Volume-08 Issue 10, Oc obe - 2024 ISSN: 2456-9348
Impac Fac o : 7.936
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [91]
LAND USE CHANGE AND PLANNING COMPLIANCE IN LAGOS: ASSESSING
ZONING RULES AND DRIVERS OF NON-COMPLIANCE
Kayode A. Adepa usi
Depa men o Ea h, A mosphe ic, & Geog aphic In o ma ion Sciences (EAGIS)
Wes e n Illinois Uni e si y, USA.
kayodeAdepa [email protected]
ABSTRACT
Rapid u baniza ion in Lagos has p oduced p o ound shi s in land use, o en di e ging om o icially designa ed
zoning ules. This s udy in es iga es he na u e and ex en o land use change be ween 2000 and 2025 and examines
he socio-economic, ins i u ional, and spa ial d i e s o zoning non-compliance. Using supe ised classi ica ion o
Landsa 8 and 9 and Sen inel-2 ime se ies, pos -classi ica ion change de ec ion echniques, and o e lays wi h o icial
zoning maps and building pe mi eco ds, we iden i y bo h complian and non-complian land de elopmen s. Spa ial
econome ic models a e employed o analyze he de e minan s o non-compliance ac oss Lagos’ local go e nmen
a eas. Findings e eal ha land use con e sion om we lands and pe i-u ban ag icul u al zones in o esiden ial and
mixed-use de elopmen s is he dominan end, wi h nea ly 42% o obse ed de elopmen s occu ing ou side zoning
alloca ions. D i e s o non-compliance include weak en o cemen capaci y, high popula ion densi y, p oximi y o
anspo co ido s, and specula i e land ma ke s. The s udy unde sco es he c i ical need o mo e adap i e u ban
planning amewo ks, enhanced en o cemen , and he in eg a ion o emo e sensing and geospa ial moni o ing in
policy implemen a ion.
Keywo ds:
Land use change, zoning compliance, Lagos, supe ised classi ica ion, spa ial econome ics, u ban planning
1. INTRODUCTION
1.1 Backg ound and Ra ionale
U ban land use planning is a co ne s one o sus ainable ci y managemen . Zoning egula ions, which speci y he
alloca ion o land o esiden ial, comme cial, indus ial, and conse a ion uses, a e designed o p omo e o de ly
de elopmen , mi iga e con lic s be ween incompa ible ac i i ies, and sa egua d en i onmen al esou ces. Howe e , in
many apidly u banizing A ican ci ies, zoning compliance is unde mined by in o mal se lemen s, weak go e nance,
and specula i e p essu es on land ma ke s [1,2].
Lagos, Nige ia’s comme cial hub, epi omizes he pa adox o apid u ban g ow h amid agile planning en o cemen .
Wi h an es ima ed popula ion exceeding 20 million esiden s, Lagos has expanded d ama ically o e he pas h ee
decades [3]. D i en by u al-u ban mig a ion, na u al popula ion inc ease, and economic concen a ion, he ci y’s
buil -up a ea has sp awled in o we lands, ag icul u al hin e lands, and ecologically sensi i e zones [4]. O icial zoning
amewo ks, howe e , con inue o designa e la ge po ions o land o conse a ion, low-densi y esiden ial use, o
es ic ed indus ial pu poses. The disjunc u e be ween planning p esc ip ions and ac ual land use ou comes aises
u gen ques ions abou he d i e s and consequences o non-compliance.
Recen ad ances in Ea h obse a ion echnologies, pa icula ly he a ailabili y o ee high- esolu ion sa elli e
image y om he Landsa and Sen inel missions, ha e e olu ionized he abili y o moni o land use change a scale.
Coupled wi h spa ial econome ic models, such da a can illumina e no only wha changes a e occu ing bu also why.
This is especially c i ical in Lagos, whe e in o mal se lemen s and un egula ed con e sions cons i u e a signi ican
sha e o u ban de elopmen [5,6].
1.2 U baniza ion and Zoning in Lagos
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [92]
Zoning as a planning ins umen in Lagos da es back o he colonial pe iod, when planning o dinances es ablished
sepa a e qua e s o Eu opeans and A icans [7]. Pos -independence, he Lagos S a e U ban and Regional Planning
Boa d codi ied zoning egula ions in o mas e plans, no ably he Lagos Me opoli an Mas e Plan (1980–2000) and
subsequen de elopmen amewo ks [8]. These plans en isioned a polycen ic ci y s uc u e, wi h esiden ial,
indus ial, and comme cial zones clea ly delinea ed.
