil up
Land
Ba eland
Tes
To al
Wa e body
28
0
0
0
0
28
Fa mland
0
43
0
1
0
44
Vege a ion
0
1
50
0
0
51
Buil up
Land
0
0
1
57
0
58
Ba eland
0
0
0
0
19
19
To al
28
44
51
58
19
200
O e all
Accu acy
0.985
Kappa
Coe icien
0.980
Table 6: E o ma ix and accu acy esul s o 2000 image classi ica ion.
CLASS
2022
Wa e body
Fa mland
Vege a ion
Buil up
Land
Ba eland
Tes
To al
Wa e body
14
0
1
0
0
15
Fa mland
0
52
0
0
0
52
Vege a ion
0
0
56
0
0
56
Buil up
Land
0
0
0
46
0
46
Ba eland
0
0
0
0
21
21
To al
14
52
57
46
21
190
O e all
Accu acy
0.994
Kappa
Coe icien
0.993
Table 7: E o ma ix and accu acy esul s o 2022 image classi ica ion.
An o e all accu acy o 0.979, 0.985, and 0.994 achie ed in he image classi ica ions ep esen s
a sa is ac o y esul . Also, a Kappa coe icien o 0.973, 0.980, and 0.993 e lec s a be e
ag eemen be ween he classi ica ion and he eali y based on he es ing sample used o he
accu acy classi ica ion.
23
4.3 LULC Change De ec ion
Classi ica ion esul s show signi ican changes o LULC makeup in he s udy a ea o e he
pe iod unde s udy. T ansi ions om landuse o ano he ha e also been no iced o e he yea s.
Table 8 below shows mo e de ails o he LULC makeup o each yea .
Table 8: LULC Makeup o di e en yea s in Sq.km
Figu e 7: Cha o LULC makeup (1986 - 2022)
LULC
Classes
Yea 1986
%
Yea 2000
%
Yea 2022
%
Wa e body
1.172
0.218
0.838
0.156
2.228
0.415
Fa mland
200.233
37.304
148.956
27.751
178.449
33.245
Vege a ion
286.366
53.350
325.896
60.715
196.608
36.628
Buil up Land
44.492
8.289
53.559
9.978
143.055
26.651
Ba eland
4.504
0.839
7.516
1.400
16.426
3.060
To al
536.7654
536.7654
536.7654
0
50
100
150
200
250
300
350
Wa e body Fa mland Vege a ion Buil up Land Ba eland
A ea (SqKm)
LULC Makeup (1986 - 2022)
Yea 1986 Yea 2000 Yea 2022
24
As shown in he able abo e, Buil up Land has con inuously inc eased om 44.492 sq.km in
he yea 1986 (8.2%) o 53.559 sq.km (9.9%) in 2000, and o 143.055 sq.km (26.6%) in 2022.
Be ween he yea 1986 and 2000, Fa mland dec eased om 200.232sq.km (37.3%) o
148.956sq.km (27.7%), and i inc eased o 178.4Sq.km (33.2%) in he yea 2022. Wa e body
and Ba eland show a gene al inc ease o e he yea s while Vege a ion shows a gene al dec ease
o e he yea s.
Figu e 8: LULC Map o 1986
25
Figu e 9: LULC Map o 2000
Figu e 10: LULC Map o 2022
26
LULC ansi ion in o ma ion ob ained by o e laying and in e sec ing he classi ica ion esul s
o he di e en yea s p o ides he de ails on unde s anding wha has changed om one landuse
o ano he o e he yea s.
