Access to this full-text is provided by Springer Nature.
Content available from Environmental Monitoring and Assessment
This content is subject to copyright. Terms and conditions apply.
Vol.: (0123456789)
Environ Monit Assess (2025) 197:464
https://doi.org/10.1007/s10661-025-13921-x
RESEARCH
Land use/land cover changes due togold mining
intheSingida region, central Tanzania: environmental
andsocio‑economic implications
AzariaStephanoLameck· BrianRotich· AbdalrahmanAhmed·
HarisonKipkulei· SilvesterRaymondMnyawi· KornelCzimber
Received: 19 December 2024 / Accepted: 17 March 2025 / Published online: 25 March 2025
© The Author(s) 2025
was conducted using the random forest (RF) classifier
to generate LULC maps with five classes (bareland,
agricultural land, forest, built-up, and shrubs and
grasses), followed by an analysis to identify LULC
change trends. The results showed an overall increase
in agricultural land 168.51 km2 (587.55%), bareland
7.70 km2 (121.45%), and built-up areas 0.55 km2
(134.15%), while forest and shrubs and grasses areas
declined by 97.67 km2 (− 72.59%) and 79.09 km2
(− 43.49%), respectively. A social survey assessment
revealed residents perceived environmental (defor-
estation, biodiversity loss, land degradation, water,
air, soil pollution), social (occupational hazards,
land use conflicts, negative effects on livelihoods
and culture, discrimination, child labor, community
displacement), and economic (improved housing,
infrastructural development, job creation, economy
boost, improved access to services) impacts result-
ing from mining activities. Our findings underscore
Abstract This study explored the land use and land
cover (LULC) changes (1995–2023) in the gold min-
ing hotspots of Mang’onyi, Sambaru, and Londoni in
the Singida region of Tanzania. The study integrated
remote sensing (RS) to evaluate the LULC transi-
tions with social survey assessments (83 respond-
ents) to determine the resident’s perceptions of the
environmental, social, and economic implications of
mining bridging technical data with socio-economic
realities. Supervised classification of Landsat images
Highlights • Land use/land cover changes in the Singida
region, Tanzania, were analyzed from 1995 to 2023.
• Agricultural land, bareland, and built-up areas expanded,
while areas under forests, shrubs, and grasses declined.
• Residents reported diverse impacts of mining, including
environmental degradation, social challenges, and
economic benefits.
• Promoting sustainable mining practices is crucial for
balancing economic growth, community well-being, and
environmental conservation.
A.S.Lameck(*)· B.Rotich
Doctoral School ofEnvironmental Science, The Hungarian
University ofAgriculture andLife Sciences, Páter Károly
U. 1, Gödöllő2100, Hungary
e-mail: azariastephano@gmail.com; Lameck.Azaria.
Stephano@phd.un-mate.hu
B. Rotich
e-mail: rotich2050@gmail.com
A.S.Lameck
Department ofEarth Science, Mbeya University
ofScience andTechnology, PO BOX 131, Mbeya,
Tanzania
B.Rotich
Faculty ofEnvironmental Studies andResources
Development, Chuka University, P.O. Box109-60400,
Chuka, Kenya
A.Ahmed· K.Czimber
Institute ofGeomatics andCivil Engineering, Faculty
ofForestry, University ofSopron, Bajcsy-Zs 4,
Sopron9400, Hungary
e-mail: omda33sd@gmail.com
K. Czimber
e-mail: czimber.kornel@uni-sopron.hu
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 2 of 20
Vol:. (1234567890)
the importance of balancing the economic benefits of
gold mining with the imperative to protect the envi-
ronment and support sustainable livelihoods in the
mining regions.
Keywords LULC change· Mining sites· Gold
rush miner· Environmental implications· Social-
economic implications
Introduction
Mining has been a key player in most nations eco-
nomic development, infrastructural development,
employment, and supply of essential raw materials
(Worlanyo & Jiangfeng, 2021). It has served as a via-
ble route to economic transformation and industriali-
zation in resource-rich developed countries and is an
equally significant component in many low- and mid-
dle-income nations since mineral resources are neces-
sary for the development of their economies (Ericsson
& Löf, 2019; Kuzevic etal., 2022). Many low-income
sub-Saharan Africa (SSA) countries are large produc-
ers and exporters of lucrative minerals such as dia-
monds, crude oil, bauxite, gold, chromite, platinum,
cobalt, titanium, and rare earth elements (Pokorny
et al., 2019). Social and economic development
indicators have shown signs of progress for African
countries that are rich in mineral resources (Erics-
son & Löf, 2019). It is evident that mining operations
and exports have played a pivotal role in the econo-
mies of SSA countries like the Democratic Repub-
lic of Congo (DRC), Malawi, Ghana, South Africa,
and Guinea. It is reported that 25% of Guinea’s and
5.9% of South Africa’s gross domestic profits (GDP)
as well as the majority of their foreign revenues are
mining-related (Aryee, 2001). Mineral exports consti-
tute 86% of total exports in the DRC and contribute
to 13% of the country’s GDP (Ericsson & Löf, 2019).
In Ghana, gold mining contributes about 5.7% of the
national GDP (Mensah et al., 2015), while uranium
mining operations in Malawi spurred the contribution
of the mining sector to 10% of the GDP in 2010 from
less than 3% in 2005 (Haundi etal., 2021). While it
is evident that mining has transformed many econo-
mies, it has also had negative impacts on the environ-
ment and society (Worlanyo & Jiangfeng, 2021). The
extraction of minerals, especially through opencast
mining, impacts the land, soil, water, and vegeta-
tion. Artisanal and small-scale mining is associated
with several negative impacts, including, loss of min-
eral revenue due to smuggling, food insecurity, con-
tamination and pollution of surface and underground
water sources, air and noise pollution, and biodiver-
sity loss (loss of natural flora and fauna) (Gbedzi
etal., 2022; Suglo et al., 2021; Yu & Zahidi, 2023).
A notable example is the increasingly alarming lev-
els of water pollution caused by illegal artisanal and
small-scale mining (Galamsey) ‘Menace’ in Ghana,
which have sparked serious debate among policymak-
ers, environmentalists, and local communities (Eduful
etal., 2020; Kuffour etal., 2018; Suglo etal., 2021).
Recently, some illegal miners were holed up in the
vast tunnel network of the Stilfontein mine in South
Africa. Negative health risks associated with illegal
mining include fatal accidents, injuries, respiratory
and skin diseases, noise-induced hearing loss, physi-
cal and psychological stress, malaria, HIV, and other
infectious diseases (Al Rawashdeh etal., 2016; Suglo
etal., 2021; Worlanyo & Jiangfeng, 2021).
Tanzania is rich in minerals like diamond, gold,
bauxite, Tanzanite, limestone, tin, and copper, con-
tributing significantly to its economy (Yager, 2023).
The country is the third-largest gold producer in
Africa, after South Africa and Ghana, accounting for
about 1% of the world’s gold output in 2019 (Lugoe,
2011; Yager, 2023). Gold mining, driven by sec-
tor liberalization and high gold prices, has become a
major economic contributor, representing over 41.3%
of export earnings and 3.6% of GDP (Lugoe, 2011).
In 2017, Tanzania’s large-scale mines employed
around 12,000 workers, with an estimated 1.5 million
artisanal miners also involved (Yager, 2023). Gold
A.Ahmed
Department ofForest andEnvironment, Faculty
ofForest Science andTechnology, University ofGezira,
WadMedani, Sudan
H.Kipkulei
Institute ofGeography, Faculty ofApplied Computer
Sciences, University ofAugsburg, Alter Postweg 118,
86159Augsburg, Germany
e-mail: hkipkulei@gmail.com
S.R.Mnyawi
Department ofNatural Science, Mbeya University
ofScience andTechnology, PO BOX 131, Mbeya,
Tanzania
e-mail: mnyawisr@gmail.com
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 3 of 20 464
Vol.: (0123456789)
accounts for more than 41.3% of Tanzania’s export
earnings, 75% of foreign direct investment (FDI), and
an increasing share of taxes, representing 3.6% of the
gross domestic product (GDP) (Lugoe, 2011). The
liberalization and privatization of the mining sector
and the soaring gold prices during the last decades
have made gold mining increasingly attractive and
triggered a gold boom in Tanzania (Hammond etal.,
2007; Lugoe, 2011; Phillips etal., 2001).
