









































Pa
ge

 
1



Pa
ge

 
84

American Journal of  
Geospatial Technology (AJGT)

Geospatial Technique and Forest Management of  Coastal Areas in Bagamoyo Tanzania
Ally Khalfan Kyengya1*, Nickson Alnkiza Ernest1, Charles Allan Mtambo1

Volume 4 Issue 1, Year 2024
ISSN: 2833-8006 (Online)

DOI: https://doi.org/10.54536/ajgt.v4i1.4865
https://journals.e-palli.com/home/index.php/ajgt

Article Information ABSTRACT

Received: March 24, 2025

Accepted: April 30, 2025

Published: August 14, 2025

This article assesses the occurrence of  forest crimes mostly in Bagamoyo coastal region us-
ing qualitative and quantitative approaches. The probability sampling along with non-prob-
ability sampling, was applied when selecting the participants who were used in the study, 
the collection of  data was done through a community information system (participatory 
GIS), documents investigation, GPS survey, remote viewing as well as observation. Arc GIs 
software was used to create hotspot maps, forest crime patterns and visual representing of  
data. The outcomes revealed that areas which are highly vulnerable to the crime increase 
overtime from 2010 to 2022, it is due to the various factors like the family status of  people 
along the study area, economic status, education, poor infrastructure around the area, low 
living standard and more, availability of  forest markets and more.

Keywords
Forest Crimes, Geospatial 
Techniques, GIS, Participatory 
Mapping Crime Reduction 

1 University of  Dar es Salaam, United Republic of  Tanzania
* Corresponding author’s e-mail: binkhalfani@gmail.com

INTRODUCTION 
Over 30 percent of  the world’s earth’s surface is covered 
by forests, which are termed the earth’s air sacs. Forests 
provide oxygen gas in the atmosphere, which saves 
humans’ and all other species’ lives (WWF, 2011).
The 75 % of  the biosphere of  the earth’s total key 
productivity is contributed by the forest as well as 80 
% of  the forestry residues (plant biomass) of  the earth 
(Keenan et al., 2015). The distribution of  forests is that 
around half  of  the resources of  the world’s forests are 
hosted in South America and Asia (Pan et al., 2013). The 
world’s Food and Agriculture Organization (FAO) data 
in the year 2013 shows that the forests of  the world 
were unevenly dispersed, but Southern America and Asia 
hold the most of  the forests of  the world; Also there 
is a notable global forest share in the North, South, and 
Central regions of  America (FAO, 2013, 2015). Moreover, 
African forested area is 17% of  the world’s forest areas 
(Keenan et al., 2015). The mainland part of  Tanzania 
dwells forests with high biodiversity. The country has 
totality of  over 10,000 species of  plants, where only 
hundreds of  the species are nationally managed by 
using Geospatial techniques. There is the identification 
of  Over 305 species of  plants as the threatened species 
in the IUCN List Red, also about 276 plant species are 
categorized as at-risk species (endangered) (IUCN, 2013). 
The protection and management of  the major forests is 
done by using cartographic (geospatial) systems. Across 
the western, central, and south zones of  Tanzania, there 
is forest kinds which are deciduous Miombo woodlands 
but in the northern regions there is Acacia-Commiphora 
woodlands, coastal forests together with along the Indian 
Ocean coast there is the mangrove forests, the eastern 
zone is covered with woodland mosaics and the closed-
canopy forests are growing across the Eastern Arc old 

mountains , they also found along the Albertine Rift, 
moreover in the west zone along the Lake Tanganyika, 
as well as across the central and north regions of  the 
country on the juvenile volcanic mountains (Burgess et 
al., 2004). The most spacious Woodlands that are always 
deteriorated, it leads to the grass and shrubs undergrowth. 
The activities done by human beings, such as agriculture, 
expose Woodlands to frequent grass fires. there is areas 
that are mostly forested around the Bagamoyo district 
in the Fukayosi ward and Makurunge ward contrary to 
the other less forested wards found in Bagamoyo, there 
are legally conserved reserve areas of  forest which are 
Kikoka forest in the mainland and Mangrove forest to 
the coastline, they are under the management of  TFS 
since 2010/2011 when it was established. The Reserve 
Forest of  Kikoka that found across the Fukayosi ward 
and Makurunge ward covers about of  1654.8 hectare 
according to G.N. number 399 of  1958 and map Jb 
No. 415. Kikoka forest is the representative forest of  all 
coastal Forests of  Tanzania, coastal forests are important 
in the world for supporting endemic plants, birds, and 
other vertebrates with high potential for investigation, 
schooling, and recreation.
Mangrove Forest Reserve found along the shoreline of  the 
Indian Ocean in Bagamoyo covered 5636ha according to 
G.N. number 324 of  1928 in various areas but concentrate 
in Mankurunge ward streets including Mkunga, Shanga, 
Kijitokamba, Dago la ukindu, kwa mwarabu, Batini, 
Chamamba and along the river mouth of  Ruvu and Wami 
river that cross the areas of  Sadani National Park and 
the Ranch of  Zanzibar found in Bagamoyo. Therefore, 
the management of  forest appears as either durable or 
unstable in some areas, the majority of  mangrove specie 
decreases, at the same time in Bagamoyo, the forests are 
seriously affected by harvesting, mostly people exploit 



