Progress 10/01/02 to 09/30/07
Outputs
OUTPUTS: The major product developed during this project is a published conceptual model that can be used to predict the occurrence and characteristics of unstable flow during redistrition of water following infiltration. The model can predict the width of fingers or plumes that form at the wetting front when an instability develops, and how deep into the soil the fingers will move before stopping as a function of measurable soil properties. Dissemination outlets were of two types: three peer-reviewed techincal journal articles published in top journals in the professional literature; and three professional talks given to sholars. Two of these weree invited talks at universities, and the third was a keynote lecture at an international symposium in Switzerland.
PARTICIPANTS: Post-doctoral student Zhi Wang, currently Asst. Professor at CSU Fresno
TARGET AUDIENCES: Scientists, water managers, and agronomists
Impacts
The focus of this investigation has been to identify the mechanisms responsible for the initiation and persistence of unstable flow at the wetting front of an infiltration or drainage (redistribution) event in soil, and to develop a mechanistic model for describing the process. Experiments used to analyze unstable flow included both extensive laboratory studies in a 1m by 1m by 0.025m Hele-Shaw cell and a field study previously conducted on two different soil types. Models developed were of two types: conceptual models that recreate the main features of unstable flow; and modifications of the classical Richard flow equation to allow simulation of unstable as well as stable water flow. The conceptual model is able to calculate the width of the unstable plumes or fingers that form and propagate ahead of the wetting front, and to determine how deep the fingers penetrate into the soil as a function of initial conditions and soil properties. The numerical model has proven
to be of limited utility because of the extreme nonlinearity of flow permeability at the wetting front and the difficulty in performing numerical calculations without generating numerical errors. There are five important findings from this research. First is the fact that all soils are inherently unstable during redistribution, because the pressure behind the draining front is less than that at the front, thereby causing any irregularity at the front to expand into a finger. The second finding is that the width of a finger is determined uniquely by the value of the soil's water-entry value and the downward flow rate at the time of formation of the finger relative to the saturated hydraulic conductivity. The third finding is that the depth reached by the finger before stopping can be calculated from the soil's wetting loop of the water characteristic function and the water content of the soil in the dry zone. As a practical matter, the fingers formed in finer-textured soils are very
wide and contain nearly as much water as the zone behind the front, so that they penetrate only a small distance. In contrast, sandy soils form narrow fingers that can easily move 1 m or so below the front before stopping. The fourth finding is that once a finger forms in soil, it leaves a memory that persists indefinitely unless washed out by an intense wetting event that saturates the soil. As a consequence, the finger flow paths remain the same with subsequent rainfall or infiltration events. The fifth finding is that the key to modeling unstable flow numerically is to create a disequilibrium between water content and water pressure at the wetting front that encourages water to flow laterally to feed fingers rather than downward. Although it is straightforward to create such a rate-limited process, the rate coefficient is so nonlinear a function of water pressure that it renders numerical solution extremely difficult. As a final note, we believe it is a priori impossible to predict
the exact locations where the first instability will develop from any known set of measurable soil properties.
Publications
- Wang Z. Wu LS. Harter T. Lu JH. Jury WA, 2003 A field study of unstable preferential flow during soil water redistribution . Water Resour. Res. 39 (4): Art. No. 1075 Apr. 2003
- Jury, W. A., Zhi Wang, and Atac Tuli, 2003. A conceptual model of unstable flow during soil water redistribution. Vadose Zone J. 2003 2: 61-67.
- Wang, Z., A. Tuli, and W. A. Jury, 2003. Unstable Flow during Redistribution in Homogeneous Soil, Vadose Zone Journal. Vadose Zone J. 2003 2: 52-60.
|
Progress 01/01/06 to 12/31/06
Outputs
Major progress on this project during the past year consisted of development and presentation of a keynote address at an international conference on preferential flow in Switzerland, synthesizing all the results from this project, summarizing the status of knowledge of preferential flow under field conditions, and making recommendations for modeling approaches. The presentation made the following major points: unstable flow always occurs whenever water redistributes in soil following redistribution; in coarse-textured soils, downward flow ceases in the soil matrix during redistribution, which occurs entirely in the fingers; the water characteristic function determines whether the unstable flow will be significant or trivial, and can be used to predict the maximum depth of penetration of a finger; the primary mechanism responsible for unstable flow is the water-entry matric potential , which is measurable; the water-entry matric potential depends strongly on the
antecedent water content of the soil; fingers remain narrow because of hysteresis in the water characteristic function; finger memory can persist in soils for weeks to months, and will determine the pathway of subsequent flows; finger size can be estimated fairly accurately from knowledge of the water-entry matric potential ; finger sizes become significantly larger as the antecedent water content becomes large; and, in sandy soils, fingers can move to great depths (i.e. > 1m) below the infiltration front during redistribution. Conventional water flow models cannot predict unstable flow, because their formulation does not include a threshold water-entry matric potential of an invading water front, below which water will not enter a dry region of soil. It is suggested that the classic Richards equation be modified to remove the assumption of instantaneous equilibrium between the water content and matric potential and substitute a kinetic expression whose rate constant mimics the
threshold water-entry matric potential. Such a model may present difficulties in solving the resulting highly nonlinear numerical equations, however.
