M. Járó - L. Költő szerk.: Archaeometrical research in Hungary (Budapest, 1988)

Prospecting - SŐRÉS László: Geophysical measurements at the site of a Roman homestead at Balácapuszta

Um vAA/ o f <•«*•» o in«»* o i «o * » JUL VW RBI 0 S • * X 6 s « « io *» .AAAA/ Ai S 0 §«*»•« 0 J « « *> •» 0 J « * » • 1- •#! R E fi B 0H0H ft WALLS DISTANCE 1 2 4tn 2 3 3 m 3 4 i m 4 4 2 n 5 2 6 m 6 00 3 m 7 oo 4m 8 9 NOISE Fig. 4. Residual filter analysis p: theoretical resistivity profiles calculated by convolution above several models; f: profiles after smoothing and residual filterings; a: position of walls along the profiles. The effects of the walls being different distances from each other are shown on the theoretical profiles after filtering. then the filter can separate their peaks reliably. With a wall distance of 4 m (curve 7), in the case of an infinite wall series, a resonance phenomenon appears. The distance of the effects is just equivalent to the wavelength of the filter, thus, the amplitude of the response function increases to 20 times as much, despite the value of the change in resistivity not being significant. This phenomenon may appear over a long building which consists of a series of 4 m long rooms. Fig.4 also shows the effect of the already mentioned surface disturbance (curve 9). As can be seen, the extent of the noise compared to the effects of the walls is negligible on the filtered map. Finally, it must be noted that the result of these calculations is only of an informative character since we applied a 7—element filter instead of infinite ones. Moreover, it is also questionable with what accuracy the convolution procedure produces the resisitivity profiles of the models.