Explore the vast range of titles available.

The size and spatial distribution of soil structural macropores impact the infiltration, percolation, and retention of soil water. Despite the assumption often made in hydrologic flux equations that these macropores are rigid, highly structured soils can respond quickly to moisture variability‐induced shrink‐swell processes altering the size distribution of these pores. In this study, we use a high‐resolution (180 μ${\\upmu }$ m) laser imaging technique to measure the average width of interpedal, planar macropores from intact cross sections and relate it to matrix water content. We also develop an expression for unsaturated hydraulic conductivity (K)$(K)$that accounts for dynamic macropore geometries and propose a method for partitioning sensor soil water content data into matrix and macropore water contents. The model was applied to a soil in northeastern Kansas where soil monoliths had been imaged to quantify macropore properties and continuous water content data were collected at three depths. Model‐predicted macropore width showed significant sensitivity to matrix water content resulting in changes of ∼${\\sim} $ 15%–50% of maximum width over the 15‐month period of record. Transient saturated hydraulic conductivity predicted from the model compared favorably to a previously developed model accounting for moisture‐induced changes to structural unit porosity. Following periods of low soil moisture, infiltrating meteoric water filled highly conductive macropores increasing K$K$by several orders of magnitude which subsequently decreased as water was absorbed into the matrix and macropores drained. This model offers a means by which to combine measurable morphological data with soil moisture sensors to monitor dynamic hydraulic properties of soils susceptible to shrink‐swell processes.

Many soils and other porous media exhibit dual- or multi-porosity type features. In a previous study (Seki et al., 2022) we presented multimodal water retention and closed-form hydraulic conductivity equations for such media. The objective of this study is to show that the proposed equations are practically useful. Specifically, dual-BC (Brooks and Corey)-CH (common head) (DBC), dual-VG (van Genuchten)-CH (DVC), and KO (Kosugi) BC -CH (KBC) models were evaluated for a broad range of soil types. The three models showed good agreement with measured water retention and hydraulic conductivity data over a wide range of pressure heads. Results were obtained by first optimizing water retention parameters and then optimizing the saturated hydraulic conductivity ( ) and two parameters ( , ) or ( , ) in the general hydraulic conductivity equation. Although conventionally the tortuosity factor is optimized and ( , ) fixed, sensitivity analyses showed that optimization of two parameters ( + , ) is required for the multimodal models. For 20 soils from the UNSODA database, the average for log (hydraulic conductivity) was highest (0.985) for the KBC model with = 1 and optimization of ( , , ). This result was almost equivalent (0.973) to the DVC model with = 1 and optimization of ( , , ); both were higher than for the widely used Peters model (0.956) when optimizing ( , , , ω). The proposed equations are useful for practical applications while mathematically being simple and consistent.

Recent advances in machine learning (ML) techniques show promise for estimating soil hydraulic properties from soil data sets. Pedo‐transfer functions (PTFs) can facilitate the mapping of the complex relationship between soil properties and soil hydraulic properties, for example, lateral hydraulic conductivity—a necessity for estimating lateral subsurface flow in distributed hydrological models. In wflow_sbm model, the horizontal‐to‐vertical saturated hydraulic conductivity ratio fKh0$\\left({f}_{\\text{Kh0}}\\right)$is a sensitive parameter, but no established PTF exists. Our objective is to investigate the potential of ML algorithms in estimating PTFs for fKh0${f}_{\\text{Kh0}}$prediction. In this study, publicly available calibrated fKh0${f}_{\\text{Kh0}}$(i.e., optimized) across Great Britain were utilized to train two ML algorithms: Random Forest (RF) and Boosted Regression Trees (BRT), employing SoilGrids data set. Both algorithms effectively predicted fKh0${f}_{\\text{Kh0}}$in 92 of the 115 tested sub‐basins (i.e., 80%), demonstrating a high correlation with the optimized values, with RF slightly outperforming BRT. As a next step, we compared wflow_sbm simulated discharge results using uncalibrated fKh0${f}_{\\text{Kh0}}$(default value) and our predicted values. The predictions notably improved discharge simulations, with a median Kling‐Gupta Efficiency (KGE) increasing from 0.55 to 0.75. Subsequently, we generated two globally distributed fKh0${f}_{\\text{Kh0}}$maps to investigate the transferability of the ML‐based PTFs in the Loire basin, France. ML‐based PTFs improved performance in 75% of sub‐basins, with an average KGE increase of 0.06. Finally, we assessed the uncertainty in fKh0${f}_{\\text{Kh0}}$predictions, confirming the robustness of the ML‐based PTFs. Our study highlights the potential of ML methods for estimating soil hydraulic properties, aiding parameter estimation for distributed hydrological models.

