Model Of Water Movement Research Articles

Effects of agricultural practices on hydraulic properties and water movement in soils in Brittany (France)

The intensive agricultural use of soils in the Brittany region (western France) has increased the need for a better understanding of soil water dynamics. The aim of the present study is to compare quantitatively the differences produced by two agricultural practices on soil hydraulic properties (water retention curve and hydraulic conductivity) as well as the infiltration and drainage fluxes in the soils. This study was carried out on two experimental plots managed in the same way for 22 years. The two practices were continuous maize fertilized with mineral fertilizer, denoted as MX, and pasture within a ray-grass/maize rotation (3/1 year) with organic fertilization (pig slurry), denoted as PR. The study consisted of measuring soil physical properties in the laboratory and in the field, and estimating water infiltration in the soil of the two plots by recording water pressure heads after simulation of 2-h artificial rainfall with an intensity of 17 mm/h. We applied the van Genuchten model to describe the water retention and hydraulic conductivity curves ( θ( h) and K( h)) for each soil horizon of the two plots. Hydrus-2D and ID softwares were used to construct a numerical model of water movement in the two soils. This model was used to quantify the infiltration rate, deep drainage and actual evaporation fluxes during the artificial rainfall experiment. The vertical influence of agricultural practices in both plots appears to be limited to the uppermost 35 cm. Deeper in the B horizon, there are only very slight differences in the hydraulic properties between the two plots. In the top soil horizons (H1–H5 and H6), the two soil properties mostly affected by practices are the hydraulic conductivity and the α parameter of the van Genuchten model. At the lowest pressure head studied here (−1.5 kPa), hydraulic conductivity in a given horizon differs by more than one order of magnitude between the two plots. The model reproduces quite satisfactorily the observed pressure heads in plot PR at all depths, in the rainy period as well as in the water redistribution period (efficiency >0.77). Results are less good for the MX plot, with efficiency ranging from 0.49 to 0.84 depending on the horizon. The different sources of simulation errors are identified and discussed. For the MX plot, the soil water movement model succeeds in reproducing the infiltration excess runoff observed in the field, allowing us to calculate that it accounts for 9% of the applied rainfall. No surface runoff or ponding appears in the PR plot during the artificial rainfall experiment. In the PR plot, the simulated deep drainage flux increases more rapidly than in the MX plot. The lower hydraulic conductivity in the top soil horizon of the MX plot compared with the PR plot appears to reduce the infiltration rate as well as the deep drainage flux. It also decreases the upward flow of water to the soil surface when the water content in the top soil layer is depleted by evaporation flux. The model simulation could be improved by a more precise representation of the soil structure, particularly the location, size and frequency of clods as well as the variability of hydraulic properties. However, we need to strike a balance between improving the quality of the simulation even further and the practical constraints and efforts involved in measuring the soil hydraulic properties.

Soil hydraulic properties near saturation, an improved conductivity model

The hydraulic properties near saturation can change dramatically due to the presence of macropores that are usually difficult to handle in traditional pore size models. The purpose of this study is to establish a data set on hydraulic conductivity near saturation, test the predictive capability of commonly used hydraulic conductivity models and give suggestions for improved models. Water retention and near saturated and saturated hydraulic conductivity were measured for a variety of 81 top and subsoils. The hydraulic conductivity models by van Genuchten [ van Genuchten, 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892–898.] (vGM) and Brooks and Corey, modified by Jarvis [ Jarvis, 1991. MACRO—A Model of Water Movement and Solute Transport in Macroporous Soils. Swedish University of Agricultural Sciences. Department of Soil Sciences. Reports and Dissertations 9.] were optimised to describe the unsaturated hydraulic conductivity in the range measured. Different optimisation procedures were tested. Using the measured saturated hydraulic conductivity in the vGM model tends to overestimate the unsaturated hydraulic conductivity. Optimising a matching factor ( k 0) improved the fit considerably whereas optimising the l-parameter in the vGM model improved the fit only slightly. The vGM was improved with an empirical scaling function to account for the rapid increase in conductivity near saturation. Using the improved models, it was possible to describe both the saturated and the unsaturated hydraulic conductivity better than a previously published model by Jarvis. The pore size boundary of the macropores was found at a capillary pressure of −4 hPa corresponding to a circular pore diameter of 750 μm.