Pa
ge

 
85

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

forest resources illegally while few harvest the forests 
legally following the procedure of  business for products 
from the forest, for both domestic and commercial 
purposes as the source of  fuel (fuelwood), it is locally 
utilized though the forest products are highly needed in 
Zanzibar. The forests are continuously deteriorated for 
the generation the Income, tourism activities that lead 
to clearing forests to build hotels, opening beach areas, 
unlawful charcoal burning, clearing for agriculture and 
harvesting of  salt (Francis et al., 2011). There have been 
evidenced various forms of  forest crimes, including 
illicit lumbering, transportation, logging, as well as the 
gathering of  products from the forest illegally (UNDOC, 
2015, 2008). Among of  the listed crimes experienced in 
some countries as the statistical data exemplify Japan 14 
percent, Hungary, Denmark, and Chile49 percent. In the 
countries of  Africa, the forest crimes are associated with 
not only economic status but also social, political state, 
and human activities, where the suspected individuals of  
forest crime evidenced that they were struggling for the 
life betterment to be obtained from the forest’s products 
(FAO, 2003, 2008). In spite of  the challenge of  crimes in 
the forests, number of  forest crimes avoiding measures 
and the managerial measures has been initiated to control 
the forest crimes in the forests. Geospatial techniques 
(GPS, CCTV, Mobile phones, and other web maps 
techniques) is the established forest crime controller that 

offers data and complete information for monitoring and 
management of  crimes in the forest. The information 
accessed and acquired by using cartographic techniques 
is used to create and to provide the clear means of  
controlling protecting forests from crimes.

MATERIALS AND METHODS
The Area of  Study
This research was done across the two wards of  
Makurunge and Fukayosi, that are located in Bagamoyo 
district, which is among of  8 (eight) district councils 
of  Tanzanian political region called Coastal region 
(Pwani Region). It is 75 kilometers from Dar-es-Salaam 
city to Bagamoyo District (UNH, 2009). Bagamoyo is 
neighboring the Tanga Region northwards, the West 
side Bagamoyo is bordered by Morogoro Region, also 
the Indian Ocean to the East, as well as Districts of  
Kinondoni and Kibaha to the South side (Bagamoyo 
District profile, 2009).
According to DP, 2006; 2009, The Bagamoyo District is 
located between 370 and 390 Longitude and between 60 
and 70 Latitude. The area of  the district is about 9,842 
km2, where water area including Ocean and river covers 
855 km2, the remaining 8,987 km2 is the area of  dry lands 
(Bagamoyo District, 2009). There are two parliamentary 
provinces that are Chalinze constituent and Bagamoyo 
constituent (Bagamoyo District profile, 2006).

Figure 1: Location of  Fukayosi and Makurunge wards

Research Design and Sampling 
The study used both qualitative n and quantitative design 
to assess the distribution and factors influencing Forest 
crime using Geospatial techniques. Moreover, this design 
gives a clear awareness of  the processes, interactions, 
as well as mechanisms through which the forest crimes 
are interacted with Geospatial technique .The researcher 
utilized a quantitative approach to gather quantitative data 
around the forest crime areas using GPS and other GIS 
software (QGIS and ArcGIS) both spatial and spatial 

attributes were collected. Using quantitative approach, 
an attempt was made to understand the evidence of  
forest crime cases in Bagamoyo; Purposive sampling was 
applied in selecting Bagamoyo district as it has forest 
cover that experiencing a lot of  forest crimes. Therefore, 
the ideal target population for this research was the forest 
staffs and officers. So far, WEO and the Street chairman 
in the study area included .The probability sampling was 
used to obtain Quantitative data where 70s of  officers 
from the department of  forest was sampled. The 



Pa
ge

 
86

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

Probability sampling technique was conducted through 
listing the officers’ names from the principal forester’s 
office Tanzania forest Service (TFS). On the isolated 
small pieces of  paper, the names were written and kept 
in the bag. So, around 94.6% which is 71 names of  the 
target population were taken from the bag to make a 
sample, furthermore, the WEO and street chairman were 
purposively selected by the researcher as the respondents 
who endowed with a lot of  forest crime information 
about in their power positions.