The Singida region, known for its gold reserves,
has been a mining site since 1909, with notable sites
like Sekenke, Shelui, and Muhentiri. Discoveries in
Londoni, Sambaru, and Mang’onyi in 2004 further
highlight its gold potential. Mining became semi-
mechanized (modern equipment) in the early 2000s,
leading to environmental and social changes in local
communities (Herman & Kihampa, 2015; Lugoe,
2011). Shanta Gold Mine Company, which began
surveying in 2004, achieved commercial production
in 2023, reflecting the region’s growing mining indus-
try (Shanta Gold Limited, 2024). Mining in the area
includes local artisanal miners, gold rush miners, and
small-scale companies, whose influx has significantly
impacted the environment, local economy, and social
dynamics (Jønsson & Bryceson, 2009; Poignant,
2023). Assessment of the cumulative environmental
impacts of mining is an important aspect of sustain-
able management as it involves balancing the benefits
of resource exploitation against environmental degra-
dation (Latifovic etal., 2005). The conflict between
mining and environmental protection has intensified,
highlighting the need for better information on min-
ing impacts at regional and local levels (Latifovic
et al., 2005). Therefore, it is essential to map and
monitor the impacts of gold mining on the LULC
dynamics in the areas surrounding the mining sites of
Mang’onyi, Sambaru, and Londoni in the Ikungi and
Manyoni districts of the Singida region, Tanzania.
This study aims to (a) assess the changes in LULC
dynamics (1995–2023) due to gold mining activities
and (b) examine the societal perceived environmen-
tal and socio-economic implications of these LULC
changes. The mapping of LULC dynamics and the
subsequent analysis to identify change trends at the
study site provide comprehensive evaluations of
mining-induced alterations, offering valuable new
insights into the environmental impacts of mining
activities in the area. Integration of remote sensing
(RS) data with local community perceptions of the
environmental, social, and economic effects of min-
ing further strengthens the study’s originality, bridg-
ing technical data with socio-economic realities.
The study provide a comprehensive analysis of the
extent of LULC dynamics and community percep-
tions regarding the environmental and socio-eco-
nomic impacts of mining activities in the study site.
The study also provide valuable practical insights for
policymakers, stakeholders, and local communities
aiming to promote sustainable mining practices that
balance environmental protection, economic growth,
and the livelihoods of the local communities in min-
ing sites. These insights will help strengthen the
enforcement of environmental regulations at mining
sites. This, in turn, will ensure that mining practices
are better regulated, minimizing adverse impacts on
both the environment and local communities.
Materials andmethods
Description of the study area
This study covers three villages (Mang’onyi, Samb-
aru, and Londoni) situated on the border of the Ikungi
and Manyoni districts in the Singida region of Tan-
zania (Fig.1). These villages are located within lon-
gitudes 34°54′ and 35°7′E and latitudes 5°12′ and
5°26′S, covering an estimated area of 351.82 km2.
The region experiences a semi-arid climate, char-
acterized by a rainy season from late November to
early May and a prolonged dry season from June to
early November (Fig.2). Annual precipitation ranges
between 400 and 600 mm, while temperatures vary
from 14 to 30 °C, with a mean annual tempera-
ture of 22°C (Fig.2). The local population depends
largely on agriculture, cultivating maize, sorghum,
and sunflowers while also keeping livestock such
as goats, sheep, and cattle. However, after discover-
ing gold deposits in 2004, villagers started engag-
ing in small-scale mining as an additional source of
income. The mining operations are conducted by a
legally independent multinational company that has
obtained a permit from the Ministry of Minerals and
Energy. However, there are a reasonable number of
small miners (artisans) who illegally exploit miner-
als to sustain their livelihoods. In Londoni village,
gold mineral deposits were first discovered in June
2004 by Mr. Jumanne Mtemi, followed by active
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 4 of 20
Vol:. (1234567890)
Fig. 1 Map of the study area showing (a) the location of Tanzania in Africa, (b) the location of the study area in the Singida region
of Tanzania, and (c) the mining sites, villages, and administrative boundaries in the study area
Fig. 2 The mean annual precipitation and temperature (Tmax, Tmin, and Tmean) of the study area
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 5 of 20 464
Vol.: (0123456789)
mining activities from July 2004 (Jønsson & Bryce-
son, 2009). This accelerated the exploration and dis-
covery of more gold reserves (Mustapha, Marando,
and Yusuph mining sites) in Sambaru village from
September to October 2004. Other gold deposits,
including Mwau (Hanje), Shanta Number 1, and the
Taru Shanta mining site, were discovered in the sub-
sequent years. The mining sites particularly in Lon-
doni and Sambaru are located on a steep ridge in
the Rift Valley, with reddish iron-rich soils, thorny
plants, grasses, and tiny trees (Herman & Kihampa,
2015). In the region, gold extraction involves crush-
ing, grinding ore, washing, and capturing gold par-
ticles using sluice tables, amalgamation, burning,
and cyanide solution leaching of tailings (Herman &
Kihampa, 2015). In Mang’onyi village, particularly in
Kinyamberu and Taru, the Shanta Gold Mine Com-
pany began a survey in 2004, subsequently initiat-
ing the construction of the mining site in late 2020.
Shanta Gold Mine Company commenced mining
operations in September 2021, with ore stockpiled,
and reached commercial production in 2023 (Shanta
Gold Limited, 2024).
Methodology
Data acquisition
Landsat 5 and 8 Surface Reflectance (SR) data, with
a spatial resolution of 30m, were acquired from the
Google Earth Engine (GEE) platform due to their
accessibility and the platform’s capability for pre-
liminary data processing. We utilized the median sur-
face reflectance values of Landsat 5 and 8 imageries
from June 1995, 2004, 2014, and 2023, correspond-
ing to the study area’s driest period. This approach
minimizes cloud cover and ensures comprehensive
spatial coverage and temporal consistency. The year
1995 was selected for comparison as it represents
the period before the discovery of mining in 2004
(Lugoe, 2011; United Republic of Tanzania, 2014),
while the years 2014 and 2023 were chosen to assess
changes that occurred after the discovery of gold
mining.
Training andtesting sample datasets
Multi-temporal Google Earth aerial imagery and
various color composites of different bands (SR_B1,
SR_B2, SR_B3, SR_B4, SR_B5, SR_B6, SR_B7)
of Landsat data were utilized to identify appropriate
training sites. Training and testing samples were col-
lected through on-screen digitizing, as widely used
and reported in the literature (Ahmed et al., 2024).
Furthermore, the identification of the training sites
was corroborated with high-resolution Google Earth
imagery. Additionally, the first author’s extensive
knowledge of the study area was instrumental in
accurately determining the ground truthing for spot-
ting land cover classes. The QGIS 3.34.3 software
was employed to generate training and testing sam-
ple data for each land cover class. A total of 148,
140, 144, and 150 training samples were created for
the years 1995, 2004, 2014, and 2023, respectively.
These samples were randomly divided, with 50%
allocated for training and 50% for accuracy valida-
tion. In supervised satellite image classification, the
typical data split for training and testing is 70–30%
or 80–20%. However, some studies have successfully
used a 50–50% split achieving overall accuracy of
between 81 and 93% (Ahmed etal., 2024; Stephano
etal., 2025). This approach provides a better estimate
of generalization performance, reducing the risk of
overfitting the training data and ensuring that results
are not inflated due to insufficient test data (Yadav
etal., 2024).
Image classification
Supervised classification was performed using the
Dzetsaka plugin in QGIS, which supports various
machine-learning algorithms. To leverage Dzet-
saka’s capabilities, additional dependencies, includ-
ing the Scikit-learn 1.0.1 Python package—a stand-
ard library for machine learning (Karasiak, 2020;
Pedregosa et al., 2011), were installed. The random
forest (RF) classifier was utilized to classify the three
Landsat images due to its robustness in distinguishing
various LULC classes across different environmental
areas (Rodriguez-Galiano etal., 2012; Thonfeld etal.,
2020) and its high accuracy in this study. The RF
algorithm, a decision-tree-based ensemble learning
method, has been employed to address various envi-
ronmental problems and can process a wide range of
data, including satellite images and numerical data
(Oo etal., 2022). The algorithm also has several other
advantages including the ability to model non-linear
relations, ability to handle outliers, and sensitivity
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 6 of 20
Vol:. (1234567890)
to overfitting. This study used the Dzetsaka classifi-
cation tool in QGIS to apply RF classification to the
Landsat images. A fixed number of 100 trees was
set by default, which is an appropriate size to avoid
overfitting (Breiman, 2001; Rubo etal., 2019). Five
classes of LULC, including forestland, shrubs and
grasses, agricultural land, bareland, and built-up
areas, were successfully classified. In this study, min-
ing areas were not classified as a distinct land cover
type due to the complex nature of mining activities
in the region. Legal mining sites are often visible
as bare land in remote sensing imagery, while ille-
gal mining occurs within shrublands and grasslands,
making it difficult to delineate separately. The spec-
tral similarity between mining areas and bareland,
along with the dynamic nature of mining activities
(where abandoned sites may undergo natural reveg-
etation), posed classification challenges. To account
for mining-induced land cover changes, we analyzed
transitions in bare lands, which are closely linked to
mining expansion. This approach has similarly been
used by other researchers in classification (Gbedzi
et al., 2022). Field observations, expert knowledge,
and Google Earth imagery were additionally used
to validate mining-related changes. This approach
ensured that both legal and illegal mining impacts
were captured while minimizing classification bias.
