Pumping test is the most reliable method for determining hydraulic conductivity and tracing test is the most reliable way for determining hydraulic connection. The pumping test consists of imposing a disturbance such as pumping in a well and measuring the response of the aquifer, for example in terms of head variations (Renard et al. 2009). These data are then interpreted using analytical or numerical models in order to deduce the hydraulic properties (hydraulic conductivity, transmissivity and storage coefficient) of the aquifer (Renard et al. 2009). Tracing tests are frequently used on a regional scale to delineate the difference between hydrogeological and topographic watersheds. Watersheds with very distinct and permanent boundaries are rare in karst. For this reason, tracer tests at different hydrological stages are strongly recommended (Milanović 2011). For regional tracer testing, the most commonly used dyes are fluorescein and rhodamine B, lycopodium spores, inorganic salts and radioactive salts (Milanović 2011). Longitudinal and transverse dispersivity as well as the ratio between hydraulic conductivity and effective porosity can be determined by tracer tests. The concurrent use of these two methods, first by Laguna (1970), reported that two jugate sandy aquifers were pumped simultaneously, and forced gradient tracer tests were used to determine what proportion of the pumped water came from of each sandy aquifer. Then, the results of the pumping and tracing tests were compared for hydraulic conductivity (Niemann and Rovey 2000; Rovey and Niemann 2005; Thorbjarnarson et al. 1998). In most studies, tracing tests are considered more accurate. because they would better implement the het...... middle of paper ......characteristics of the fractured aquifer system (Lee and Lee 1999). The geometric mean of hydraulic conductivity is 30.50 m/day for the pumping test and 106.339 m/day for the tracer test in the Salman Farsi Dam site. The geometric mean of the hydraulic conductivity based on the tracer test is greater than the geometric mean of the hydraulic conductivity based on the pumping test. Because, in this area, conduit paths are abundant (Reference ????) and in tracer tests, transmission is mainly from conduit paths. But, in pumping tests, transmission is from the conduit and matrix routes. At the Salman Farsi Dam site with many conduit routes, tracers were observed following high conductivity conduits and noted that using an average K value from an aquifer pump test to predict solute transport would result in considerable underestimation of transport distances for a given period.