A forced-gradient experiment of virus and carboxylated microsphere transport was carried out at a two-well system in saturated aquifer at Borden, Ontario. The purpose of experiment was to qualitatively and quantitatively investigate bacteriophage transport relative to that of a conservative solute in porous media. A simplified plane radial advection dispersion equation coupled with reversible first-order and equilibrium mass transfer was found to be adequate to simulate the attachment and transport process. For simulating detachment and transport, all rate parameters were varied with time up/down (depending on the parameter) to reflect the changes in pH of groundwater with time from 7.4 to 8.4 then back to 7.4. Both constant and scale-dependent dispersivity were used in the modeling of the transport process. Time-moment analysis of the conservative-tracer breakthrough curves produced dispersivity values of 0.1-0.6 m, close to the macrodispersivity of 0.6 m obtained using a stochastic model to describe a previous larger-scale experiment at the site. The multiple-peak feature of all the breakthrough curves suggests that the aquifer heterogeneity may be more important than local dispersion in affecting the appearance of both electrical conductivity and phage breakthrough curves. Strack's model was found quite well to describe the hydraulic head profile during the whole period of experiment if proper values for transmissivity and cone radius are chosen. Virus traveled at least a few meters in the experiment, but virus concentrations at observation points 1-m to 2-m away were a small fraction of those injected. Though clearly not an equilibrium process, retardation involving a dynamic steady state between attachment and detachment is nevertheless a major determinant of transport versus retention of virus.

Observations that colloidal natural organic matter (NOM) enhances the migration of pollutants in groundwater have focused scientific interest towards the transport of NOM and its adsorptive properties. A small-scale tracer test was performed at a field site in Georgetown, SC, to investigate the movement of NOM under field conditions. Special emphasis was given to the hydrological heterogeneity of the site,with the idea that the flow field must be accurately known in order to distinguish adsorption from the effects of the hydrological processes. 308 slug tests were performed to characterize the spatial variability of hydraulic conductivity at the site. Using the hydraulic conductivity dataset, a three-dimensional transport model successfully reproduced the migration of a chloride plume. In this way, the uncertainties due to hydrological factors were minimized. NOM was then injected in a second tracer test. A two-site adsorption model was used to describe NOM transport. Adsorption on the first site of the model was described by a linear equilibrium isotherm, with adsorption on the second site being described by a linear time-dependent (first order kinetic) reaction. Modeling results indicated that the time-dependent process dominated the adsorption of NOM, with a fast attachment and slow detachment rates. An approximate retardation factor of 77 was estimated for NOM. Because of the high velocities created by the forced gradient, chemical equilibria was not reached during the test. Spatial variability of the chemical properties of the aquifer was identified at two different depths of aquifer. Furthermore, differences at late times between the observed and simulated NOM breakthrough curves suggested possible changes on the adsorption properties of the soil caused by continuous NOM adsorption.

The problem of analyzing and estimating transmission losses in ephemeral streams with compound channels was the focus of this research. The Kinematic Runoff and Erosion model KINEROS (Woolhiser et al., 1990) and the Soil Conservation Service SCS model (Lane, 1983) were employed in this study as they represent a range in model complexity. Initial soil moisture and channel wetness conditions were employed to modify channel infiltration capacity in the SCS model and the saturated hydraulic conductivity in KINEROS. Numerically both models have been improved to treat compound channel routing with differential over-bank and main channel infiltration. The KINEROS model was extended by coupling the over-bank and channel one-dimensional kinematic and infiltration equations through the lateral inflow terms with an assumed horizontal water surface. The SCS was extended to incorporate the Manning's equation for flow with the ordinary differential equation employed in the model. Results demonstrate that simulations by multiple trapezoidal and multiple compound cross sections further decrease computed outflow volumes and increase transmission losses as compared to using a single average trapezoidal cross section. Both models simulated observed channel losses with a comparable degree of accuracy for eight carefully checked runoff events. Over-bank losses represented 12.5% of the mean main channel loss in the case of KINEROS model and 13.4% in the SCS model. The evaluation of the performance of both models demonstrated that the best results were obtained by introducing the channel bed capacity term in the analysis to more fully treat pre-runoff channel moisture conditions.

