UNAM Gā-aisib Repository
The UNAM Gā-aisib Repository serves as a central platform to preserve and showcase the university's scholarly output.
Communities in the Gā-aisib Repository
Select a community to browse its collections.
Recent Submissions
Spatial and temporal assessment of by-catches in the Hake trawl fishery off Namibia and implications for fisheries management
(University of Namibia, 2025) Marufu, Tamuona; Hay, Clinton
This study assessed by-catches within the Namibian hake trawl fishery. As with other
fisheries, the trawling method used during commercial fishing in the hake fishery results
in by-catch. By-catch refers to the unintentional capture of non-target species during
fishing activities. Consequently, by-catch may negatively impact the marine environment
through biodiversity loss and the eventual collapse of fisheries. Therefore, understanding
the spatial and temporal distribution of by-catches off Namibia can lead to the formulation
of measures which may further mitigate by-catch within the hake trawl fishery off
Namibia. However, research on how by-catches vary spatially and temporally off
Namibia remains limited. Thus, the objectives of this study with focus on the Namibian
hake trawl fishery were to: assessthe spatial and temporal variation of the speciesrichness
and diversity of by-catches, to assess the spatial and temporal composition of by-catches
and to analyse the effect of sea bottom temperature on the spatial and temporal variation
of by-catches. At each trawl station, samples were randomly collected onboard
commercial fishing vessels. Random samples were obtained by using a conveyor belt
sampling method in both the northern and central fishing regions off Namibia, during
August and September 2023. The study revealed the following statistical results: the
PERMANOVA test showed that sea bottom temperature had a significant effect on the
spatial and temporal variation of by-catches in both regions, Pseudo-F=1.93, p=0.030.
One-way ANOSIM based on the abundance of by-catch species revealed that spatially,
there was a significant difference in the composition of the sampled by-catches between
the northern and central regions, r = 0.22 and p = 0.004. At a cut-off cumulative
dissimilarity of 67%, SIMPER showed that fifteen by-catch species contributed to the
dissimilarity between both sampled regions. The fifteen by-catch species were Brama
brama, Caelorinchus simorhynchus, Chaceon maritae, Chlorophthalmus punctatus,
Helicolenus dactylopterus, Lophius vomerinus, Macrouridae spp., Mustelus palumbes,
Pterothrissus belloci, Scomber japonicus, Suberites dandelenae, Todarodes angolensis,
Trachurus capensis, Trachyrinchus scabrus and Zeus faber. Similarly, one-way
ANOSIM based on the biomass of the sampled by-catch species showed that spatially,
there was a significant difference in the composition of by-catches between the northern
ii
and central regions, r = 0.24 and p = 0.007. At a cut-off cumulative dissimilarity of 70%,
SIMPER showed that fifteen by-catch species contributed to the dissimilarity between
both sampled regions. The fifteen by-catch species were Austroglossus microlepis, B.
brama, C. simorhynchus, C. maritae, C. punctatus, H. dactylopterus, L. vomerinus,
Macrouridae spp., M. palumbes, P. belloci, S. japonicus, S. dandelenae, T. angolensis,
T. capensis, Trachyrinchus scabrus and Z. faber. Furthermore, the weight of the trawled
by-catch was significantly higher in the northern region during daytime trawling. On the
other hand, this study may aid towards filling the gap in knowledge regarding the spatial
and temporal composition of by-catch within the Namibian hake fishery. Consequently,
this study underscores the critical need to incorporate the spatial and temporal dynamics
of by-catches into fisheries management strategies for the hake trawl fishery off Namibia.