Accuracy assessment
Accuracy assessment is a critical step in the clas-
sification process, aimed at quantitatively evaluat-
ing the effectiveness of pixel classification into the
correct land cover classes. Following the successful
classification of LULC, the Semi-automatic Classi-
fication Plugin (SCP) in QGIS was utilized to assess
the accuracy of the classified maps. A random strat-
ified sample, with an equal number of samples for
each land cover class, was employed to train and
test the classifier by computing several metrics and
presenting them using the confusion matrix (Ahmed
etal., 2024). The confusion matrix is the most com-
mon method for presenting the accuracy of classi-
fied images (Plourde & Congalton, 2003). Other
accuracies included are the overall accuracy and the
kappa index. The kappa coefficient ranges from neg-
ative values to 1, where 0 indicates no agreement
and 1 indicates perfect agreement. A kappa value
around zero suggests a fully random classification,
while negative values indicate a classification worse
than random ( Landis & Koch, 1977a, 1977b).
Intensity analysis andLULC transitions
LULC dynamics in the study region were analyzed
using intensity analysis. The analysis decomposes
changes into interval, category, and transition levels.
The analysis incorporates a mathematical framework
that compares a uniform intensity to the observed
temporal changes among various land use catego-
ries. Uniform intensity is a hypothetical intensity
where the overall change during the time interval was
uniformly distributed across the classes. Intensity
analysis helps determine the spatial distribution and
magnitude of landscape changes by identifying land
use categories that exhibit the most substantial shifts
compared to the uniform rate of change. It provides
information about the rate and extent of land cover
changes over time. It measures the degree to which
these changes are non-uniform, highlighting which
land categories are most active or relatively dormant
within a given time interval. The technique uses
cross-tabulation matrices, where each matrix summa-
rizes the LULC change at each time interval.
The intensity analysis implements this in three
levels: the interval level, the category level, and the
intensity level. The interval level analyzes the overall
change size and annual change intensity for the whole
area in each time interval. The annual change inten-
sity of the study area during time t(St) is given by:
where St is the annual change intensity, J is the num-
ber of categories, j is the index of the category at a
later time point, and i is the index of the category at
an initial time point. Ctij is the size of the transition
from category i to category j during the interval Yt
and Yt +1. Yt is the year at time point t, and Yt +1 is the
year at time point t + 1. The upper part of Eq.1 gives
the change during the time interval t, while the lower
part is the product of the study area size and the inter-
val duration.
Uniform change intensity (Ut) during interval t is
shown in the following equation:
(1)
S
t=
∑
I
j[(∑
J
i=1Ctij )−Ctij
(Yt+1−Yt)(∑I
j
∑J
i=
1Ctij
)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 7 of 20 464
Vol.: (0123456789)
The upper part of the equation computes the
change during all intervals, and the denominator
gives the product of the study area size and the study
duration.
The category level provides information on the
variation in size and intensity of gross gains and
losses across categories during each period. There-
fore, loss intensity (Altieri et al., 2015) from cat-
egory i during the time interval t corresponds to the
percentage lost at the beginning of the time interval
t (Eq.3)(Allaire etal., 2022). On the other hand, the
gain intensity (Gtj) is the percentage of the end size of
category j gained during the time interval t (Eq.4). If
Lti < St or Gtj < St, then the loss from category i or gain
to category j during the time interval t is dormant.
The reverse indicates activeness between the catego-
ries in the time interval t.
The intensity analysis was implemented using the
OpenLand package in the R statistical software (Exa-
vier & Zeilhofer, 2020). The package uses the LULC
classifications in the studied periods as input data in
the contingency table to generate a quantity of change
from one category to another between two points. The
changes between various class categories were also
represented using a Sankey plot. The Sankey plot vis-
ualizes flows through a network. It incorporates the
cross-tabulation matrices, which contain information
on the sizes of categorical differences between two
maps to describe the amount and type of land cover
change that has occurred between two points in time.
Perceived implications of gold mining
A social survey was conducted from 2nd to 28th of
September 2024 to support LULC change results from
the satellite data. Household (HH) questionnaires with
open and close-ended questions were administered
(2)
U
t=
∑T
−1
t=1(∑J
j=1∑J
j=1Ctij )
(YT−Y1)(∑J
j
∑J
i=
1Ctij )
×
100
(3)
L
ti =∑
J
i=1Ctij −Ctii
(Y
t+
1−Y
t
)∑J
i=1
C
tij
∗
100
(4)
G
tj =∑
J
i=1Ctij −Ctjj
(Y
t+
1−Y
t
)∑J
i=1
C
tij
∗
100
to the residents using the ODK collect application to
solicit information. The questions focused on house-
hold demographics, types of mining, previous LULC
types, and the impacts (environmental, economic, and
social) of mining activities in the study area. Strati-
fied sampling was used to distribute the questionnaires
among 83 households in the three mining villages. Key
informant interviews (Gbegbelegbe etal., 2018) were
also conducted with environmental officers, commu-
nity elders, mining companies, and non-governmental
organizations (Pinto etal., 2013) in the study area to
capture information that might have been overlooked
in the HH questionnaires and simultaneously enrich
information gathered through HH surveys. Field obser-
vations were also made at the mining sites during the
surveys.
Results
Accuracy result table
Land use classification revealed high accuracies
(Table 1). The overall accuracies ranged between 80
and 88%, while the kappa index ranged between 0.68
and 0.82 for the studied years. Since these results fell
within the acceptable range, we proceeded with utiliz-
ing the classification outputs (Ahmed etal., 2024; Lan-
dis & Koch, 1977a, 1977b; Patil & Nataraja, 2020).
Distributions of different LULC classes and statistics
The land cover statistics and the land cover maps show-
ing the distribution of the different LULC classes
across the study area are presented in Table 2 and
Fig. 3. Statistical analysis shows that in 1995, shrubs
and grasses were the dominant LULC class at 181.84
km2 (51.68%) while forest had the second largest cov-
erage of 134.55 km2 (38.25%). Built-up areas had the
least area coverage at 0.41 (0.12%) (Table2). In 2004,
shrubs and grasses still covered most parts of the study
area at 161.50 km2 (45.90%), agricultural land had the
second largest area coverage of 90.42 km2 (25.70%),
Table 1 LULC classification accuracies
1995 2004 2014 2023
Overall accuracy 80.96 88.41 81.77 80.59
Kappa hat 0.68 0.82 0.72 0.69
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 8 of 20
Vol:. (1234567890)
while forest had the third largest area at 89.44 km2
(25.42%) (Table2). Bareland had the second least area
of 9.87 km2 (2.81%), while built-up areas occupied the
least area of our study region at 0.59 km2 (0.17%). By
2014, agricultural land had dominated the study area
with a coverage of 178.46 km2 (50.73%) over shrubs
and grasses, with the second largest area at 86.74 km2
(24.65%). Forest came in third with a reduced area of
73.85 km2 (20.99%). Despite increasing size, built-up
areas still had the smallest area coverage of 0.66 km2
(0.19%). In 2023, agricultural land coverage increased
to 197.19 km2 (56.05%), shrubs and grasses covered
102.75 km2 (29.21%), forest 36.88 km2 (10.48%), bare-
land 14.04 km2 (3.99%), and built up made up 0.96 km2
(0.27%) of the study area (Table2).
From the maps (Fig. 3), the dominant LULC
classes in the study area were agricultural lands,
shrubs and grasses, and forest lands. Forested areas
dominating in the central and southern parts of the
study anddeclined over time notably the area around
Chipinga, with agricultural areas, shrublands, and
bareland areas occupying formerly forested areas.
There was a continuous expansion of agricultural
areas, with grasslands and shrublands contribut-
ing much of this transformation. Bareland areas in
the northwestern part of the study area around the
Mang’onyi trading centre slowly diminished, with
much of the area being occupied by agricultural land.
Bareland areas also expanded, especially in the east-
ern parts of the study area where most of the min-
ing sites (Yusuph, Marando) and villages (Londoni,
Sambaru) are located (Fig. 4). There was a surge in
built areas between 2014 and 2023 notably around the
Shanta Singida mining site in the northern part of the
study area (Figs.3 and 4).