Impacts
Current models of water flow in soil cannot predict unstable flow, which has been shown by this research to be widespread and triggered by redistribution following infiltration. The research provides a method for estimating the importance and impact of unstable flow in a given soil, as a function of parameters that can be measured or estimated. The research also indicates the gaps in current understanding and proposes a new direction for modeling research.
Publications
- Jury, William, 2006. Pore and Darcy scale physics of preferential flow arising from fluid instability in homogeneous soil - Keynote lecture at International Conference on Preferential Flow and Transport Processes in Soil, 4-9 November, 2006, Centro Stefano Franscini, Monte Verita, Ascona, Switzerland. Online at http://www.ito.ethz.ch/conferences/preferential-flow/program.htm
|
Progress 01/01/05 to 12/31/05
Outputs
During the past year the model HYDRUS representing the Richards flow equation was modified to allow it to represent various candidate hypotheses for the initiation and propagation of instabilities in soil. We are attempting to develop a numerical representation of the transformation of a wetting front during redistribution when one part of the front gets slightly ahead of the main draining front. Because the water potential decreases behind the front during redistribution, and the matric potential of the front is the same everywhere, the region that is slightly deeper than the main body of the front draws water from adjacent regions due to gravitational potential differences. The loss of water from the main part of the front lowers the pressure below what is required to advance the front and only the fingers move thereafter. The classical Richards equation cannot represent this phenomenon. The most promising hypothesis for modifying the Richards equation involves
relaxing the requirement of instantaneous equilibrium between the matric potential and water content during transport, and making the rate coefficient describing the speed of restoration of equilibrium a strong function of matric potential.
Impacts
It is now generally conceded that preferential flow of water and solutes occurs under a wide variety of conditions in natural soil. At the present time, no model exists for describing this phenomenon, despite the widespread use of transport modeling. Preferential flow is important in a number of practical applications, including pesticide and fertilizer movement in agricultural soils and waste spills. development of a transport model describing preferential flow will allow improved management and celanup strategies to be developed.
Publications
- No publications reported this period
|
Progress 01/01/04 to 12/31/04
Outputs
The Newport Bay/San Diego Creek Watershed has had a total maximum daily load (TMDL) established for the allowable loading of nitrogen into the bay. Although agriculture is identified in the TMDL as a major contributor of nitrogen in the watershed, observations from a monitoring study and estimates from a conceptual model show that agriculture is contributing far less than assumed. As a result, limitations on loading from agricultural sources to be obtained by 2007 have already been met. Agricultural land use in the region has been greatly reduced, from approximately 20% of the watershed area in the 1980s to a current 2% of land area. We developed a simple nitrogen transport model that superimposes contributions made by all of the sources contributing to the flow of the surface water. The model assesses the travel times from the sources and their mass estimates. It predicts that agriculture now contributes only 2% of N loading directly to surface waters. An additional
6% from agriculture is contributed if estimates of leaching to shallow groundwater and subsequent mixing with surface water are also considered. The time lag involved in groundwater mixing suggests that even if all remaining agricultural production in the watershed were ceased today, the effects of nitrate loading from fertilization would continue to be seen for another 10 to 30 years. The greatest input to current surface water conditions is not agriculture, but increasing urban development. Nurseries also contribute a significant portion of nitrate, and have the highest contribution per unit area of any N source.
Impacts
This study clearly demonstrates the influence of travel time on surface water quality and the need to include time estimates in TMDL assessments. Our research shows that agriculture is not a significant contributor to nitrogen levels in the watershed, and that changes in surface management by agriculture will have no discernable effect on water qualify in the region.
