ANN Model Development for Air Permeability in Biochar Amended Unsaturated Soil

Passage of municipal waste induced greenhouse gases such as carbon di oxide (CO₂) and methane in landfill covers, majorly depends on the state of unsaturation and compaction of soil biochar composite (SBC). The unsaturated state of SBC can be identified by measuring soil suction and volumetric water content (VWC) that can affect the air permeability of landfill covers. To design the landfill covers, it is required to propose a model which can forecast the air permeability up to a certain degree of accuracy. The aim of this study is to investigate the effect of soil suction and moisture content on gas permeability for different biochar application percentages at high degree of compaction and develop an artificial neural network (ANN) based model to predict the gas permeability and obtain optimized value of soil suction and moisture content for extreme gas passage. In this study, results represent that presence of biochar can decrease the gas permeability significantly. For 5% and 10% biochar application percentages decrement in gas permeability is around 50% and 65% with respect to bare soil. Developed ANN model shows that in the presence of biochar, gas permeability of SBC is more sensitive to VWC than soil suction i.e. a small change in VWC can change the gas permeability, significantly. Optimization analysis also shows that addition of biochar can increase the optimized VWC for biochar amended soils which can help to design most effective soil cover to provide required nutrients and water for vegetation growth.