The area changes of the different LULC classes
The study area underwent significant transformations
in its major LULC categories, displaying diverse pat-
terns and varying degrees of change (Fig.5). The first
study period (1995–2004) experienced an expansion
in agricultural land, bareland, and built-up areas by
61.74 km2 (215.27%), 3.53 km2 (55.68%), and 0.18
Table 2 Land use and land cover statistics
LULC class 1995 (km2) % 2004 (km2) % 2014 (km2) % 2023 (km2) %
Forest 134.55 38.25 89.44 25.42 73.85 20.99 36.88 10.48
Bareland 6.34 1.80 9.87 2.81 12.11 3.44 14.04 3.99
Shrubs and grasses 181.84 51.68 161.50 45.90 86.74 24.65 102.75 29.21
Agricultural land 28.68 8.15 90.42 25.70 178.46 50.73 197.19 56.05
Built up 0.41 0.12 0.59 0.17 0.66 0.19 0.96 0.27
Total 351.82 100.00 351.82 100 351.82 100 351.82 100
Fig. 3 LULC class coverage in the study area in 1995, 2004, 2014, and 2023
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 9 of 20 464
Vol.: (0123456789)
km2 (43.90%), respectively. In the second decade
(2004–2014), agricultural land increased signifi-
cantly by 88.04 km2 (97.37%). Similarly, bareland
increased by 2.24 km2 (22.70%) and built-up areas
by 0.07 km2 (11.86%). On the other hand, shrubs
and grasses declined by 74.76 km2 (− 46.29%) and
forests by 15.59 km2 (− 17.43%). In the third period
(2014–2023), agricultural land, shrubs and grasses,
Fig. 4 The LULC map of 2023 shows active mining sites and the study area villages
Fig. 5 LULC class change
(1995–2004, 2004–2014,
2014–2023, and 1995–
2023)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 10 of 20
Vol:. (1234567890)
bareland, and built-up areas all increased by 18.73
km2, 16.01 km2, 1.93 km2, and 0.3 km2, respectively,
while forest areas diminished by 36.97 km2. Over-
all (1995–2023), there was an increase in agricul-
tural land 168.51 km2 (587.55%), bareland 7.70 km2
(121.45%), and built-up areas 0.55 km2 (134.15%),
while forest and shrubs and grasses areas declined
by 97.67 km2 (− 72.59%) and 79.09 km2 (− 43.49%),
respectively (Fig.5).
LULC change intensity analysis
The intensity analysis of the LULC dynamics in
the study region is presented in Fig. 6. The results
Fig. 6 Category inten-
sity analysis. a Intensity
gain area outcomes and b
intensity loss area outcomes
for three-time intervals:
1995–2004, 2004–2014,
and 2014–2023
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 11 of 20 464
Vol.: (0123456789)
revealed varied dynamics of the LULC classes driven
by anthropogenic forces due to mining operations in
the region. The category level analysis (Fig.6) shows
the size and annual intensity of change of each cat-
egory’s gain relative to the size of the category and
the interval end time point. The plot shows that built-
up, bareland, and agricultural land gained the most
during all the time intervals. The result revealed that
bareland, built-up areas, and agricultural land had
active changes during the two intervals. The right
side of Fig.6a shows that the bar for gain of bareland,
built-up areas, and agricultural land extends to the
right of the uniform line in both time intervals except
agricultural land in the first interval. In terms of loss,
shrubland areas and forestland areas were the major
losers (Fig.6b). The right side of Fig.6b the forested
area extends to the right of the uniform rate during
the first and the third period indicating that the inten-
sity loss of the forest area was active during these
periods. Similarly, the bar for loss of shrubland areas
extends to the right of the uniform rate in the first and
second intervals; however, the intensity of loss was
more active during the second period (Fig. 6b). The
gross changes of the LULC category revealed that the
forest, shrubs, and grasses experienced net losses dur-
ing the study period.
Land use transitions
A Sankey plot (Fig.7) illustrated the relative changes
in the LULC classes of the study area over three dis-
tinct years: 1995, 2004, 2014, and 2023. The plot
provides a robust visualization of transitions between
different land cover types, highlighting how various
Fig. 7 Sankey plot showing the LULC transitions between 1995, 2004, 2014, and 2023 in the study area
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 12 of 20
Vol:. (1234567890)
LULC classes have evolved following the discov-
ery and initiation of mining operations. It is evi-
dent from the results that much of the forestland in
the first period transitioned to agricultural land and
shrubs and grasses. A similar pattern was observed in
the second period (Fig.7). Shrubs and grasses in the
first period lost much of their coverage to agricultural
lands (82.57 km2), with a small portion converting to
forestlands (9.86 km2).
In the second period, agricultural lands further
expanded, but part of the gains from the first epoch
(37.89 km2) were lost to shrublands. Nonetheless,
agricultural lands maintained a steady growth across
the two analyzed intervals. Forests reduced further
in the second decade, losing most of its coverage
to shrubs and grasses (20.74 km2) and agricultural
land (19.25 km2). Similarly, bareland areas steadily
increased in area over the two decades, gaining much
of its coverage from shrubs and grasses, and forest
(Fig.7). Built-up areas expansion from 1995 to 2014
was slow; however, from 2014 to 2023, the expansion
surged, reflecting rapid urbanization and infrastruc-
tural development in the region. This trend under-
scores ongoing urbanization due to mining activities
as the result of the conversion of natural or semi-natu-
ral landscapes into urban areas.
Perceived environmental, social, and economic
implications of gold mining
Household socio‑economic characteristics
Table3 summarizes the socio-economic characteris-
tics of the population in the study area. The survey
revealed that most household heads in the study area
were males, constituting about 69.9%, while 30.1%
were females. Most respondents (48.2%) had attained
primary-level education, while a few (9.6%) had no
formal education. Most residents (38.6%) in the area
engaged in farming, whereas 15.7% were involved
in mining. The average age of the respondents was
37years, while the mean household size in the study
area was 6 (Table3).
Perceived environmental, social, andeconomic
implications ofmining
Linking the observed LULC changes from satellite
images to the perceived social, economic, and envi-
ronmental implications of the changes on the liveli-
hoods of local communities is essential in LULC
studies (Basommi etal., 2016a, 2016b). The majority
of respondents acknowledged the impacts of mining
Table 3 Summary of the
sampled households’ socio-
economic characteristics
Variables Characteristics (n = 83) (%) Min Max Mean Std. dev
Gender Female 25 30.1
Male 58 69.9
Age (years) 83 18 75 37.1 10.8
Marital status Single 14 16.9
Married 61 73.5
Separated 6 7.2
Divorced 1 1.2
Widowed 1 1.2
Education level No formal education 8 9.6
Primary 40 48.2
Secondary 17 20.5
Tertiary 18 21.7
Occupation Skilled labor 4 4.8
Businessperson 9 10.8
Casual labor 13 15.7
Civil servant 12 14.5
Farmer 32 38.6
Miner 11 15.7
Household size Number of members 83 1 25 5.5 3.7
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 13 of 20 464
Vol.: (0123456789)
in the study area, with 98% reporting social and eco-
nomic implications, and 96% observing environ-
mental consequences. The perceived environmental
impacts of mining included air pollution (92%), bio-
diversity loss (90%), deforestation (88%), land degra-
dation (84%), water pollution (63%), and soil pollu-
tion (27%). The top three social impacts of mining in
the study area were occupational hazards, especially
for the mine workers (88%), land use conflicts (83%),
and negative impacts on livelihood and culture (72%).
The leading perceived economic impacts of mining
were improved housing (92%), infrastructure devel-
opment (84%), and job creation (82%). The details of
the perceived impacts of mining in the study area are
presented in Fig.8.
Discussion
LULC trends
The LULC analysis was performed with an over-
all accuracy exceeding 80% and a kappa coefficient
greater than 0.68 (Table1), which meets the recom-
mended thresholds for LULC classifications. These
accuracy levels surpass the classification accuracy
assessment reported by Patil and Nataraja, (2020),
reinforcing the reliability of the generated LULC
maps. The LULC analysis and social survey results
indicate that mining and associated activities are the
key drivers of LULC dynamics in the Mang’onyi,
Sambaru, and Londoni areas of the Singida region.