Data Collection
The study applied the primary source and secondary 
source of  data. The Primary sources of  data included 
inclusive mapping, field observation and in-depth 
interview. The remote sensing and scanning of  raster 
data and digitization forest management journals, 
reports, and magazine were the secondary sources of  
data , obtained from Tanzania Forest Agency office 
(TFS) coastal zone. The participatory mapping technique, 
gave the participants an opportunity to contribute their 
experiences, information, and knowledge concerned 
with location and the intensity of  forest crime in the area 
of  study; this process resulted into the production of  
maps of  distribution of  the forest Crimes. participatory 

mapping was used to execute a base map in form of  
the high-resolution Google earth images was used to 
map locations of  different levels of  forest crime using a 
systematic procedure. Firstly, the identification of  ten (10) 
participants engaged in participatory mapping was made. 
Eight (8) of  the identified participants were forest staffs 
and other officers 2 WEO. All of  these have more than 6 
years of  experiences about forest crime occurrence along 
their areas. Next, the chosen participants were involved in 
the process of  mapping that was done through analyzing 
the past and the present forest crime occurrence. it was 
fulfilled through indicating the places of  the mapped area 
and approximating the magnitude of  the forest crimes 
on the study area and labeled the places with different 
colored gravels guided by the prepared questions. Third 
step, the labeled area that is regarded as the forest crime 
area were calculated to analyses hotspot to get distance 
and relationship between one point to another 
The images with high resolution showing the area 
encompassed by a single image cell on ground were for 
participatory mapping. With a scale of  1:3600. Features 
such as the roads, Indian Ocean, Schools and churches 
were used to orient participants during the participatory 
mapping process. the obtained Data was visualizing using 
the ArcGIS 10.7 software’s

Figure 2: Participatory Mapping with research participants

This study utilized Global Positioning System (GPS) 
surveying and systematic field observations to validate data 
obtained through interviews and participatory mapping 
exercises. GPS surveying proved instrumental in accurately 
verifying the spatial locations of  reported forest crime 
incidents and key land use features, such as religious 
institutions, commercial areas, educational facilities, and 
other relevant sites. To further enhance data reliability, a 
transect walk was conducted, serving to complement both 
the GPS data and the observational findings. This approach 
enabled direct, on-the-ground verification of  forest 
crime locations and associated land use activities, thereby 
strengthening the validity of  the spatial data collected from 
community mapping and key informant interviews.
 
Data Analysis
Hotspot analysis was employed in this study to identify 
areas prone to forest crimes by mapping the spatial and 
socio-economic factors associated with the occurrence 
of  these crimes. The primary objective was to generate 
maps illustrating the locations and distribution of  forest 
crime incidents, including illegal lumbering, Charcoal 
production, unauthorized timber harvesting, agricultural 
encroachment, and illegal grazing activities.
Key underlying drivers of  forest crime—namely 

economic pressures, educational attainment levels, and 
the availability of  supportive markets—were identified 
and spatially referenced using GPS coordinates. Point 
data were collected with a handheld GPS device offering 
an accuracy range of  0–5 meters. These geospatial data 
points were subsequently overlaid onto high-resolution 
Google Earth imagery to cross-validate the accuracy of  
locations reported by participants. This integrated analysis 
provided deeper insights into the spatial distribution 
of  forest crime and contributed to an enhanced 
understanding of  participants’ awareness regarding the 
geographic patterns of  such offenses.
The hotspot analysis aimed to highlight the density of  
forest crime incidents and pinpoint areas with frequent 
occurrences of  these illegal activities. Using Arc Map 
software, the hotspot mapping was carried out following 
a series of  specific steps. First, a 200-meter fishnet grid 
(polygon grid) was created to cover the area of  analysis, 
using the “Create Fishnet” tool in the Data Management 
Tools (Sampling). The grid cells served as the spatial units 
for the analysis. Next, the point data layer representing 
individual crime incident locations was joined to the 
polygon grid cells based on spatial proximity, using the 
“Join” function. This allowed each grid cell to be linked 
with the corresponding forest crime points. To calculate 



Pa
ge

 
87

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

crime hotspots, the “Hotspot Analysis” tool in the 
Geographical Statistics toolbox was applied, using the 
Z-score method. This tool identifies clusters of  statistically 
significant forest crime incidents .The results of  the 
hotspot analysis were visualized, with areas of  high crime 
density represented by a high Z-score and small P-values, 
indicating significant hotspots. Conversely, areas with low 
negative Z-scores and small P-values were classified as 
cold spots, representing areas of  lower crime activity. The 
intensity of  clustering was measured by the magnitude of  
the Z-score: higher positive Z-scores indicate stronger 
clustering, while Z-scores near zero suggest random 
distribution of  incidents. For a significance level of  0.05 

(95% confidence), Z-scores below -1.96 were classified 
as statistically insignificant, while those above 1.96 were 
considered statistically significant. Finally, additional steps 
were taken in Arc Map to refine the analysis and ensure 
the accuracy of  the results, leading to the identification 
of  key forest crime hotspots within the study area. 
Statistically important, for the comparison of  a specific 
confidence values level range. For instance, if  0.05 level of  
significance (95 percent level of  confidence), the “z” score 
is below -1.96. Statistically it is insignificantly classified 
also, when the z score is between 1.96 and above is said 
as factually significant. To finish this study, the other steps 
were performed in the Arc map Computer program.