By identifying patterns in by-catch composition, abundance and biomass, this research
offers valuable insights into the need to include targeted spatial regulations and adaptive
fishing strategies, in order to minimize negative ecological impacts and optimize resource
use
Preparatory studies for the Africa millimetre telescope
(University of Namibia, 2025) Frans, Lott Ndeyanale; Backes Michael
The Event Horizon Telescope (EHT) is a network of antennas across the globe
currently used to image supermassive black holes (SMBHs) at a frequency of
230 GHz. Since the release of the image of M87* in 2019 and, subsequently,
that of Sgr A* in 2022 by the EHT collaboration, the focus has shifted towards
dynamically imaging SMBHs. This has led to a search for potential sites to extend and fill the gaps within the EHT network. The Gamsberg Mountain and
the H.E.S.S. site are both located within the Khomas highlands and have been
identified as potential sites for the Africa Millimetre Telescope (AMT). Precipitable water vapour (PWV) in the atmosphere is the primary source of opacity and
noise from atmospheric emissions, when observing at millimetre to sub-millimetre
wavelengths. This study aims to establish the PWV content and meteorological
conditions at the potential sites of the AMT using Global Navigation Satellite System (GNSS) measurements. Using conservative specifications and potential dish
sizes of 13, 14 and 15 m for the AMT, the System equivalent flux density (SEFD)
and Signal to Noise ratio (S/N) to the reference stations ALMA and NOEMA
were also assessed from simulated observations of M87* and Sgr A*. The EHT
window PWV had a 25th percentile of 12.21 mm at the H.E.S.S. site and 7.54 mm
at the Gamsberg Mountain. The simulated results of M87* and Sgr A* showed
the 15 m and 14 m dishes at the Gamsberg Mountain to have the lowest SEFD,
with the performance of the 13 m dish being slightly comparable to the 15 m
dish at the H.E.S.S. site. For M87* and Sgr A* observations, the 13 m dish at
Gamsberg Mountain achieved a higher average S/N output for the ALMA-AMT
i
baseline than the 15 m dish at the H.E.S.S. site. For M87*, the S/N output from
Gamsberg Mountain was higher for the NOEMA-AMT baseline, with the 13 m
dish providing 41 additional minutes of S/N ≥ 4 compared to the 15 m dish at
H.E.S.S. For Sgr A*, the AMT at Gamsberg Mountain achieved the highest S/N
for the ALMA-AMT baseline, yielding an additional 1h32 of S/N ≥ 4 for the 13 m
dish compared to the 15 m dish at the H.E.S.S. site. The PWV and simulation results indicate that the Gamsberg Mountain is the more suitable site for the AMT
Relative importance of sources of plant recruitment and the role of plant-to-plant interactions as early indicators of grazing-induced desertification tipping points
(University of Namibia, 2025) Hamunyela, Ndamonenghenda; Nesongano, Wellencia C.
The soil seed bank (SSB) and seed dispersal (seed rain) within a population play a crucial
role in plant recruitment and regeneration. Seed dispersal, both in space (true dispersal)
and in time (persistent soil seed banks), confers resilience and resistance to vegetation
change, acting as a buffer against disturbances like climatic extremes and overgrazing.
However, through the removal of the aboveground biomass, continuous
overgrazing often causes a shift in vegetation structure and composition, directly affecting
seed production, seed dispersal, and subsequently the replenishment of the soil seed bank.
The depletion of the soil seed bank affects aboveground plant regeneration potential, often
driving once productive land over a tipping point to a degraded state. Land degradation is
a serious concern in semi-arid savannas, affecting herbaceous forage productivity,
impacting livestock production, and threatening human livelihood and economic
development. Despite its importance in rangeland stability, there have been few
investigations into the significance of seed-related plant demographic processes in the
development of tipping points. This study investigated the influence of livestock
grazing on the sources of plant recruitment, focusing mainly on the soil seed bank in
central Namibia, with the aim of identifying potential indicators of grazing-induced
desertification tipping points. Grazing gradients (based on a space-for-time substitution
approach), originating and radiating away from active artificial water points, were
established within two land tenure systems, namely, freehold farms and communal areas.
