Quantitative evaluation of leachate using pilot scale landfill lysimeter

Abstract

The population and the economic growth and the fast urbanization lead to the generation of increasing quantities of solid as well as liquid wastes which have severe consequences to the environment. More than eight million tons of solid waste is produced per day in developing countries. Over 95% of this waste is disposed off in landfills, open dumps, on riverbanks, directly into the sea, or just combusted on site because of insufficient waste collection and final disposal systems. Meanwhile Europe and industrialized countries go for high-tech solutions (e.g. modern incineration technologies) there is still a huge demand for proper landfilling in developing countries. Landfilling is likely to be the most appropriate and costeffective final disposal option for solid waste in developing countries. The emissions of leachate and gas from landfills are main threat to environment. One of the main concerns is water contamination by leachate from landfills. The environmental risks associated to landfilling are related to leachate generation and risks of surface and groundwater pollution and offensive odour. The biodegradable portion of waste is largely responsible for the production of leachate and landfill gas. The leachate management is thus utmost important in terms of both quantity and quality. The leachate collection and treatment system in a landfill has to be designed carefully keeping the leachate production minimum. The leachate generation depends upon the climatic conditions of specific location and can be controlled by variation of hydraulic properties of materials of layers used in any landfill and analyses of water balance. This particular research presents the outcome of the study on the water balance of landfill in Nepal using a pilot scale landfill lysimeter at Kathmandu University. The related leachate production (percolation) as an effect of variation of properties of layer materials and climatological factors has been assessed. The Hydrologic Evaluation of Landfill Performance (HELP), a computer model has been used to estimate the water balances and comparison with the actual leachate (percolation) measurement was done. The local weather data (daily values of rainfall, temperature and solar radiation), vegetative growths were collected as required and variable soil and waste data (total porosity, field capacity, wilting point, initial moisture content and saturated hydraulic conductivity of layers and materials) have been determined at laboratories and some default data were used from the model. A set of simulations were done viz; A, B and C with variations in Field Capacity values of 0.2, 0.292 and 0.35 vol./vol. for hydraulic conductivity values of 0.1, 0.01 and 0.001 cm/s of waste materials respectively and also another set of simulations using various hydraulic conductivity values at the exponential orders of E-3 to E-5 and E-4 to E-9 cm/s for cover soil and barrier soil liners respectively. vii It was observed that the percolation generally follows the rainfall trend. With the results of simulations carried out, it indicates that the evapotranspiration (ET) do not exactly follow the rainfall & percolation trend and ET is more on horizontal trend based on average years. The annual percolation rate is high in lysimeter (78 to 86% of rainfall), which is due to small area of lysimeter. There seems to be more percolation than evapotranspiration as more infiltration occurred before evapotranspiration could take place. The daily average percolation rate is as low as 5.8 mm (only 5.4% of daily average rainfall) when the higher rainfall events are considered compared to the annual values. The percolation response is observed only after few days of rainfall instead of immediate response. This is an important design consideration for landfill. Thus, the design of landfill leachate treatment system should be done on annual leachate generation basis rather than daily data. The leachate should be directed to collection and treatment system rather than allowing percolation through barrier soil liner and to ground water or surface water bodies. The model has been calibrated for the local situation with the observed data (from June to December 2006) of leachate generation from the pilot scale landfill lysimeter. However, the trend of leachate generation on HELP simulation and actual percolation seem to be similar during October to December season, but from June to September, the trend shows higher actual percolation rate compared to the model. This may be due to the higher value (in the range of E-5 cm/s) of hydraulic conductivity of barrier soil liner, which should be generally lower value (in the range of E-7 cm/s or more), though difficult to achieve naturally. Also higher actual percolation may be due to the rainy season (June-September) when soil is wet at most of the time. The response of average percolation and evapotranspiration with change of hydraulic conductivity values of barrier soil liner is very important. With the change of order of E-6 to E-7 cm/s in hydraulic conductivity of barrier soil liner, there is significant change in the results. With lesser values, there is no percolation and there is significant increment in ET value. This provides an important design consideration of landfill, where hydraulic conductivity of barrier soil liner is deciding parameter and should be in the order of E-7 cm/s or lesser. When less or no percolation is observed, there will be a leachate mound in the layers above barrier soil liner, which should be collected from drainage layer and sent for treatment. Another important parameter observed is Field Capacity of waste, which has been simulated under three conditions A, B and C as mentioned earlier. The FC value of 0.292 vol./vol. and viii hydraulic conductivity (HC) of 0.001 cm/s of waste seems to best fit during regression analyses. The HELP model simulations results and sensitivity analyses have given a guideline for evaluation of operation and design of landfill in developing countries like Nepal. The major design considerations are the Field Capacity & hydraulic conductivity of waste and hydraulic conductivity of barrier soil liner for water balance in terms of controlling leachate generation. The estimates of the cumulative leachate volume were strongly dependent on the variation of the above parameters. The evapotranspiration component of the water balance have been underestimated, as it is dependent on solar radiation, vegetative growth, evaporative zone depth, wind speed, and relative humidity. The runoff has been considered NIL in this research and model simulations as it is a small-scale lysimeter. The runoff would also have been percolated in this small area of lysimeter even if it has been considered. This is one of the reasons that percolation is higher than evapotranspiration. The landfill cover specification for Nepal is about 60 cm and capping of 30 cm. The top cover used for lysimeter is also 30 cm in this research. The depth of cover soil and other layers do not seem to have much impact on the quantity of leachate produced. Another important consideration is the formation of cracks in the cover soil and development of wall effect and preferential pathways. The hydraulic conductivity as determined in the laboratory scale could not be achieved at the field and possibly the actual leachate (percolation) might have been overestimated. During most dry period, there is a high possibility of development of these cracks. Cracks can also develop due to poor workmanship during construction or low compaction. These will aggravate the preferential flow from sidewalls. The model also does not take into account of such cracks and fissures, and if occurs in large scale, the model result might be much under estimated than the actual percolation. Thus, this is one of the important design parameter. Simulating with variations of other parameters of soil and waste, the performance of the HELP model could be further validated using long-term measured data. In future research, study of leachate characteristics, qualitative evaluation and leachate treatment options could be focused. Laboratory and filed scale lysimeters, lysimeters with variations in waste and other material properties, recirculation and simulations and application in real landfills could also be focused in future research works. In summary, the HELP model has been considered as a good tool for evaluation of design and planning purpose and operation of landfills in developing countries like Nepal based upon the findings of this research.