Many areas in the arid southwestern United States depend upon ground water for irrigation and municipal water supplies. The high transpiration rates of the extensive phreatophyte tamarisk (Tamarix pentandra Pall.) cause considerable depletion of ground water in this region. Eradication of tamarisk communities would appear to be a useful method for salvaging ground water; however, the vigorous reproduction and growth of tamarisk pose economic problems because of the re-treatment and maintenance work required. Also, the removal of tamarisk thickets may adversely affect wildlife habitat, esthetic values, and the ecosystem in general. The use of antitranspirant compounds on tamarisk was investigated in the laboratory, greenhouse, and field, to determine if transpiration could be reduced significantly without damaging the plant. Athel tree (Tamarix aphylla L.) was subjected to the same analyses as tamarisk to better evaluate treatment effects. Initially, several antitranspirants were tested in greenhouse studies from which the potentially useful compounds were selected for more detailed study. Phenylmercuric acetate was toxic to both species and was not used in further experiments with tamarisk. Daily transpiration rates of tamarisk, measured gravimetrically in a greenhouse, were reduced 28 to 36 percent for 20 days with spray applications of 0.01 M 8-hydroxyquinoline sulfate (8-HQS), and the combination of 150 ppm mono-methyl (MDSA) and 150 ppm mono-glyceryl (GDSA) esters of n-decenylsuccinic acid in Triton X-100 (0.5 percent) solutions. The MDSA-GDSA combination and 8_I-[QS significantly reduced stomatal apertures of tamarisk. Stomatal measurements were not obtained for athel tree, although daily transpiration rates of athel tree were reduced 31 percent for 32 days with the MDSA-GDSA combination. The addition of the filmforming Foli-gard (10 percent) did not enhance treatment effect. Treatment effects on net photosynthesis and respiration of both species were evaluated with an infrared gas analyzer. The net photosynthesis rates of tamarisk were significantly reduced 1 to 3 days by 8-HQS and MDSA-GDSA. Respiration rates, measured in the dark, were increased significantly only by 8-HQS. Net photosynthesis rates of athel tree were significantly reduced 13 days by MDSA-GDSA, but there were no effects on respiration. MDSA-GDSA and 8-HQS reduced tamarisk growth rates for 10 to 20 days following treaLment. These growth reductions may have contributed to the decrease in transpiration rates of treated plants. Analyses of foliage samples obtained from tamarisk after treatment indicated that chlorophyll and protein contents were not affected by 8-HQS or MDSA-GDSA. Similar results were noted for athel tree treated with MDSA-GDSA and MDSA-GDSA-Foli-gard. Transpiration rates of potted tamarisk in the field were reduced significantly for only five days by 8-HQS, MDSA-GDSA, and the reflective compound, kaolinite. Rainfall after the fifth day, and phenological phenomena may have contributed to the comparatively short duration of transpiration reduction. Increased foliage temperatures of 1.5 to 2.5 C were detected with a Barnes infrared thermometer 1 and 3 days after treatment. These were obtained under conditions of intense solar radiation, and high air and soil temperatures. This study indicated that tamarisk transpiration can be reduced substantially without lethal increases in foliage temperatures, but with reductions of growth. Additional studies with these antitranspirants may provide a useful management alternative for tamarisk dominated areas to salvage ground water and yet maintain plant cover for wildlife and other uses.