These land tenure systems represented two distinct grazing management systems, which
are rotational grazing and continuous grazing, respectively. The first objective of the
study was to assess how soil seed bank characteristics, including size, species richness,
ii
and composition, vary in response to increasing grazing pressure and the potential
influence of the two grazing management systems, with the main aim of investigating the
presence of sudden changes indicative of threshold behaviour. The second objective was
to assess the role of two nurse plant types, namely shrubs and established grass tussocks,
on SSB and seed rain (SR) distribution along a gradient of increasing grazing pressure by
analysing the relative interaction index (RII). The third objective was to assess the relative
contribution of seed bank and seed rain to recruitment along spatial grazing
gradients, as well as the influence of grazing management systems on the contribution of
dormancy, dispersal, and local reproduction to recruitment. A total of 1058 SSB
samples were collected from nine plots (100 m2
) established along the grazing gradients.
Additionally, for the assessment of the influence of grazing pressure on the interaction
intensity between SSB and SR and nurse plants, samples were collected from three
microhabitats, namely, woody plants, perennial grass tussocks, and open
microhabitats. The seedling emergence method was used to quantify the relative size and
composition of the germinable soil seed bank. A total of 7318 seeds from 95 species were
recorded in the SSB, while 1374 seeds from 40 species were recorded in the SR. The linear
mixed effect model (LMM) revealed that freehold farms exhibited a
significantly larger and more diverse SSB compared to communal areas. At the gradient
level, a LMM with a segmented relationship structure showed a non-linear increase in
SSB density with grazing intensity in freehold farms, while no significant variations were
found in SSB density or richness in communal areas. Significant dissimilarities in species
composition were observed between the two land tenure systems and within freehold
farms. The LMM was also used to analyse the RII with nurse plants along the grazing
gradient and found a notable negative trend in SSB density and richness in response to
iii
grazing intensity, regardless of the type of nurse plant, with a particularly
strong interaction with management. For the relative contribution of the sources of plant
recruitment to the natural population, a generalised linear mixed effects model
(GLMM) was used, and showed that the seed bank played a more substantial role in the
recruitment of both annual and perennial grass populations. Only the annual grass SR
showed a response to grazing pressure. Surprisingly, the impact of grazing on soil seed
bank patterns in this Namibian rangeland was relatively small. There were minimal signs
of thresholds beyond which the seed bank deteriorated, and variations in SSB density and
species composition emerged as potential indicators of threshold behaviour. Furthermore,
the findings demonstrated that rangelands managed under the freehold system are less
degraded compared to those under the communal system. Therefore, controlled grazing
maintains a relatively resilient and resistant SSB, while overgrazing may shift a system
beyond a degradation threshold into a highly degraded state. The study highlights the
importance of sustainable grazing under global change, particularly in dryland systems
Utilising herbaceous plant population patterns as indicators of desertification tipping points: An experimental and space-for-time approach
(University of Namibia, 2025) Chambara, Faith; Nesongano, Wellencia
Although it is known that various factors such as overgrazing and its combined
effect with rainfall deficiency may shift rangelands from a healthy to a degraded
state, the sensitivity of semi-arid vegetation, including the use of herbaceous
vegetation attributes that can easily be identified and measured to detect possible
irreversible rangeland degradation when a desertification tipping point (DTP) is
crossed, have not been adequately explored. For this reason, the study sought to
assess possible changes in attributes of pre-selected perennial grass species as well
as common annual grass and forb species under different conditions. The
following variables were determined per species: density, basal area, height,
biomass allocation, bio-volume, and seedling recruitment, using either a spacefor-time substitution or field experiment approach. To implement the space-fortime approach, four freehold (rotational grazing commonly practiced) and four
communal (generally characterized by continuous grazing) rangelands were
selected, which are the two main land management systems in and around the
study area (the Greater Waterberg Landscape, east-central Namibia). At each site,
two transects were laid out radiating away from a cattle water point and along
each transect, nine plots of 10 m × 10 m were logarithmically demarcated. The
experimental approach was implemented through a fully randomized factorial
block design with two precipitation treatments (ambient rainfall and severe
drought) and three grazing treatments at two sites assumed to be either far or
closer to DTP). Grazing was simulated through fencing and clipping at varying
frequencies over two growing seasons (light: clipped once, moderate: clipped
twice, and severe: clipped four times) while drought was simulated with rainout
shelters. The response of two key perennial grass species, Eragrostis rigidior Pilg.