Probabilistic Modeling of One-Dimensional Water Movement and Leaching from Highway Embankments Containing Secondary Materials

Predictive methods for contaminant release from virgin and secondary road construction materials are important for evaluating potential long-term soil and groundwater contamination from highways. The objective of this research was to describe the field hydrology in a highway embankment and to investigate leaching under unsaturated conditions by use of a contaminant fate and transport model. The HYDRUS2D code was used to solve the Richards equation and the advection–dispersion equation with retardation. Water flow in a Minnesota highway embankment was successfully modeled in one dimension for several rain events after Bayesian calibration of the hydraulic parameters against water content data at a point 0.32 m from the surface of the embankment. The hypothetical leaching of Cadmium from coal fly ash was probabilistically simulated in a scenario where the top 0.50 m of the embankment was replaced by coal fly ash. Simulation results were compared to the percolation equation method where the solubility is multiplied by the liquid-to-solid ratio to estimate total release. If a low solubility value is used for Cadmium, the release estimates obtained using the percolation/equilibrium model are close to those predicted from HYDRUS2D simulations (∼10–4–10–2 mg Cd/kg ash). If high solubility is used, the percolation equation over predicts the actual release (0.1–1.0 mg Cd/kg ash). At the 90th percentile of uncertainty, the 10-year liquid-to-solid ratio for the coal fly ash embankment was 9.48 L/kg, and the fraction of precipitation that infiltrated the coal fly ash embankment was 92%. Probabilistic modeling with HYDRUS2D appears to be a promising realistic approach to predicting field hydrology and subsequent leaching in embankments.

Tillage Effects on Subsurface Drainage

Tillage near the soil surface may greatly influence drainage discharge and pressure head values in the subsurface zone. In this study, a controlled comparative experiment was conducted in the field under natural weather conditions, using a tilled and an untilled column, to evaluate the effects of tillage on subsurface drainage discharge and pressure head values. There were no significant differences between the two columns in subsurface drainage before tillage treatment before the study, as checked by a preliminary experiment. After tilling one column of the two columns, cumulative subsurface drainage discharge was larger and occurred earlier for the tilled column than for the untilled column. The measured drainage discharge and pressure head values were evaluated using the water movement model, HYDRUS version 6.0, which numerically solves the Richards equation. The HYDRUS model reproduced measured values well for subsurface drainage discharge and pressure head values after the tillage, as determined by root mean squared error (RMSE), mean bias error (MBE), and R2 Therefore, it is concluded that the model, which includes the effects of tillage on the hydraulic properties, can explain the reasons for differences in observed drainage in the two columns. Moreover, the effects of tillage depth on subsurface drainage discharge were simulated. The results implied that the effect of tillage on subsurface drainage were induced by the conditions of pressure head just before the first rainfall, and this effect was equal to, or greater than, the effects of changes in hydraulic properties due to tillage.

The problem of data assimilation for soil water movement

The soil water movement model governed by the initial-boundary value problem for a quasilinear 1-D parabolic equation with nonlinear coefficients is considered. The generalized statement of the problem is formulated. The solvability of the problem is proved in a certain class of functional spaces. The data assimilation problem for this model is analysed. The numerical results are presented.

Interpreting INEEL Vadose Zone Water Movement on the Basis of Large‐Scale Field Tests and Long‐Term Vadose Zone Monitoring Results

A review of the results from two large‐scale field tests and vadose zone monitoring data at two facilities at the Idaho National Engineering and Environmental Laboratory (INEEL) has provided information to support development of a conceptual model of water movement in the highly heterogeneous vadose zone beneath two INEEL facilities. Both of the large‐scale field tests were ponded water infiltration tests that approached the size of the two facilities. An assembly of perched water wells, neutron access tubes, and tensiometers have been used to monitor water movement beneath these sites. Once water has percolated through the surficial sediment at these facilities, the presence of the surficial sediment–basalt interface results in some perching of the percolated water with limited lateral movement. Water movement in basalt formations is mainly gravity‐dominated in preferential fracture pathways and rubble zones and results in rapid vertical flow especially frequently under conditions of positive hydrostatic head. Detection of wetting front advance is typically not sequential with depth and has been shown to vary between episodic infiltration events. At the basalt–sediment interface of the interbeds, perched water is often detected and has resulted in more lateral spreading of recharge than at the surficial sediment–basalt interface. Gaps in the sedimentary interbeds will allow some of the recharge water to bypass the interbeds and rapidly continue toward the aquifer. This complex water movement in the vadose zone beneath the INEEL supports the need to control surface infiltration, identify gaps and the lateral continuity of the interbeds, and investigate the influence of nearby surface water sources on perched water movement beneath these two facilities.