In p ac ice, howe e , compliance wi h zoning p esc ip ions has been inconsis en . High demand o a o dable housing
has led o he p oli e a ion o in o mal se lemen s in a eas zoned o ag icul u e o open space, while comme cial
de elopmen s equen ly enc oach on esiden ial dis ic s [9]. Weak en o cemen mechanisms, inadequa e esou ces
wi hin planning au ho i ies, and poli ical in e e ence u he exace ba e non-compliance [10]. A he same ime,
specula i e land acquisi ion and en -seeking beha io by bo h eli es and local landholding amilies con ibu e o
spa ial ou comes ha di e ge sha ply om o icial zoning [11].
The ecological implica ions a e p o ound. Lagos’ we lands, which p o ide lood egula ion and biodi e si y se ices,
ha e been ex ensi ely eclaimed o housing es a es, shopping complexes, and oad in as uc u e [12]. Such land use
change inc eases lood ulne abili y, educes ecosys em esilience, and unde mines long- e m sus ainabili y [13].
Unde s anding he ex en and d i e s o hese pa e ns is essen ial o in o ming adap i e and en o ceable planning
s a egies.
1.3 Resea ch P oblem and Objec i es
Despi e he impo ance o zoning o u ban go e nance, he e is limi ed empi ical esea ch ha sys ema ically links
obse ed land use change wi h planning compliance in Lagos. P e ious s udies ha e examined land co e dynamics
using emo e sensing [14,15] and ha e documen ed he p oli e a ion o in o mal housing [16], bu ew ha e explici ly
assessed zoning non-compliance ela i e o o icial maps. Simila ly, while socio-economic d i e s o u ban g ow h
ha e been analyzed [17], he in eg a ion o zoning o e lays and econome ic models o compliance emains
unde explo ed.
This s udy add esses hese gaps by posing he cen al esea ch ques ion: How has land use in Lagos changed ela i e
o zoning ules, and wha d i e s explain zoning non-compliance?
To answe his ques ion, he s udy se s ou ou objec i es. One objec i e is o classi y land use in Lagos o selec ed
yea s using Landsa 8/9 and Sen inel-2 da a. Ano he is o quan i y land use changes h ough pos -classi ica ion change
de ec ion. In addi ion, he esea ch seeks o assess he ex en o compliance wi h zoning alloca ions by o e laying
classi ied maps wi h o icial zoning laye s and building pe mi eco ds. A u he objec i e is o model he socio-
economic, ins i u ional, and spa ial d i e s o zoning non-compliance using spa ial econome ic me hods.
1.4 Signi icance o he S udy
This esea ch o e s bo h academic and policy con ibu ions. Academically, i in eg a es emo e sensing, GIS, and
spa ial econome ic modeling in o he s udy o zoning compliance, he eby ad ancing me hodological app oaches o
u ban land use analysis. Unlike desc ip i e s udies o in o mal se lemen s, his s udy p o ides a sys ema ic amewo k
o linking ac ual land use wi h planned zoning alloca ions.
F om a policy pe spec i e, indings can in o m Lagos S a e planning au ho i ies and policymake s in Nige ia mo e
b oadly. By iden i ying he d i e s o non-compliance, he s udy p o ides an e idence base o designing a ge ed
in e en ions, whe he s eng hening en o cemen capaci y, imp o ing access o a o dable housing, o e ising
ou da ed zoning amewo ks. Mo eo e , he app oach can be eplica ed in o he apidly u banizing A ican ci ies
acing simila challenges o compliance and en o cemen .
2. METHODS
2.1 S udy A ea
Lagos S a e, loca ed in sou hwes e n Nige ia, is he coun y’s economic hub and one o he as es g owing megaci ies
in he wo ld. Co e ing app oxima ely 3,577 km², Lagos has an es ima ed popula ion o o e 20 million people and is
cha ac e ized by apid u baniza ion, high popula ion densi y, and ex ensi e in o mal se lemen s. The ci y’s coas al
loca ion and we land ecosys ems make i pa icula ly ulne able o en i onmen al p essu es such as looding and land
deg ada ion [16].
2.2 Da a Sou ces
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [93]
This s udy combined emo e sensing da ase s, o icial zoning laye s, building pe mi s, and socio-economic indica o s.