Figure 3: The forest management analytical Model

The overlaid GPS point data, with the sub wards shape 
file, then filled into an excel work sheet file and served 
as a comma-delaminate text file (CSV) configuration, 
uploaded into the Arc GIS and overlaid with sub-
wards shape file to verify the response obtained from 
participatory mapping.
 
RESULTS AND DISCUSSION 
The Mangrove Forests Distribution in the Area of  

Study 
The observation of  Mangrove Forests was carried out 
across several streets within the study area, including 
Mkunga Street, Shanga Street, Mto Ruvu Street, Dango 
Lakindu Street, Kijito Kamba Street, Batini Street, 
Kwamwarabu Street, and Bandari ya Kati Street. A variety 
of  human activities were reported within these forested 
areas, including illegal gathering of  forest resources, 
charcoal burning, land clearing, and encroachment for 

Figure 4: The Mangrove Forest distribution of  along the Makurunge ward coastline



Pa
ge

 
88

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

settlement, agriculture, tourism, and livestock grazing. 
Additionally, unlawful fishing activities have been noted, 
contributing to the degradation of  these vital ecosystems. 
These activities are largely driven by the proximity of  the 
surrounding communities, as highlighted by respondents 
from the Makurunge ward, who identified the close 
relationship between the local population and the 
mangrove forests. The mangrove forests in the Makurunge 
ward, located along the Indian Ocean coastline, are 
particularly vulnerable to these human pressures. The 
map below illustrates the distribution of  these mangrove 
forests, highlighting their importance as both ecological 
and economic resources for local communities.

The Distribution of  Kikoka (Terrestrial) Forest 
along the Area of  Study
The Kikoka Forest Reserve spans two wards within the 

Bagamoyo District, namely Makurunge and Fukayosi, 
encompassing several villages and neighborhoods, 
including Relini, Vihagata, Mwana Senga, and Usigwa, as 
shown in the map extract below. In the map, the boundaries 
of  Kikoka Forest are demarcated by green-colored lines, 
which outline the forested area. The total area covered 
by the Kikoka Forest Reserve is approximately 1,654 
hectares, as recorded in field data collected in 2023. The 
surrounding areas of  Kikoka Forest are heavily influenced 
by human activities, particularly those associated with illegal 
forest use. Reported activities include charcoal burning, 
timber harvesting, land encroachment for settlement and 
agricultural purposes, as well as illegal lumbering. These 
activities are largely driven by the proximity of  the local 
communities, with respondents from both Fukayosi 
and Makurunge wards confirming the close interactions 
between forest areas and human settlements.

Figure 5: The distribution of  Kikoka Forest around the Fukayosi and Makurunge wards

Table 1: The Distribution of  Kikoka Forest table
Boundary Name of  forest The Distribution Predominant species
Fukayosi -Koka forest -Around reline strip area -Riverine and miombo species 

Around manga strip area
Around usigwa strip area

Maku Runge -Kikoka forest -Around kifunde strip area -Riverine and miombo species 
Around the Gezaulole Street area 

Table 2: The tabulation of  Distribution Forest in Makurunge ward and Fukayosi ward 
Boundary (ward)  Forest Type Distribution of  Forest Predominant species
Makurunge -Miombo tree species and riverine 

forest vegetation
-Kifunde Geza ulole - miombo and Riverine 

species
-Mangrove forest -Makurunge streets Viagata 

Relini Usage Mwanasenga
Fukayosi -Miombo tree species and riverine 

forest vegetation
- -Mangrove Riverine and 

miombo species
-Mangrove forest -



Pa
ge

 
89

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

Mapping of  the Illegal Ports in the Research Area
The survey of  unlawful ports in this study aimed to 
identify specific coastal locations requiring monitoring 
and regulatory control due to their association with illegal 
forest product transport. The exercise was conducted in 
collaboration with local stakeholders, specifically the Beach 
Management Units (BMUs) and the Village Natural Resource 
Management Committees (VNRCs) from Fukayosi and 
Makurunge wards, who served as primary informants and 
participants in the mapping process Participatory mapping 
(community mapping) was employed as the principal data 
collection method. Participants were provided with geo-
referenced Google Earth imagery, which served as the base 
map for marking known illicit port locations. The imagery 
had been symmetrically processed and projected into the 
Arc 1960 Zone 37S coordinate reference system to align 

with the national mapping standard and ensure consistency 
with other geospatial datasets. Each participant was 
individually interviewed and asked to identify any known 
illegal ports after reviewing the reference map. Orientation 
landmarks such as rivers, coastal lines, roads, and settlement 
features were used to assist participants in spatially locating 
the ports. The participatory mapping approach ensured 
that local knowledge was accurately captured and spatially 
represented, enhancing the reliability of  the dataset. Areas 
of  Mkunga, Batini, and Chamamba were listed; it challenged 
the forest management since they transported the forest 
products including poles and fuel wood with the support 
of  the listed illegal seaports to other areas like Zanzibar. 
By using the Global Positioning Systems, the researcher 
recorded the coordinates of  the illegal ports area with as 
shown on the map extract underneath.