This is because most of the changes occurred after the
year 2004 when gold reserves exploration and min-
ing intensified in the aforementioned villages (United
Republic of Tanzania, 2014). While forests, shrubs,
and grasses were the initial dominant land cover types
in the study area in 1995 and 2004, respectively, agri-
cultural land, bareland, and built-up areas became
much more visible in 2014 and 2023 (Fig.3). Built-
up areas increased steadily over the study period due
to general population growth and population influx
into the study area after the discovery of gold reserves
in 2004 (Lugoe, 2011; Ministry of Finance and Plan-
ning, 2022). The discovery of gold mineral reserves
in the region enticed local artisanal miners, gold rush
miners, and small-scale and large-scale mining com-
panies to participate in mining operations (Jønsson &
Bryceson, 2009). The influx of small business owners
in the mining area also boosted the population den-
sity in the surrounding communities. This accelerated
the construction of residences in mining camps and
neighboring villages, thus increasing the build-up
area over the study period (Fig.4). The relocation of
Fig. 8 Perceived environ-
mental, social, and eco-
nomic impacts of mining in
the study area
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 14 of 20
Vol:. (1234567890)
the inhabitants from the Shanta mining site to a newly
planned and constructed village, particularly in the
northern part of the study area, enhanced the visibil-
ity of the build-up region in 2023 (Fig.3). Our find-
ings align with findings from other studies (Basommi
etal., 2016a, 2016b; Garai & Narayana, 2018; Gbedzi
et al., 2022), which have documented similar trends
of increasing built-up areas in the mining sites due to
the mining-induced influx of people into the mining
area, which created the demand for better housing.
A notable continuous forest decline (Table2) was
observed in the study area over the three decades
(1995 to 2023). The forested area was significantly
converted into shrublands, grasslands, agricultural
fields, barelands, and built-up areas (Fig. 7). While
the landscape underwent a significant and ongoing
reduction in forest cover, other land cover type classes
(agricultural land, bareland, and buildup) progres-
sively expanded over the years. This pattern of forest
change aligns with findings from other studies (Addo-
Fordjour & Ankomah, 2017; Garai & Narayana,
2018; Kamga etal., 2020; Kumi et al., 2021), which
have documented similar trends of declining forest
cover, often transformed into other land use classes
due to gold mining activities. The transformation
of forested areas is linked to mining-related activi-
ties and other human-induced disturbances brought
on by the influx of people to the mining site. This is
evidenced by the population surge following the gold
discovery in Londoni in 2004, where the village ini-
tially had approximately 1,600 subsistence farmers
and livestock keepers, but within a few months, its
population grew to over 10,000 as people engaged in
hard rock gold mining (Jønsson & Bryceson, 2009).
Mining operations generally require clearing vegeta-
tion (Mishra etal., 2022) to open up the mining site
and provide lumber for the mining tunnels (open
pit), leading to forest loss. The increase in population
due to the influx of rush miners boosted demand for
firewood and charcoal, exacerbating unlawful forest
resource harvesting and thus decreasing forested areas
over time. Additionally, mining activities in areas
where farmlands existed led to the shifting of farming
activities into the nearby forest zones. The intensifica-
tion of agricultural operations towards the adjoining
forest zone due to the great demand for food among
the surrounding populations resulted in a decline in
the forest areas. A study in Western Ghana similarly
showed a substantial loss of farmlands within mining
areas and widespread spill-over effects as displaced
farmers expand farmland into forest areas (Schueler
etal., 2011).
Interestingly, there was a constant rise in agricul-
tural acreage (Table2 and Fig.7) over the years in the
study area. The expansion of agricultural land in the
study area is likely driven by the rising demand for
food, spurred by population growth resulting from the
influx of people attracted by mining activities. This
increased population places additional pressure on
land resources, necessitating the conversion of natural
landscapes into agricultural fields to meet the grow-
ing food requirements. This aligns with the observa-
tions from other studies (Garai & Narayana, 2018;
Kanianska, 2016; Kumi et al., 2021) that reported
human activities significantly accelerate land-use
change, driven by rapid population growth and the
escalating demand for food to sustain the growing
population. Changes in economic priorities (from
mining to agriculture) among gold rush miners and
population growth might explain the steady increase
in agricultural land in the study area. This converted
forest, bareland, shrub, and grassland areas into agri-
cultural fields (Fig. 7). This is seen in the increase
of agricultural fields in the Chipinga region in 2023
(Fig.4), as most of the residents of Londoni Village
established new farms in the Chipinga area.
Bareland areas, which are predominantly created
through mining and human activities, similarly kept
increasing throughout the study period since min-
ing operations in the study area primarily involve
surface (alluvial mining) and hard rock gold mining
(underground mining tunnels) via excavation methods
(Jønsson & Bryceson, 2009). Surface mining involves
the removal of ground vegetation and soils, result-
ing in bareland (Schueler etal., 2011). Therefore, the
continuous expansion of bareland can be attributed to
the expansion of small-scale mining sites, large-scale
production of gold by commercial companies, and the
construction of roads for gold and agricultural prod-
ucts transportation within and out of the study area.
The changes in grasses and shrubs over time can
be linked to conversion to other land cover types,
such as bareland and agricultural (Gbedzi et al.,
2022). Between 2004 and 2014, the incoming popu-
lation converted many of the shrubs and grasses into
agricultural lands to cater to their food production
needs. Mining in the forest also led to forest degra-
dation and conversion of forests to grass and shrubs.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 15 of 20 464
Vol.: (0123456789)
After the exhaustion of minerals in most of the min-
ing areas between 2014 and 2023, the bareland was
abandoned, allowing shrubs and grasses to grow and
subsequently expand in area coverage.
Environmental and socio-economic implications
Results from the social survey of the study area resi-
dents backed by relevant existing literature suggest
that gold mining activities and the resultant LULC
changes in the Singida region of central Tanzania
have had significant environmental and socio-eco-
nomic implications.
Environmental implications
The social survey established water and soil pollution
as a key environmental impact of mining in the study
area (Fig.8). This finding is validated by a study car-
ried out around the small-scale mines of Londoni
and Sambaru, which revealed heavy metals (Hg, Pb,
Zn, and Cu) contamination of water and soils above
the permissible maximum Tanzanian limits (Her-
man & Kihampa, 2015). Sources of the heavy metal
contamination were from mining-related activities
including discharge of mine water to the surround-
ings, amalgamation and burning activities, improper
waste rocks and tailings disposal, and leachates
from waste rocks and tailings (Herman & Kihampa,
2015). The respondents also listed land degradation,
deforestation, and biodiversity loss as environmental
impacts of mining (Fig.8). Surface mining interrupts
ecosystem service flows, removes ground vegetation
and soils, often causes irreversible loss of farmlands
(Schueler et al., 2011). Surface mining also causes
deforestation and habitat loss, which harms native
wildlife and biodiversity. Clearing vegetation on the
mining sites promotes soil erosion, which can further
deteriorate the land. These findings align with that of
Mencho (2022), who reported water pollution, defor-
estation, land degradation, and biodiversity loss due
to gold mining operations in the Shekiso district, Guji
zone, Ethiopia.
Air pollution, primarily generated by mine blasts
and the resulting dust, is a major environmental con-
cern in the region, particularly near the Shanta min-
ing site. Many respondents voiced concerns about
the spread of dust particles in the air following each
blast, particularly noting its impact on visibility
for drivers near the site. The dust becomes so thick
that drivers are often forced to turn on their head-
lights even during the day to improve visibility and
ensure safer driving conditions. Additionally, resi-
dents reported multiple instances of bird deaths and
foxes around the Shanta mining site, suspecting that
the birds drank untreated, uncovered wastewater. This
can pose a significant health risk to children who may
collect and consume these birds when their parents
are not at home. Without proper supervision, expo-
sure to contaminated birds could pose significant
health risks, including cancer, as these toxic chemi-
cals can accumulate and transfer to higher trophic
levels. This underscores the urgent need for imme-
diate action to mitigate this issue. In Sambaru, near
the Yusuph Mwandami vat leaching plant, residents
reported the deaths of 11 cows in 2022 and two more
in 2023, attributing these fatalities to the consumption
of wastewater from gold processing operations. Sev-
eral cases of goats, sheep, and other animals dying
have been reported due to insufficient fencing around
the mining areas, which allows animals to access the
wastewater sites. This lack of proper barriers has led
to the animals’ exposure to hazardous waste, result-
ing in their deaths and eventually leading to biodiver-
sity loss. The study underscores that the environmen-
tal impact of these mining operations in the area is
severe and must be addressed immediately.
Social implications
Occupational hazards for mine workers in the form of
exposure to dust, chemicals, and unsafe working con-
ditions were listed as a key social impact in the study
region (Fig.8). Results from a similar study carried
out at Londoni village in the Manyoni District of Sin-
gida region revealed that mineworkers were exposed
to health and safety risks due to alack of protective
gears and rudimentary technologies used in gold
extraction (Ringo & Kingu, 2018). These risks lead to
diseases (diarrhea, flu, and backbone pains), injuries,
loss of lives, and property loss due to exposure to
chemicals and machine accidents. Negative impacts
on local livelihoods and the culture of residents were
also attributed to the mining activities. Ringo and
Kingu (2018) established social vices resulting from
the existence of mining activities in the study area.