Yea 2022 in Sq.km
Ba eland
Buil up
Land
Fa mland
Vege a ion
Wa e body
G and
To al
Yea 1986 in Sq.km
Ba eland
1.164
2.623
0.603
0.046
0.065
4.503
Buil up
Land
3.229
35.213
4.888
0.979
0.180
44.491
Fa mland
8.205
59.544
78.575
53.471
0.435
200.232
Vege a ion
3.810
45.465
94.314
141.760
1.014
286.365
Wa e body
0.016
0.207
0.067
0.349
0.531
1.171
G and
To al
16.425
143.055
178.449
196.607
2.227
536.765
Table 9: LULC T ansi ion Table (1986 – 2022)
Ba eland
(%)
Buil up Land
(%)
Fa mland
(%)
Vege a ion
(%)
Wa e body
(%)
Ba eland
25.85
58.25
13.38
1.03
1.45
Buil up Land
7.25
79.14
10.98
2.2
0.4
Fa mland
4.09
29.73
39.24
26.7
0.21
Vege a ion
1.33
15.87
32.93
49.5
0.35
Wa e body
1.38
17.74
5.76
29.8
45.31
Table 10: Pe cen age LULC T ansi ion Table (1986 – 2022)
T ansi ion ables abo e show he ex en o LULC con e sions ha ha e aken place wi hin he
s udy pe iod in he s udy a ea. Fa mland con e ed mos o Buil up land by 29.7%, ollowed
by con e sion o Vege a ion by 26.7%. I can also be seen ha Vege a ion is he highes
con ibu o o a mland expansion o e he yea s. I con ibu es a bi mo e han i akes om
he a mland a ea. This shows he u ban expansion is he majo land use ha akes up Fa mland
spaces in he s udy a ea.
2022
1986
27
Figu e 11: Map o LULC T ansi ions o Buil up Land
Figu e 12: Map o LULC T ansi ions o Fa mland
28
The ansi ion maps abo e show ha g ea e a eas o a mlands close o he u ban a eas in he
yea 1986 we e con e ed o buil up land in he yea 2022. Looking a he ansi ions o
a mlands, he map also shows ha a ming ac i i ies ad anced mo e in o he u al a eas while
aking up mos ly ege a ion a eas among o he landco e .
4.4 No malized Di e ence Vege a ion Index (NDVI) Analysis
NDVI, de i ed in his s udy se es as an indica o o ege a ion heal h and densi y. In he
con ex o ou s udy, NDVI analysis p o ides a quan i a i e means o assess changes in land
co e , especially he enc oachmen o u ban de elopmen s in o ag icul u al a eas. In he
con ex o his s udy, i p o ided al e a ions in ege a ion pa e ns associa ed wi h u ban
expansion.
Figu e 13: NDVI Maps o 1986 and 2000
29
Figu e 14: NDVI Maps o 2022
Figu e 15: NDVI Time Se ies cha 1986 – 2022
A dec ease in NDVI o e he yea s as shown in igu e 15 abo e signi ies he con e sion o
e ile ag icul u al land in o impe ious su aces, o e ing a angible jus i ica ion o a mland
loss. And i enables he iden i ica ion o a mlands expe iencing s ess o deg ada ion due o
u baniza ion, aiding in he iden i ica ion o ulne able zones.
0
0.05
0.1
0.15
0.2
0.25
0.3
1986-12-19 2000-12-17 2022-12-30
NDVI Time Se ies o he S udy A ea
36
Dis inguishing be ween subsis ence and comme cial a ming p o ided a undamen al
unde s anding o he economic o ien a ion o he economic dynamics in he s udy a ea, and
how i impac s di e en ace s o he communi y’s li elihood. A g ea e pe cen age o he
a me s p ac ice subsis ence a ming, so i can be a gued ha a mland loss due o u ban
expansion in he s udy a ea may ha e di ec implica ions on household ood secu i y.
Figu e 25: Cha showing a me s ha ha e expe ienced loss.
Fa m yield is di ec ly linked o he economic well-being o a me s and he communi y.
Changes in yield, whe he posi i e o nega i e, ha e inancial implica ions o he a me s and
he communi y a la ge. A g ea e numbe o a me s (13 o 22) epo ed an inc ease in a m
yields, while 4 epo ed a dec ease, and he o he s expe ienced insigni ican changes. The
inc ease in yield i espec i e o he u ban expansion and i s enc oachmen in o a mlands
shows ha a me s implemen ed adap i e s a egies o cope wi h he impac .
Figu e 26: Cha showing a me s who acqui ed new a mland.
No
23%
Yes
77%
Fa me s Who Acqui ed New Fa mland
0
2
4
6
8
10
12
14
Dec eased Inc eased No Signi ican
Coun
Fa me s Tha Expe ienced Change in Fa m Yields
37
Acquisi ion o new a mland is an adap i e s a egy o cushion he impac o a mland loss. 77
pe cen o pa icipa ing a me s acqui ed new a mlands o make up o hei a mland loss o
u ban expansion. This helps us unde s and ha a me s a e ac i ely aking p oac i e measu es
o coun e he impac o u ban expansion.