Figure 6: The distribution of  illicit boatyards along the Makurunge coastline

The Location-Based Trend of  the Forest Crimes 
Across the Research Area 
As condemned in different literatures such as the 2002 
forest act no .14 describes the forest offences including 
the human economic activities performed within a reserve 
area of  forest without a legal license (authorization) from 
an authorized body. The human activities conducted 
are cutting of  Tree, burning of  Charcoal, harvesting of  
Timber, Agriculture, Settlement, Mining, Fishing, animal 
keeping and unpermitted Entering within a forest reserve 
area. Thus, this study part confers about the forest offences 
geographical trend in the Fukayose ward and Makurunge 

ward for ten years from the year 2003 to the year 2023. 
There have been the negative and positive presentation 
of  forest changes from the official patrol reports and 
documents and several relevancies were verified by 
the participants from the area of  the field. Participants 
used various referring records including their everyday 
documentation, reports and journals of  district council that 
contained various information and public stories about of  
forest crimes along Fukayosi ward and Makurunge ward 
in Bagamoyo district and the Coastal region (PWANI)at 
large, this provided a critical room for the understanding 
of  forest crimes along Fukayosi and Makurunge wards.

Table 3: The tabulation of  the location -based trend of  Forest offences in the Fukayosi ward and the Makurunge ward
Year The Kikoka forest reserve Mangrove forest Total
2003-2005 129 63 192
2006-2008 102 78 180
2009-2011 99 84 183



Pa
ge

 
90

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

As presented in the table above, data derived from various 
official documents indicate a clear trend in the decline 
of  forest-related crimes within the Makurunge and 
Fukayosi wards, particularly in the Kikoka Forest Reserve. 
The analysis reveals a positive trajectory in forest crime 
reduction between the years 2003 and 2023. Specifically, 
between 2003 and 2005, a total of  129 forest offence 
cases were recorded. This number declined to 102 cases 
between 2006 and 2008, followed by a further reduction 
to 99 cases from 2009 to 2011. The downward trend 
continued with 70 cases recorded between 2012 and 2014, 
and subsequently, only 44 cases were reported during the 
2015–2017 period. Between 2018 and 2020, the number 
of  forest crime incidents further decreased to 29, and most 
notably, only 11 offences were recorded between 2021 and 
2023.This consistent decline over the past two decades 
signifies a substantial improvement in forest governance 
and protection efforts within the Kikoka Forest Reserve. 
The findings suggest that the area has been gradually 
recovering and experiencing increased forest cover, which 
can be attributed to the implementation of  strategic 
forest management interventions. These include the 
establishment of  effective monitoring plans, enforcement 
of  forest conservation laws, and active involvement of  
the Tanzania Forest Service Agency (TFS). A notable 
milestone was the enforcement operation conducted by 
TFS in 2015, which included significant banishments 

and restrictions within the forest reserve. The sustained 
decline in forest crimes demonstrates the positive 
outcomes of  these interventions and underscores the 
critical role of  institutional enforcement in achieving 
long-term forest conservation goals. above, from the 
year 2003, the Mangrove Forest along the Indian Ocean 
coast across the area of  Makurunge ward witnessed the 
negative variations as per trend, indicates that between 
the year 2003 and 2005 the forest offences experienced 
were 63, then between 2006 to 2008 the area witnessed 
about 78 forest offence cases. Also, from 2009 to 2011 the 
forest crime increased too up to 84 incidents; also from 
2012-2014 forest incidents witnessed 99. Between the year 
2015 and 2017 the number of  cases of  the forest crimes 
affecting the forests of  mangrove species raised to 139.
furthermore, between the year 2018 to 2020 the crime 
cases increased to 144. Finally, from the year 2021 to 2023 
the mangrove forest crimes increased contrary to the other 
years to 151 cases. For The Mangrove forests, the crimes 
increase contrary to the previous two decades and this 
may be because of  the presence of  illicit boat yards, this 
influences unpermitted exploitation and transportation of  
the products from the forest and extension of  tourism 
activities including building of  new hotels and settlement 
adjacent to the mangrove forest boundary. The activity 
that needs both intensive and extensive track purposively 
to undermine and solve issues of  forest crimes increments.

