Hydraulic conductivity curves (HCCs) are important parameters in land surface modeling. The general way for predicting HCC from soil water retention curve (SWRC) requires an additional input of the saturated hydraulic conductivity. The time-consuming in measurement and more importantly, the macro-effect near saturation, however, often result in difficulty and poor performance in predicting the conductivity. In this study, we provided a physically based method for predicting the HCC fully from SWRC requiring no additional parameters. This is achieved by applying an estimated conductivity (from SWRC) in the dry range as new matching point, in together with modifying the HCC model developed by Wang et al. (2018) that accounts for both capillarity and adsorption forces. Testing with a total of 159 soil samples yielded that the new model significantly improved the predictions of HCC, with R2 being 0.74 and root mean value being 0.84 cm d−1, nearly double and half of the value predicted with the input of the saturated hydraulic conductivity, respectively. The abrupt drop near saturation of the HCC model that provided by Wang et al. (2018) for soils with small n values close to 1, a parameter in shaping the SWRC, was also overcome by introducing a non-zero air-entry value.

The significant impact of soil structure on soil hydraulic properties and then on the associated water and solute transport is well recognized. However, existing soil hydraulic models that account for the effect of soil structure are often at the cost of overparameterization, hindering further application on large scales. In this study, we developed a new model that considers the effect of soil structure in hydraulic conductivity prediction when introducing no new free parameters. Testing with 152 soil samples that include different soil types shows that the new model considerably improves the prediction of conductivity, with an average root-mean-square-value (RMSE) of 0.65 cm d−1. When applying the new model in fitting observations, the model showed excellent performance, with an RMSE of only 0.30 cm d−1 remaining. This new soil hydraulic model provides a simple and practical way to incorporate the influence of soil structure in water and solute transport simulations.

The role of groundwater in maintaining streamflow in the alpine area with distribution of permafrost and seasonal frost is a poorly studied topic of considerable interest. The stream and groundwater interactions and groundwater contributions to the Heihe River were investigated during this study in a representative subcatchment in the headwater region of the Heihe River Basin, the northeastern Qinghai-Tibet Plateau of China. The hydraulic, chemical and isotopic data as well as Bayesian mixing model results show that groundwater-stream water interactions were both spatially and temporally variable. The tributaries were primarily recharged by springs within the permafrost zone during the frozen period when the water source and sediments were frozen and the groundwater discharged to the mainstream within the seasonal frost zone to maintain the streamflow. The groundwater contribution to mainstream discharge decreased from 95% during the frozen period to 80-90% in the thawing period due to the inflow from tributaries. However, the stream and groundwater interactions vary several times along altitude during the thawed period from June to early September, due to the increased glacial/snow meltwater volume, deepened active layer, and melted seasonal frost. Groundwater contribution decreased to ~40-60% of the mainstream discharge during the thawed period because tributary streams contribution largely increased. As shown by ~70-90% contribution from groundwater to the mainstream discharge, the mainstream flow mainly sourced from the release of groundwater in aquifers in the freeze-back period. These data indicate that the variations in groundwater-surface water interactions were largely influenced by the distribution and freeze-thaw cycle of permafrost and seasonal frost. The importance of groundwater storage in maintaining streamflow in the Heihe headwater region was highlighted by this study.

The commonly applied pedotransfer functions (PTFs), which predict soil hydraulic properties (SHPs) from easily measured soil properties such as texture information, often account only for capillary forces. Recent advances in soil hydraulic modeling suggest that, to improve the prediction of SHPs under dry conditions, the impact of adsorption forces has to be taken into account. However, the lack of observations in particularly dry conditions, due to the difficult and time-consuming measurement, hinders the development of PTFs that predict SHPs from saturation to oven dryness. In this paper, we first present a simple method for predicting complete SHPs with limited measurements that cover only a relatively high matric potential range. With this method, we extended a public dataset to cover dry conditions, and then applied it to develop PTFs that can predict SHPs from saturation to oven dryness. This was achieved by applying the complete soil hydraulic model proposed by Wang et al. (2021), which accounts for both capillary and adsorptions forces and overcomes the unrealistic decrease near saturation for fine-textured soils. The impact of vapor diffusion was also considered. We further applied this method in extending an existing capillary-based PTF to dry conditions. The results showed that: 1) the proposed method performs very well in describing SHPs over the entire moisture range; 2) the PTFs developed with the extended observations and the complete model show a superior prediction performance, especially for the hydraulic conductivity; and 3) the extended capillary-based PTF improves the performance in describing SHPs under dry conditions.