Figu e 27: Fa me s who employed new a ming ac i i ies.
Fa me s employing new a ming p ac ices also shows he adap i e capaci y o he a me s
which indica es esilience and inno a ion in he ace o challenges posed by u ban expansion
in o he communi ies. Among he a me s in e iewed, 72.7% a es ed inco po a ing new
p ac ices, om he use o mechanized a ming, use o e ilize s, o iding, among o he s.
Conside ing he in o ma ion om he su ey abou he inc ease o a m yields, i means ha
hese p ac ices a e p o ing success ul. Policies can be ailo ed o suppo and incen i ize he
adop ion o hese s a egies ac oss he b oade communi y while also in oducing mode n
s a egies.
Fu he mo e, on he su ey esul , en (10) ou o he wen y- wo (22) a me s a gued ha he
go e nmen does no p o ide suppo o a ec ed a me s; six (6) said hey a e no su while he
emaining posi ed ha a ec ed a me s ecei e such suppo . This p o ides insigh s in o he
adequacy o exis ing suppo sys ems and iden i ies a eas whe e imp o emen is needed.
0
1
2
3
4
5
6
7
8
No Yes Yes -
Mechanized
Fa ming
Yes - Ridging
and Use o
Animal
Dungs
Yes - Use o
Animal
Dungs
Yes - Use o
Fe ilize s
Yes - Use o
Imp o ed
C ops
Yes - Use o
O ganic
Manu e
Fa me s Who Employed New Fa ming P ac ices
38
Asking he a me s o how hey a e conce ned abou he u u e o a ming in he communi ies,
77 % epo ed ha hey a e e y conce ned while 23% epo ed ha hey a e somewha
conce ned. This in o ma ion gauges he a me s' le el o conce n abou he sus ainabili y o
a ming in he long e m.
On whe he he a me s a e happy wi h u ban expansion going on in he s udy a ea, ele en (11)
o wen y- wo (22) posi ha hey a e e y happy, while i e (5) said hey a e “no happy”. The
emaining said hey a e “somewha happy”. This in o ma ion explo es he emo ional and
a i udinal aspec s o a me s' esponses. I cap u ed a holis ic iew o hei sen imen s owa ds
u ban expansion, conside ing no only he challenges bu also po en ial bene i s o posi i e
pe cep ions.
In gene ali y he app oach o communi y engagemen aligns wi h p inciples o pa icipa o y
and sus ainable de elopmen , p omo ing collabo a i e decision-making o he bene i o bo h
u ban and u al popula ions.
39
5. CONCLUSION & RECOMMENDATION
The use o Cloud-based machine lea ning app oaches demons a ed e iciency in moni o ing
and mapping u ban expansion pa e ns in he s udy a ea and i s impac on a mlands, le e aging
Ea h obse a ion da a. The machine lea ning me hodologies employed ha e no only enhanced
he p ecision o ou analysis bu ha e also se he s age o u u e esea ch endea o s in he
ealm o ea h obse a ion s udies.
The iden i ica ion and mapping o u ban expansion using geoin o ma ion ools ha e yielded
insigh ul esul s. The spa ial ex en o u ban g ow h, quan i ica ion o a mland loss, and
isualiza ion o he e ol ing landscape we e success ully assessed. The spa ial analysis has
p o ided a comp ehensi e o e iew o he impac on ag icul u al land, allowing o he
iden i ica ion o ulne able zones and he assessmen o changing a ming p ac ices. A
comp ehensi e unde s anding o he e ol ing u ban expansion landscape and ends o he
s udy a ea we e d awn h ough his s udy whe e i was no iced ha s ip/linea along he oads
and sca e ed/leap og pa e ns o u ban expansion a e aking place, which can be e e ed o
as an u ban sp awl.
Reduc ion o a mland a ea o e he pe iod o s udy was no iced, which has no able
consequences on local ood p oduc ion and long- e m ood secu i y. The decline in a e age
NDVI alue o he s udy a ea o e he yea s is an indica ion ha g een a eas including
a mlands a e expe iencing s ess, deg ada ion, o comple e con e sion in o impe ious
su aces.