Figure 7: The forest crimes geographical trend along the Fukayosi ward and the Makurunge ward

Satellite imagery used in this study was obtained from the 
United States Geological Survey (USGS) Earth Explorer 
platform. A temporal scope of  ten years, beginning from 
2003, was considered for the analysis. Two datasets were 
utilized: a 2003 Landsat Thematic Mapper (TM) image 
and a 2023 high-resolution thematic map. Both images 
provided a spatial resolution of  30 meters by 30 meters, 
which is appropriate for detecting land cover changes 

over time. The 2003 Landsat image was selected to 
assess forest cover in the Makurunge and Fukayosi 
wards, while the 2023 imagery was employed to analyze 
recent forest cover dynamics in the same locations. 
In addition to standard-resolution data, higher spatial 
resolution imagery (i.e., finer than 30 meters) was also 
utilized to enhance the accuracy of  forest cover analysis. 
Satellite imagery was critical for evaluating changes in 

2012-2014 70 99 169
2015-2017 44 139 183
2018-2020 29 144 173
2021-2023 11 151 162
SUM 486 758 1,242



Pa
ge

 
91

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

Table 4: Collected geographical data for forest cover changes detection
S/N Type of  Data The Path The Date of  Acquisition 
1. Landsat 7 ETM+ 166/064 & 166/065 2003
2 Landsat 8-9 OLI/TIRS 166/064 2013
3 Landsat 8-9 OLI/TIRS 166/064 2023

forest ecosystems and has been widely applied in similar 
studies involving more complex landscapes than those 
of  the Makurunge and Fukayosi wards (Basommi et al., 
2015; Acheampong et al., 2018). To support the image 
classification and accuracy assessment, approximately 
108 Geo-referenced checkpoints were collected using 
a Global Positioning System (GPS) receiver. These 

ground control points were used as training sites 
during supervised image classification. Additionally, a 
topographical map sheet from 2008 was incorporated to 
supplement the forest cover assessment, particularly in 
supporting the spatial analysis and management of  forest 
resources in Bagamoyo District, with specific attention to 
the Makurunge and Fukayosi wards.

Table 5: The Seven classes of  LULC 
N LULC illustration
1 Water The waterlogged Areas throughout the year, may be not included among the temporary 

water areas (sporadic water or ephemeral water areas).
2. Forest The classification (clustering) of  long (15 m or higher) and thick vegetation, naturally 

characterized with the thick - closed canopy.
3 Wetland The Areas with any kind of  vegetation and covered with water all over the year; the 

seasonally flooded area that includes grasslands, shrubs, and bare ground. 
4 Agriculture The artificial or Humans planted crops, grasses, and cereals that are not tree-like height.
5 Built Area The Human-constructed infrastructures including vital roads and rail ways, also the large 

correspondent water-resistant structures such as the office buildings and household 
buildings (residential housing).

To facilitate effective visualization and analysis, a band 
combination process was carried out by integrating multiple 
Landsat satellite images into a single dataset. For each year 
analyzed, seven spectral bands from the respective Landsat 
datasets were combined using ArcGIS Desktop version 10.5. 
This step was essential for enhancing the interpretability of  
land cover features across the study period. Additionally, 
a mosaic operation was performed to merge individual 
raster datasets into a seamless image. This was necessary 
particularly for the 2003 imagery, which comprised scenes 
from different paths and rows. These were mosaicked to form 
a single continuous raster dataset that fully covered the study 
area. This data preparation ensured spatial continuity and 
completeness of  the area under investigation. Subsequent 
to the mosaic process, image processing and projection 
were conducted. The datasets were clipped according to 
the defined boundaries of  the study area to isolate relevant 
spatial information and minimize processing time for 
subsequent analysis shape file (Fukayosi and Makurunge 
wards). Likewise, image projection was performed to give 
images the actual reference system (projected coordinate 
system). The images used were projected into Arc1960 
UTM Zone 37S. Lastly, Land Use/Land Cover Classification 
was done as a classification providing information on land 
cover and the types of  human activity involved in the use of  
land. The classification helped on the assessment of  impacts 
to the environment, as well as the latent alternative land 
uses. The sets of  information were added to the Arc Map 
windows for the classification. So, with the use of  Maximum 
Likelihood Classification, the seven classes of  LULC were 

classified they are; Forest, Wetland, Water, Agriculture, Area 
with Buildings, Rangeland, and the open ground. shape 
file (Fukayosi and Makurunge wards). Likewise, image 
projection was performed to give images the actual reference 
system (projected coordinate system). The images used 
were projected into Arc1960 UTM Zone 37S Lastly, Land 
Use/Land Cover Classification was done as a classification 
providing information on land cover and the types of  human 
activity involved in the use of  land. The classification helped 
on the assessment of  impacts to the environment, as well as 
the latent alternative land uses. The sets of  information were 
added to the Arc Map windows for the classification. So, with 
the use of  Maximum Likelihood Classification, the seven 
classes of  LULC were classified they are; Forest, Wetland, 
Water, Agriculture, Area with Buildings, Rangeland, and the 
open ground Shape file (Fukayosi and Makurunge wards). 
Likewise, image projection was performed to give images 
the actual reference system (projected coordinate system). 
The images used were projected into Arc1960 UTM Zone 
37S. Lastly, Land Use/Land Cover Classification was done 
as a classification providing information on land cover and 
the types of  human activity involved in the use of  land. The 
classification helped on the assessment of  impacts to the 
environment, as well as the latent alternative land uses. The 
sets of  information were added to the Arc Map windows for 
the classification. So, with the use of  Maximum Likelihood 
Classification, the seven classes of  LULC were classified 
they are; Forest, Wetland, Water, Agriculture, Area with 
Buildings, Rangeland, and the open ground.



