Water Table Drawdown Research Articles

Effect of Drain Tube Openings on Water-Table Drawdown

Conventional methods assume that the drain tube is completely open and offers no resistance to the entry of water. An approximate method of determining the effect of drain tube openings on water-table drawdown is presented, based on the use of Hooghoudt's equivalent depth to account for convergence near the drain during water-table drawdown. Data from the literature are used to define an effective drain tube radius, r e , for drains with a finite number of openings. The r e value is then used to define the Hooghoudt equivalent depth which is in turn used in solutions to the Boussinesq equation for water-table drawdown. The use of an envelope wrap material will permit a somewhat wider drain spacing for all cases with a larger increase in spacing for the deeper profile. The allowable increase in spacing for both cases was relatively small and the total system cost with the envelope wrap was about 6% higher than without the wrap for the shallow profile and about 7% lower for the deep profile.

Subsurface Drainage in Soils with High Hydraulic Conductivity Layers

ABSTRACT THE Richards equation for two-dimensional, saturated-unsaturated flow during drainage was solved for layered soils using finite difference meth-ods. Solutions were obtained for soils in which the deeper layer has a higher hydraulic conductivity than the surface layer. Solutions were presented to show the distribution of equipotential lines and position of the water table during drainage processes. When the drain depth is increased so that it is closer to the high conductivity layer, head loss due to convergence near the drain is reduced. Thus, deeper drains may significantly increase drainage rates even if the hy-draulic head at the outlet remains unchanged. This effect is larger for deep profiles and narrow drain spacings than for shallow profiles and wide spacings. The effect of increasing the drain depth also increases with the hydraulic conductivity of the bottom lay-er. When outlet conditions are limiting, the most efficient drain depth is that of the layer interface. Solutions for various cases showed that further in-creasing the depth had only a slight effect on water table drawdown.

Experimental evaluation of theoretical solutions for subsurface drainage and irrigation

The Richards equation for two‐dimensional saturated‐unsaturated water movement was solved for drainage and subirrigation boundary conditions by using finite difference methods. The solutions were compared to results of experiments conducted for various initial and boundary conditions in a large laboratory soil tank. The soil water characteristic and hydraulic conductivity function were measured by conventional methods and used as inputs in the numerical solution to the Richards equation. Although predicted water table drawdown was somewhat slower than was observed, agreement between the numerical solutions and experimental observations was generally good. Approximate methods were also used to characterize drainage and subirrigation and were found to be in good agreement with solutions to the more exact Richards equation and with experimental results when a variable drainable porosity is considered. In view of significant field variability of soil properties and the difficulty of measuring those properties, required by the Richards equation, it appears that simpler, approximate methods will continue to be more useful than the exact approach for predicting water table movement under drainage and subirrigation conditions.

Drainage to a water table analysed by the Green-Ampt approach

By means of the Green-Ampt analysis, it is shown that the drainage from an initially saturated column of porous material after the water table is suddenly lowered to a given level from the surface on which infiltration is maintained at a rate q during the drainage, may be described to a good approximation by the equation: 1- Q′ Q ′ ∞ = exp [-(F 0-q)t Q ′ ∞ ] where Q′ and Q′ ∞ are the quantities of water drained at time t and at infinite time, respectively, and F 0 is the initial flow rate from the column. Experimental confirmation of this equation was obtained with measurements on the drainage from a sand column. Experiments also showed that the equation described the drainage from a column when a steady infiltration rate on the surface was changed to a lower one, although the Green-Ampt analysis could not be applied to this case. The Green-Ampt approach was also used, with supporting experimental results from a sand column, to estimate the drainage from a column of porous material with a moving water table. This analysis was applied to investigate the delayed yield effect during the water-table drawdown over a field-drainage installation.