Extreme rainfall events pose a severe challenge to soil and water conservation, even in areas with high vegetation cover on the Loess Plateau. In this study, the artificial extreme rainfalls with cumulative rainfall of 270 mm and intensity of 60 mm · hr−1 were conducted on in‐situ experimental plots (20 × 2.5 m) on a loess gully–slope with gradients of 35°–40° that were treated with different grass coverage: (0%, 30%–40%, 70%–80%, >90%). The ephemeral gully/rill and shallow landslide occurred in plots were analyzed. Revegetation changed the erosion type on gully–slope, reducing gully erosion but promoting shallow landslide due to the change from infiltration–excess runoff to saturation–excess runoff. Under grass coverage of >90%, over 95% of rainfall seeped into the soil, and subsurface flow was generated due to the lower saturated hydraulic conductivity of underlying soil, which increased the possibility of landslides. The average erosion rate (0.36–3.29 g · m−2 min−1; no obvious erosion) in plots with 70%–80% coverage was 95.5% lower than that in bare land plots (27.8–47.5 g · m−2 min−1; ephemeral gully erosion), while due to landslides the average erosion rate in plots with >90% coverage (135.1–184.3 g · m−2 min−1) was 86.5 times higher than that in plots with 70%–80%. For grass, a coverage of 70%–80% was most effective in preventing soil erosion, controlling gully erosion and preventing landslides under extreme rainfall. These results deepen the understanding of the complex relationship between vegetation, gully erosion, and landslides in ecologically sensitive areas. Key Points Vegetation changed the erosion type on slope from water erosion to gravity erosion High‐coverage vegetation promoted shallow landslides under extreme rainfall For grass cover, a coverage of 70%–80% was most effective in preventing soil erosion

Exact solutions for one‐dimensional steady‐state and transient liquid flow toward a water table in heterogeneous unsaturated soils are critical in predicting saturation profiles in several real‐world applications including interpretation of climate change effects on the subsurface and impacts on slope stability. In this study, vertical heterogeneity in saturated hydraulic conductivity with depth is characterized by an exponential decay function. A steady‐state solution is derived based on Darcy's law and the water table depth, and two transient‐state solutions are obtained from Richards' equation using the Laplace transform and the modified Bessel equation under common upper boundary conditions, that is, flow rate and pressure head, following the initial steady‐state flow condition and a water table at a specified depth. The transient and steady‐state solutions are compared with numerical solutions obtained from a multi‐layered configuration and an analytical solution for homogeneous soils, demonstrating their reliability and efficacy. Various hypothetical heterogeneous soils with differing parameters are employed to assess the flow behaviors, illustrating that vertical heterogeneity impacts the pressure head profiles. More pronounced heterogeneous soils exhibit lower pressure head and effective saturation during flow compared to homogeneous soils, which depend on the air‐entry value and the water table depth, irrespective of the upper boundary conditions. The solution has been employed to predict volumetric water content profiles in the Loess Plateau of China and evaluate the effect of infiltration on the stability of a shallow landslide in Japan based on the infinite slope model and the suction stress concept.

• Amazonian droughts are increasing in frequency and severity. However, little is known about how this may influence species-specific vulnerability to drought across different ecosystem types. • We measured 16 functional traits for 16 congeneric species from six families and eight genera restricted to floodplain, swamp, white-sand or plateau forests of Central Amazonia. We investigated whether habitat distributions can be explained by species hydraulic strategies, and if habitat specialists differ in their vulnerability to embolism that would make water transport difficult during drought periods. • We found strong functional differences among species. Nonflooded species had higher wood specific gravity and lower stomatal density, whereas flooded species had wider vessels, and higher leaf and xylem hydraulic conductivity. The P50 values (water potential at 50% loss of hydraulic conductivity) of nonflooded species were significantly more negative than flooded species. However, we found no differences in hydraulic safety margin among species, suggesting that all trees may be equally likely to experience hydraulic failure during severe droughts. • Water availability imposes a strong selection leading to differentiation of plant hydraulic strategies among species and may underlie patterns of adaptive radiation in many tropical tree genera. Our results have important implications for modeling species distribution and resilience under future climate scenarios.

Bioclogging in porous media is common and affects many engineering projects. The temperature of recharge water could significantly affect the process of bioclogging, thus impacting the hydraulic conductivity of porous media. In this study, a series of laboratory percolation experiments was conducted to understand the effects of recharge water temperature. The results of these experiments showed that bioclogging evolved in phases, and the gradual reduction (attenuation) of hydraulic conductivity caused by bioclogging could be described by an inverse logistic model. Analysis of microbial growth suggested that the bioclogging phases were strongly correlated with microbial growth stages. Both the clogging rate and degree of clogging through the seepage column decreased with distance from the inlet. Within the range of 10–25 ℃, the degree of clogging decreased with the increasing recharge water temperature; however, the degree of clogging increased with recharge water temperature within the range of 25–35 ℃. The relative hydraulic conductivity values decreased by 86.9% at a recharge water temperature of 10 ℃, 76.0% at 15 ℃, 65.1% at 20 ℃, 44.9% at 25 ℃, 82.5% at 30 ℃ and 98.7% at 35 ℃. Investigation by scanning electron microscopy found that the microorganism micromorphology differed at different recharge water temperatures, which made a significant difference in terms of clogging degree. A comprehensive model that describes hydraulic conductivity attenuation with varying recharge water temperature has been developed.