The influence of geologic factors in controlling flow patterns in hydraulic systems is evaluated by hydraulic models which are constructed as a porous consolidated media that simulates the interstitial geometry of consolidated rocks. Colored inks are injected into the flow system and are observed through the transparent sides of the model case. Visual analysis of the flow system, leads to salient conclusions concerning fundamental aspects of complex flow systems. Empirical experiments were conducted on the following aspects: Refraction of flow lines across lithologic interfaces: The law of streamline refraction, as described by King Hubbert, was found to be correct where boundary conditions do not interfere. Continuity of flow around and through highly permeable and impermeable lenses of different lithologies: The flow system around impermeable lenses indicates the nonexistence of stagnant areas where a hydraulic gradient is imposed on a saturated ground water system. Flow net system caused by a single pumping well: Transient changes of individual flow vectors, within the immediate area of influence, were analyzed at the moment pumping began. The absence of a transition phase indicates a rapid adjustment of the flow system to the pumping condition. Flow net system of mutual interference of depression cones caused by pumping multiple wells: This permitted an analysis of the ground water divide. Flow bands divided into flow paths which moved in opposite directions. Effects of emplacing pumping wells in highly permeable media: The increased area of influence of water movement to wells was clearly illustrated. This analysis demonstrates the capture of partially confined flow from great dths. The relation of the shape of artificial recharge pits to infiltration rate: Variations of flow net systems of rectangular and wedge-shaped pits were analyzed. When all other factors were held constant, the shape of the recharge pit was found to have no important effect upon recharge rate. Flow toward an effluent stream: The potential head of ground water beneath the stream was found to increase with depth. The relationship between hydraulic gradient and flow net configuration: The hydraulic gradient was found to have no effect on the flow net of a confined system, but a definite effect upon unconfined flow systems, in as much as it alters the water table which is the upper flow boundary. Flow pattern through tilted and faulted sedimentary structures: The geometric convergence of the aquifer boundaries caused the convergence of flow lines through a brecciated fault zone. A possible genesis of a hydrothermal vein ore was suggested by this flow pattern. Formation of perched water tables: The mechanisms by which saturated ground water mounds can be formed on a low permeability lens was demonstrated. Evidence was found which indicates that perched ground water probably escapes through the perching body as well as around its extremities. Confluence of gravity water and saturated flow: Unsaturated flow arriving at the water table of the saturated ground water body becomes an integral part of that body which acts as a single hydrodynamic system. Artesian ground water systems: A model illustrating the classic artesian aquifer situation was constructed, andflowing and non-flowing artesian wells were studied. A ground water mound was formed in the water table aquifer by water discharging naturally from the artesian aquifer through a fault in the confining layer. Subsidence around a pumping well: Water was pumped from a simulated artesian well, the piezometric surface in the vicinity of the well was lowered and the overburden was observed to subside while compressing the artesian aquifer. Cone of depression formed around single and multiple well systems: The drawdown at any point within the area of influence of a multiple well system was shown to be equal to the sum of the individual drawdowns of each well in the multiple well system, provided recharge and evaporation are neglected. Information resulting from these studies will provide guides for scientific development and exploitation of ground water supplies. Contributions were made that will advance the use of hydraulic models as exploratory tools in scientific hydrology. Further, this work brought into focus the importance of hydraulic models as couinunication media for interpreting cause-effect relationships in highly complex flow systems, of the type that so often are involved in regional problems of water resource development and management.

Soil and grass filtration of a domestic sewage effluent for trace element removal was investigated by applying oxidation-pond-treated waste water to twelve 2.44-meter-long, 30.5-cm-diameter, lysimeter columns and to a half-acre plot planted to common bermudagrass. Bermudagrass was also planted on the soil columns to aid in the removal of accumulated metals, thus prolonging the filtering capacity of the soil system. Four irrigation patterns, continuous flooding, alternate day flooding, one day wet-three days dry cycles and two days wet-five days dry cycles, were employed during the eleven week lysimeter test. Water samples were extracted from the soil columns at eight sampling depths to determine the soil depth at which the various trace metals were removed from the filtrate by the processes of adsorption, absorption and biological assimilation. Atomic absorption spectrophotometric techniques were used to determine the concentrations of iron, manganese, hexavalent chromium, nickel, copper, zinc, lead, cadmium, cobalt and strontium in the applied effluent, filtrate, soil and grass clippings. Filtrate samples for trace metal analysis were also extracted at shallow depths and from two groundwater zones at 9.1 and 15.2 meters depth under the grass plot. In the lysimeter study, iron, manganese, nickel, copper, zinc, lead and cadmium were removed from the filtrate at, or near, the soil surface. Some copper, zinc and cobalt were found in the deeper filtrate samples during periods of high infiltration rates and in the presence of anaerobic conditions. Smaller amounts of manganese, nickel and lead were also found at the deeper sampling points. Strontium was not removed from the water percolating through the calcareous soil employed in this study. Chromium and cobalt were not present in measurable quantities in the applied effluent. Iron and manganese were removed from the soil system in the grass clippings in the greatest amounts. Traces of copper, zinc and cadmium were also detected in the grass clippings. Larger quantities of iron, manganese and copper were detected in the filtrate at the grass plot than at comparable depths of the lysimeter columns, probably due to effluent applications at the grass plot over a period of two years. Lower quantities of nickel and lead in the filtrate at the grass plot were attributed to leaching by effluent during grass filtration tests, prior to the trace metal experiments. The concentrations of trace metals in the filtrate at the bottom of the lysimeter columns and in the groundwater zones at the grass plot were irsignificant by United States Public Health Service drinking water standards (1962) and would meet most other water supply requirements. The total nitrogen and nitrate contents were successfully reduced when a combination of aerobic and anaerobic environments were present in the soil system. Total nitrogen and nitrate levels were not reduced by grass filtration through 304.8 meters of dense grass. Fewer total coliform organisms were detected in the aerobic lysimeter columns than in the continuously flooded columns. The majority of coliform organisms detected in the filtrate at the 61 cm depth were believed to be of non-fecal origin. The chemical oxygen demand of the applied effluent was reduced to as low as 105 mg/L by grass filtration. A substantial portion of the remaining COD was attributed to high algal concentrations. The most effective treatment employed in this study was the one day wet-three days dry irrigation cycles. This treatment provided the necessary aerobic environment for trace metal removal from the effluent, nitrification of reduced nitrogen compounds and coliform organism elimination. The flooding period provided the anaerobic conditions required for denitrification losses of nitrate.

