Abstract: Soil hydraulic properties play an important role in modelingwater and solute movement within the vadose zone of soils. Direct measurement of hydraulic parameters at a wide range of scales involves considerable time, labor, and money. Pedotransfer functions (PTF) may provide an alternative way of estimating these parameters indirectly from easy-to-measure soil properties. The Loess Plateau of China lacks large databases of hydraulic parameters and also the PTF that could determine them accurately enough for scientists and policy makers to address many of the region’s related problems, such as severe soil erosion. In this study, new PTF for saturated hydraulic conductivity (Ks), field capacity, and saturated soil-water content were developed. Multiple linear regression was used to analyze 252 data sets of the hydraulic and basic soil properties, as well as altitude, to derive the PTF. A further 130 data sets were used for validation. The predictive capabilities of the PTF were the best for saturated soil-water content (Radj2 = 0.78) and least for log Ks (Radj2 = 0.36). Bulk density, soil organic carbon, and soil particle composition were identified as significant input variables for the PTF. The inclusion of a topographic factor (altitude) significantly improved the predictive capability of the PTF for log Ks. Compared with established PTF, the PTF developed in this study predicted the hydraulic parameters more accurately as indicated by higher R2 and lower RMSE values when predicted, and measured parameter values were compared, and the greatest improvement was obtained for log Ks. The new PTF are the first set of PTF based on data from the Loess Plateau. Their better performance makes them applicable for a variety of purposes in the Plateau region and possibly in other loess regions around the world.