Publications
- Liu WP, Gan JY, Schlenk D, and W. A. Jury, 2005. Enantioselectivity in environmental safety of current chiral insecticides. Proc. Nat. Acad. Sci. 102 (3): 701-706
- Kim SB, Jo KH, Kim DJ, and W. A. Jury, 2004. Determination of two-dimensional laboratory-scale dispersivities . Hydrological Processes 18 (13): 2475-2483.
|
Progress 01/01/03 to 12/31/03
Outputs
Experimental work from our field study and Hele Shaw cell was analyzed and used to develop a comprehensive model of the fingering process, showing that all soils produce fingers during redistribution, but that the most important instabilities develop in coarse-textured soils. The soil parameter responsible for fingering is the water-entry pressure, which may be roughly estimated from the wetting curve of the moisture release function. The width of the fingers is controlled both by hysteresis and by the rate of water flow during redistribution. Under the same conditions, a coarse-textured soil may develop fingers of a few cm-width that travel a meter or more below the main wetting front, while a finer textured soil will develop wide fingers that penetrate only a few cm. Other important features of fingering we have observed include persistence of finger memory in soils for a month or more, and development of fingers in soil that is initially moist. Several candidate
models have emerged for describing fingering. We will test all of them on data in the coming year.
Impacts
Characterization of preferential flow remains an unsolved dilemma in water flow and chemical transport modeling. This work will shed light on the dominant mechanisms responsible for this process, and will allow us to establish its prevalence and importance in different soils. The model will be useful in characterizing which soils are most susceptible to unstable flow and how great the effect is.
Publications
- Jury,W. Zhi Wang, and Atac Tuli. A Conceptual Model of Unstable Flow in Unsaturated Soil during Redistribution.Vadose Zone J. 2003 2: 61-67.
- Wang Z. Wu LS. Harter T. Lu JH. Jury WA. A field study of unstable preferential flow during soil water redistribution - art. no. 1075. Water Resources Research. 39(4):1075, 2003 Apr 1. Wang Z. Lu JH. Wu LS. Harter T. Jury WA. Visualizing preferential flow paths using ammonium carbonate and a pH indicator. Soil Science Society of America Journal. 66(2):347-351, 2002 Mar-Apr.
- Wang,Z., Atac Tuli, and William A. Jury, Unstable Flow during Redistribution in Homogeneous Soil.Vadose Zone J. 2003 2: 52-60 Wang,Z., Atac Tuli, D-J Kim, and William A. Jury. Unstable flow during redistribution: Controlling factors and practical implications., Vadose Zone J. 2004 (in press)
|
Progress 01/01/02 to 12/31/02
Outputs
Experimental work continued in the Hele-Shaw cell described last year. Our 2-D and 3-D experiments confirmed that unstable flow forms during redistribution following infiltration in a homogeneous soil under both dry and wet initial conditions, which is not predicted by the Richards equation. Fingers form and propagate rapidly when the soil is dry, but form more slowly and are larger when the soil is wet. The porous medium retained a memory of the fingers formed previously. A conceptual model was created and published that represents the development of unstable flow in uniform soils during redistribution. The flow instability results in the propagation of fingers that drain water from the wetted soil matrix until equilibrium is reached. The model uses soil retention and hydraulic functions, plus relationships describing finger size and spatial frequency. The model assumes that all soils are unstable during redistribution, but shows that only coarse-textured soils will
form fingers capable of moving appreciable distances. Once it forms, the finger moves downward at a rate governed by the rate of loss of water from the soil matrix, which can be predicted from the hydraulic conductivity function. Fingers are assumed to stay narrow as a result of hysteresis, which prevents lateral diffusion. The draining front in the soil matrix between the fingers is assumed to cease downward movement because water pressure drops below the threshold water-entry matric potential. The threshold potential is not present in the conventional Richards equation of soil water flow, which explains why unstable flow is not predicted by widely used simulation model codes.
Impacts
Characterization of preferential flow remains an unsolved dilemma in water flow and chemical transport modeling. This work will shed light on the dominant mechanisms responsible for this process, and will allow us to establish its prevalence and importance in different soils. The model will be useful in characterizing which soils are most susceptible to unstable flow and how great the effect is.