Landsa 8 and 9 OLI/TIRS mul ispec al da a, wi h a spa ial esolu ion o 30 m, we e acqui ed a i e-yea in e als
be ween 2013 and 2023. These da ase s we e selec ed o hei long empo al co e age and consis en calib a ion,
which enable his o ical compa isons [17]. To complemen hese, Sen inel-2 MSI image y wi h 10 m esolu ion was
ob ained o he pe iod 2016–2025, p o iding highe spa ial de ail ha suppo s he alida ion o Landsa
classi ica ions [18].
Adminis a i e da ase s we e ob ained om he Lagos S a e Minis y o Physical Planning and U ban De elopmen
(MPP&UD). O icial zoning maps delinea ing esiden ial, comme cial, indus ial, ag icul u al, conse a ion, and
mixed-use zones we e digi ized in o a GIS da abase [19]. Building pe mi eco ds o he yea s 2010–2024 we e also
collec ed o di e en ia e be ween au ho ized and unau ho ized de elopmen s, hough co e age was incomple e [20].
Socio-economic da a included popula ion densi y, po e y indices, and housing demand s a is ics, sou ced om he
Nige ian Na ional Bu eau o S a is ics (NBS) and he Lagos S a e Bu eau o S a is ics [21]. T anspo in as uc u e
da a, including oad ne wo ks and ansi hubs, we e acqui ed om OpenS ee Map and c oss- alida ed agains
da ase s p o ided by he Lagos Me opoli an A ea T anspo Au ho i y (LAMATA) [22].
2.3 Image P ep ocessing
All sa elli e image y unde wen p ep ocessing o ensu e compa abili y ac oss senso s and ime pe iods. A mosphe ic
co ec ion was applied using he LEDAPS algo i hm o Landsa da a and he Sen2Co p ocesso o Sen inel-2 da a
[23]. Geome ic co ec ion was pe o med by p ojec ing all images o WGS84/UTM Zone 31N, main aining a oo
mean squa e e o (RMSE) o less han 0.5 pixels. Clouds and shadows we e masked using he Fmask algo i hm,
ensu ing ha only cloud- ee pixels we e e ained [24]. To achie e ull co e age o Lagos, mul iple sa elli e scenes
we e mosaicked o each s udy yea , c ea ing seamless da ase s o classi ica ion.
2.4 Land Use Classi ica ion
A six-class land use/land co e (LULC) scheme was employed, aligned wi h Lagos’ zoning ca ego ies. These included
esiden ial/u ban buil -up, comme cial/indus ial, ag icul u al land, we lands/mang o es, open space/conse a ion,
and wa e bodies. T aining da a we e gene a ed using s a i ied andom sampling om mul iple sou ces: high-
esolu ion Google Ea h image y be ween 2013 and 2023, g ound- u h poin s collec ed du ing 2023 ield su eys (n
= 450), and digi ized polygons o e i ied land uses om planning documen s. Valida ion samples (n = 600) we e
wi hheld om aining o independen accu acy assessmen .
Classi ica ion was ca ied ou using he Random Fo es (RF) algo i hm, which has been shown o pe o m well in
he e ogeneous u ban en i onmen s [25]. Model pa ame e s we e op imized h ough 10- old c oss- alida ion, se ing
he numbe o ees o 500 and adjus ing maximum dep h o minimize ou -o -bag e o . To ha monize esul s ac oss
senso s, Sen inel-2 image y was esampled o 30 m and me ged wi h Landsa classi ica ions using a majo i y il e .
Accu acy was hen assessed using con usion ma ices, epo ing o e all accu acy, p oduce ’s accu acy, use ’s
accu acy, and he Kappa coe icien . Classi ica ions wi h less han 85% o e all accu acy we e e ained using
addi ional samples [26].
2.5 Pos -Classi ica ion Change De ec ion
Land use change was quan i ied h ough pos -classi ica ion compa ison be ween ime pe iods (2013–2018, 2018–
2023). T ansi ion ma ices we e gene a ed o iden i y dominan con e sions. The annual a e o change was calcula ed
as:
R = A 2 – A 2 ×100
A 1 × ( 2 – 1)
whe e A 1 and A 2 ep esen land use a ea a imes 1 and 2.
Spa ial ho -spo analysis (Ge is-O d Gi*) was used o iden i y clus e s o in ense land use change.