This s udy has also highligh ed he quali a i e dimension o a me s’ conce ns. In eg a ion o
a me s su ey added a human pe spec i e owa ds unde s anding he impac o u ban
expansion in he s udy a ea, he eby se ing as a b idge be ween scien i ic inqui y and he li ed
expe iences o he local a me s. The ou comes o his esea ch ha e di ec ele ance o u ban
planning and policymaking while emphasizing he need o sus ainable de elopmen s a egies.
The s udy ecommends ailo ed in e en ions, sus ainable land-use planning, communi y-
speci ic adap i e s a egies, economic suppo p og ams, policy ad ocacy o go e nmen
suppo , and long- e m sus ainabili y p og ams o os e ha monious coexis ence be ween
ag icul u e and u ban de elopmen . To gain a de ailed unde s anding o he implica ion on
40
a ming ac i i ies in he s udy a ea, i is ecommended ha subsequen public pa icipa ion
in e en ions should each ou o much la ge numbe o a me s as ha will p o ide mo e
in o ma ion o be e unde s anding o he economic impac o u ban expansion on he s udy
a ea.
41
Bibliog aphic Re e ence
A Cen u y o Moni o ing U ban G ow h in Meno ya Go e no a e, Egyp , Using Remo e
Sensing and Geog aphic In o ma ion Analysis. (2015). Re ie ed Feb ua y 10, 2024,
om h ps://www.sci p.o g/jou nal/pape in o ma ion?pape id=58837
Al-Alola, S. S., Alogayell, H. M., Alkadi, I. I., Mohamed, S. A., & Ismail, I. Y. (2021).
Recogni ion and P edic ion o Land Dynamics and I s Associa ed Impac s in Al-
Qu ayya Ci y and along Al-Shamal T ain Pa hway in Saudi A abia. Sus ainabili y,
13(17), A icle 17. h ps://doi.o g/10.3390/su13179913
Al-Bilbisi, H. (2019). Spa ial Moni o ing o U ban Expansion Using Sa elli e Remo e
Sensing Images: A Case S udy o Amman Ci y, Jo dan. Sus ainabili y, 11(8), A icle 8.
h ps://doi.o g/10.3390/su11082260
Alha hi, B., & El-Dama y, T. A. (2022). S udy he U ban Expansion o Tai Ci y Using
Remo e Sensing and GIS Techniques o Decision Suppo Sys em. Ad ances in
Remo e Sensing, 11(1), A icle 1. h ps://doi.o g/10.4236/a s.2022.111001
A e , I., Ahmed, W., & Abdel-Maguid, R. H. (2023). Modelling o land use land co e
changes using machine lea ning and GIS echniques: A case s udy in El-Fayoum
Go e no a e, Egyp . En i onmen al Moni o ing and Assessmen , 195(6), 637.
h ps://doi.o g/10.1007/s10661-023-11224-7
Bai, T., Li, D., Sun, K., Chen, Y., & Li, W. (2016). Cloud De ec ion o High-Resolu ion
Sa elli e Image y Using Machine Lea ning and Mul i-Fea u e Fusion. Remo e Sensing,
8(9), 715. h ps://doi.o g/10.3390/ s8090715
Busho, S. W., Wendimagegn, G. T., Mule a, A. T. (2021). Quan i ying Spa ial Pa e ns o
U baniza ion: G ow h Types, Ra es, and Changes In Addis Ababa Ci y F om 1990 To
2020. Spa . In . Res., 5(29), 699-713. h ps://doi.o g/10.1007/s41324-021-00388-4
Da ids, R., Rouge , M., Boon, R., Robe s, D. (2018). Spa ial Analyses o Th ea s To
Ecosys em Se ice Ho spo s In G ea e Du ban, Sou h A ica. Pee J, (6), e5723.
h ps://doi.o g/10.7717/pee j.5723
42
Da ids, R., Rouge , M., Boon, R., Robe s, D. (2018). Spa ial Analyses o Th ea s To
Ecosys em Se ice Ho spo s In G ea e Du ban, Sou h A ica. Pee J, (6), e5723.
