Representative sampling of landfills: a robust procedure for selecting trench locations based on electromagnetic induction survey data
Abstract
In densely populated areas, (re)development projects often encounter subsurface structures of previous urban land use, such as waste deposits. Mobile surveys with multi-receiver frequency-domain electromagnetic induction (FDEM) have been proven efficient tools for the investigation of landfills in support of site remediation, including an evaluation of possible resource recovery through landfill mining. The derived high-resolution maps of subsurface variations in electrical conductivity and magnetic susceptibility allow for accurate detection of the (lateral) landfill boundaries and useful qualitative interpretation with respect to the nature of the landfilled waste. However, FDEM alone - as any other geophysical exploration method - cannot provide a unique solution for the multivariate distribution of subsurface properties, including both widely diverse waste material characteristics and strongly variable landfill conditions (e.g. moisture content). Initial interpretations require validation with information from conventional investigations such as borehole drillings and trench excavations. For landfill characterization, the latter is usually preferred to guarantee sufficient volume for sample analysis and a representative support for interpretation.
We present a modification of the conditioned Latin Hypercube sampling (cLHS) strategy to determine locations for landfill trenching based on high-resolution survey data collected with a multi-receiver FDEM sensor. Trench excavation areas are selected providing maximum spread over the range of the apparent electrical conductivity (ECa) measurement signals and the lateral extent of the landfill, with the extra condition of minimum overlap in the ECa range covered by different trenches. The procedure was tested on a municipal solid waste landfill constructed in a former gravel quarry in Flanders, Belgium. Afterwards, trench excavations were conducted on the selected locations. Trench profile descriptions and layered waste sampling and analysis (including extract conductivity) allowed to verify the representativeness of the designed trenching configuration, with respect to achieving an effective coverage of the present variation in types of waste. We conclude that the modified cLHS-based procedure provides a robust statistical tool to select trench excavation locations using exhaustive FDEM data and the thus obtained set of trench samples effectively summarizes the spatial variability in waste composition and landfill conditions.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMNS24A..06V
- Keywords:
-
- 0439 Ecosystems;
- structure and dynamics;
- BIOGEOSCIENCES;
- 1625 Geomorphology and weathering;
- GLOBAL CHANGE;
- 1829 Groundwater hydrology;
- HYDROLOGY;
- 1835 Hydrogeophysics;
- HYDROLOGY