Interpreting INEEL Vadose Zone Water Movement on the Basis of Large-Scale Field Tests and Long-Term Vadose Zone Monitoring Results

A review of the results from two large-scale field tests and vadose zone monitoring data at two facilities at the Idaho National Engineering and Environmental Laboratory (INEEL) has provided information to support development of a conceptual model of water movement in the highly heterogeneous vadose zone beneath two INEEL facilities. Both of the large-scale field tests were ponded water infiltration tests that approached the size of the two facilities. An assembly of perched water wells, neutron access tubes, and tensiometers have been used to monitor water movement beneath these sites. Once water has percolated through the surficial sediment at these facilities, the presence of the surficial sediment–basalt interface results in some perching of the percolated water with limited lateral movement. Water movement in basalt formations is mainly gravity-dominated in preferential fracture pathways and rubble zones and results in rapid vertical flow especially frequently under conditions of positive hydrostatic head. Detection of wetting front advance is typically not sequential with depth and has been shown to vary between episodic infiltration events. At the basalt–sediment interface of the interbeds, perched water is often detected and has resulted in more lateral spreading of recharge than at the surficial sediment–basalt interface. Gaps in the sedimentary interbeds will allow some of the recharge water to bypass the interbeds and rapidly continue toward the aquifer. This complex water movement in the vadose zone beneath the INEEL supports the need to control surface infiltration, identify gaps and the lateral continuity of the interbeds, and investigate the influence of nearby surface water sources on perched water movement beneath these two facilities.

Interpreting INEEL Vadose Zone Water Movement on the Basis of Large-Scale Field Tests and Long-Term Vadose Zone Monitoring Results

A review of the results from two large-scale field tests and vadose zone monitoring data at two facilities at the Idaho National Engineering and Environmental Laboratory (INEEL) has provided information to support development of a conceptual model of water movement in the highly heterogeneous vadose zone beneath two INEEL facilities. Both of the large-scale field tests were ponded water infiltration tests that approached the size of the two facilities. An assembly of perched water wells, neutron access tubes, and tensiometers have been used to monitor water movement beneath these sites. Once water has percolated through the surficial sediment at these facilities, the presence of the surficial sediment–basalt interface results in some perching of the percolated water with limited lateral movement. Water movement in basalt formations is mainly gravity-dominated in preferential fracture pathways and rubble zones and results in rapid vertical flow especially frequently under conditions of positive hydrostatic head. Detection of wetting front advance is typically not sequential with depth and has been shown to vary between episodic infiltration events. At the basalt–sediment interface of the interbeds, perched water is often detected and has resulted in more lateral spreading of recharge than at the surficial sediment–basalt interface. Gaps in the sedimentary interbeds will allow some of the recharge water to bypass the interbeds and rapidly continue toward the aquifer. This complex water movement in the vadose zone beneath the INEEL supports the need to control surface infiltration, identify gaps and the lateral continuity of the interbeds, and investigate the influence of nearby surface water sources on perched water movement beneath these two facilities.

Peritoneal Function: The Role of Aquaporins

The aquaporins (AQPs) constitute a large family of water channels that play a part in transcellular water movement in many tissues. They are particularly abundant and important in the kidney and the lung. Derangement of AQP structure or function leads to a variety of water-balance disorders. Aquaporin-1 (AQP1) is constitutively expressed in the endothelial cells of the capillaries and venules that provide the ultrasmall pores predicted by the three-pore model of water movement during peritoneal dialysis. The ultrasmall pores are critical for ultrafiltration. In addition, AQP1 and AQP3 are constitutively expressed in peritoneal mesothelial cells. The expression of those AQPs is upregulated by hyperosmolality. Reduced expression or function of AQP1 may be responsible for some cases of ultrafiltration failure, but further evidence is required to establish that link.