2.6 Zoning Compliance Assessmen
The classi ied land use maps we e o e laid wi h o icial zoning maps o assess compliance. A de elopmen was
conside ed complian when i s obse ed land use class ma ched he designa ed zoning ca ego y, such as esiden ial
uses wi hin esiden ial zones. Non-compliance was de ined as misma ches, o example, esiden ial o comme cial
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [94]
de elopmen in a eas zoned o ag icul u e o conse a ion. Building pe mi eco ds we e also in eg a ed o
di e en ia e be ween au ho ized and unau ho ized de elopmen s. Compliance a es we e calcula ed a bo h he s a e
le el and he Local Go e nmen A ea (LGA) scale o highligh spa ial a ia ion [27].
2.7 Spa ial Econome ic Modeling
To in es iga e he d i e s o zoning non-compliance, spa ial econome ic models we e es ima ed. The dependen
a iable was he p opo ion o non-complian land wi hin each LGA, calcula ed as he a io o non-complian a ea o
o al land a ea. Explana o y a iables included popula ion densi y (expec ed posi i e), median household income
(expec ed nega i e), dis ance o majo oads (expec ed nega i e), p oximi y o he cen al business dis ic (expec ed
nega i e), p esence o we lands (expec ed posi i e), pe mi app o al a e (expec ed nega i e), and a poli ical in luence
sco e ep esen ing eli e-domina ed LGAs (expec ed posi i e) [28].
Th ee model speci ica ions we e es ed. An O dina y Leas Squa es (OLS) eg ession p o ided he baseline. A Spa ial
Lag Model (SLM) accoun ed o spillo e e ec s ac oss LGAs, while a Spa ial E o Model (SEM) cap u ed
unobse ed spa ially co ela ed e o e ms. Spa ial weigh ma ices we e cons uc ed using he queen con igui y
me hod a he LGA le el. Model selec ion was guided by Akaike In o ma ion C i e ion (AIC) alues and log-
likelihood sco es [29].
2.8 Model Diagnos ics
Diagnos ic es s we e conduc ed o alida e he models. Va iance in la ion ac o s (VIF) we e used o assess
mul icollinea i y, applying a h eshold o less han 5. He e oskedas ici y was es ed using he B eusch–Pagan es ,
while esidual spa ial au oco ela ion was checked using Mo an’s I s a is ic. Robus ness was u he es ed by e-
es ima ing models wi h al e na i e spa ial weigh ma ices, speci ically using k-nea es neighbo s wi h k se o 4 [30].
2.9 E hical Conside a ions
This s udy elied p ima ily on publicly a ailable emo e sensing and adminis a i e da a. G ound- u h da a collec ion
in ol ed ield isi s bu did no include pe sonal iden i ie s o household su eys.
3. RESULTS
3.1 Classi ica ion Accu acy
The supe ised classi ica ions o Landsa and Sen inel image y achie ed s ong pe o mance ac oss all ime pe iods.
Table 1 summa izes he accu acy me ics o 2013, 2018, and 2023 classi ica ions.
Table 1. Classi ica ion accu acy me ics (2013–2023).
Yea
O e all Accu acy
(%)
Kappa
Coe icien
P oduce ’s Accu acy Range
(%)
Use ’s Accu acy Range
(%)
2013
87.3
0.82
80.1–91.6
79.4–92.8
2018
88.9
0.84
82.5–93.2
81.7–94.0
2023
90.4
0.86
83.4–94.7
82.1–95.6
We land and ag icul u al classes showed sligh ly lowe accu acies due o spec al con usion du ing he ainy season,
bu accu acies we e consis en ly abo e he 80% h eshold ecommended o eliable land co e mapping. Buil -up and
wa e classes had he highes accu acies ac oss all yea s, suppo ed by s ong spec al sepa abili y.
3.2 Land Use Change T ends
3.2.1 O e all Land Use Dynamics
Be ween 2013 and 2023, Lagos expe ienced subs an ial land use ans o ma ion (Figu e 1). Buil -up a eas expanded
om 760 km² in 2013 o 1,085 km² in 2023, ep esen ing a 42.8% inc ease in a decade. Ag icul u al land declined
sha ply om 960 km² o 655 km², while we lands sh ank om 285 km² o 173 km². Conse a ion a eas we e
enc oached upon, dec easing by 19.4%. Wa e bodies emained ela i ely s able due o na u al cons ain s.