and Stipagrostis uniplumis (Licht.) De Winter var. uniplumis to different grazing
and rainfall treatments was monitored. The space-for-time approach revealed that
density, basal area, height, seedling recruitment, and reproductive allocation of
perennial grasses varied significantly either at aggregated or species-specific
levels and were, as expected, influenced by land management and grazing
intensity. While polynomial nonlinear responses were observed on freehold
ii
rangelands, piecewise regression further identified breakpoints in some grazingsensitive and grazing-tolerant species in both land management systems. The
breakpoint for Stipagrostis uniplumis was identified closer to water points on
freehold farms than on communal areas, indicating that this species is potentially
suitable as an early-warning indicator for desertification. Total seedling
recruitment decreased towards high grazing on freehold farms (p<0.01) with no
significant response in communal areas. In terms of both bio-volume and biomass
allocation to the reproductive parts, there were either linear responses or no clear
trends detected toward the water points. On freehold farms, forbs decreased in
total bio-volume towards high grazing (p<0.05). Biomass allocation to the
reproductive parts was higher in annual grasses than perennial grasses in both land
management systems (p<0.05). The initial experiment findings showed all the
main effects and their interactions (site, species, grazing, and precipitation
treatments) to have a significant influence on the cumulative basal area of the two
key grass species (p<0.05). The seedling density of Eragrostis rigidior was higher
under light than moderate and heavy grazing during the peak growing season
while S. uniplumis had generally very low numbers in all treatments and sites
throughout the years. Overall, both aggregated and species-level results were
either significant or indifferent between land management systems or in response
to grazing pressure. The results also show the suitability of using both grazingsensitive and grazing-tolerant perennial grass species, and the use of aggregated
grazing niche groups as early warning indicators for looming desertification.
Although this study could not, with the space-for-time substitution method,
demonstrate clear responses in most species, there is potential to detect tipping
points by aggregating species and when more grazing indicator species are
screened, as well as through long-term monitoring of the experiment.
Mycotoxicological safety of food commodities and health implications in children from rural households in Oshana region, Namibia
(University of Namibia, 2025) Angula, Maria A.; Misihairabgwi, Jane
Contamination of agricultural foods by fungi and exposure to secondary metabolites
of fungi, (mycotoxins) poses a significant global health risk especially for vulnerable
children. Given limited data regarding mycotoxicological safety and health risks of
foods consumed by Namibian children, this study determined commonly consumed
foods, nutritional status of children, fungal and mycotoxin contamination in processed
cereal foods, and exposure of non-breastfeeding children under 5 years to mycotoxins.
Additionally, exposure of children to the carcinogenic mycotoxin, aflatoxin B1
(AFB1) through urinary aflatoxin M1 (AFM1) levels was determined and associated
with growth impairment indicators in children under 5 years from rural households in
the Oshana region of Namibia.
This study employed a cross-sectional quantitative design, involving 277 children
under 5 years. Validated dietary diversity and food frequency questionnaires were used
to gather demographic and food consumption pattern data. Anthropometric
measurements were used to assess nutritional status of the children using Emergency
Nutrition Assessment (ENA) software. Processed cereal foods (n=207) were analysed
for multi-mycotoxins via liquid chromatography mass spectrometry (LC-MS/MS) and
fungal species using molecular techniques. Urinary aflatoxin M1 was quantified in 126
urine samples using Enzyme Linked Immunosorbent Assay (ELISA) and child
exposure to various mycotoxins was determined using the deterministic approach.
Descriptive and inferential statistics were computed using IBM® SPSS® Statistics.