RAPID SOIL MOISTURE RECHARGE TO DEPTH BY ROOTS IN A STAND OF ARTEMISIA TRIDENTATA

The temporal patterns of soil water potential in a stand of Artemisia tridentata in central Utah, USA, were monitored during the summer, which included small periodic rainfall events, and over the winter, when most of the soil recharge occurs in this environment. The pattern of recharge, when compared to an area cleared of aboveground vegetation, strongly indicated that the downward movement of water to 1.5 m was primarily conducted via roots by the process known as hydraulic redistribution. Rainwater was moved rapidly downward shortly after the rain event and continued over a period of a few days. For rainwater reaching a 0.3–1.5 m depth, the portion redistributed by roots was estimated to range from 100% for small rainfall events (<8 mm) to 74% for a 36-mm event. Simulations with a model of soil water movement that compared situations with and without hydraulic redistribution by roots, indicated that during the fall–spring recharge period, 67% of all water moved downward below 0.1 m was via roots, while 87% of the water moved below 0.3 m was via roots. These results indicate that rapid downward movement of rainwater by roots can be a significant mechanism of soil water recharge to depth in arid and semiarid ecosystems. Corresponding Editor: F. C. Meinzer

Marked differences between van Genuchten soil water-retention parameters for temperate and tropical soils: a new water-retention pedo-transfer functions developed for tropical soils

All of the physical, chemical and soil water-retention data suitable for the derivation of a Pedo-Transfer Functions (PTF) for water retention for tropical soils (771 suitable horizons) were extracted from the IGBP-DIS soil database. The parameters θ s, θ r, α and n of the van Genuchten (vG) [Soil Sci. Soc. Am. J. 44 (1980) 892] equation were derived and compared with parameter values from two published data sets of temperate region soils. Thirty-five percent of the soils in the tropical (IGBP/T) database were classified as clays, compared with only 4% and 7% in the temperate databases. The IGBP/T soil bulk densities were significantly lower ( p<0.001) for nine textural classes. For the more clayey textural classes, the value of α was higher for the IGBP/T soils, implying more large pores and more structure. For kaolinitic soils in the IGBP/T database, α was typically around 0.4 kPa −1 compared with 0.04 kPa −1 for montmorillonitic soils. Mineralogy is clearly important and should be included in PTFs (if available). The value of n for clay soils in the IGBP/T data set was significantly higher than for temperate soils, and θ s was significantly higher for nine textural classes ( p<0.001) reflecting their lower bulk density. θ r was also significantly higher for all textural classes except sand. For clay, the mean value of θ r was 0.27 m 3 m −3, compared to only 0.11 m 3 m −3 for temperate soils. For 17% of the IGBP/T data, θ r exceeded 0.3 m 3 m −3, a value often used as a fitting constraint. A tropical soil PTF was developed using multiple regression techniques. This predicted water-retention curves more reliably than either class PTFs or “soil class” PTFs, but was not reliable for low-density soils such as Andosols. The IGBP/T data set includes some groups of soils, e.g., Andosols and Ferralsols, whose properties are extremely different from those of most temperate soils. These differences emphasise the need to develop separate PTFs for tropical soils, or perhaps for specific groups of soils. There were few water-retention data from below 1.5-m depth and a complete lack of hydraulic conductivity data in the IGBP/T database. This is a major impediment to the modelling of water movement and uptake in deep-rooted ecosystems, for example, tropical forest and savannah.

Numerical analysis of the transport and restoration schemes of aqueous oil in soils

On the basis of field investigations and tests, the authors developed a coupling model of water and solute movement to quantitatively analyse the effects of climate condition, irrigation mode, chemical reactions and restoration schemes on aqueous oil (effective oil) distribution using the finite difference method. It is concluded from this study that the effect of strong adsorptive forces between aqueous oil and the soil matrix is the distribution of aqueous oil in the plough layer rather than its migration to groundwater. In addition, restoration schemes involving clean-wastewater mixing irrigation and crop-type adjustment may greatly reduce the aqueous oil concentrations, but clean surface water or groundwater irrigation must be used for complete restoration of the contaminated soils.

Effect of temporal variability in infiltration on contaminant transport in the unsaturated zone

