Cesium-137 is an artificial radionuclide with a half-life of 30.2 years and strong gamma radiation that due to nuclear explosions in the 1950s and 1960s entered the Earth's surface by the rain. This element was rapidly absorbed to fine particles of soil surface horizon and its distribution is basically accompanied by the physical changes of soil such as erosion and plowing. The basis of calculating of erosion by radioactive-cesium method is based on a comparison between the amount of cesium-137 in soil exposed to erosion and a reference surface. The values attributed to the reference inventory of cesium-137 plays an important role in converting measurements to estimates of rates of soil erosion or sediment rate. Incorrect and uncertain values of Cs-137 in reference site will be caused to biased and unreliable estimates of soil erosion and sediment deposition. This study was conducted to evaluate the variability of the content of cesium-137 regarding to the sampling program including sample size, distance and sampling method. This study was conducted in a reference site located in vicinity of Fereydan city, located at western of Isfahan province. Throughout two network sampling (3×8) comprising big network (35 m length and 8 m width), and small network (24 m length and 6 m width), 96 soil samples were collected from 0-15 cm and 15-30 cm depths. Moreover, for evaluation of vertical distribution of Cs-137 in two networks two profiles were dug and soil samples were collected from 0-5, 5-15, 15-25, and 25-35 cm depths. Gamma-ray spectrometry device was used for Cesium-137 analysis. Magnetic susceptibility values were measured using Bartington susceptimeter at low (0.46 kHz) and high (4.6 kHz) frequencies and magnetic susceptibility dependent was calculated by low and high frequency values. Soil characteristics including pH, EC, texture, gravel, organic material, total nitrogen, cation exchange capacity, calcium carbonate equivalent, and bulk density of all the samples and the types of clay in the two soil samples were determined. To evaluate spatial variability of Cs in two networks, variography analysis was done using Variowin software. R software was employed for evaluating the sampling method (regular network, simple random, and transect) and selecting the number of samples for covering the variability of the content of cesium-137. The results of study showed that among the soil physicochemical and magnetic properties, bulk density had the highest correlation with cesium-137 (r =0.46**, p 0.01). A negative correlation was observed between the clay content and cesium-137 inventory (r=-0.52**). Clay mineralogy showed that clay minerals included illite, kaolinite and chlorite, were existed in the reference site and the results confirmed that nature of mineralogy is effective in absorption of cesium-137 in soil. The results of multiple regression analysis indicated that bulk density, clay content and magnetic susceptibility at low frequency explained 43% of total variability of Cs-137. No significant differences were observed for the values of sill and range of the two networks, and spherical models was fitted for two studied networks. The content of cesium-137 in the reference site showed relatively low variability (15%). The distance of sampling and the sampling method was not significantly on variability of the content of cesium-137 in the studied reference site. The existence of microtopography in the big network resulted lower coefficient of variation of cesium-137 compared to small network. The number of samples affected the variability of cesium-137 in the studied reference site. Also, as the variability of the cesium-137 content and the accuracy increased, the number of samples required for a reliable estimate of average cesium-137 activity increased. With permissible error of 10% and confidence levels of 90, 95 and 99%, respectively, 4, 6 and 11 samples in the small network, and 5, 7, and 13 samples in the big network are required. In overall, it is suggested that similar studies would be done in reference sites of western Iran to identify the relevant and reliable numbers in reference sites for Zagros zone. Keywords : Cesium-137, Variability, Reference site, Magnetic susceptibility