A procedure for estimating runoff from convective storms in the semiarid Southwest is needed for the design of small hydraulic structures. The aim of this study was to develop and test rainfall and runoff relationships based on the analysis of 12 years of hydrologic data for an 18-square mile experimental watershed. rhe experimental area is divided into four subwatersheds ranging in size from 0.5 to 7.8 square miles, Vegetation and soils are typical of what is encountered in the valley floors of southern Arizona. Rainfall is measured at 29 locations. Isohyetal maps were prepared for all of the storms which lead to the development of a rainfall model that describes the distribution of rainfall in space. An exponential relationship was found to adequately represent the spatial variation of each storm. A single equation for all storms was developed by using a parameter that is related to the storm center depth. The Kolmogorov-Smirnov procedure was used to test the hypothesis that storm center location is governed by chance in areas not influenced by topographic changes. It was found that the assumption which states that convective storm cells are randomly located within valley floors is acceptable. An equation was derived for calculating point rainfall probabilities from raingage network data, The results were based on the random location of storm centers and on an extremal distribution function fitted to storm center depths. The calculated probabilities were found to be significantly higher than the observed probabilities determined from a nearby, long-term U. S. Weather Bureau station. The volume of runoff from small, semiarid watersheds was found to be a function primarily of mean rainfall. In a multiple linear regression model, mean rainfall accounted for 67 to 82 percent of the variance. The use of a time distribution factor which includes the maximum 15-minute intensity reduced the unexplained variance to 11 to 16 percent. Inserting a space distribution variable into the model indicated that storm center location on the watershed was not a significant factor in predicting runoff. An antecedent rainfall index did not produce any significant correlation with runoff from convective storms. For winter frontal storms, however, a four-day antecedent rainfall index was found to be an important factor in oxplaining runoff. It appears that the commonly used Soil Conservation Service method underestimates convective storm runoff for most storm center depths below about three inches. A direct comparison with the multiple regression equation was not possible as this method does not take into account the variability of convective rainfall in time and space. As a means for estimating runoff volumes for ungaged watersheds, a runoff coefficient was defined as the ratio of runoff to effective rainfall (mean rainfall less initial abstractions). It appears that as a first approximation, the runoff coefficient can be considered as being equal to the coefficient in the well known rational formula. There is some evidence to the belief that the runoff coefficient is affected by a storm's time distribution factor. It was demonstrated that runoff volume recurrence intervals can be determined adequately from the rainfall and runoff relationships developed in this study.

This study consists of a theoretical analysis of the energy balance equation for a partially covered body of water, and experimental analyses of the energy balances of partially covered insulated evaporation tanks. The theoretical analysis indicates that surface reflectance for solar radiation and infrared emittance are the most important cover properties. White colored materials were found to satisfy the requirement that both these parameters be as large as possible. Experiments were conducted using covers of foamed wax, lightweight concrete, white butyl rubber, and styrofoam. A variety of shapes and sizes were tested. Cover radiative properties were again noted to be most important, and thin covers proved to be slightly more efficient than thick insulated covers of the same size. Evaporation reduction was found to be proportional to the percent of surface area covered, the constant of proportionality depending upon the color and type of material used. For the white, impermeable materials tested, the constant of proportionality was near unity. It was also noted that reduction in evaporation and reduction in net radiation, as compared to an open tank, were highly correlated. Evaluation of two Dalton-type expressions, the Bowen ratio method and the combination method, for predicting evaporation from an open water surface, showed the combination method to be better under conditions of this experiment. Based on this finding, a modified combination method was derived. This modified equation proved valid for predicting evaporation from a partially covered body of water. The use of insulated evaporation tanks also provided an easy and accurate method of predicting net radiation over other surfaces, and long-wave atmospheric radiation.

