International Journal of Sustainable Energy and Environmental Research

Daily and spatial generation of palm oil mill effluent in determining volumetric capacity for bio-methane planning

E B Tambe, A U Okonkwo, I E Mbuka-Nwosu, S E Unuafe, C O Cookey, M F Umunna — Thu, 06 Jun 2024 00:00:00 -0500

Cumulative determination of wastes generated per day and in different periods is a necessary prerequisite for planning waste recovery and averting environmental degradation. This study examined palm oil mills (small, medium and large-scaled mills) in the oil palm (Elaeis guineensis) industry at Ohaji/Egbema Local Government Area (LGA), Imo State, Nigeria for daily and periodic generation of palm oil mill effluent (POME). Survey research design was used. Eight catchment communities of the large-scaled mill, Agricultural Development Authority Palm (ADAPALM) were categorised into three strata in relation to the number of small-scaled mills in each community. In each stratum, a community was randomly sampled. Nine small-scaled mills were sampled from the three communities using proportional sampling. The lone medium and large-scaled mill in the study area represented the other scales of milling. For small and medium-scaled mills, the volume of POME generated was measured from the dimensions of the vessels where POME was stored, while that of large-scaled mill was obtained from industrial records. Independent sample T-test revealed an average of 17.574±0.408m3 POME/day and 15.509±0.465m3 POME/day in wet and the dry seasons respectively (p˂0.01). Similarly, within milling-scales, ANOVA and T-test showed that significant variations occur (p<0.01). Wilcoxon’s test revealed that there was no significant difference (p˃0.05) in the median functional small-scaled mills across seasons, hence functionality did do not influence the volume of POME generated across seasons. The findings revealed the copious volume of POME generated in the area and the required volumetric requirements to planning its collection and transformation to wealth.

Review of the environmental and human health impacts of mercury use in gold mining in the Santurban Paramo area

Daniela Garcia Moreno, Angie Tatiana Ortega-Ramirez — Thu, 06 Jun 2024 00:00:00 -0500

Illegal mining is a problem that currently affects the vegetation cover of Colombia due to the environmental liabilities and the damage it causes to the environment in the eagerness to obtain gold in an easy way by using mercury. For this reason, this article presents a detailed bibliographical review of the environmental, social and economic problems of artisanal mining in the Santurban Paramo due to the inappropriate use of mercury and its impact on both the environment and the health of the people who perform this activity, describing first of all the importance of the Paramo in relation to the protection of its ecosystem and biodiversity; secondly, mercury as an imminent danger due to its use and the effects of the activity taking into account the policies and regulations carried out for the preservation and reduction of environmental contamination generated by the bad practices of artisanal mining and finally, the environmental and human health impacts due to direct and indirect contact with mercury. As a result of this research, although artisanal mining represents an important role for the survival of the community, it is necessary to implement continuous improvement strategies to reduce the use of mercury due to the problems caused by this dangerous metal both for the environment and the water source fundamental for the life of the Santurban Paramo and for the health of the people, where education and environmental awareness gain more strength for the protection and preservation of this natural reserve in Colombia.

A machine learning approach to forecast wind speed based on geographical location in Bangladesh

Elmeeh Hasan Shipra, Md Sujaur Rahaman, Tasnim Ara, Saeed Mahmud Ullah — Thu, 20 Jun 2024 00:00:00 -0500

This paper examines a learning approach to forecasting wind speed based on geographical location in Bangladesh. Most people use wind energy, a rapidly expanding renewable energy source, to produce electricity, replacing traditional fossil fuel-based electricity generation. The variable nature of wind speed necessitates an accurate estimate for planning wind power generation and grid integration. Machine learning models, based on various methods, commonly make wind speed predictions. This study implements machine learning algorithms, namely Random Forest Regression, XGBoost Regression, Multi-Layer Perceptron Regression, Ridge Regression, and Lasso Regression, to determine wind speed prediction accuracy. Firstly, we divide the dataset on wind records of Bangladesh into seven categories, each encompassing different geographical locations in Bangladesh, to account for the change in wind characteristics based on location, and then apply the algorithms to each category. The performance metrics Mean Absolute Error, Mean Squared Error, and Root Mean Square Error are utilized to draw comparisons. The findings showed that the performance of XGBoost makes it the most reliable tool for predicting wind speed in Bangladesh. The near proximity to large water bodies causes considerable variation in accuracy, i.e., it showed higher accuracy compared to the other three algorithms after applying the ensemble learning approach, which is more effective and accurate. This research will help in identifying optimal power plant locations and efficient linking methods to Bangladesh's power grid, ensuring smooth electricity access and efficient utilization of the country's energy resources.