2.1 Site description
Alto Paraíso de Goiás is the main city within the Chapada dos Veadeiros National Park which still disposes of solid waste in the open (dump). The APGD is located approximately 6 km from the urban area, on the left bank of the BR010 highway (Figure 1). The collection of solid waste in the municipality and consequent deposition in the APGD began in 1993 (Oliveira Jr 2017). The solid urban waste was disposed of in the open until 2016. After that, the city authorized the excavation of ditches and the burial of residues without any type of waterproofing. This increases the flow of leachate into the ground, which also increases the risk of pollution as compared to the surficial through of wastes (Figure 2).
The national solid waste policy (LEI No. 12,305), sanctioned (on 2nd of August, 2010), that the dumps be extinguished and replaced by landfills within four years after the publication of the Law. However, after expiration the Brazilian government continues to extend this limit. The estimated production of solid waste in the municipality is 2.4 tons per day (Pfeiffer et al. 2017). If superficial or underground leachate flows out of the APGD area, the drainage to be compromised is the São Bartolomeu River that passes through the eastern part of the city (Figure 1).
The APGD is situated on the PP4ts2 lithofacies of the Arraias Formation (Campos et al. 2013; Figures 3A and 3B). The authors individualized these lithofacies according to the presence of micaceous quartzites with intercalations of metasiltites. Quartzites are characteristic of an interlaced fluvial environment, since they are medium to thick quartzites, have a low degree of selection, and presence of cross-tabular and channeled stratifications (Alvarenga et al. 2007).
Few boreholes were drilled for the installation of water level monitoring wells indicating the presence of a quartzarenic neossol (medium quartz sand) with a thickness of up to 7.5 m (Figure 3C). This soil type in Chapada dos Veadeiros region has an average hydraulic transmissivity of 10-5 to 10-6 m / s (Almeida et al. 2006). According to NBR 13896 (non-hazardous waste landfills design criteria, implementation, and operation), the implantation of areas for the disposal of solid waste is restricted to places with soils or geotechnical structures that have hydraulic conductivity less than 5x10-7 m / s (ABNT 1997).
The photographs along with the field visits are used to divide the APGD into different zones such as trash piles, trenches, sewage lagoons, solid waste trenches, and the proximity to houses and civil engineering structures as roads (Figure 2). These marked features aid the planning of the geophysical surveys in the later acquisition stage.
Thus, due to the geological, pedological, and hydrogeological characteristics of the APGD, it has a high potential for the development of a contamination plume. In the research of subsoil and groundwater contamination, the Brazilian Association of Technical Standards (ABNT) recommends the use of geoelectric methods (NBR, 15935). These consist of electrical resistivity, induced polarization, spontaneous potential, and GPR, the latter being used for shallow investigations. The shallow term is generic since the depth of investigation of the GPR depends mainly on the electrical properties of the medium (electrical conductivity and dielectric permittivity).
2.2 Data acquisition and processing
In order to verify the efficiency of the GPR method in the mapping of the soil thickness, buried residues and leachate percolation zones, the GPR profiles were taken at, peripheries and adjoining areas of APGD (Figure 4A).
Data were acquired using SIR-3000 equipment (manufactured by Geophysical Survey Systems Inc.) consisting of an acquisition module connected to a 200 MHz shielded antenna (Figures 4B, 4C, 4D, and 4E). The GPR configuration parameters were: time window of 400 nanoseconds (ns), time sampling interval (Δt) of 0.195 ns, spatial sampling interval (Δx) of 0.05 meter, 2048 samples per trace, and sampling frequency of 2550 MHz. The choice of the best GPR data processing routine to identify contaminated areas must take into account the maintenance of the recorded amplitudes since the application of gains or normalization filters can hide the electromagnetic signal attenuation zones, and consequently induce errors in interpretation. Thus, for this work, it was decided only to adjust the zero time, remove the gain applied during data acquisition (header gain), and apply a constant 1D linear gain. All routines for processing, visualizing, and generating 2D sections occurred in the ReflexW software (Sandmeier 2019).
The data were acquired along with the eleven GPR profiles (Figure 4). These profiles can be divided into three classes: (i) on dumpsite (L3, L6, L7, L8, and L10), (ii) its peripheries (L1, L5, L9, L11) and (iii) the adjoining areas (L2, L4). This dense coverage is deployed to achieve the research objectives of leachate detection as well as tracking its mobility in the surrounding areas.
To obtain the medium velocities on the APGD, information on the depths of the rocky top inferred from the monitoring wells was used (Figure 5). The intersection of the GPR L5 section and the PM02 well (50 m section position), shows a strong section reflector at 137 ns (Figure 5B). Similarly, at PM01 (position 300 m from the section), there is also a strong reflector at 38 ns (Figure 5C). Thus, the speeds of the GPR wave in the medium were calculated from the travel times recorded at TPM01 and TPM02 as 0.102 m/ns and 0.105 m/ns, respectively. Thus, an average speed value of 0.10 m/ns was used for time to depth conversion.