Publications
- No publications reported this period
|
Progress 01/01/01 to 12/31/01
Outputs
Results from the field experiment described in last year's report were analyzed, and a Hele-Shaw sand tank was constructed in the laboratory to continue the study of the onset of preferential flow during redistribution. We observed in our field experiments that fingered flow began at the wetting front at the onset of the redistribution phase following infiltration, and that the fingers grew during the subsequent 48 hours. The location of the fingers did not reveal any relation to soil properties, suggesting that the phenomenon was a fluid instability. The effect was largest in a coarse loamy sand and did not occur at all in a sand. Laboratory experiments have confirmed that the fingering process occurs after the succession of infiltration and is related to the reversal of the matric potential gradient above the wetting front. Fingers which are established in one draining cycle will cause subsequent infiltration and drainage patterns to follow the same pathways, even
as much as 30 days after water flow ceases. The width of the fingers formed is a function of the soil properties, and of the rate of infiltration prior to redistribution. Typical diameters in the loamy sand were of the order of 10 cm, and about half that size in the sandy soils within the laboratory tank. The fraction of the total area occupied by fingers depends on soil properties and flow rate also, and is typically 20-50 percent. A theoretical model has been constructed to describe the size of fingers and how deep they will penetrate ahead of the wetting front before stopping. The model assumes that the finger is fed from the soil matrix surrounding it, and that the matrix flow ceases at the time of finger formation. The model cannot represent the dynamics of the flow process, but does predict the final profile shape and extent.
Impacts
Characterization of preferential flow remains an unsolved dilemma in water flow and chemical transport modeling. This work will shed light on the dominant mechanisms responsible for this process, and will allow us to establish its prevalence and importance in different soils.
Publications
- Wang, Z., Lu, W., Wu, L., Harter, T., and Jury W., 2002. Visualizing preferential flow paths using a pH-indicator. Soil Sci Soc Am J 66 (2):347-351.
- Gan, J., Q. wang, S. Yates, W. Koskinen, and W. Jury, 2002. Dehalogenation of Chloroacetamide Herbicides by Thiosulfate Salts. Proc. Nat. Acad. Sci. (in press)
- Wang, Z., Harter, T., Wu, L., and Jury W., 2001. A field study of preferential flow during soil water redistribution. Water Resour. Res. (provisionally accepted)
|
Progress 01/02/00 to 12/31/00
Outputs
Two fields comprised of sandy loam and loamy sand textures were chosen to represent a range of conditions favorable to preferential flow during redistribution. The experiment consisted of the uniform addition of infiltrating water under unsaturated flow conditions to the field surface by a specialized low impact spray boom that adds the water to the surface at a spatially and temporally uniform flow rate. Within the infiltrating water are a pulse of potassium bromide and ammonium carbonate so that the pulse may be monitored in the soil by both analysis of soil samples and by visible dye tracing both at the end of the infiltration phase and during four days of subsequent redistribution. A series of preliminary studies were performed on the fields over a period of 4 months at a variety of input flow rates, until we were able both to induce and prevent preferential flow in each field. Following these preliminary studies, a flow rate was chosen that maximized the
appearance of preferential flow and also contrasted its characteristics on the two fields. The progress of the infiltrating and redistributing front was monitored on the two fields using a combination of soil sampling and dye trace photography. The monitored region of the study area was a soil cube 1.2 m on each side and 1 m deep that was photographed by shaving a trench face in successive 10 cm increments to 50 cm from the end over several days following the end of infiltration, spraying the face with a special solution to stain the regions that contained ammonium, photographing the region with a precision digital camera over a prescribed grid, and then taking 120 soil samples on a 10 cm unit grid along the entire face. At the conclusion of the sampling, the remaining undisturbed 50 cm half of the soil block was intensively sampled vertically by soil coring. In all, some 3000 samples were taken and are now being analyzed for water content, nitrogen, bromide, and select hydraulic
properties. Until the samples are analyzed, it will not be possible to assess fully the degree to which we were able to create preferential flow during the redistribution phase. However, from the evidence obtained from the dye trace studies, we are optimistic that we have detected preferential flow and can associate it with fluid and soil characteristics as indicated in our proposal.
Impacts
Characterization of preferential flow remains an unsolved dilemma in water flow and chemical transport modeling. This work will shed light on the dominant mechanisms responsible for this process.
Publications
- Gan, J., S. R. Yates, F. F. Ernst, and W. A. Jury,2000. Degradation and volatilization of the fumigant chloropicrin after soil treatment. J. Environ. Qual. 29 1391-1396.
|
Progress 01/01/99 to 12/31/99
Outputs
A field experiment has been performed measuring the influence of water application rate on the occurrence of preferential flow. At the lower flow rate, the infiltration and redistribution process were completely uniform, whereas when the rate was doubled the front was observed to be highly skewed and preferential when it was excavated. Further work is planned to isolate the flux rate at which the process becomes unstable and to determine whether it occurs during infiltration or redistribution or both. As part of the long-term project on the behavior of methyl bromide in soil, we characterized the effect of moisture and temperature variations on its degradation rate in the soil. The research on pesticide sorption and degradation reported previously was extended to include the effects of rate-limited mass transfer between sorbed and dissolved phases on the interpretation of the degradation rate in each phase. Results obtained by this analysis showed that the degradation
rate in the sorbed phase, while small compared to that in the dissolved phase, was significant enough that it must be included in the analysis of dissipation in soils with substantial sorbing capacity.