According to Ringo and Kingu (2018), there has been
a notable rise in sexually transmitted diseases (STDs)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 16 of 20
Vol:. (1234567890)
from unsafe sexual intercourse and increased drug
and substance abuse in and around the mining camps.
Some respondents, particularly those with high blood
pressure, voiced concerns about the noise pollution
caused by blasting rock minerals, especially around
the Shanta mining site. In some cases, individuals
even collapsed due to the extreme noise levels during
the blasts. The problem is more severe for individu-
als with critical conditions, who are notified of blast-
ing days in advance and evacuated to safer areas away
from the blasting site. This highlights the severity of
the issue, as the high-pitched noise from the blasting
poses serious health risks, particularly for those with
pre-existing conditions.
Community displacement and land use conflicts
were other outstanding social impacts of mining in
the area. Although the intention of evacuating the
surrounding community, particularly near the Shanta
mining site, was well-meaning, respondents still
expressed concerns about the lack of essential ser-
vices such as access to clean water, hospitals, and
schools in their new settlement. The absence of these
critical services has left the displaced communities
struggling to meet these services despite the evacu-
ation efforts. The inadequate compensation and lack
of public education about the terms and pricing of
land compensation have intensified land use con-
flicts between the Shanta mining company and the
surrounding community. This lack of clarity and
fair compensation has increased tensions and dis-
putes over land ownership and usage. Previous stud-
ies have likewise reported land use conflicts between
miners due to the large operation areas demand, and
the surrounding communities who depend largely
upon the land for their livelihoods (Hilson, 2002;
Schueler et al., 2011). Child labor and discrimina-
tion of women in the allocation of mining jobs were
also mentioned as other social impacts. Child labor
is common in the small-scale gold mines in Tanza-
nia. Household poverty is the main factor pushing
children to work in the mines due to their house-
holds’ inability to provide for their basic needs (Metta
etal., 2023). The unequal participation of males and
females in mining activities has a social impact con-
tributing to financial disparities among household
members. Our study aligns with the study conducted
in the Prestea–Huni Valley Municipality of Ghana,
which reported that women involved in artisanal gold
mining were correspondingly discriminated against
via cultural marginalization, poor work support ser-
vices for women with children, poor working envi-
ronment, and inter-ethnic discrimination by employ-
ers (Arthur-Holmes & Abrefa Busia, 2021).
Economic implications
The economic implications of gold mining were
mostly positive as they comprised improved hous-
ing, infrastructural development, job creation, and
improved access to social services. Gold mining
and processing activities contribute to the economic
development of the study region and Tanzania as a
whole since it offers employment and an alternative
source of income to the mine workers (Merket, 2018;
Phillips etal., 2001; Ringo & Kingu, 2018). Respond-
ents reported experiencing economic growth follow-
ing the discovery of gold deposits in the area in 2004,
which provided an alternative source of income and
significantly improved their livelihoods. This newly
discovered resource led to greater financial stability
and opportunities for the local community, as dem-
onstrated by improved housing conditions (Fig. 8).
The influx of gold miners, mining companies, mine
workers, and business people in the area created a
heightened demand for better housing. This influx of
people significantly increased the need for improved
living accommodations to support the growing pop-
ulation. Infrastructure improvements throughout
the year have significantly boosted the import and
export of goods, particularly benefiting small busi-
ness owners and farmers, resulting in better trade and
economic opportunities. The increasing population
also led to a higher demand for food, which boosted
agricultural activities, such as irrigation practices in
nearby villages, ultimately improving the livelihoods
of the surrounding community. This rise in agricul-
tural engagement provided greater food security and
enhanced economic opportunities for residents. This
finding is echoed by Haundi etal. (2021) who assert
that artisanal and small-scale gold mining are key
sources of informal employment for local commu-
nities and an alternative source of income for rural
farmers, especially during the dry season when farm-
ing becomes unfavorable. A study by Pokorny etal.
(2019) in Burkina Faso likewise associated better
wages, infrastructural development, improved social
programs, and new business opportunities with min-
ing operations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 17 of 20 464
Vol.: (0123456789)
Conclusions
This study demonstrates that gold mining has been
a significant driver of LULC changes in the Singida
region, particularly in the Mang’onyi, Sambaru,
and Londoni mining sites within the Ikungi and
Manyoni districts. Over the period from 1995 to
2023, mining activities contributed to the expan-
sion of agricultural land, bareland, and built-up
areas, while forest cover, shrublands, and grasslands
have declined. These findings suggest that mining-
induced land transformations could have long-term
environmental consequences, including habitat
degradation, soil erosion, and water contamination.
Beyond the environmental impacts, the study fur-
ther reveals socio-economic challenges faced by
local communities, such as land displacement,
loss of traditional livelihoods, and conflicts over
resource use. Community perceptions highlight
concerns about livelihood sustainability, inadequate
compensation mechanisms, and environmental deg-
radation, emphasizing the urgent need for integrated
land management policies.
While mining has economic benefits, its sustain-
ability remains in question if environmental regula-
tions and rehabilitation measures are not adequately
enforced. Therefore, policymakers, mining com-
panies, and local communities must collaborate to
establish responsible mining practices. This includes
enforcing stronger environmental regulations, imple-
menting land restoration programs, improving com-
pensation mechanisms, and integrating local com-
munities into decision-making processes. Balancing
economic gains from gold mining with environmental
conservation and social equity is imperative. With-
out proactive interventions, the long-term sustain-
ability of land resources and community well-being
in thestudy area could be compromised. This study
underscores the need for science-driven policy inter-
ventions to ensure that mining remains a catalyst for
development without jeopardizing ecosystem health
and local livelihoods. The study acknowledges limi-
tations such as the need for long-term monitoring to
capture finer-scale land changes and their cascading
effects. Future research should explore quantitative
assessments of soil and water quality, biodiversity
losses, and socio-economic livelihood transitions to
provide a more comprehensive understanding of min-
ing’s long-term effects.
Acknowledgements The authors of this article would like to
express their sincere gratitude to Mr. Muhsin Yusuph and Eng.
Petro Elia Maidi for their assistance in conducting the socio-
economic survey. The authors also thank the reviewers for their
valuable time, expertise, insightful comments, and constructive
criticism, which greatly contributed to improving the quality of
this work.
Author contribution A.S.L, Conceptualization, data col-
lection, Methodology, Analysis, Writing-original draft,
Writing-Review and editing, B.R, Conceptualization, data
collection, Methodology, Analysis, Writing-original draft,
Writing-Review and editing, A.A,Conceptualization, data
collection, Methodology, Analysis, Writing-original draft,
Writing-Review and editing H.K,Conceptualization, data col-
lection, Methodology, Analysis, Writing-original draft, Writ-
ing-Review and editing, S.R.M, data collection, Methodology,
K.C. Writing-Review and editing:
Funding Open access funding provided by Hungarian Uni-
versity of Agriculture and Life Sciences. This research was
funded by the “TKP2021-NVA-13 project,” implemented
with support from the Ministry of Culture and Innovation of
Hungary. The funding was provided by the National Research,
Development and Innovation Fund under the TKP2021-NVA
funding scheme.
Data Availability No datasets were generated or analysed
during the current study.
Declarations
Ethics approval The authors confirm that this research was
conducted per ethical guidelines and regulations.
Competing interests The authors declare no competing inter-
ests.
Open Access This article is licensed under a Creative Com-
mons Attribution 4.0 International License, which permits
use, sharing, adaptation, distribution and reproduction in any
medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Crea-
tive Commons licence, and indicate if changes were made. The
images or other third party material in this article are included
in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not
included in the article’s Creative Commons licence and your
intended use is not permitted by statutory regulation or exceeds
the permitted use, you will need to obtain permission directly
from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 18 of 20
Vol:. (1234567890)
References
Addo-Fordjour, P., & Ankomah, F. (2017). Patterns and driv-
ers of forest land cover changes in tropical semi-deciduous
forests in Ghana. Journal of Land Use Science, 12(1),
71–86. https:// doi. org/ 10. 1080/ 17474 23X. 2016. 12413 13
Ahmed, A., Rotich, B., & Czimber, K. (2024). Assessment of
the environmental impacts of conflict-driven internally
displaced persons: A Sentinel-2 satellite based analysis of
land use/ cover changes in the Kas locality, Darfur, Sudan.