Staple cereal foods were commonly consumed by children. Child feeding practices:
minimum dietary diversity (MDD) (36.5%), minimum meal frequency (MMF)
(52.3%), minimum milk feeding frequency (MMFF) (32.1%) and minimum
acceptable diet (MAD) (7.2%) were low among children. Stunting (28.9%),
underweight (13%) and wasting (4.7%) was prevalent among the studied population.
The identified fungal genera included Aspergillus being the most prevalent (72.3%),
followed by Fusarium (15.2%), Penicillium (5.9%), Paecilomyces (2.3%), Alternaria
(2.7%), Monascus (0.8%), and Asordaria (0.8%). Fungal secondary metabolites
quantified in processed cereal foods included major mycotoxins such as AFB1
(36.7%), zearalenone (24.2%), fumonisin B1 (FB1, 21.2%), citrinin (CIT, 16.4%) and
deoxynivalenol (8.2%). Food samples (35.7% (n=74) and 4.8% (n=10)) exceeded the
ii
0.1 µg/kg AFB1 and 200 µg/kg FB1 EU limit for children’s food, respectively. Several
emerging mycotoxins including the neurotoxic 3-nitropropionic acid, moniliformin,
and tenuazonic acid were quantified in over 50% of all samples. Co-occurrence of
AFB1, FB1 or CIT was in 7.2% (n=15) and could heighten the food safety concerns.
Regarding exposure assessment and risk characterization, average probable dietary
intake for AFB1 from all ready-to-eat-foods was 0.036µg/kgbw/day, which resulted in
margin of exposures (MOE) of 11, far below 10000 and 0.65 risk cancer
cases/year/100000 people, indicating a risk of chronic aflatoxicosis. High tolerable
daily intake values for FB1, and low MOE for CIT exceeded reference values
suggesting potential health risks. Additionally, AFM1 was present in 13% of the
children’s urine with a mean concentration of 0.23±0.15 ng/ml, indicating the recent
exposure of children to AFB1. However, no association was found between stunting,
underweight and wasting with urinary AFM1 exposure. Appropriate child feeding
practices are required to combat malnutrition among rural children and targeted
interventions including appropriate food handling or storage practices to mitigate
fungal contamination of foods must be prioritised
Epidemiological measures and multivariate time series application to investigate trends of Covid-19 cases in Namibia
(University of Namibia, 2025) Shifidi, Tuyakula; Oyedele, Opeoluwa
The outbreak of the novel COVID-19 has caused a global health crisis since its emergence
in December 2019. Namibia, like many other countries, has been affected by the COVID19 pandemic, with the virus spreading rapidly across all 14 regions of the country. The
impact of COVID-19 on the health and economy of Namibia has been significant, with
many individuals being infected, and a significant number of deaths reported. However,
not many studies have been conducted on the trend and epidemiology of the virus in
Namibia. This lack makes it challenging to understand the pattern and distribution of the
virus's spread and how to handle future pandemics, hindering the development of effective
strategies to prevent and control the spread of COVID-19 in the country. Therefore, this
study aimed to provide an epidemiological examination of COVID-19's impact in
Namibia from March 2020 to May 2023. Utilizing data from the Ministry of Health and
Social Services (MoHSS), this study analyzed key epidemiological measures, including
Case Fatality Rate (CFR), Infection Fatality Rate (IFR), Mortality Rate (MR), and
Incidence Rate (IR), investigated the impact of comorbidities in influencing COVID-19
outcomes and employed a multivariate time series analysis to understand the dynamics
between COVID-19 cases, deaths, and recoveries in the country. Results showed that the
national average CFR was 2.38%, with the Omaheke Region having the highest rate at
5.78% while the Khomas Region had the lowest at 1.41%. The estimated IFR, adjusted
using a multiplier of 3 to 5 to reflect likely underreporting, ranged from 0.47% to 1.93%
under the ×3 scenario and from 0.28% to 1.16% under the ×5 scenario, with a national