Impacts
The results on pesticide degradation resolve an old controversy over whether degradation occurs in the sorbed phase. This result has serious implications for transport and fate modeling.
Publications
- Flury, M., M.V.Yates, W.Jury 2000. Numerical analysis of the effect of the lower boundary condition on solute transport in lysimeters. Soil Sci. Soc. Am. J. (in press)
- Guo, Lei, W. Jury, R. Wagenet, and M. Flury 2000. Dependence of pesticide degradation on sorption: nonequilibrium model and application to batch reactors. J. Contaminent Hydrol. (in press)
- Gan, J., S.K. Papiernik, S.R. Yates, and W.A. Jury 2000. Temperature and moisture effects on fumigant degradation in soil. Journal of Environmental Quality 28(5): (in press).
|
Progress 01/01/98 to 12/01/98
Outputs
A laboratory and theoretical investigation was conducted to measure separately the degradation rate coefficients for pesticide degradation in the dissolved and sorbed phases. Using a series of soil columns amended with different amounts of activated carbon, we were able to create an environment in which the extent of pesticide sorption changed without altering the rate co-efficients for degradation under the different conditions. As a result, we were able to separately fit the rate coefficients for the two processes contained in our transport model, finding that degradation in the dissolved phase was approximately 70 times faster than in the sorbed phase. A theoretical investigation was conducted comparing the use of a free drainage lower boundary condition, such as is found in field environments, with a constant head lower boundary condition commonly found in field lysimeters. Significant differences were observed between the break-through of solutes when lysimeters
were used compared to their behavior under free drainage. This finding suggests that careful interpretation will have to be made of lysimeter experiments used in policy regulation. A new theoretical framework was developed that combines the Lagrangian and Eulerian frameworks for expressing solute concentrations in a form that can be solved by a differential equation. This mathematical framework will allow the solution of problems by numerical methods that involve rate processes that are functions of time after application rather than elapsed time.
Impacts
(N/A)
Publications
- FLURY, M., M. V. YATES, W. JURY, and DAMAAN ANDERSON, 1998. Variability of solute transport in field lysimeters. Amer. Chem. Soc.
- SKAGGS, TODD H., Z. J. KABALA, and WILLIAM A. JURY, 1998. Development of a nonparametric transfer function for solute transport in soils. J. of Hydrol. 207:170-178.
- FLURY, M. , M. V. YATES, and W. JURY, 1999. Solute Transport with Residence-Time-Dependent Sink/Source Reaction. Water Resour. Res. (in
|
Progress 01/01/97 to 12/01/97
Outputs
Generally, both soil water flow and solute transport properties must be known to model chemical contamination of ground water on a regional scale. The variability of these properties make deterministic modeling unfeasible, necessitating some form of approximate statistical approach that extrapolates from limited samples of properties and input parameters. There are several options for exercising this strategy, but most of them involve using a local model representation that is averaged by using a number of discrete one-dimensional simulations in parallel. With this strategy, the important question becomes what type of local model to use, and how complex to make it. We explored options for local representation in modeling the water flow regime, ranging from full simulation using the Richards flow equation, to steady flow using only the field capacity estimate of water content. Simulations of flow and transport to ground water were run on a hypothetical field with
variable climatic data and properties generated by geometric scaling theory, using data from 20 sites averaged in parallel to represent field-scale movement to ground water for a conservative and reactive chemical pulse. Although the transient flow model was necessary to achieve accurate representation of the position of the pulse within the profile, mass loading of ground water was represented quite accurately with a simple flow regime assuming steady state flow and uniform, water content. The field capacity estimate was greatly out of agreement with the other methods, however.