A general methodology has been developed for investigating the effect of short-term temporal variability in infiltration on the long-term transport of contaminants in soils. A one-dimensional, unsaturated transport model was used to simulate the transport of a sorbing, nonvolatile solute, using either steady-state or randomly varying infiltration. Concentration breakthrough curves are plotted against time and cumulative infiltration for constant rainfall, and for five, random-rainfall realizations. The observed time-based breakthrough curves for an individual year depend significantly on the actual rainfall pattern for that year. The average breakthrough curve, generated from many years of randomly generated rainfall, approaches the constant infiltration time breakthrough curve. For an individual year, the cumulative infiltration breakthrough curves for variable and constant infiltration match closely, as suggested by Wierenga, P.J. [Wierenga, P.J., 1977. Solute distribution profiles computed with steady-state and transient water movement models. Soil Sci. Soc. Am. J., 41, 1050]. This indicates that, for the conditions examined, cumulative rainfall can be used to predict adequately contaminant transport for a given time period. Under more severe conditions, increased variability in infiltration is expected to increase dispersion. A dimensional analysis of the governing equations indicates two additional explanations for the influence of infiltration variability on contaminant transport. First, the hydraulic conductivity (and therefore, the velocity) and retardation factor depend on the soil water content, which depends on the infiltration pattern. Second, molecular diffusion dominates transport during dry periods. The impact of this diffusion on the overall contaminant transport depends on the duration of dry periods.

Determining coastal mixing rates using radium isotopes

Coastal waters contain elevated dissolved activities of four radium isotopes. These elevated activities arise through desorption of Ra from particle surfaces and the input of submarine groundwaters enriched in Ra. The input of Ra near the coast is balanced by a flux of each Ra isotope toward the open ocean. Two of the Ra isotopes decay almost completely before they reach the edge of the continental shelf; the other two decay hardly at all. These differences in decay rates provide a powerful constraint on models of water movement and mixing on the shelf. These models are used to assess the factors that control the export of radium. Understanding the factors that control the export of radium allows an assessment of physical processes that regulate fluxes of other dissolved constituents in the coastal ocean. In this paper I use observations that were made during vertically stratified conditions in the South Atlantic Bight. Offshore transects of the long-lived 226Ra and 228Ra indicate that eddy diffusion controls their distributions within 50 km of shore. The short-lived 223Ra and 224Ra distributions in this region yield an eddy diffusion coefficient of 360–420 m 2 s −1. The offshore fluxes of 226Ra and 228Ra derived from their across-shelf activity gradients and the eddy diffusion coefficient require a substantial volume of groundwater discharge to balance Ra removal.

Intensification of the regime of thin film desalination plants

The paper considers the process of hydrodynamic and heat transfer of thin film thermal desalination plants with different methods of organization of the movement of seawater in the film (horizontal, vertical and jet-impact spraying). The regimes of these plants influence plant productivity, fresh water quality, decreasing the heating energy required for desalination. The investigation includes the development of a mathematical model of thin film water movement by heating surfaces. The equations for velocity values and flow stability conditions were obtained. The influence of the liquid stabilizing action on surface heating by hydrodynamic and heat transfer was established. Analysis of regimes of all plant types was made and equations for coefficients of heat transfer and hydrodynamic resistance were derived.

Unconventional flow of water from dry to wet caused by hysteresis: A numerical experiment

We demonstrate with the help of a numerical simulation model for hysteretic water movement that for horizontal redistribution three types of flow can occur. Raats and Van Duijn [1995] previously identified these types as a reaction to a specific analysis by Philip [1991]. Depending on the initial conditions in two parts of an infinitely long soil column, there can be no flow, conventional flow from the wet part to the dry part, or unconventional flow from the dry part to the wet part. When flow occurs, the cumulative wetting and cumulative drying both show the expected linear relationship with the square root of time. Hysteresis is described according to the modified dependent domain model of Mualem [1984]. We also indicate that Kool and Parker [1987], who used this model as well, presented a erroneous expression for the differential moisture capacity of the scanning drying curve.

Pollution history of a tropical estuary revealed by combined hydrodynamic modelling and sediment geochemistry

Hydrodynamic modelling of water movement in Hunts Bay, a protected part of Kingston Harbour, Jamaica, shows that depth averaged tidal flows are very low. In the northeast corner of Hunts Bay, water is essentially stagnant. Even under high flow conditions, much of the Bay bottom water is `bypassed' by buoyant, lower salinity surface flows. The muddy sediments of Hunts Bay reflect these sluggish to stagnant conditions; sediment cores from the northeast corner of the Bay contain progressively higher amounts of organic matter in their upper parts (∼last 15–20 years sedimentation). Combined C/N ratios and stable carbon isotope compositions of this organic matter imply a sewage origin. Both lead and chromium metal concentrations and enrichment factors relative to average crustal shales show geographically related patterns that reflect hydrodynamic circulation predicted by modelling. In particular, metal concentrations and enrichment factors are highest at the northern end of the bay, especially in the northeast corner. Modelling confirms that stagnant conditions would occur in the northeast part of the bay even without the presence of a major causeway. The causeway may contribute to low flow conditions, but is not the principal cause of organic contamination, which is simply an excessive input of sewage.

