*3.3. Experimental Surveys for the Trial Pit*

Figure 15 presents the GPR B-scans collected around the trial pit. As can be seen in Figure 7a, several rectangular tiles formed a joint at one of the pit's sides. Therefore, during the acquisition of the C-1 scan (Figure 15a), the GPR antenna was moved along the joint, thus the bottom faces of the tiles were not clearly imaged. However, the analysis of the upper part of the scan allowed identifying two different rows of tiles; the typical tile (43 cm × 43 cm) row was replaced at the edge of the trial pit with a narrower one (43 cm × 10 cm) laid with the overlap (cf. Figure 2a). The overlap length was about 0.16 m and it could be seen the clearest at the distance of 0.42–0.58 m. The ground under the C-1 scan had many inclusions, resulting in numerous irregularly distributed reflections (cf. the concentrated inclusions in Figure 14c,d). The C-2 scan (Figure 15b) clearly presents the line patterns being the reflections from the joints. The bottom faces of the tiles were identified at dissimilar levels, therefore the tiles (or the mortar layers below them) had different thicknesses. The hyperbola located at the distance of 0.7 m and the depth of 0.18 m denoted the presence of a metallic (steel or aluminum) pipe (cf. Figure 2). The ground on the right side of the scan (around the pipe, at the distance of 0.5–1.0 m) seemed less heterogeneous than the remaining part. This might be caused by the fact that the original ground was removed and replaced with another, more homogeneous one during placing the pipe under the floor. The C-3 scan (Figure 15c) shows the clear image of two tiles, one with the regular bottom face and the second with an unbalanced shape. The ground seemed to be original because it is highly heterogeneous, such as in the C-1 scan. The C-4 scan (Figure 15d) also clearly shows two typical tiles together with the narrow one (located at the distance of 0.88–0.98 m). The shape of the bottom face of all tiles was irregular. The shift was observed in the scan (at the distance of 0.85 m), caused by the slip of the antenna at the joint between the standard and the narrow tile (cf. Figure 2d). The reflection from the pipe was visible at the distance of 0.25 m and the depth of 0.18 m (the same as observed in the C-2 scan). The ground around the pipe was less inhomogeneous (the distance of 0.0–0.5 m) compared with the remaining part. Generally speaking, it was difficult to say whether there were any air gaps under the floor. The subtle difference between the image of the tiles with and without the air gap (observed in the numerical results) was here blurred by the ground inhomogeneities and a signal noise.

**Figure 14.** Numerical GPR B-scans for the floor: (**a**) model #1.1; (**b**) model #1.2; (**c**) model #1.3; (**d**) model #1.4.

**Figure 15.** GPR B-scans for the trial pit: (**a**) edge 1 (trace C-1); (**b**) edge 2 (trace C-2); (**c**) edge 3 (trace C-3); (**d**) edge 4 (trace C-4).

Figure 16 contains the UT B-scans corresponding to the above-discussed GPR B-scans. The C-1 scan (Figure 16a) was performed through the narrow tiles, as can be seen in Figure 7a, thus the joints of the standard tiles were not observed. In the top part of the scan (at the depth of about 0.03 m), there was a straight line denoting the regular bottom face of the narrow tiles. However, there was no regularity in the lower part of the scan, where many multiple reflections occurred. Compared with the results for the concrete slabs, it might be stated that the air gap with an irregular shape was present in this area. The non-uniform settlement of the original ground probably led to the appearance of the air gap. The C-2 scan (Figure 16b) shows the image of two different tiles. The right tile, unlike the left one, was clearly visible at the top of the scan; it was possibly removed and then placed again with the use of a different (stronger) mortar. The multiple reflections appeared under the left tile, indicating the presence of the air gap (this tile laid on the original ground). On the contrary, there were no reflections under the right tile. The pipe detected in the GPR scans was not visible, which stays in agreement with the results for the concrete slabs where the UT scans did not reveal the table tennis ball. This observation allowed concluding that concentrated inclusions could not be detected by the UT scanning. It also should be mentioned that, according to the UT antenna instruction, the diameter of the pipe did not exceed 30 mm. In the C-3 scan (Figure 16c), two tiles were also visible. The left one was much more pronounced: it could have been replaced during some renovation works. Additionally, there were multiple reflections under both tiles, indicating the presence of the air gaps (resulting from

the ground settlement). However, it needs to be noted that the reflections were less intensive under the right tile because it was weakly identified itself. The possibility of detecting the tiles in the C-4 scan (Figure 16d) was also distinguished: the left one was much better imaged. Although the ground below the left tile was replaced, the multiple reflections indicating the presence of an air gap were visible under both tiles. Like in the C-3 scan, the reflections under the right tile were weaker because it was weakly imaged itself. It also needs to be added that the pipe was not observed like in the C-2 scan.

**Figure 16.** Ultrasonic B-scans for the trial pit: (**a**) edge 1 (trace C-1); (**b**) edge 2 (trace C-2); (**c**) edge 3 (trace C-3); (**d**) edge 4 (trace C-4).