Impacts
(N/A)
Publications
- TSENG, P. W. JURY, and M. FLURY, 1997. Stochastic Solute Transport Modeling of Regional Non-point Source Pollution. AGU/ SSSA Special Publication from 1997 Chapman Conference, Riverside, CA October,
- WU, LAOSHENG, W. JURY, A. C. CHANG, and R. R. ALLMARIS, 1997. Time series analysis of water content variations in a field soil Soil Sci Soc Amer. J. 61: 736-742.
|
Progress 01/01/96 to 12/30/96
Outputs
We are developing a model of selenium dissipation in soil that embodies the transport and reactions that are important in describing its fate under natural remediation. The model has been useful in explaining the trends in a long-term remediation demonstration at Kesterson Wildlife Reservoir. The field study, which has been going on for seven years, focuses on the use of various amendments to accelerate the natural dissipation of selenium by methylation and subsequent volatilization. Our model has been valuable in separating out the volatilization process from leaching, which also removes selenium for the surface zone where field monitoring is occurring. We are conducting an experimental and theoretical study to analyse the effectiveness of using solute transport measurements to estimate the unsaturated hydraulic conductivity. The method is approximate, but theoretically sound as long as the travel time of the solute is long compared to the lateral mixing time. The
theory is straightforward and allows a significant range of the unsaturated conductivity-water content curve to be estimated from a single solute pulse leaching experiment. Current research is testing the theory at a variety of scales from the laboratory to the field.
Impacts
(N/A)
Publications
- FLURY, M. W. FRANKENBERGER, and W. JURY, 199x. Selenium removal at Kesterson reservoir by volatilization. Science of the Total Environment. (in press).
|
Progress 01/01/95 to 12/30/95
Outputs
Methyl bromide (MB) is the world's best fumigant, but it is also an ozone depleter and is scheduled for elimination in the year 2001 by the USA. According to farmers its elimination will cause enormous economic damage in a variety of industries, particularly strawberries and almonds. Moreover, at the present time the US is the only country planning on banning it. In standard operation as a preplant or post harvest fumigant, MB is injected into the subsurface by a moving tractor and then the field is covered with a 1 mil polyethylene tarp for 1-2 days. We conducted model calculations and a field plot experiment of methyl bromide loss under various management strategies We found through model calculations that a water addition to the surface after methyl bromide injection in the subsurface can have a substantial affect in reducing methyl bromide loss to the atmosphere. Calculations of the concentration-time index showed that this strategy actually improves the efficacy
of a given addition of fumigant, because it is held in the soil longer. This prediction was verified in plot studies which showed less loss to the atmosphere and better control of pathogens, weeds, and nematodes with water additions than without when methyl bromide was used as a fumigant. A statistical analysis was performed on data from a seven year field study of selenium removal at the Kesterson reservoir.
Impacts
(N/A)
Publications
|
Progress 01/01/94 to 12/30/94
Outputs
Twelve soils in Switzerland comprising a wide range of texture, structure, and layering characteristics were evaluated for preferential flow by a standard experiment in which dye was sprinkled on prepared plots with a specified amount of water and then a trench was dug and the pattern of water flow recorded. With one exception, all of the Swiss soils showed preferential flow, ranging from moderate in the structureless sandy soils to extreme in the structured clay soils. The single exception was a sandy loam soil devoid of structure that appeared to have all of its large pores filled with silt, which prevented any localized flow from developing to any extent. A series of experiments were conducted in our volatilization chamber and in the field under standard agricultural conditions to determine the factors governing methyl bromide release to the atmosphere when it is injected into soil as a fumigant. We found that under typical agricultural conditions, about 30-40% of
applied methyl bromide reaches the atmosphere. The polyethylene tarp used to prevent release to the atmosphere is ineffective at preventing methyl bromide emission, because it dissolves into the film and diffuses through it. We showed that adding a small amount of irrigation water after the injection and before the tarp is put on substantially reduces the loss to the atmosphere, because the water reduces the diffusion rate. Further, water will not penetrate the tarp so that the barrier remains effective for as long as the tarp is on the field.
Impacts
(N/A)
Publications
|
Progress 01/01/93 to 12/30/93
Outputs
The bromide and simazine concentrations from the field plot experiment reported on last year were analyzed by a two-stage linear or nonlinear adsorption model which used laboratory measurements of the sorption and desorption loops of the isotherm, together with the field-measured bromide curve to predict the final simazine concentrations. In contrast to the single-stage model tested and reported on last year, this model predicted essentially perfect agreement with the data without any calibration or fitting procedures on the data itself. The two-stage model also showed that laboratory sorption hysteresis is an experimental artifact caused by using too short a period of shaking time to reach a true equilibrium. In a second study, a fundamental analysis was performed on stochastic-convective processes using a stream tube network as the framework for deriving initial and boundary value problems in a medium with stochastic velocity caused either by flux or water content
variations. Through the analysis it was shown that there are two types of stochastic-convective problem, depending on how the solute is applied, and that in either case it is possible to derive a relation between the pdfs of the initial and boundary value problems. Consequently, a single field experiment is sufficient to calibrate the entire process for an arbitrary input or boundary condition.