average of 0.79% (×3) and 0.48% (×5) respectively. The MR averaged 0.00163
nationally, with the highest in Omaheke Region (0.00425) and the lowest in Kavango
West Region (0.00025). The IR was highest in the //Kharas Region at 0.119 and lowest
in Kavango West Region at 0.009. The comorbidities analysis showed that individuals
with pre-existing conditions had a significantly higher mortality rate (6.26%) compared
to those without (2.08%). The risk ratio indicated that individuals with comorbidities were
about three times more likely to die from COVID-19 (RR = 3.012, p < 0.001, 95% CI:
2.79 - 3.25), with odds of death being 3.15 times higher compared to those without
comorbidities (OR = 3.1452, p < 0.001, 95% CI: 2.90 - 3.41). The multivariate time series
ii
analysis using Vector Autoregression (VAR) models revealed that past deaths had a
significant positive effect on current cases, recoveries, and deaths, accentuating a
persistent pattern in the pandemic’s progression. The VAR model estimation for the cases
equation indicated that the first lag of deaths had a significant positive impact on current
COVID-19 cases (coefficient = 64.3926, p = 0.0139), suggesting that an increase in deaths
in the previous period is associated with a significant rise in current cases. The model also
showed that approximately 76% of the variation in current cases was explained by the
included variables (multiple R2 = 0.7576). Furthermore, the Impulse Response Function
(IRF) analysis showed that shocks to deaths had a stabilizing effect over time, influencing
subsequent case and recovery trends. The Variance Decomposition analysis highlighted
that the variance in COVID-19 cases explained by their own shocks decreased from 100%
in the first period to 71.53% by the tenth period, with deaths and recoveries contributing
15.93% and 12.54%, respectively. For deaths, the variance explained by cases dropped
from 83.93% to 64.31% by the tenth period, while contributions from deaths and
recoveries rose to 18.52% and 17.17% respectively. These findings reflect the dynamic
interplay among these variables and emphasize the importance of managing mortality to
control the spread of the virus. Thus, it is recommended that the Namibian government
utilize these findings to strengthen preparedness by focusing on improving healthcare
systems, addressing comorbidities, and tailoring interventions to different regions, age
groups, and genders. This can further help to ensure a more effective and inclusive
response to future pandemics
An assessment of factors influencing housing demand in Namibia
(University of Namibia, 2025) Mutakalilumo, Daphne; Tingum, Ernest
A house is one of humanity's fundamental needs, providing shelter and sustenance for
family life. Therefore, having a "dream home" is a goal that many people have for their
lives, but this requires taking many factors into account. The aim of this study was to
examine the nexus between macroeconomic factors and housing demand in Namibia
for the period 2000Q1 to 2022Q1. The methodology employed relied on secondary
time series data for macroeconomic variables. The dependent variable, the quantity of
houses demanded was analysed against the independent macroeconomic variables
such as GDP, Household Disposable Income, Household Debt, Inflation Rate,
Mortgage Lending Rate, and Real Exchange Rate. The study adopted a bound test to
cointegration approach under an Auto Regression Distributive Lag Model (ARDL)
and pairwise Granger causality. The ARDL results revealed the existence of a long run
association among the variables, thereby necessitating the estimation of an Error
Correction Model. Further analysis showed that only inflation and household debt
were found to have a negative effect on housing demand in Namibia. The results of
the Granger causality test confirmed that there is a unidirectional causality relationship
between housing demand and the majority of the macroeconomic variables.
