2647 research outputs found
Sort by
Optimization of goat manure against use of N-Minjingu Nafaka Plus fertilizer for improved lablab growth and yield in semi - arid areas of Northern Tanzania
Get PDFA research article was submitted to Environmental Challenges Volume 13, December 2023The study was conducted in Northern Tanzania from the 2022–2023 cropping seasons to develop a cropping model that farmers can use to improve their lablab production. The field experiments were conducted at the Tanzania Agricultural Research Institute, Selian station and was laid down in a Split Plot Designs with main factor being tillage system and sub-factor being fertilizer type (Goat manure and Minjingu Nafaka Plus). The trials consisted of three treatments (goat manure, Minjingu Nafaka Plus and the control-no fertilizer) replicated thrice. There were significant differences (p=<0.001) among the number of pods per plant, plant height and lablab grain yield. The highest grain yield (3952 g and 3933 g) with net benefit of (1,290,210Tshs and 1,249,175Tshs) was recorded in a plot treated with (Minjingu Nafaka Plus) fertilizer in Conventional and Zero tillage system compared with (Goat manure) which had (3944 g and 3928 g) with net benefit of (1,306,710Tshs and 1,292,675Tshs) in Conventional and Zero tillage practices. Conclusively, the study revealed that there were high net benefits under zero tillage compared to conventional tillage based on input-output costs analysis at both TARI - Selian and Saweni sites. Zero tillage was more economically viable than conventional tillage practices as becomes more friendly for the resource-constrained farmers in increasing their potential yield
The hydrochemical evolution and water balance of the emakat lake in the northern crater highland of Tanzania
Get PDFA Dissertation Submitted in Partial Fulfilment of the Requirements for the Degree of
Master’s in Hydrology and Water Resources Engineering of the Nelson Mandela
African Institution of Science and TechnologyThis study aimed to ascertain the hydrochemical evolution and water balance of Emakat Lake,
of the Empakaai Crater. Water and rock samples were collected from the lake and springs on
the inner and outer crater rims, and at the foot of the Empakaai Crater. The results showed that
the lake is a highly alkaline (pH > 10) and saline (electrical conductivity (EC) = 28,860 - 29,460
μs/cm) with the concentration of total dissolved solids (TDS) ranging from14,432 to 14723
mg/L. Springs exhibited lower pH (6.85 - 8.69), EC (562 - 1584 μs/cm) and TDS (276 - 1016
mg/L). The dominant ions in Emakat Lake were Na+
and CO3
2-+HCO3
- which occupy about
80% and 85% of the cation and anion phases with ion distribution of Na+ > K+ > Ca2+ > Mg2+
and (CO3
2- + HCO3
-
) > Cl- >SO4
2- > F- > NO3- > PO4
3-
. Piper, chloro-alkaline indices, Chadha,
and Gibbs plots revealed that Na-K-HCO3 water type dominated Emakat Lake, and a majority
of springs exhibited mixing characteristic water type. Base ion-exchange dominated the
hydrochemical evolution of both lake and springs, influenced by evaporation and water-rock
interaction for the lake and springs respectively. The water balance of Emakat Lake was highly
influenced by groundwater flow which accounted for 49% of the inflow and 56% of the
outflow. This suggest that Emakat Lake plays a major role in the hydrological system in the
area alongside the springs which are the sources of the major rivers of Engaruka and
Engaresero
Removal of arsenic in a sand filter coupled with zero valent iron
Get PDFThis research article was published in Hydro-research Journal, Volume 6, 2023.Arsenic (As) in wastewater has negative effects on the environment and human health, hence As containing wastes must be handled properly. Given the accessibility of metallic iron, studies investigating into the potential application of zerovalent iron in the removal of arsenic are promising. In this study, the performance of sand filter blended with several kinds of zero valent iron (Fe0), such as iron wool, iron fillings, and iron nails, were compared. These materials were combined in a sand filter column, and the efficiency was calculated using the As concentrations in the influent and effluent samples. Experiments were carried out in order to compare performance as a function of Fe0 dose and contact time. The outcome of this investigation showed that sand filter containing iron wool had a better removal efficiency of arsenic removal than iron filings and iron nails. The results in all columns showed that as dosage was increased, removal efficiency of arsenic increased significantly. In case of contact time the results revealed that arsenic can effectively be removed from water in the first 48 h. The early adsorption response is quick in all columns, but get slower as time goes on. The highest removal efficiency was 99.6% and the lowest removal efficiency was 82.7%