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [95]
Figu e 1. Land use maps o Lagos o 2013, 2018, and 2023 de i ed om supe ised classi ica ion.
3.2.2 Dominan Con e sions
T ansi ion ma ices (Table 2) e eal he majo land use con e sions.
Table 2. Majo land use ansi ions, 2013–2023 (km²).
F om → To
Residen ial/Comme cial
Ag icul u al
We lands
Conse a ion
Ag icul u al
242
–
16
47
We lands
65
7
–
12
Conse a ion
88
23
9
–
The la ges single con e sion was ag icul u al o esiden ial/comme cial, accoun ing o 242 km² (36.9% o all
change). We lands we e disp opo iona ely con e ed o esiden ial es a es and indus ial complexes, e lec ing
widesp ead eclama ion. Conse a ion a eas we e also con e ed o bo h esiden ial and ag icul u al uses.
3.2.3 Spa ial Ho spo s o Change
Ho spo analysis (Figu e 2) iden i ied h ee clus e s o in ense change:
1. Lekki Peninsula – apid esiden ial and comme cial expansion in o we lands and conse a ion zones.
2. Alimosho and Agege LGAs – ag icul u al land con e ed in o dense housing es a es.
3. Badag y co ido – pe i-u ban sp awl d i en by highway cons uc ion.
Figu e 2. Ho spo map o land use change in ensi y, 2013–2023 (Ge is-O d Gi).
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [96]
3.3 Zoning Compliance Assessmen
3.3.1 Ex en o Compliance
O e lay analysis e ealed ha 41.7% o new de elopmen s be ween 2013 and 2023 occu ed in zones inconsis en
wi h o icial zoning alloca ions (Table 3).
Table 3. Zoning compliance a es, 2013–2023.
De elopmen Type
Complian (%)
Non-Complian (%)
Residen ial
62.4
37.6
Comme cial/Indus ial
54.9
45.1
Mixed-use
57.8
42.2
Comme cial and indus ial de elopmen s showed he highes le els o non-compliance, o en loca ed in esiden ial o
conse a ion zones. Residen ial enc oachmen in o ag icul u al and conse a ion zones was also widesp ead.
3.3.2 Spa ial Va ia ion in Non-Compliance
Non-compliance a ied ma kedly ac oss LGAs (Figu e 3).
• High non-compliance (>50%): E i-Osa, Ibeju-Lekki, Amuwo-Odo in, Badag y.
• Mode a e non-compliance (30–50%): Alimosho, Iko odu, Su ule e.
• Low non-compliance (<30%): Ikeja, Lagos Island, Apapa.
Pe iphe al LGAs exhibi ed highe non-compliance due o weak en o cemen and apid pe i-u ban g ow h.
Figu e 3. Zoning compliance map by LGA, 2023.
3.3.3 Building Pe mi Analysis
C oss-checking wi h building pe mi eco ds e ealed ha only 58% o complian de elopmen s had o icial pe mi s,
while 74% o non-complian de elopmen s we e unpe mi ed. This highligh s he limi ed each o o mal planning
app o al p ocesses, especially in pe iphe al LGAs.
3.4 Econome ic Model Resul s
3.4.1 O dina y Leas Squa es (OLS)
The baseline OLS eg ession (Table 4) indica ed signi ican posi i e associa ions be ween popula ion densi y,
p oximi y o oads, we land p esence, and non-compliance a es. Howe e , spa ial au oco ela ion es s e ealed
signi ican clus e ing o esiduals (Mo an’s I = 0.219, p < 0.01), sugges ing ha OLS was misspeci ied.
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [97]
Table 4. OLS eg ession esul s o d i e s o zoning non-compliance.
Va iable
Coe icien
S d. E o
-s a
p- alue
Popula ion densi y
0.014
0.004
3.52
0.001
Median income
-0.008
0.003
-2.69
0.011
Dis ance o majo oads
-0.021
0.006
-3.36
0.002
Dis ance o CBD
-0.007
0.005
-1.44
0.163
We land p esence (dummy)
0.118
0.042
2.81
0.009
Pe mi app o al a e
-0.036
0.012
-3.00
0.006
Poli ical in luence sco e
0.051
0.019
2.68
0.012
Adjus ed R² = 0.41.