Transient free surface flow to a well: An analysis of theoretical solutions

Several simplifying assumptions are in general required to solve analytically the set of differential equations controlling transient free surface flow to a fully penetrating gravity well. Results based on the linearization technique as well as on the delayed yield concept provide the most advanced theoretical developments to date. A review of available solutions is presented together with some complementary interpretations of their physical behavior and an analysis of the interrelations between the corresponding basic theories. If the elastic storage coefficient S8 is markedly less than the specific yield Sy, the vertical fluid transport is predominant in a cylindrical zone of the water table aquifer included between two regions where flow is essentially radial and is controlled by two different Theis equations. Its boundary surfaces move in time far from the well. If S8 is of the same order of or greater than Sy, the vertical components always tend to become subordinate throughout the aquifer, which at any instant basically behaves as an artesian formation. The physical meaning of Boulton's (1955) delay index 1/α is extensively discussed. It is shown that, contrary to Neuman's (1975) conclusions, the true physical α does not vary linearly with the logarithm of the radial distance r from the pumping well, but it is almost linearly related to 1/r and becomes practically constant at some distance from the well (equal to twice the aquifer thickness if the medium is isotropic). An empirical delay index is also found by equating Boulton's (1963) solution and the linearized average drawdown. Its expression enables us to use the classical delayed yield solution to compute the average piezometric decline with an accuracy equal to that of the linearization approach whose final outcome is, however, more complex and needs more computational time for its numerical evaluation. The two major assumptions underlying the present solutions, which are generally more accurate than the Dupuit‐Boussinesq equation, essentially require small water table drawdowns and an aquifer of infinite areal extent.

Drawdown Solutions for Simultaneous Drainage and ET

Numerical methods were used to solve the nonlinear Boussinesq equation for water table drawdown between parallel drains. The solutions were compared to other methods of predicting drawdown for initially horizontal, parabolic, and elliptical profiles and good agreement was found with methods that considered the nonlinear aspects of the governing equation. Solutions were presented in graphical form for simultaneous drainage and ET or deep seepage. Example problems were solved to show the use of these solutions for evaluating the effect of ET or deep seepage on water table drawdown and for considering this effect when predicting drain spacings to satisfy a predetermined drawdown rate. The solutions were also used in an example to explain apparent discrepancies in field tests of drain spacing equations.

Water table drawdown due to a pumped well in an unconfined aquifer

A general solution of the water table drawdown for unsteady radial flow to a pumped well is considered. Tabulated values for the free surface function are calculated for the range of parameters occurring in practical situations. A comparison with the available solution of Boulton is made, an appraisal of the assumptions being introduced in the derivation of the solutions. A numerical example of unsteady radial flow in an unconfined aquifer with details of the main part of the computer program is presented.

Water table drawdown around an open ditch in organic soils

Water level control is a potential method of improving the amount and distribution of runoff from peatlands as well as their capability to produce timber, agricultural crops, and wildlife. The effectiveness of this practice varies with the hydraulic conductivity of the peat materials through which the water must flow. An open ditch hastened the drainage of water from surface or near surface horizons of fibric (moss) peat on a typical lake-filled organic soil. However, it had little influence on the water table beyond 5 m from the ditch when the water table was in deeper horizons of moderately decomposed (hemic) peat. In an organic soil with less decomposed (fibric) peat throughout several m depth at the surface a similar ditch influenced the water table as far as 50 m from the ditch.

Effect of Soil Moisture Hysteresis on the Water Table Profile Around a Gravity Well

Gravity wells, pumped intermittently, can have different degrees of drainage in the unsaturated zone at the onset of water table recovery. Each part of the aquifer profile follows a specific scanning curve of rewetting. The measured hysteresis phenomenon of Oso Flaco fine sand and the hydraulic conductivity as a function of moisture content are incorporated in the finite difference form of the partial differential equations for steady and unsteady radial flow to solve for the soil moisture pressure distribution of one water table drawdown and three different kinds of recovery. The method of independent domains describes the hysteresis phenomenon and accounts for a gradual release and uptake of water rather than constant changes based on the concept of specific yield. Small differences between the computed values of the three water table recoveries, each one starting at a different degree of drainage in the zone above the water table, are probably due to the relatively small unsaturated portion of the aquifer profile.

A comparison of plastic and tile drain pipes with differing cover materials

A comparison was made between three types of plastic drain pipes and conventional earthen tile drains combined with cover materials of washed river sand, fibreglass, and tar paper. Rate of drawdown of a water table was measured for each combination of pipe and cover material over a period of 220 hours. Soil material that accumulated over three years in the observation pits and over four years in the pipes was measured. All combinations of pipes and cover materials provided satisfactory drainage and there were no differences of practical significance between them in terms of rate of drawdown. The river sand used in the experiment caused blockage of slot openings in corrugated plastic pipe, and one type of smooth walled plastic pipe was badly distorted by soil pressure leading to a widening of the slots, entry of soil material and breaking the fibreglass covering. Another type of plastic pipe was unsatisfactory because the perforation method left plastic material as hairs within the pipe. The fibreglass, 0.2 mm thick, was too thin to be entirely satisfactory, but was, under favourable conditions, a most effective cover material. Tar paper was quite unsatisfactory as material to prevent soil entry to pipes.

