摘要：Core Ideas Estimation of K s using established PTFs for one specific field was evaluated. Selected PTFs exhibited unsatisfactory prediction of K s at the field scale. Taking macropore volume effect into account improved the prediction of K s with PTFs. Saturated hydraulic conductivity ( K s ) is one of the crucial hydraulic properties for assessing water and solute transport in soils. However, direct measurement of K s is time consuming and arduous. Alternatively, pedotransfer functions (PTFs) have been developed to estimate K s indirectly through more easily measurable soil properties that are part of regional, national, or international databases. These PTFs are usually based on datasets collected from large regions. However, their validity for a specific site remains unclear. The objectives of this study were to evaluate the performance of established PTFs in estimating K s in a specific field and improve PTFs to arrive at a locally adapted estimation result for K s . Forty‐one soil samples were collected from 10 locations at five depths at a farmland in western Kentucky for hydraulic conductivity and physical property measurements. The performance of seven PTFs in estimating K s was evaluated using the root mean square error (RMSE), Nash–Sutcliffe efficiency (NSE), and the coefficient of determination ( R 2 ). At this scale, all the selected PTFs exhibited unsatisfactory prediction of K s (high RMSE, low NSE and R 2 ). In the field studied, approximately 60% of variance in K s could be explained by soil texture and macropore components based on factor analysis. Clay content and macroporosity were identified as the most representative variables for each component. The performance of a PTF in estimating K s for the field site investigated was significantly improved by including macroporosity (pores with diameter >75 μm) as a predictor. The results confirmed that soil structure was crucial in characterizing soil hydraulic conductivity.
关键词：PCA; principal component analysis; PTF; pedotransfer function; RMSE; root mean square error; NSE; Nash–Sutcliffe efficiency