A study was conducted at Tucson, Arizona to determine the variation in water-use efficiency among 150 genotypes of Medicago sativa L. cultivar ‘Mesa-Sirsa' grown under three soil moisture regimes and to determine if there were physiological and anatomical parameters related to efficiency of water use and growth components of alfalfa genotypes at three different growth stages. Information was obtained that would contribute to the development of techniques in the selection of alfalfa genotypes efficient in water use. Mesa-Sirsa genotypes varied in their water requirement. Some plants required three times more water to produce one gram of dry matter than other plants. Soil-moisture content significantly affected the water-use efficiency in the seedling stage but not in the mature stage of growth. The amount of available soil. moisture significantly influenced the production of dry forage, stems, and leaves at both seedling and mature stages of growth. The relationship between water-use efficiency and growth components studied was significantly influenced by the amount of available soil moisture at the 'seedling stage but not at the mature stage of growth. The water-use efficiency of Mesa-Sirsa seedlings grown under the medium soil moisture regime increased when dry matter production from forage, stems, and leaves increased. There was also a significant relationship between the total transpired water per plant and dry matter produced, number of stems and leaflets. Total water transpired was not significantly associated with the water-use efficiency of alfalfa seedlings. At the mature stage of growth, there was a significant correlation between the water-use efficiency and each of the following factors: dry weight of forage and stems, number of leaflets, and stem-leaf ratio. The amount of water transpired per plant was significantly and positively correlated to dry matter production from forage, stems, and leaves, and to number of stems and leaflets per plant. There was no association between the total water transpired from a plant and the water requirement. The relationship between the water requirement and transpiration rate significantly varied with the three growth stages. The most efficient alfalfa genotypes at floral bud or flowering stage transpired significantly less water per unit leaf area per unit time compared to the less efficient genotypes. The rates of net photosynthesis and respiration varied among alfalfa genotypes and with growth stages. The rates of both decreased as plants progressed toward maturity. At the floral bud stage plants with the highest water requirement also had the highest rates of photosynthesis and transpiration. At the vegetative stage of growth, respiration was directly associated with leaf-stem ratio and transpiration rate. The number of xylem vessels was significantly associated with the water-use efficiency of alfalfa genotypes at the vegetative stage. The most efficient alfalfa plants had significantly less xylem vessels in the leaf midvein when compared to the least efficient plants. The thickness of palisade layer was also significantly related to water requirement of alfalfa genotypes at the floral bud stage but not at other stages. Efficient alfalfa genotypes developed a thinner palisade layer.

The purpose of the study was to develop a method to determine the coefficient of storage of a water-table aquifer by correlating change in gravitational field intensity with change in groundwater storage. In theory, this purpose may be accomplished by modifying the Bouguer slab equation to coefficient of storage equals 78.3 times the ratio of change in gravity in milligals to change in water-table elevation in feet. Errors which result from the Bouguer slab assumptions may be corrected through analysis of tilted finite slabs. Field investigations were made to test the theory. The study area is located in the northern Tucson basin, Pima County, Arizona, and lies on unconfined basin-fill deposits and flood-plain alluvium aquifers. The basin-fill aquifer overlies less permeable Rillito beds and is overlain by the flood-plain alluvium. The two upper aquifers are flat-bedded heterogeneous deposits of sand and gravel. The water table through these aquifers slopes westward at a rate of approximately 0.5 degree. Estimates of the coefficient of storage for the basin-fill deposits and the flood-plain alluvium have been previously made by others from laboratory and field tests and by model analyses. The most reliable determinations of the coefficient of storage range from 0.15. to 0.30. The significance of the gravity method lies in determination of the coefficient of storage by measuring the quantities which define it: rise or decline in head and weight of water placed into or removed from storage. Change in gravity was determined by repeated gravity surveys using the same set of field stations through the period, October 1968 to June 1970. Water levels in wells were recorded for the same period. The relationship between change in gravitational field intensity and change in head was determined using a straight line solution method, and the coefficient of storage was computed from the slope of the straight line. At the conclusion of the field investigations, coefficients of storage were computed for 17 field stations. After correction for limited area of water-level decline and for water-table slope, the values of the coefficients ranged from 0.11 to 0.41. An error analysis indicates a maximum probable error in gravity data of ± 26 microgals. This error may be reduced by modifying the survey and reduction procedures and by using a more sensitive gravimeter. Analysis of changes in gravitational field intensity resulting from change of amounts of water in the unsaturated zone indicates that the coefficient of storage computed for field stations near Rillito Creek, the source of the unsaturated-zone water, are too low. Using data from stations least affected by gravity increases after stream recharge, a probable range of 0.25 to 0.29 was determined for the coefficient of storage in the study area. The range for values of the coefficient of storage using the gravity method confirms the larger coefficient of storage estimation made by others for the same area. The study indicates that the gravity method may be used with success over aquifers which have high coefficients of storage and in which the water table rises or declines 20 feet or more. However, large changes in the water content of the unsaturated zone cause gravity data to show large scatter with respect to water-level data. For this reason the gravity method is more suitable for analysis of those portions of a water-table aquifer which are recharged by underflow than for the portions recharged by infiltration from surface sources.