Impacts
(N/A)
Publications
|
Progress 01/01/92 to 12/30/92
Outputs
A field plot study was conducted in which solute and water fluxes were measured at high spatial resolution using a multi-cell solution sampler. This device was installed via a trench into an undisturbed soil layer through which water was added in a series of irrigations along with a single chemical application of bromide, simizine and MS2 virus. The study revealed a high degree of spatial variability in water flux at the 30-cm depth, even when the surface water application was essentially without spatial variation. Solutes tended to move vertically downward without experiencing lateral mixing or lateral transport. Bromide pulses in the effluent were used to construct a transfer function model which was subsequently used to project the spatial distribution of pesticide in the undisturbed soil layer. Results showed that the pesticides underwent rate-limited adsorption in the field. In a related study, a linear transport model was developed to predict rate-limited
pesticide transport using measured tracer distributions to monitor the water flow and independently measured chemical properties to predict pesticide flow. Comparison with data from a set of 36 experiments leaching chloride and pesticide through undisturbed soil produced better agreement between predicted and observed pesticide outflow than could be obtained from conventional simulation models.
Impacts
(N/A)
Publications
|
Progress 01/01/89 to 12/30/89
Outputs
In the past year the transfer function model for solute movement developed at UCR was extended to cover simulation of transient water flow. To achieve this extension, the convolution integral defining the steady-state model in terms of net applied water is modified by replacing the net applied water variable by cumulative drainage past the appropriate depth, and adding a water storage term to the impulse response function. To operate the model, one characterizes field-scale water transport in terms of area-averaged drainage and water content functions, with the drainage assumed to be a unique function of the water stored above it. The model was field tested by using the steady-state impulse response function together with a field scale water balance model constructed from local air-entry permeameter measurements to predict transient solute movement through the top 2 m of a 0.6j ha loamy sand field during a winter of erratic rainfall. Excellent agreement was found
between the area-average solution samples and the model predictions of concentration. In a second story, a transfer function model describing solute movement from the soil surface through the unsaturated and saturated zones to a tile drain was constructed. The model was used with various representations of the unsaturated zone to predict pesticide migration to tiles.
Impacts
(N/A)
Publications
|
Progress 01/01/88 to 12/30/88
Outputs
Data from the three-dimensional solute transport experiment reported during the previous year were analyzed using the method of spatial concentration moments. In all plots in which the soil was initially moist, the massive solute pulses moved downward with little observed lateral spreading. The piston flow model adequately described the position of the center of mass of the pulses at all locations in the field. A fine-textured soil layer between 3 and 4 m caused the pulse to contract in the vertical direction, thereby decreasing its dispersion coefficient. All components of the solute velocity and dispersion tensor were measured using concentration moments. Data from the pesticide field mass balance study reported last year were analyzed and interpreted using a screening model. Volatilization losses of all compounds agreed well with prediction, but leaching losses did not, because of preferential flow of pesticides within the top 1 M. A field study of the effect of
irrigation method and surface preparation on pesticide leaching was performed in the previous year. Atrazine, napropamide and prometryn were added to 1(subscript m)(superscript 2) plots and leached by flood or sprinkler irrigation. The pesticides were in three formulation states and the soil surface was either tilled or undisturbed. Data obtained by soil coring are currently being analyzed.
Impacts
(N/A)
Publications
|
Progress 01/01/87 to 12/30/87
Outputs
During the previous year a three dimensional solute transport study was conducted on 1.4 ha of a loamy sand field. In this study the field was pre-irrigated daily with high quality water for two weeks to establish a steady water flow condition, after which seven 1.5m x 1.5m plots and one 2m x 2m plot were irrigated daily with water containing high concentrations of Cl, NO(3) and B while the surrounding field was irrigated with high quality water at the same rate. After the saline plots received 20 cm of water, saline water input was terminated and the entire field was irrigated with fresh water every other day. The plots were sampled by soil coring with a grid of approximately 20 cores per plot, to a depth of 4 to 6m. The smaller plots were sampled once at various times during the experiment, while the larger plot was sampled three times. Data will be analyzed by the method of spatial concentration moments to determine solute velocity and macrodispersion tensor
components for each plot and for each tracer solute. In a second experiment, five pesticides were sprayed on the surface of two 4m x 4m plots and were monitored for volatilization by portable surface flux chambers and for persistance by replicated soil coring. During the 45 day experiment, the field was irrigated weekly and sampled four times by soil coring. Data are being analyzed to determine the loss pathways for the five pesticides.