Lime treatment of loess in foundation engineering modifies the soil structure, leading to changes in mechanical and hydraulic properties of soil, which in turn will affect the flow of water and transport of contaminants in the loess. In light of this, it is essential to identify the dominant effects of different lime treatments on hydraulic conductivity, and to ascertain the optimum lime treatment. For this purpose, we investigated the effects of dry density and lime content on changes in hydraulic conductivity and microstructure of loess in Yan’an City, China. The results indicate that hydraulic conductivity has a log negative correlation with dry density, and lime addition can result in a decrease of hydraulic conductivity of loess at the same dry density. Under a given degree of compaction, however, lime addition can lead to a decrease in dry density due to an increase in flocculation and aggregations. The significant decrease of dry density leads to an increase in hydraulic conductivity when lime content (in mass percentage) is lower than 3%. Nevertheless, when lime content is higher than 3%, the reactions between loess particles and lime will be intensified with an increase in lime content, and become the primary factors affecting pore characteristics. These reactions can further decrease the hydraulic conductivity of lime-treated loess, and the lowest hydraulic conductivity was obtained for lime-treated loess with 9% lime content. The excess lime (above 9% lime content) dramatically increased pore size, leading to a significant increase in hydraulic conductivity. Therefore, 9% is the optimum lime content for loess treatment, and the degree of compaction in engineering should be higher than 95%. In addition, statistical analysis of microstructure of lime-treated loess shows that the distribution trends of macro- and meso-pores coincided with that of saturated hydraulic conductivity, which indicates that lime content affects saturated hydraulic conductivity of lime-treated loess by changing the soil structure, especially the properties of pores larger than 8 µm.

The water retention and hydraulic conductivity curves of unsaturated soils are important parameters for seepage analysis. Experimental results in the literature generally show that with increasing density, the air‐entry value and adsorption/desorption rate of the water retention curve increase and the relative hydraulic conductivity (kr) at a given degree of saturation changes. The above phenomena, except the density‐dependency of air‐entry value, have not been considered in existing models. This study aims to address these problems by developing new hydraulic models based on experimental evidence from microscopic analysis. First of all, a new equation was proposed to model the evolution of pore size distribution with soil density. For a given pore, the ratio of its initial to final sizes is higher when the initial size is larger and when there is a greater increase in density. Based on this equation, a new and simple water retention equation was derived to predict the increase in air‐entry value (resulting from the reduction in pore size) and the adsorption/desorption rate (due to a more uniform pore size distribution) as density increases. Then, a new equation for kr was developed by incorporating the evolution of pore size distribution and tortuosity upon soil deformation, and therefore it can capture the changes of kr. To validate the above equations, test data from several soils with distinct properties were used. The measured and calculated results are well‐matched. Key Points A new equation was proposed to model the evolution of pore size distribution with soil density The water retention model captures the increase of not only air‐entry value but also adsorption/desorption rate with increasing density The hydraulic conductivity model considers the evolution of pore size distribution and tortuosity upon soil deformation

Machine learning (ML) has been extensively applied in various disciplines. However, not much attention has been paid to data heterogeneity in databases and number of samples used to train ML models in hydrology. In this study, we addressed these issues and their impacts on the accuracy and reliability of ML models in the estimation of saturated hydraulic conductivity, Ks. We selected 17,990 soil samples from the USKSAT database and created random subsets N = 2,000, 4,000, 6,000, 8,000, 10,000, 12,000, 14,000, 16,000, and 17,990, 80% of which were used for training. The random subset selection was repeated 50 times. The extreme gradient boosting (XGBoost) algorithm was used to estimate Ks from other soil properties, such as bulk density, soil depth, texture, and organic content. For each subset, we conducted the learning curve analysis on the training and cross‐validation data sets. Results showed that for all training sample sizes the number of samples was not enough for the training and cross‐validation curves to reach a plateau. We also applied the concept of representative elementary volume by plotting the average coefficient of determination, R2, and root mean square log‐transformed error, RMSLE, against the training sample size. For the testing data set, as the number of training sample size increased from 1,600 to 14,392 the average R2 value increased from 0.74 to 0.90, while the average RMSLE value decreased from 1.08 to 0.69. Either the learning curve or representative sample size analysis is required to investigate whether the number of samples is enough or not. Key Points Learning curves were applied to address effects of data heterogeneity and number of samples on machine learning‐based model estimations Concept of representative elementary volume was used to determine the representative sample size in machine learning The number of samples was not enough for the training and cross‐validation curves to reach a plateau