Optimal management of water movement in irrigation bays

Salinisation of Australia's agricultural lands and inland waterways is a widespread problem and its management involves significant financial resources. A key component in understanding catchment salt export is interpreting the on-land processes which determine runoff and groundwater recharge/discharge. This paper presents a decision support system for irrigation management based around a predictive model for water and salt movement within irrigation bays. The “bay model” employs a physically-based formulation, where: groundwater movement is explained by Darcy's equation; capillary rise by a quasi–equilibrium approach and overland flow by a Mannings type equation. The bay model is tested through application to a field site where water and salt transport processes were intensively monitored over a two-year period. Model predictions of overland flow and transport, groundwater and unsaturated water movement are shown to be in close agreement with field observations. An objective function is defined which incorporates irrigation efficiency (drain flow), the soil moisture deficit after irrigation, and the duration of surface ponding during irrigation events. The response of the constraints which make up this objective function are investigated with respect to irrigation practice and, as a result, the objective function is refined. This objective function is then incorporated into an optimisation algorithm (Powell's method) which is used to identify the management arrangements which lead to efficient water use.

Kinematic wave model for water movement in municipal solid waste

The movement of water in a large (3.5 m3) undisturbed sample of 22‐year‐old municipal solid waste has been modeled using a kinematic wave approximation for unsaturated infiltration and internal drainage. The model employs a two‐parameter power expression as macroscopic flux law. The model parameters were determined and interpreted in terms of the internal geometry of the waste medium by fitting the model to one set of infiltration and drainage data. The model was validated using another set of data from a sequence of water input events. The results of the validation show that the model performs satisfactorily, but further development of the model to incorporate spatial variability would increase its capability.

Water flow through temperate glaciers

Understanding water movement through a glacier is fundamental to several critical issues in glaciology, including glacier dynamics, glacier‐induced floods, and the prediction of runoff from glacierized drainage basins. To this end we have synthesized a conceptual model of water movement through a temperate glacier from the surface to the outlet stream. Processes that regulate the rate and distribution of water input at the glacier surface and that regulate water movement from the surface to the bed play important but commonly neglected roles in glacier hydrology. Where a glacier is covered by a layer of porous, permeable firn (the accumulation zone), the flux of water to the glacier interior varies slowly because the firn temporarily stores water and thereby smooths out variations in the supply rate. In the firn‐free ablation zone, in contrast, the flux of water into the glacier depends directly on the rate of surface melt or rainfall and therefore varies greatly in time. Water moves from the surface to the bed through an upward branching arborescent network consisting of both steeply inclined conduits, formed by the enlargement of intergranular veins, and gently inclined conduits, spawned by water flow along the bottoms of near‐surface fractures (crevasses). Englacial drainage conduits deliver water to the glacier bed at a limited number of points, probably a long distance downglacier of where water enters the glacier. Englacial conduits supplied from the accumulation zone are quasi steady state features that convey the slowly varying water flux delivered via the firn. Their size adjusts so that they are usually full of water and flow is pressurized. In contrast, water flow in englacial conduits supplied from the ablation area is pressurized only near times of peak daily flow or during rainstorms; flow is otherwise in an open‐channel configuration. The subglacial drainage system typically consists of several elements that are distinct both morphologically and hydrologically. An upglacier branching, arborescent network of channels incised into the basal ice conveys water rapidly. Much of the water flux to the bed probably enters directly into the arborescent channel network, which covers only a small fraction of the glacier bed. More extensive spatially is a nonarborescent network, which commonly includes cavities (gaps between the glacier sole and bed), channels incised into the bed, and a layer of permeable sediment. The nonarborescent network conveys water slowly and is usually poorly connected to the arborescent system. The arborescent channel network largely collapses during winter but reforms in the spring as the first flush of meltwater to the bed destabilizes the cavities within the nonarborescent network. The volume of water stored by a glacier varies diurnally and seasonally. Small, temperate alpine glaciers seem to attain a maximum seasonal water storage of ∼200 mm of water averaged over the area of the glacier bed, with daily fluctuations of as much as 20–30 mm. The likely storage capacity of subglacial cavities is insufficient to account for estimated stored water volumes, so most water storage may actually occur englacially. Stored water may also be released abruptly and catastrophically in the form of outburst floods.