3.4.2 Spa ial Lag Model (SLM)
The SLM (Table 5) imp o ed model i (AIC educ ion o 22 poin s) and showed a signi ican posi i e spa ial lag
coe icien (ρ = 0.26, p < 0.01), con i ming spillo e e ec s: LGAs wi h high non-compliance in luenced neighbo ing
LGAs.
Table 5. Spa ial lag model esul s.
Va iable
Coe icien
p- alue
Popula ion densi y
0.011
0.003
Median income
-0.007
0.018
Dis ance o majo oads
-0.018
0.004
We land p esence
0.097
0.012
Pe mi app o al a e
-0.029
0.009
Poli ical in luence
0.046
0.020
Spa ial lag (ρ)
0.26
0.000
3.4.3 Spa ial E o Model (SEM)
The SEM also pe o med well, bu he SLM p o ided supe io i based on log-likelihood. In bo h models, popula ion
densi y, income, anspo access, we lands, and poli ical in luence emained signi ican d i e s.
3.5 Robus ness Checks
• Mul icollinea i y: All VIF alues < 3, indica ing no se ious mul icollinea i y.
• Residual au oco ela ion: Co ec ed unde SLM speci ica ion.
• Al e na i e weigh ma ices: Resul s consis en when using k-nea es neighbo s (k=4).
3.6 Summa y o Findings
The analysis shows ha land use change in Lagos is cha ac e ized by apid u ban expansion, wi h g ow h occu ing
la gely a he expense o ag icul u al land and we lands. This pa e n e lec s bo h he in ensi y o u baniza ion and he
ulne abili y o ecologically sensi i e a eas o con e sion.
Zoning non-compliance was ound o be widesp ead, wi h mo e han 40 pe cen o new de elopmen s occu ing
ou side o icially designa ed zones. The p oblem is mos acu e in pe i-u ban LGAs such as Lekki and Badag y, whe e
weak en o cemen and specula i e de elopmen p essu es a e especially p onounced.
The spa ial econome ic analysis u he e ealed ha popula ion p essu e, he a ailabili y o we lands, p oximi y o
majo oads, low pe mi app o al a es, and poli ical in luence a e he s onges p edic o s o zoning non-compliance.
These indings unde sco e he complex in e play o demog aphic, en i onmen al, in as uc u al, and ins i u ional
ac o s shaping land use ou comes in Lagos.
4. DISCUSSION
4.1 P incipal Findings
This s udy p o ides sys ema ic e idence o he ex en o which land use in Lagos has di e ged om o icial zoning
p esc ip ions, while also iden i ying he socio-economic, spa ial, and ins i u ional d i e s o non-compliance. The
esul s show ha u ban expansion be ween 2013 and 2023 occu ed p ima ily a he expense o ag icul u al land and
we lands, wi h buil -up a eas inc easing by nea ly 43 pe cen . O e 40 pe cen o new de elopmen s we e loca ed in
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [98]
zones inconsis en wi h planning alloca ions, wi h non-compliance pa icula ly concen a ed in pe i-u ban LGAs such
as E i-Osa, Ibeju-Lekki, and Badag y. Spa ial econome ic modeling e ealed ha popula ion densi y, oad
accessibili y, we land p esence, weak pe mi app o al sys ems, and poli ical in luence we e he s onges p edic o s
o zoning iola ions, while spillo e e ec s ac oss LGAs sugges ed ha non-compliance ends o sp ead
geog aphically. Toge he , hese indings highligh he deep misalignmen be ween Lagos’ u ban planning amewo ks
and i s li ed de elopmen pa e ns [31].
4.2 Compa ison wi h Exis ing Li e a u e
The esul s a e consis en wi h p io esea ch documen ing Lagos’ apid and la gely un egula ed u baniza ion. S udies
using Landsa image y ha e shown ha buil -up land in Lagos inc eased by mo e han 600 pe cen be ween he 1980s
and he la e 2010s, much o i occu ing on we lands and ag icul u al land [32,33]. Simila ly, Akinmoladun and
Adejumo iden i ied widesp ead we land con e sion in he Lekki co ido , a inding echoed in his s udy’s ho spo
analysis [34].