PLoS ONE, 19(5 May), 1–19. https:// doi. org/ 10. 1371/
journ al. pone. 03040 34
Al Rawashdeh, R., Campbell, G., & Titi, A. (2016). The socio-
economic impacts of mining on local communities: The
case of Jordan. Extractive Industries and Society, 3(2),
494–507. https:// doi. org/ 10. 1016/j. exis. 2016. 02. 001
Allaire, J. J., Ellis, P., Gandrud, C., Owen, J., Russell, K.,
Rogers, J., Sese, C., & Yetman, C. (2022). Package
‘networkD3.’
Altieri, M. A., Nicholls, C. I., Henao, A., & Lana, M. A.
(2015). Agroecology and the design of climate change-
resilient farming systems. Agronomy for Sustainable
Development, 35(3), 869–890. https:// doi. org/ 10. 1007/
s13593- 015- 0285-2
Arthur-Holmes, F., & Abrefa Busia, K. (2021). Occupying the
fringes: The struggles of women in artisanal and small-
scale gold mining in Rural Ghana—Evidence from the
Prestea-Huni Valley Municipality. Gender Issues, 38(2),
156–179. https:// doi. org/ 10. 1007/ s12147- 020- 09261-4
Aryee, B. N. A. (2001). Ghana’s mining sector: Its contribution
to the national economy. Resources Policy, 27(2), 61–75.
https:// doi. org/ 10. 1016/ S0301- 4207(00) 00042-8
Basommi, L. P., Guan, Q., Cheng, D., & Singh, S. (2016a).
Dynamics of land use change in a mining area: A case
study of Nadowli District, Ghana. Journal of Mountain
Science, 13(2), 633–634.
Basommi, L. P., Guan, Q. F., Cheng, D. D., & Singh, S. K.
(2016b). Dynamics of land use change in a mining area: A
case study of Nadowli District, Ghana. Journal of Moun‑
tain Science, 13(4), 633–642. https:// doi. org/ 10. 1007/
s11629- 015- 3706-4
Breiman, L. (2001). Random forests. Machine Learning, 45,
5–32. https:// doi. org/ 10. 1007/ 978-3- 030- 62008-0_ 35
Eduful, M., Alsharif, K., Eduful, A., Acheampong, M., Edu-
ful, J., & Mazumder, L. (2020). The illegal artisanal and
small-scale mining (Galamsey) ‘Menace’ in Ghana: Is
military-style approach the answer? Resources Policy,
68(July 2019), 101732. https:// doi. org/ 10. 1016/j. resou
rpol. 2020. 101732
Ericsson, M., & Löf, O. (2019). Mining’s contribution to
national economies between 1996 and 2016. Gornaya
Promyshlennost, 2019(6), 84–93. https:// doi. org/ 10.
30686/ 1609- 9192- 2019-6- 84- 93
Exavier, R., & Zeilhofer, P. (2020). OpenLand: Software for
quantitative analysis and visualization of land use and
cover change. R Journal, 12(i), 1–16. https:// doi. org/ 10.
32614/ rj- 2021- 021
Garai, D., & Narayana, A. C. (2018). Land use/land cover
changes in the mining area of Godavari coal fields of
southern India. Egyptian Journal of Remote Sensing and
Space Science, 21(3), 375–381. https:// doi. org/ 10. 1016/j.
ejrs. 2018. 01. 002
Gbedzi, D. D., Ofosu, E. A., Mortey, E. M., Obiri-Yeboah, A.,
Nyantakyi, E. K., Siabi, E. K., Abdallah, F., Domfeh, M.
K., & Amankwah-Minkah, A. (2022). Impact of mining
on land use land cover change and water quality in the
Asutifi North District of Ghana West Africa. Environmen‑
tal Challenges, 6(August 2021), 100441. https:// doi. org/
10. 1016/j. envc. 2022. 100441
Gbegbelegbe, S., Serem, J., Stirling, C., Kyazze, F., Radeny,
M., Misiko, M., Tongruksawattana, S., Nafula, L., Gakii,
M., & Sonder, K. (2018). Smallholder farmers in eastern
Africa and climate change: A review of risks and adapta-
tion options with implications for future adaptation pro-
grammes. Climate and Development, 10(4), 289–306.
https:// doi. org/ 10. 1080/ 17565 529. 2017. 13742 36
Hammond, D. S., Gond, V., De Thoisy, B., Forget, P. M., &
DeDijn, B. P. E. (2007). Causes and consequences of
a tropical forest gold rush in the Guiana Shield. South
America. Ambio, 36(8), 661–670. https:// doi. org/ 10. 1579/
0044- 7447(2007) 36[661: CACOAT] 2.0. CO;2
Haundi, T., Tsokonombwe, G., Ghambi, S., Mkandawire, T.,
& Kasambara, A. (2021). An investigation of the socio-
economic benefits of small-scale gold mining in Malawi.
Mining, 1(1), 19–34. https:// doi. org/ 10. 3390/ minin g1010
003
Herman, A., & Kihampa, C. (2015). Heavy metals contamina-
tion in soils and water in the vicinity of small scale gold
mines at Londoni and Sambaru, Singida Region, Tanza-
nia. International Journal of Environmental Monitoring
and Analysis, 3(6), 397. https:// doi. org/ 10. 11648/j. ijema.
20150 306. 13
Hilson, G. (2002). An overview of land use conflicts in mining
communities. Land Use Policy, 19, 65–73.
Jønsson, J. B., & Bryceson, D. F. (2009). Rushing for gold:
Mobility and small-scale mining in East Africa. Devel‑
opment and Change, 40(2), 249–279. https:// doi. org/ 10.
1111/j. 1467- 7660. 2009. 01514.x
Kamga, M. A., Nguemhe Fils, S. C., Ayodele, M. O., Olatu-
bara, C. O., Nzali, S., Adenikinju, A., & Khalifa, M.
(2020). Evaluation of land use/land cover changes due to
gold mining activities from 1987 to 2017 using Landsat
imagery, East Cameroon. Geojournal, 85(4), 1097–1114.
https:// doi. org/ 10. 1007/ s10708- 019- 10002-8
Kanianska, R. (2016). Agriculture and its impact on land-
use, environment, and ecosystem services. Landscape
Ecology‑The Influences of Land Use and Anthropogenic
Impacts of Landscape Creation, 1–26. https:// doi. org/ 10.
5772/ 63719
Karasiak, N. (2020). Museo ToolBox: A Python library for
remote sensing including a new way to handle rasters.
Journal of Open Source Software, 5(48), 1978. https:// doi.
org/ 10. 21105/ joss. 01978
Kuffour, R. A., Tiimub, B. M., & Agyapong, D. (2018).
Impacts of illegal mining (Galamsey) on the environment
(water and soil) at Bontefufuo area in the Amansie West
District. Journal of Environment and Earth Science, 8(7),
98–107. https:// www. iiste. org.
Kumi, S., Addo-Fordjour, P., Fei-Baffoe, B., Belford, E. J.
D., & Ameyaw, Y. (2021). Land use land cover dynam-
ics and fragmentation-induced changes in woody plant
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464 Page 19 of 20 464
Vol.: (0123456789)
community structure in a mining landscape, Ghana. Trees,
Forests and People, 4(September 2020), 100070. https://
doi. org/ 10. 1016/j. tfp. 2021. 100070
Kuzevic, S., Bobikova, D., & Kuzevicova, Z. (2022). Land
cover and vegetation coverage changes in the mining
area—A case study from Slovakia. Sustainability (Swit‑
zerland), 14(3), 1180. https:// doi. org/ 10. 3390/ su140 31180
Landis, R., & Koch, G. (1977b). An application of hierarchical
kappa-type statistics in the assessment of majority agree-
ment among multiple observers. Biometrics, 33(2), 363–
374. Koch Published by: International Biometric Society
Stable URL: https:// www. jstor. org/ stab.
Landis, J. R., & Koch, G. G. (1977a). The measurement of
observer agreement for categorical data. Biometrics,
33(1), 159–174. https:// doi. org/ 10. 2307/ 25293 10
Latifovic, R., Fytas, K., Chen, J., & Paraszczak, J. (2005).
Assessing land cover change resulting from large surface
mining development. International Journal of Applied
Earth Observation and Geoinformation, 7(1), 29–48.
https:// doi. org/ 10. 1016/j. jag. 2004. 11. 003
Shanta Gold Limited. (2024). Q4 2023 production and opera-
tional update. In Report.
Lugoe, F. (2011). The Land Factor in Mining Gold Reserves
in Tanzania. Policy briefing 34, Governance of Africa
Resource Programme.
Mencho, B. B. (2022). Assessing the effects of gold mining
on environment: A case study of Shekiso district, Guji
zone. Ethiopia. Heliyon, 8(12), e11882. https:// doi. org/ 10.