Pa
ge

 
92

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

Using Land Use/Land Cover (LULC) prediction 
techniques, the MOLUSCE plugin was employed to 
project potential LULC changes for the year 2033. The 
analysis was conducted using version 3.0 of  the QGIS 

MOLUSCE 22 plugin, applying the Artificial Neural 
Network (ANN) model—specifically the Multilayer 
Perceptron (MLP) approach. The resulting predictive 
map and associated outputs are presented as follows:

6 Bare Ground The exposed surface Areas where either rock or soil is not covered with vegetation 
throughout the year; more over the large sand areas with no plants.

7 Rangeland The homogeneous grass - covered open areas with less or no longer (plants) vegetation; 
wild cereals and grasses, a mix of  small clusters of  plants or single plants dispersed; scrub-
filled clearings in the forests that are not taller than trees.

Table 6: Seven LULC (Ha) matrix table 
N C.name 2003 2013 2023 2033 Total 
1 water 740.5 516.1 586.6 612.5 4911.3
3 Forest 27542.9 32953.7 23708 22988.7 214386.4
4 Wetland 808.5 380.7 305.31 345.6 3580.14
5 Agriculture 2350.4 4192.3 7464.1 8824 45661.66
6 Built areas 20.5 684.2 1288.4 1516.1 7018.39
7 Bare ground 10.8 58.25 33.2 7.1 218.05
8 Rangeland 57760 49847.45 55247.33 54438.84 434586.92
 Total 89233.6 88632.7 88632.94 88732.84 710362.86

The matrix table above shows the land use, From the year 
2003 to 2023 at Makurunge and Fukayosi wards found 
in Bagamoyo district. The table displays that the forest 
changes with time from 2003 to 2023 due to different 
factors. The study identified seven major Land Use and 
Land Cover (LULC) classes in the Bagamoyo District, 
namely: water bodies, forest areas, wetlands, agricultural 
land, built-up areas, rangelands, and bare ground. 
Forecasts based on temporal geospatial analysis predict 
both partial and complete changes in land use across 
these categories from 2023 to 2033, particularly within 
the Fukayosi and Makurunge forest zones. In 2003, 
water bodies covered approximately 740.5 hectares. This 
coverage decreased to 516.1 hectares by 2013, before 
increasing again to 586.6 hectares in 2023. Projections 
suggest that by 2033, water bodies will expand further 
to approximately 612.5 hectares. Forest cover in the 
study area was estimated at 27,542.9 hectares in 2003. 
This increased to 32,953.7 hectares in 2013, indicating 
improved conservation efforts during that period. 
However, by 2023, forest cover had declined significantly 
to 23,708 hectares. The study predicts a continued decline 
to 22,988.7 hectares by 2033, potentially due to increasing 
human pressures and land use changes. Nevertheless, 
enforcement of  forest protection laws in Fukayosi and 
Makurunge wards appears to have contributed to some 
level of  forest preservation. Agricultural land use has 
shown a notable increase over the study period. In 2003, 
agricultural land occupied approximately 2,350.4 hectares. 
This expanded to 4,192.3 hectares in 2013 and further 
to 7,464.1 hectares by 2023. Projections estimate that 
agricultural land will extend to 8,824 hectares by 2033. 
This expansion is largely attributed to population growth 

and the consequent rise in demand for food and related 
agricultural resources. However, this trend is considered 
a negative factor in relation to forest conservation, as it 
contributes to the continued reduction of  forest cover. 
Built-up areas have also experienced significant growth. 
In 2003, built environments covered only 20.5 hectares. 
This increased to 684.2 hectares in 2013 and further to 
1,288.4 hectares in 2023. By 2033, it is projected that 
built-up areas will cover 1,516.1 hectares. This expansion 
correlates with the rising population, which drives the 
demand for both permanent and temporary human 
settlements. Bare ground—land without significant 
vegetation or use—was approximately 11 hectares 
in 2003, increased to 58.3 hectares in 2013, and then 
declined to 33.2 hectares in 2023. Forecasts suggest 
this will further reduce to 7.1 hectares by 2033. This 
reduction is likely due to the increased establishment 
of  forest cover, rangeland, and water bodies, all of  
which reduce the extent of  unused land. Rangeland, 
defined as land used for grazing and natural vegetation 
cover, is expected to increase considerably. By 2033, it 
is projected that rangeland will cover approximately 
54,438.84 hectares, partly due to the emergence of  
new vegetation, albeit predominantly juvenile. Overall, 
the study concludes that changes in LULC patterns—
driven by population growth, increasing food demand, 
settlement expansion, and forest product utilization—
pose significant challenges to sustainable forest and land 
resource management in Bagamoyo District, particularly 
in the Fukayosi and Makurunge wards. These trends 
underscore the importance of  integrated land use 
planning, enforcement of  environmental regulations, 
and the promotion of  sustainable livelihood alternatives.