Furthermore, a bidirectional causality was confirmed between few variables, mostly
at 1 percent level of significance. The findings of the study have policy implications
hence the study recommends the central bank to be extra cautious as they institute the
monetary policy in controlling inflation and household debts. Additionally, investors
are also cautioned to closely monitor developments of the macroeconomic
environment prior to any investment decision in the housing market. Finally, the
government is also encouraged to provide home ownership incentives to those who are
not able to finance homes through mortgage loans
Modelling co-occurrence of risk factors for non communicable diseases in selected urban areas of Namibia
(University of Namibia, 2025) Sakaria, Naphtal Mwiishuule; Indongo, Nelago
It is observed that the global burden of diseases has shifted from infectious diseases to
Non-Communicable Diseases (NCDs), with an accumulative trend in developing
countries. NCDs share key modifiable behavioural risk factors like unhealthy diet and lack
of physical activity that are typically established during adolescence or young adulthood
and will set the stage for NCDs development later in life. The target population for the
study was people residing in urban areas, aged 30 years and older. Of these population a
sample of 1533 respondents were selected based on the 2011 Population and Housing
Census sampling frame with 3 per cent at the 95 per cent confidence interval (CI) was
used for the sample calculation. The study used secondary data based on a cross-sectional
survey conducted in 2017 by the University of Namibia to assess the influence of lifestyle
related diseases on the well-being of people living in urban towns in Namibia.
This study aimed to explore the co-occurrence of risk factors for NCDs among persons
aged 30 years and above in relation to the socio-demographic and behavioral factors in
selected urban areas of Namibia. Specifically, the study focused on investigating the
prevalence of NCDs in relation to sociodemographic, behavioral and attitude
factors. Simple descriptive statistics were used to describe the socio-demographic
characteristics of the population and the prevalence of NCDs among the respondents.
Pearson Chi-square for association was used to assess the association between variables
using cross-tabulation with chi-square to examine the prevalence of NCDs and association
with socio-demographic characteristics of the respondents. A multinomial regression
model was used to predict factors associated with the co-occurrence of risk factors and
i
their effects. The results showed that overall, the prevalence of NCDs increases with age,
and it is more prevalent among respondents with better education levels and those
employed, which may be a result of a lack of physical activity and the uptake of an
unhealthy diet. The study concluded that there is a high co-occurrence of risks, and most
risk factors are positively associated with NCDs. Hence, the study recommends the need
for the government and all sectors to play a role in the well-being of urban people through
initiatives such as better urban planning with facilities that promote healthy lifestyles; the
promotion of urban agriculture, as well as introduction of weekend vegetable markets to
encourage residents to consume health foods and promote healthy diet. Furthermore,
preventative measures should be prioritised to reduce medical costs and workplaces
should actively promote wellness programs for staff members.
A statistical analysis of factors influencing first teenage pregnancy in Namibia
(University of Namibia, 2025) Nghikongwa, Rosalia; Indongo, N.
The World Health Organization (WHO) estimated that about 21 million girls aged
15-19 years in developing countries became pregnant in 2019. In Namibia
Teenage Pregnancy has been a national concern because of the implications to
education attainment and individual socio-economic development. The 2013
Namibia Demographic and Health Survey (NDHS) indicated that 19% of women
aged 15-19 had begun childbearing which represented a 4% increase in teenage
pregnancies since 2006.
This study aimed to identify factors associated with teenage pregnancy in Namibia.
Specifically, the study also aimed to describe the regional variation of teenage
pregnancy in Namibia; and to assess the influence of socio-demographic and
behavioural factors on teenage pregnancy. The study was mainly quantitative and
based on secondary analysis of data. The target population was all female
teenagers aged 15-19 and the main source of data were the 1992, 2000, 2006 and
2013 NDHS. The methods of analysis used were descriptive, bivariate, binary
logistic regression model and cox proportional hazards model. Regional variations
was used to examine the trend and extent of teenage pregnancy. Descriptive and
Chi-square tests were performed to highlight the prevalence and regional
variations of teenage pregnancy, while the binary logistic regression was used to
estimate the probability of experiencing teenage pregnancy taking into account
socio-economic and demographic factors. The Cox Proportional Hazards model
was further used to model the survival time to first pregnancy amongst teenagers
with different background characteristics.