A Multi-Modal Wireless Sensor System for River Monitoring: A Case for Kikuletwa River Floods in Tanzania
Get PDFA research article was submitted to Sensors 2023, volume 23Reliable and accurate flood prediction in poorly gauged basins is challenging due to data scarcity, especially in developing countries where many rivers remain insufficiently monitored. This hinders the design and development of advanced flood prediction models and early warning systems. This paper introduces a multi-modal, sensor-based, near-real-time river monitoring system that produces a multi-feature data set for the Kikuletwa River in Northern Tanzania, an area frequently affected by floods. The system improves upon existing literature by collecting six parameters relevant to weather and river flood detection: current hour rainfall (mm), previous hour rainfall (mm/h), previous day rainfall (mm/day), river level (cm), wind speed (km/h), and wind direction. These data complement the existing local weather station functionalities and can be used for river monitoring and ext reme weather prediction. Tanzanian river basins currently lack reliable mechanisms foraccurately establishing river thresholds for anomaly detection, which is essential for flood prediction models. The proposed monitoring system addresses this issue by gathering information about river depth levels and weather conditions at multiple locations. This broadens the ground truth of river
characteristics, ultimately improving the accuracy of flood predictions. We provide details on the monitoring system used to gather the data, as well as report on the methodology and the nature of the data. The discussion then focuses on the relevance of the data set in the context of flood prediction,the most suitable AI/ML-based forecasting approaches, and highlights potential applications beyond flood warning systems
Automated Optimization-Based Deep Learning Models for Image Classification Tasks
Get PDFThis research article was published by Computers 2023Applying deep learning models requires design and optimization when solving multi-
faceted artificial intelligence tasks. Optimization relies on human expertise and is achieved only
with great exertion. The current literature concentrates on automating design; optimization needs
more attention. Similarly, most existing optimization libraries focus on other machine learning
tasks rather than image classification. For this reason, an automated optimization scheme of deep
learning models for image classification tasks is proposed in this paper. A sequential-model-based
optimization algorithm was used to implement the proposed method. Four deep learning models, a
transformer-based model, and standard datasets for image classification challenges were employed in
the experiments. Through empirical evaluations, this paper demonstrates that the proposed scheme
improves the performance of deep learning models. Specifically, for a Virtual Geometry Group
(VGG-16), accuracy was heightened from 0.937 to 0.983, signifying a 73% relative error rate drop
within an hour of automated optimization. Similarly, training-related parameter values are proposed
to improve the performance of deep learning models. The scheme can be extended to automate the
optimization of transformer-based models. The insights from this study may assist efforts to provide
full access to the building and optimization of DL models, even for amateurs
Monitoring Kikuletwa river levels in northern Tanzania: A data set unlocking insights for effective flood early warning systems
Get PDFA research article was submitted to Data in Brief Volume 49, August 2023Floods are a recurring natural disaster that pose significant risks to communities and infrastructure. The lack of reliable and accurate data on river systems in developing countries has hindered the development of effective flood early warning systems. This paper presents a data set collected using ultrasonic distance sensors installed at two locations along the Kikuletwa River in the Pangani River Basin, Northern Tanzania. The dataset consists of hourly measurements of river water levels, providing a high-resolution time series that can be used to study trends in water level changes and to develop more accurate flood early warning systems.
The Kikuletwa River dataset has significant potential applications for flood management, including the calibration and validation of hydrological models, the identification of critical thresholds for flood warning, and the evaluation of flood forecasting techniques. The dataset can also be used to study the hydrological processes in the basin, such as the relationship between rainfall and river discharge, and to develop more efficient and effective flood management strategies.