1016/j. heliy on. 2022. e11882
Mensah, A. K., Mahiri, I. O., Owusu, O., Mireku, O. D.,
Wireko, I., & Kissi, E. A. (2015). Environmental impacts
of mining: A study of mining communities in Ghana.
Applied Ecology and Environmental Sciences, 3(3),
81–94. https:// doi. org/ 10. 12691/ aees-3- 3-3
Merket, H. (2018). Mapping artisanal and small‑scale mining
in northwest Tanzania: A survey on its nature, scope, and
impact. International Peace Information Service vzw.
Metta, E., Abdul, R., Koler, A., & Geubbels, E. (2023). Eco-
logical aspects shaping child labour in Tanzania’s arti-
sanal and small-scale gold mines: A qualitative inquiry.
Heliyon, 9(3), e14417. https:// doi. org/ 10. 1016/j. heliy on.
2023. e14417
Ministry of Finance and Planning, National Bureau of Statistics
Tanzania (2022). Age and sex distribution Census report.
Mishra, M., Augusto, C., Victor, T., Kumar, M., & Marques, R.
(2022). Mining impacts on forest cover change in a tropi-
cal forest using remote sensing and spatial information
from 2001–2019: A case study of Odisha (India). Journal
of Environmental Management, 302, 114067. https:// doi.
org/ 10. 1016/j. jenvm an. 2021. 114067
Oo, T. K., Arunrat, N., Sereenonchai, S., Ussawarujikulchai,
A., Chareonwong, U., & Nutmagul, W. (2022). Compar-
ing four machine learning algorithms for land cover classi-
fication in gold mining: A case study of Kyaukpahto Gold
Mine, Northern Myanmar. Sustainability (Switzerland),
14(17), 10754. https:// doi. org/ 10. 3390/ su141 710754
Patil, N. S., & Nataraja, M. (2020). Effect of land use land
cover changes on runoff using hydrological model: A case
study in Hiranyakeshi watershed. Modeling Earth Systems
and Environment, 6(4), 2345–2357. https:// doi. org/ 10.
1007/ s40808- 020- 00808-8
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., &
Thirion, B. (2011). Scikit-learn: Machine learning in
Python. Journal OfMachine Learning Research, 12(9),
2825–2830. https:// doi. org/ 10. 1289/ EHP47 13
Phillips, L. C., Semboja, H., Shukla, G. P., Swinga, R., Mutag-
waba, W., Mchwampaka, B., Wanga, G., Kahyarara, G.,
& Keller, P. C. (2001). Tanzania’s precious minerals
boom: Issues in mining and marketing. In Sustainable
Development.
Pinto, R., de Jonge, V. N., Neto, J. M., Domingos, T., Marques,
J. C., & Patrício, J. (2013). Towards a DPSIR driven inte-
gration of ecological value, water uses and ecosystem
services for estuarine systems. Ocean and Coastal Man‑
agement, 72, 64–79. https:// doi. org/ 10. 1016/j. oceco aman.
2011. 06. 016
Plourde, L., & Congalton, R. G. (2003). Sampling method
and sample placement: How do they affect the accuracy
of remotely sensed maps? Photogrammetric Engineering
and Remote Sensing, 69(3), 289–297. https:// doi. org/ 10.
14358/ PERS. 69.3. 289
Poignant, A. (2023). Small-scale mining and agriculture: Evi-
dence from northwestern Tanzania. Resources Policy,
83(December 2022), 103694. https:// doi. org/ 10. 1016/j.
resou rpol. 2023. 103694
Pokorny, B., von Lübke, C., Dayamba, S. D., & Dickow, H.
(2019). All the gold for nothing? Impacts of mining on
rural livelihoods in Northern Burkina Faso. World Devel‑
opment, 119, 23–39. https:// doi. org/ 10. 1016/j. world dev.
2019. 03. 003
Ringo, J., & Kingu, R. (2018). A status of health and safety
to small scale mineworkers in Singida Region, Tanzania.
International Journal of Scientific and Research Publi‑
cations (IJSRP), 8(3). https:// doi. org/ 10. 29322/ ijsrp.8. 3.
2018. p7550
Rodriguez-Galiano, V. F., Ghimire, B., Rogan, J., Chica-Olmo,
M., & Rigol-Sanchez, J. P. (2012). An assessment of the
effectiveness of a random forest classifier for land-cover
classification. ISPRS Journal of Photogrammetry and
Remote Sensing, 67(1), 93–104. https:// doi. org/ 10. 1016/j.
isprs jprs. 2011. 11. 002
Rubo, R. A., de Carvalho Carneiro, C., Michelon, M. F., & dos
S Gioria, R. (2019). Digital petrography: Mineralogy and
porosity identification using machine learning algorithms
in petrographic thin section images. Journal of Petroleum
Science and Engineering, 183(June), 106382. https:// doi.
org/ 10. 1016/j. petrol. 2019. 106382
Schueler, V., Kuemmerle, T., & Schröder, H. (2011). Impacts
of surface gold mining on land use systems in Western
Ghana. Ambio, 40(5), 528–539. https:// doi. org/ 10. 1007/
s13280- 011- 0141-9
Stephano, A., Rotich, B., Ahmed, A., Kipkulei, K., Akos, P.,
& Boros, E. (2025). Influence of environmental vari-
ables and anthropogenic activities on soda-saline lakes
chemistry in northern Tanzania: A remote sensing and
GIS approach. Heliyon, 11(2), e41691. https:// doi. org/ 10.
1016/j. heliy on. 2025. e41691
Suglo, P., Effah, P., Amponsah Acheampong, A., Sunkari, R.,
& Yeboah, A. (2021). Effects of illegal mining on the
environment, economy, and agricultural productivity. Bio‑
chemistry and Molecular Biology, 6(4), 79. https:// doi.
org/ 10. 11648/j. bmb. 20210 604. 11
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Environ Monit Assess (2025) 197:464
464 Page 20 of 20
Vol:. (1234567890)
Thonfeld, F., Steinbach, S., Muro, J., & Kirimi, F. (2020).
Long-term land use/land cover change assessment of the
Kilombero catchment in Tanzania using random forest
classification and robust change vector analysis. Remote
Sensing, 12(7), 1057. https:// doi. org/ 10. 3390/ rs120 71057
United Republic of Tanzania. (2014). Mineral sector environ‑
mental impact assessment guideline. Ministry of Energy
and Minerals. Govt Printer, Dar es Salaam.
Worlanyo, A. S., & Jiangfeng, L. (2021). Evaluating the envi-
ronmental and economic impact of mining for post-mined
land restoration and land-use: A review. Journal of Envi‑
ronmental Management, 279(November 2020), 111623.
https:// doi. org/ 10. 1016/j. jenvm an. 2020. 111623
Yadav, A., Kansal, M. L., & Singh, A. (2024). Land use and
land cover dynamics in the Upper Ganga Riverine Wet-
land: Unraveling ecosystem services over two decades.
Environmental Monitoring and Assessment, 196(6).
https:// doi. org/ 10. 1007/ s10661- 024- 12748-2
Yager, T. R. (2023). The mineral industry of Tanzania. Tanza-
nia-2019 (ADVANCE RELEASE), October, 49.1–49.13.
https:// miner al. usgs. gov/ miner als/ pubs/ count ry/ 2008/
myb3- 2008. pp. pdf
Yu, H., & Zahidi, I. (2023). Spatial and temporal variation of
vegetation cover in the main mining area of Qibaoshan
Town, China: Potential impacts from mining damage,
solid waste discharge and land reclamation. Science of the
Total Environment, 859(November 2022), 160392. https://
doi. org/ 10. 1016/j. scito tenv. 2022. 160392
Publisher’s Note Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional
affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”),
for small-scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are
maintained. By accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use
(“Terms”). For these purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or
a personal subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or
a personal subscription (to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the
Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data
internally within ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking,
analysis and reporting. We will not otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of
companies unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that
Users may not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to
circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil
liability, or is otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by
Springer Nature in writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer
Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates
revenue, royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain.
Springer Nature journal content cannot be used for inter-library loans and librarians may not upload Springer Nature journal
content on a large scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any
information or content on this website and may remove it or features or functionality at our sole discretion, at any time with or
without notice. Springer Nature may revoke this licence to you at any time and remove access to any copies of the Springer Nature
journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express
or implied with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or
warranties imposed by law, including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be
licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other
manner not expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
Content uploaded by Brian Rotich
Author content
All content in this area was uploaded by Brian Rotich on Mar 25, 2025
Content may be subject to copyright.
