Investigation of groundwater flow in karst areas using component separation of natural potential measurements

Abstract

The natural (electrical) potential (NP) method – also known as self-potential, spontaneous potential and streaming potential (SP) – has been used to locate areas of groundwater flow in karst terrane. NP is the naturally occurring voltage at the ground surface resulting from ambient electrical currents within the earth. The measurement of NP can be used to characterize groundwater flow in karst terrane because electrical potential gradients are generated by the horizontal flow of water along fractures or conduits and the vertical infiltration of water into fractures or shafts. NP data from a site on the Mitchell Plain of southern Indiana, USA, revealed that NP data can be decomposed into three components: topographic effect, residual NP and noise. At this site, NP was inversely proportional to elevation, but the correlation varied with time. The topographic correction factor varied from –2.5 to –1.2 mV/m (NP change per unit elevation increase), with an average linear correlation coefficient (R) of 0.95. Because the site slopes toward an adjacent creek that is the local groundwater discharge zone, one possible explanation for this effect is a streaming-potential mechanism generated by groundwater movement toward the creek. The residual NP data revealed three negative anomalies at the survey area. Two of them coincide with sinkholes. A part of the third anomaly is coincident with a small valley, and concentrated infiltration does occur at this elevation in other valleys at the site, as evidenced by the existence of sinkholes. However, the dispersed, low-magnitude nature of the third anomaly does not prove the existence of concentrated groundwater recharge activity.