Overall the study revealed regional variations of teenage pregnancy whereby
Kavango region recorded the highest prevalence across all the years whilst Hardap
region recorded the lowest prevalence across all the years. The results also showed
that age of respondent, age at first intercourse, education, current contraceptive
method, total children ever born, relationship of last sexual partner, employment
status, marital status, region and place of residence were significantly associated
with teenage pregnancy. Furthermore, the results also revealed that teenagers had
a survival chance of experiencing their first birth at 17 years.
This study concluded that a number of socio-economic and demographic factors
were associated with teenage pregnancy in Namibia and recommended that
ii
adolescent friendly services at hospitals and clinics should be strengthened and
teenagers should be given relevant information on contraceptives to make
appropriate and informed decisions on contraceptive usage. The life skills
curriculum for the Junior and Secondary learners should also be improved to
include topics relating to awareness and prevention of teenage pregnancy. There
is also need for robust awareness campaigns on prevention of teenage pregnancy
targeted to teenagers in more affected regions such as in Kavango and Zambezi
regions
Impact of human activities and environmental variables on habitat use of Heaviside’s dolphin (Cephalorhynchus heavisidii) and bottlenose dolphin (Tursiops truncatus) in Walvis Bay: A passive acoustic monitoring approach
(University of Namibia, 2025) Nghaangulwa, Priskilla P. N.; Iitembu, Johannes
The rapid expansion and scope of development along marine coastlines threaten marine
mammals with cumulative effects from human practices. such as vessel strikes, by-catch,
entanglement, debris ingestion, petroleum exploration, underwater shipping noise, and
other noise pollution. This study aimed to understand the impacts of environmental and
anthropogenic factors on the habitat use patterns of two dolphin species; Heaviside’s
dolphins (Cephalorhynchus heavisidii) and common bottlenose dolphins (Tursiops
truncatus) which are a small, localized population, at two key sites in Walvis Bay: Afrodite
Beach and Pelican Point. The specific objectives were to: (a) Investigate the influence of
anthropogenic factors (boat sonar and a period harbor expansion and their associated noise)
on the diel present patterns and habitat use of C. heavisidii at Pelican Point and T. truncatus
at Afrodite Beach. (b) Assess whether environmental factors (time of day, season, moon
illumination, temperature, wind direction and wind speed) influence habitat use of C.
heavisidii at Pelican Point and T. truncatus at Afrodite Beach. (c) Investigate the
interspecific influence of T. truncatus on the habitat use patterns and diel presence of
heavisidii at Pelican Point. (d) Assess the impacts of harbor expansion on the seasonal
distribution and habitat use of C. heavisidii and T. truncatus, comparing patterns of
presence and area usage before and during the development process. The study used data
obtained from the Namibian Dolphin Project and collected using a Passive Acoustic
Monitoring (PAM) over a long-term study period of 2009-2014 (pre construction) and
2015-2018 (during construction). Acoustic data was collected using C-POD hydrophones
with sounds extracted and processed using the C-POD developer software CPOD.exe
v2.033. A hurdle modelling approach was used to investigate sonar, time of day, season,
i
period of study, and environmental parameters predictor variables (water temperature, wind
speed and direction and moon illumination) potentially influencing dolphin presence and
behavior at these sites. The results showed that boat-related noise had a significant impact
on the diel presence ((p ˂ 0.001) at Pelican Point and (p ˂ 0.001) at Afrodite Beach) and
habitat utilization (p= 0.0433 at Pelican Point and p = 0.0153 at Afrodite Beach) of both
species. Environmental elements such as moon illumination (p ˂ 0.001), wind direction (p
= 0.025), wind speed (p = 0.0156), and temperature (p ˂ 0.001) have a significant impact
on their behaviors. Furthermore, inter-species interactions were observed, with T. truncatus
having a considerable influence on C. heavisidii, especially at seasonal time scales. As a
result, the study proposes targeted noise reduction from boats and the incorporation of
seasonal dynamics into conservation planning