h ps://doi.o g/10.7717/pee j.5723
A geog aphic in o ma ion sys em-based logical u ban g ow h model o p edic ing spa ial
g ow h o an u ban a ea—Ripan Debna h, ATM Nu ul Amin, 2016. (n.d.). Re ie ed
Feb ua y 10, 2024, om
h ps://jou nals.sagepub.com/doi/abs/10.1177/0265813515618565?jou nalCode=epba
De ibew, K. T. (2020). Spa io empo al analysis o u ban g ow h on o es and ag icul u al
land using geospa ial echniques and Shannon en opy me hod in he sa elli e own o
E hiopia, he wes e n inge o Addis Ababa ci y. Ecological P ocesses, 9(1). Scopus.
h ps://doi.o g/10.1186/s13717-020-00248-3
Dhana aj, K., & Angadi, D. P. (2022). Land Use Land Co e Mapping and Moni o ing U ban
G ow h Using Remo e Sensing and GIS Techniques in Mangalu u, India. GeoJou nal,
87, 1133–1159. h ps://doi.o g/10.1007/s10708-020-10302-4
Didichenko, M., Bulakh, I., & Kozako a, O. (2019). Spa ial and Tempo al P inciples and
Me hods o he His o ical U ban En i onmen Composi ion T ans o ma ions. U ban
and Regional Planning, 4(4), A icle 4. h ps://doi.o g/10.11648/j.u p.20190404.13
Dolui, S., & Sa ka , S. (2023). Assessmen o LULC Changes and I s Impac on Ag icul u al
Landscape in Pe i-u ban Space o Bolpu Town, Wes Bengal (India). In U. Cha e jee,
N. Bandyopadhyay, M. D. Se iawa i, & S. Sa ka (Eds.), U ban Commons, Fu u e
Sma Ci ies and Sus ainabili y (pp. 341–373). Sp inge In e na ional Publishing.
h ps://doi.o g/10.1007/978-3-031-24767-5_16
Fabolude, G., & Aighewi, I. T. (2022). E alua ion o he Ex en o Land Use-Land Co e
Changes o Benin Ci y, Edo S a e, Nige ia om 1987-2019. Jou nal o Applied
Sciences and En i onmen al Managemen , 26(8), A icle 8.
h ps://doi.o g/10.4314/jasem. 26i8.18
Geo ab ik Download Se e . (n.d.). Re ie ed Feb ua y 17, 2024, om
h ps://download.geo ab ik.de/a ica/nige ia.h ml
Ge u, K., & Bha , H. G. (2021). Analysis o spa io- empo al dynamics o u ban sp awl and
g ow h pa e n using geospa ial echnologies and landscape me ics in Bahi Da ,
No hwes E hiopia. Land Use Policy, 109, 105676.
h ps://doi.o g/10.1016/j.landusepol.2021.105676
43
Home | Na ional Bu eau o S a is ics. (n.d.). Re ie ed Feb ua y 17, 2024, om
h ps://www.nige ians a .go .ng/
Kadhim, N., Mou shed, M., & B ay, M. (2016). Ad ances in emo e sensing applica ions o
u ban sus ainabili y. Eu o-Medi e anean Jou nal o En i onmen al In eg a ion, 1(1),
7. h ps://doi.o g/10.1007/s41207-016-0007-4
Landsa Missions | U.S. Geological Su ey. (n.d.). Re ie ed Feb ua y 17, 2024, om
h ps://www.usgs.go /landsa -missions
Lee, J., Han, D., Shin, M., Im, J., Lee, J., & Quackenbush, L. J. (2020). Di e en Spec al
Domain T ans o ma ion o Land Co e Classi ica ion Using Con olu ional Neu al
Ne wo ks wi h Mul i-Tempo al Sa elli e Image y. Remo e Sensing, 12(7), 1097.
h ps://doi.o g/10.3390/ s12071097
Liu, S., & Shi, Q. (2020). Mul i ask Deep Lea ning Wi h Spec al Knowledge o
Hype spec al Image Classi ica ion. IEEE Geoscience and Remo e Sensing Le e s,
17(12), 2110–2114. h ps://doi.o g/10.1109/LGRS.2019.2962768
Mos a a, E., Li, X., Sadek, M., & Dossou, J. F. (2021). Moni o ing and Fo ecas ing o U ban
Expansion Using Machine Lea ning-Based Techniques and Remo ely Sensed Da a: A
Case S udy o Gha bia Go e no a e, Egyp . Remo e Sensing, 13(22), 4498.