The ultrasonic distance sensors were configured to record river level data at hourly intervals, providing a continuous time series of river levels. The data was subjected to quality control procedures to ensure accuracy and consistency, and missing or erroneous data was corrected or removed where necessary
Are electric vehicles economically viable in sub-Saharan Africa? The total cost of ownership of internal combustion engine and electric vehicles in Tanzania
Get PDFThis research article was published by Transport Policy Volume 141, September 2023,The prevalence of internal combustion engine vehicle (ICEV) fleets globally has resulted in various environmental issues, such as the emissions of greenhouse gases, reliance on imported petroleum products, significant degradation of air quality, and adverse health impacts on people. To address these challenges, the adoption of electric vehicles (EVs) is viewed as a sustainable solution. This study analyzed the Total Cost of Ownership (TCO) of EVs in sub-Saharan Africa to determine if they are viable options for consumers from Tanzania. Contrary to previous studies on the competitive position of EVs that focused on Europe, Asia, and other regions with high EV diffusion, and are more advanced in terms of EV manufacturing capacity and promoting policies, this study focused on Tanzania, a country with low EV diffusion and no EV manufacturing capacity. We compared the economics of electric cars and electric two-wheelers (e2Ws) and their ICE counterparts. The findings show that the TCO per km of electric cars is higher than that of their ICE car counterparts, while the TCO of e2W was less than that of their petroleum counterparts. Importing taxes charged to all vehicles imported into the country significantly hike the upfront cost of EVs. For electric cars, particularly battery electric vehicles, to reach TCO parity with ICE car counterparts, the current import taxes have to be reduced by 40% or more, which is equivalent to removing all import duty or value-added taxes. In this regard, electric cars are still not economically viable for Tanzanian automotive consumers, unless economic incentives are introduced. With EVs being in the early stage in the country, it is recommended to start by promoting e2Ws, which are economically viable for many consumers in the Tanzanian context
Catalytic hydrothermal liquefaction of orange peels into biocrude: An optimization approach by central composite design
Get PDFThis research article was published in the Journal of Analytical and Applied Pyrolysis, Volume 173, August 2023.Global instability, persistent increase in pump-price, inflation and depletion of fossil fuel resources amidst the continuous discharge of greenhouse gases from fossil fuels combustion call for urgent attention. The application of novel catalyst towards an improved biocrude production and enhanced biomass conversion are required for effective performance of hydrothermal liquefaction of biomass feedstocks. In the present study, iron supported carbon nanospheres (Fe/CNSs) catalyst has been developed using wet impregnation approach and explored for its catalytic potency in improving yield of biocrude using an optimization approach; central composite design. Hydrothermal liquefaction (HTL) was adopted as a process route where the catalyst dosage (3–6 wt%) and the reaction temperature (330–430 °C) were optimized on the constant weight of orange feedstock (10 g), reaction time of (15 mins) and solvents’ ratio of 3:1 (acetone to ethanol). The performance of the catalyst was found to be aided with the even dispersion of the Fe on the surfaces of the CNSs support and improved surface area. The effects of reaction temperature were observed to be more progressive on biocrude yield formation over the catalyst loading. The optimum biocrude yield, solid residue, biomass conversion and gas yield were obtained to be 71.09 wt%, 28.18 wt%, 71.82 wt% and 40.14 mL/g at 430 °C and 3 wt% catalyst loading. The obtained analysis of variance suggests a good correlation between the experimental and predicted data in all responses. The selectivity of Fe/CNSs catalyst to high yield of phenolics and aromatic compounds in the biocrude suggest that, the biocrude can be further be upgraded into transportation fuel through hydrodeoxygenation process. The findings have further suggested the viability of the developed Fe/CNSs as an effective catalyst for improved HTL products in a batch reactor
Data from the batch adsorption of ciprofloxacin and lamivudine from synthetic solution using jamun seed (Syzygium cumini) biochar: Response surface methodology (RSM) optimization
Get PDFThis research article was published by Elsevier, 2023This dataset expresses the experimental data on the batch adsorption of ciprofloxacin and lamivudine from synthetic solution using jamun seed (JS) (Syzygium cumini) biochar. Independent variables including concentration of pollutants (10-500 ppm), contact time (30–300 min), adsorbent dosage (1-1000 mg), pH (1-14) and adsorbent calcination temperature (250,300, 600 and 750 °C) were studied and optimized using Response Surface Methodology (RSM). Empirical models were developed to predict the maximum removal efficiency of ciprofloxacin and lamivudine, and the results were compared with the experimental data. The removal of polutants was more influenced by concentration, followed by adsorbent dosagage, pH, and contact time and the maximum removal reached 90%
Data Balancing Techniques for Predicting Student Dropout Using Machine Learning
Get PDFThis research article was published MDPI, 2023Predicting student dropout is a challenging problem in the education sector. This is due to an imbalance in student dropout data, mainly because the number of registered students is always higher than the number of dropout students. Developing a model without taking the data imbalance issue into account may lead to an ungeneralized model. In this study, different data balancing techniques were applied to improve prediction accuracy in the minority class while maintaining a satisfactory overall classification performance. Random Over Sampling, Random Under Sampling, Synthetic Minority Over Sampling, SMOTE with Edited Nearest Neighbor and SMOTE with Tomek links were tested, along with three popular classification models: Logistic Regression, Random Forest, and Multi-Layer Perceptron. Publicly accessible datasets from Tanzania and India were used to evaluate the effectiveness of balancing techniques and prediction models. The results indicate that SMOTE with Edited Nearest Neighbor achieved the best classification performance on the 10-fold holdout sample. Furthermore, Logistic Regression correctly classified the largest number of dropout students (57348 for the Uwezo dataset and 13430 for the India dataset) using the confusion matrix as the evaluation matrix. The applications of these models allow for the precise prediction of at-risk students and the reduction of dropout rates