Pa
ge

 
93

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

The availability of  wetland and drained areas in the study 
region is projected to be significantly influenced by the 
presence of  the Ruvu River, which primarily sources 
its water from the Indian Ocean and serves as a major 
hydrological contributor. Additionally, the existence of  
unauthorized ports along the Indian Ocean coastline 
facilitates the illegal transportation of  forest products. 
This activity is expected to intensify over time, and 
projections indicate that by the year 2033, the volume of  
forest products transported through these illicit channels 
will surpass that of  previous years. As a consequence, the 
mangrove forest is increasingly transitioning into a high-
risk degradation zone, as indicated by its classification 
within the “RED” (critical) category. Conversely, Kikoka 
Forest is forecasted to exhibit a positive vegetative trend, 
with projections indicating an increase in forest density. 
This improvement is attributed to effective protection 
measures implemented under various legal frameworks 
and environmental regulations.

CONCLUSION
The findings of  this study confirm that forested areas within 
the study region are adversely affected by forest-related 
criminal activities. The study aimed to explore strategies 
for mitigating these crimes through the application of  
geospatial techniques. Using spatial analysis, crime-prone 
forest zones were delineated, and hotspot areas were 
successfully identified. The study further established that 
socioeconomic factors significantly influence individuals’ 
involvement in forest crimes, often as a means of  

economic survival. Key contributing factors include 
employment status, educational attainment, and income 
level. Among these, the unemployment rate emerged 
as the most influential determinant in the decision to 
engage in forest-related criminal behavior. Moreover, 
the study demonstrated the critical role of  cartographic 
techniques in supporting forest crime prevention and 
management. These techniques proved effective in 
spatially visualizing areas of  high criminal activity, 
thereby enabling enforcement agencies to take targeted 
and evidence-based interventions. The results underscore 
the necessity for enforcement bodies to adopt proactive 
measures informed by the spatial distribution of  criminal 
activities and their underlying drivers. Finally, the study 
recommends incorporating community stakeholders 
into policy formulation processes. Given their localized 
knowledge and lived experiences regarding forest crime 
incidents and affected locations, community members 
can provide valuable insights that enhance the relevance 
and efficacy of  forest crime mitigation strategies.

Acknowledgements
The authors thank United States Geological Survey 
(USGS) for providing the Landsat 7 ETM+ imageries 
and SRTM DEM. We extend our gratitude to the 
reviewers for their critical observations and to everyone 
who contributed to the improvement of  this paper. 
 
REFERENCES
Food and Agriculture Organization of  the United Nations 

Figure 8: Land use land cover from 2003-2033 at Makurunge and Fukayose 



Pa
ge

 
94

https://journals.e-palli.com/home/index.php/ajgt

Am. J. Geo Spat. Technol. 4(1) 84-94, 2025

(FAO). (2011). Framework for assessing and monitoring 
forest governance. The Program on Forests (PROFOR), 
FAO. https://shorturl.at/DZhqK 

FAO. (2013). Sustainable Forest Management in a Changing 
Climate; FAO‐Finland Forestry Programme TANZANIA. 
A Fire Baseline for Tanzania. Dar es Salaam, Tanzania. 

Food and Agriculture Organization of  the United 
Nations (FAO). (2014). State of  the world’s forests 2014: 
Enhancing the socioeconomic benefits from forests. FAO. 

Food and Agriculture Organization of  the United Nations 
(FAO). (2015). State of  the world’s forests 2015. FAO. 

Ishengoma, R. (2015). National Trends in Biomass Energy in 
Tanzania. Presentation to Workshop on Exploring the 
Evidence, mapping the way forward, and Planning 
for Future Actions for Developing Biomass Energy 
in Tanzania Hyatt Regency Hotel 26 – 27 February 
2015 Dar es Salaam, Tanzania. 

International Union for Conservation of  Nature (IUCN). 

(2011). The IUCN Red List of  Threatened Species (Version 
2011.2). http://www.iucnredlist.org 

Tanzania Forest Conservation Group. (2016). A review of  
policy instruments relevant to the integration of  sustainable 
charcoal production in community-based forest management 
in Tanzania (Technical Paper No. 51, p. 56). Tanzania 
Forest Conservation Group. 

United Republic of  Tanzania. (2016). Tanzania’s Forest 
Reference Emission Level Submission to the UNFCC. 
Division of  the Environment-Vice President’s Office, 
National Carbon Monitoring Centre (NCMC), Dar es 
Salaam, Tanzania.

United Nations Office on Drugs and Crime (UNODC). 
(2013). UNODC homicide statistics. http://www.unodc.
org/unodc/en/data-and-analysis/homicide.html

World Bank .(2016). World Bank Group Forest Action Plan 
FY16-20. Washington. DC. The USA.