Effect of Opening on Design of Subsurface Drains

Theoretical effects of joint spacing on flow into drain tile and effects of perforations are presented in a form useful for the design engineer. The influence of these openings on the rate of water table drawdown between parallel drains is presented. Experimental field data support the theoretical results.

Analytical and Computer Solutions of Transient Water Tables for Drainage of Sloping Land

Transient water table drawdown in sloping land lying between parallel open ditches was obtained by solving Boussinesq's equation on the analog computer and by analytical methods for zero and time‐variable end conditions at the ditches. Comparison of the results with observations made on a viscous flow model indicated that the solution with time‐variable end conditions described the flow region beter than the solution with zero end conditions.

A field test of drain spacing equations for agricultural land

A field experiment to test six chosen drain spacing equations is described. A method is shown for the determination of the value of the varying drainable porosity, occurring during water table drawdown, from the pF curves of soil samples. From the result of the experiment three equations are selected for design purposes.

Predicted and experimental water table drawdown during tile drainage

Comparisons were made between experimental and predicted positions of the water table during drawdown by drains. Experimental positions were obtained from tank drainage data. The predicted water table positions were determined with a modified form of the Kirkham-Gaskell equation for drawdown. An electrical resistance network was used to obtain the components of the potential gradient which are required by this equation. Two modifications were made in the Kirkham-Gaskell equation: The capillary fringe replaced the water table as the upper boundary of the region in which flow occurred. Instead of using a constant drainable porosity, one was used which related drainable pore space to the water table depth. The agreement between experimental and predicted positions of the water table was good. When either of the aforementioned modifications was not made, significant discrepancies were noted. There was considerable deviation in both shape and mean position of the water table when a constant porosity was used in the theory. The proposed equation for drawdown offers a sound basis for studying drawdown by drains. An electrical resistance network as used in this study is convenient for eliminating the tedious calculations which are required by the equation. Because of extensive calculations, the theory is still not suited for design purposes. It will serve its greatest role in evaluating more practical drawdown equations.

Laboratory Research on Interceptor Drains

Results of large scale model study on design of interceptor drains; previously developed analytical relationships for determining shape of resultant water table drawdown curve are analyzed; new relationships are developed for estimating flow from drains; examples given.

Laboratory Research on Interceptor Drains

Results of large scale model study on design of interceptor drains; previously developed analytical relationships for determining shape of resultant water table drawdown curve are analyzed; new relationships are developed for estimating flow from drains; examples given.

AN EVALUATION OF SOME TILE-DRAIN DEPTH AND SPACING FORMULAE FROM THE PHYSICAL PROPERTIES OF SOME ONTARIO SOILS

Soil physical properties determined on six fine-textured soil types near Brampton and Chatham were used to solve and evaluate formulae for the depth and spacing of tile drains proposed by Hooghoudt, Walker, Slater, and Dumm. At the Chatham site, measurements of water-table drawdown between tile drains were recorded to compare with Neal's water-table observations of adequate drainage.A wide difference in soil properties was exhibited between the two study sites. None of the formulae results of spacing agreed precisely. Causes for discrepancies between the results were shown to be due to technical differences in applying data to the formulae and to inherent differences in derivation of the formulae. Data are presented showing the need for a physical criterion of adequate drainage.

Seepage into ditches in the case of a plane water table and an impervious substratum

It is shown how the quantity of water seeping from a plane water table into equally spaced ditches with vertical walls may be computed by considering the ditch problem as a special case of an auger hole problem, already solved. Streamlines, distribution of hydraulic head, and the seepage rate into unit length of ditch are computed. The computed seepage rate does not account for drawdown of water table around the ditch and therefore represents a maximum value. Applications are discussed and a result compared with the work of Nelson‐Skorniakov [1940].