h ps://doi.o g/10.3390/ s13224498
Nige ia—Adminis a i e bounda ies—Humani a ian Da a Exchange. (n.d.). Re ie ed
Feb ua y 17, 2024, om h ps://da a.humda a.o g/da ase /nige ia-admin-le el-0-1-
and-2-adminis a i e-bounda ies
Nige ia: Adminis a i e Di ision (S a es and Local Go e nmen A eas)—Popula ion
S a is ics, Cha s and Map. (n.d.). Re ie ed Feb ua y 17, 2024, om
h ps://www.ci ypopula ion.de/en/nige ia/admin/
Rahman, G., Chandio, N. H., Moazzam, M. F. U., & Al Ansa i, N. (2023). U ban expansion
impac s on ag icul u al land and he mal en i onmen in La kana, Pakis an. F on ie s
in En i onmen al Science, 11.
h ps://www. on ie sin.o g/a icles/10.3389/ en s.2023.1115553
Samiullah, Khan, M. A., Rahman, A.-U., & Mahmood, S. (2019). E alua ion o u ban
enc oachmen on a mland: A h ea o u ban ag icul u e in Peshawa Ci y Dis ic ,
44
Pakis an. E dkunde, 73(2), 127–142. Scopus.
h ps://doi.o g/10.3112/e dkunde.2019.02.04
Singh A, Mai han S, Begum A, Kuma V (2019) Analysis o he u ban sp awl dynamics using
Shannon’s en opy in RooRkee planning a ea. In J Tech Inno Model Eng 5(2):1–11
S e an Leh ne . (2022). Sus ainable and Fu u e-O ien ed U ban De elopmen in Vienna &
Ba celona. In e na ional Jou nal o Sus ainable De elopmen Resea ch, 8(2), 25-32.
h ps://doi.o g/10.11648/j.ijsd .20220802.11
Supe ised Classi ica ion | Google Ea h Engine. (n.d.). Google o De elope s. Re ie ed
Feb ua y 9, 2024, om h ps://de elope s.google.com/ea h-engine/guides/classi ica ion
Sus ainable De elopmen unde Popula ion P essu e: Lessons om De eloped Land
Consump ion in he Con e minous U.S. | PLOS ONE. (2015). Re ie ed Feb ua y 10,
2024, om h ps://jou nals.plos.o g/plosone/a icle?id=10.1371/jou nal.pone.0119675
Tao, Z. (2019). Resea ch on he Deg ee o Coupling be ween he U ban Public In as uc u e
Sys em and he U ban Economic, Social, and En i onmen al Sys em: A Case S udy in
Beijing, China. Ma hema ical P oblems in Enginee ing, 2019, e8206902.
h ps://doi.o g/10.1155/2019/8206902
T ao e, A., & Wa anabe, T. (2017). Modeling De e minan s o U ban G ow h in Conak y,
Guinea: A Spa ial Logis ic App oach. U ban Science, 1(2), A icle 2.
h ps://doi.o g/10.3390/u bansci1020012
Ribei o, F. L. (2021). Unplanned U ban De elopmen : A Neglec ed Global Th ea . Cu en
U ban S udies, 9(3), A icle 3. h ps://doi.o g/10.4236/cus.2021.93027
Liu, X., Fa oyinbo, T. E., Thomas, N. M., Guan, W. W., Zhan, Y., Mondal, P., Lagomasino,
D., Sima d, M., T e in, C. C., Deo, R., & Ba enbli , A. (2021). La ge-Scale High-
Resolu ion Coas al Mang o e Fo es s Mapping Ac oss Wes A ica Wi h Machine
Lea ning Ensemble and Sa elli e Big Da a. F on ie s in Ea h Science, 8.
h ps://www. on ie sin.o g/a icles/10.3389/ ea .2020.560933
Remo e Sensing Image Wa e Body Recogni ion Algo i hm Based on Deep Con olu ion
Gene a ing Ne wo k and Combined Fea u es. (2022). Re ie ed Feb ua y 10, 2024,
om h ps://www.hindawi.com/jou nals/wcmc/2022/9932251/
45
……