The p e alence o zoning non-compliance aligns wi h obse a ions om o he A ican ci ies, whe e zoning
amewo ks a e o en aspi a ional a he han en o ceable. Kombe has desc ibed how weak ins i u ions and apid
popula ion g ow h c ea e a “planning–implemen a ion gap” in which egula o y amewo ks ha e li le bea ing on
u ban ou comes [35]. Wa son also emphasizes ha A ican ci ies ace a disconnec be ween u ban isions and on- he-
g ound eali ies, a pa e n s ongly e iden in Lagos [36].
The d i e s o non-compliance iden i ied he e also mi o pa e ns epo ed elsewhe e. Popula ion g ow h has been
shown o push in o mal housing expansion in bo h Lagos and Nai obi [37], while oad access has been linked o
specula i e de elopmen in Da es Salaam and Acc a [38,39]. We land enc oachmen has eme ged as a c i ical
en i onmen al issue in coas al A ican megaci ies, especially whe e land sca ci y and eal es a e specula ion in e sec
[40].
The ole o poli ical in luence unco e ed in his s udy esona es wi h Nige ian li e a u e documen ing eli e-d i en
ezoning and selec i e en o cemen o planning ules [41]. The nega i e ela ionship be ween pe mi app o al a es
and non-compliance sugges s ha s eng hening ins i u ional e iciency could se e as a co ec i e mechanism,
echoing indings om Ghana and Uganda ha e ec i e pe mi sys ems educe unau ho ized cons uc ion [42,43].
4.3 Implica ions o U ban Planning and Go e nance
The e idence om Lagos aises impo an implica ions o planning p ac ice. To begin wi h, he high a es o non-
compliance challenge he ele ance o adi ional mas e planning. While o icial plans ha e long en isioned o de ly,
polycen ic g ow h, ac ual de elopmen has been d i en by in o mal p ac ices, specula ion, and ins i u ional weakness.
This ein o ces Wa son’s a gumen ha many A ican ci ies ace a undamen al disconnec be ween u ban isions and
li ed eali ies [36].
Equally signi ican is he inding ha 74 pe cen o non-complian de elopmen s lacked pe mi s, which unde sco es
he limi ed en o cemen capaci y o Lagos’ planning ins i u ions. En o cemen is no only hampe ed by inadequa e
echnical s a bu also unde mined by poli ical in e e ence and co up ion [44]. Add essing hese ins i u ional
weaknesses equi es a combina ion o inc eased unding, expanded echnical aining, and he deploymen o
geospa ial echnologies o imp o e moni o ing.
Ano he impo an dimension is he ole o anspo in as uc u e, which eme ges as a double-edged swo d. P oximi y
o majo oads was a signi ican p edic o o non-compliance, sugges ing ha in as uc u e expansion can
inad e en ly s imula e unau ho ized de elopmen . This highligh s he need o in eg a e anspo in es men s wi h
land use planning, ensu ing ha oad co ido s a e p e-zoned and subjec o s ic moni o ing [38,39].
Equally u gen a e he en i onmen al consequences o zoning iola ions. We land eclama ion o esiden ial es a es
and indus ial acili ies inc eases lood isks, unde mines biodi e si y, and educes clima e esilience. The ca as ophic
looding in Lagos du ing 2022 illus a es he human and economic cos s o neglec ing ecological p o ec ion zones
[45]. S onge en o cemen o conse a ion a eas, suppo ed by eal- ime moni o ing, is he e o e c i ical.
A u he implica ion is he de ec ion o spillo e e ec s, which unde sco es he need o me opoli an-scale
go e nance. Since non-compliance in one LGA in luences pa e ns in neighbo ing ju isdic ions, agmen ed local
en o cemen is unlikely o succeed. A me opoli an planning au ho i y wi h c oss-bounda y ju isdic ion could p o ide
he necessa y coo dina ion o managing g ow h mo e cohe en ly [46].
Volume-09 Issue 10, Oc obe -2025 ISSN: 2456-9348
Impac Fac o : 8.232
In e na ional Jou nal o Enginee ing Technology Resea ch Managemen
(IJETRM)
h ps://ije m.com/
IJETRM (h p://ije m.com/) [99]
4.4 Policy Implica ions
Se e al policy implica ions a ise om hese indings. Re o ming zoning amewo ks o e lec p esen -day eali ies is
essen ial, pa icula ly in apidly u banizing pe i-u ban a eas. Adap i e and lexible zoning ha accommoda es mixed-
use de elopmen can educe p essu es o non-compliance, p o ided ecological zones a e s ic ly p o ec ed.
En o cemen mus be mode nized h ough he use o d ones, sa elli e moni o ing, and au oma ed change de ec ion
sys ems, while pe mi app o al p ocesses should be digi ized and in eg a ed wi h geospa ial da abases o imp o e
anspa ency and educe bu eauc a ic delays [47].
T anspo and land use planning mus be igh ly coo dina ed, wi h in as uc u e p ojec s accompanied by p oac i e
zoning en o cemen o p e en oppo unis ic de elopmen . Ins i u ional e o ms a e also needed o insula e planning
agencies om poli ical in luence and co up ion, including independen e iew boa ds and anspa en amendmen
p ocesses [41,44]. P o ec ing we lands and conse a ion zones mus emain a op p io i y, suppo ed no only by legal
sanc ions bu also by communi y-based s ewa dship p og ams [40,45]. Finally, pa icipa o y app oaches o planning
a e i al. In ol ing esiden s in zoning decision-making enhances legi imacy and inc eases compliance, a lesson d awn
om success ul communi y-d i en ini ia i es ac oss A ica [36,42].
4.6 Limi a ions
While his s udy p o ides obus e idence, se e al limi a ions mus be acknowledged. One impo an limi a ion ela es
o da a gaps. Building pe mi eco ds we e incomple e, which cons ained he abili y o comp ehensi ely dis inguish
be ween au ho ized and unau ho ized de elopmen s. As a esul , some in o mal o unpe mi ed ac i i ies may no ha e
been ully cap u ed in he analysis.
Ano he limi a ion s ems om classi ica ion unce ain y. Al hough he o e all accu acy o land use classi ica ion
exceeded 85 pe cen , he e was occasional con usion be ween ag icul u al and we land classes, pa icula ly in a eas
we e seasonal looding blu ed dis inc ions. This could ha e in oduced mino e o s in he de ec ion o land use
ansi ions.
The ea men o zoning maps as s a ic benchma ks also posed a challenge. In eali y, zoning egula ions in Lagos
ha e been pe iodically e ised, some imes in o mally o wi hou clea documen a ion. By elying on o icial zoning
maps as ixed e e ences, he s udy may no ha e ully accoun ed o such un eco ded adjus men s, po en ially
a ec ing compliance assessmen s.
A u he limi a ion is he p esence o unobse ed a iables. The econome ic models did no explici ly cap u e ac o s
such as land enu e complexi y, communi y esis ance o planning di ec i es, o in o mal land ma ke dynamics.
Al hough di icul o quan i y, hese ac o s a e highly in luen ial in shaping compliance pa e ns.
Finally, he empo al scope o he s udy was es ic ed o he pe iod 2013–2023 due o he a ailabili y o consis en
sa elli e image y. Ea lie pe iods o land use change we e no examined, which limi s he abili y o analyze longe -
e m his o ical ajec o ies o compliance and non-compliance.
4.7 Di ec ions o Fu u e Resea ch
Fu u e esea ch should build on hese indings while add essing he s udy’s limi a ions. One p io i y is he
inco po a ion o dynamic zoning maps ha accoun o empo al changes in egula ions. Such an app oach would
p o ide a mo e nuanced unde s anding o compliance, especially in con ex s whe e zoning ules a e equen ly e ised,
o en in o mally. Simila ly, he in eg a ion o cadas al da a a he pa cel le el would enhance analysis by linking land
use pa e ns mo e di ec ly o enu e and owne ship s uc u es, he eby cla i ying how p ope y igh s shape compliance
beha io .
Me hodological inno a ion also o e s p omising a enues. Agen -based modeling, o example, could be used o
simula e he beha io o de elope s, landowne s, and egula o s, cap u ing he mic o-le el decision-making p ocesses
ha d i e non-compliance. Compa a i e s udies ep esen ano he impo an di ec ion. Applying he same
me hodological amewo k o o he A ican ci ies such as Nai obi, Acc a, o Da es Salaam would help o iden i y
bo h commonali ies and di e ences in compliance dynamics, en iching he b oade discou se on u ban go e nance.
Finally, mo e esea ch is needed on he social and en i onmen al dimensions o zoning en o cemen . Communi y
engagemen s udies ha explo e how esiden s pe cei e zoning ules and en o cemen p ac ices could in o m mo e
pa icipa o y app oaches o compliance. A he same ime, in eg a ing clima e isk modeling wi h land use compliance