Impacts
(N/A)
Publications
|
Progress 01/01/86 to 12/30/86
Outputs
During the previous year a 1.4 ha field study of bromide movement on a loamy sand was completed. In this experiment asingle concentrated pulse of KBr was added to the soil surface by sprinkler irrigation and was monitored over 8 months of bidaily irrigation averaging 1.5 cm/d with a network of solution samplers at 16 sites, 7 depths per site (0.3, 0.6, 0.9, 1.2, 1.8, 3.0, 4.5 m). Solute recovery at each depth ranged from 86 to 117% of the mass applied, and the apparent longitudinal macrodispersivity increased linearly between 0 and 3 m., and decreased between 3 and 4.5 m. A final replicated core sampling to 30 m showed continued evidence of convective movement not consistent with a constant longitudinal macrodispersion coefficient. Two one dimensional models, the convection dispersion equation CDE and the convective lognormal transfer function CLT were tested on the data set. The CLT consistently outperformed the CDE after calibration and adequately described pulse
shape through 3 m.
Impacts
(N/A)
Publications
|
Progress 01/01/85 to 12/30/85
Outputs
In the previous year, a major study was completed looking at the ability of mathematical screening models using chemical benchmark properties to predict the relative mobility of pesticides applied under field conditions. Fourteen plots 1m were sprayed with a mixture of 4 pesticides in 4 different mobility categories, together with potassium bromide which acted as a water tracer. The plots were irrigated over a 2 week period and were covered between irrigations. At the end of this period, a single core was taken to a depth of 3 m in the center of each plot and was analyzed in 10 cm increments for concentrations of each chemical. The bromide moved to the greatest depth followed in order by the chemicals in the respective mobility categories. However, the most strongly adsorbed compounds exhibited split behavior; part of their mass was located very near the surface as expected whereas the rest of the peak migrated to depths approaching those reached by the more mobile
chemicals. This result reinforced an earlier experiment conducted on a single pesticide in which movement to great depths was observed for a portion of the applied chemical across a 1.5 hectare field.
Impacts
(N/A)
Publications
|
Progress 01/01/84 to 12/30/84
Outputs
The pesticide transport experiment reported last year was completed and written up into a Ph.D. thesis. The principal findings of this study are: that batch equilibrium and flow through methods of measuring the distribution coeffcient do not correlate well on individual field samples; the two methods showed similar mean and variance values (N=36) over the field; the mean values predicted pesticide confinement to the top 20cm in the field experiment, whereas 22% of the applied pesticide was found between 30 and 180 cm. The second pesticide leaching experiment was performed during 1984 on the same field. In this experiment 5 chemicals of variable adsorbtion were sprayed onto 12 1my plots and leached for two weeks prior to soil core sampling. Analysis of the data is in progress.
Impacts
(N/A)
Publications
|
Progress 01/01/83 to 12/30/83
Outputs
The two field experiments reported earlier in this project, involving detection of soil applied bromide by solution samplers and detection of soil applied chloride by soil core samplers have been analyzed together to determine the relative accuracy of solution and soil core monitoring and to determine the best way to apply the model validation to spatially variable data. A further experiment has been conducted in the last year studying movement of bromide on a fine-textured soil in contrast to the coarse-textured soil where all previous experiments have been conducted. A theoretical pesticide screening model has been developed for use in determining the potential environmental danger of an organic chemical as a function of easily obtained laboratory benchmark properties. This screening model has been applied to 35 organic chemicals which have been classed into categories of relative volatility, mobility and persistence. A literature search has revealed significnt
experimental support for the predictions of the model. A field experiment leaching a mildly adsorbed pesticide has been partially analyzed. At the present time it appears that the leaching variability cannot be explained by taking into account variability in pesticide adsorption properties.
Impacts
(N/A)
Publications
|
Progress 01/01/82 to 12/30/82
Outputs
During the previous year a major field experiment was conducted on a 1.5-ha. sandy loam field to determine the spatial variability of adsorbed chemical movement. The adsorbed chemical, napropamide, was assumed to have a linear adsorption cefficient. This adsorption coefficient was obtained in three ways. Thirty-six batch equilibrium determinations on the 1.5-ha. field were made and taken back to the laboratory and the simultaneous leaching of chloride and napropamide pulses were used to define a dynamic adsorption coefficient. In addition a pulse of chloride and napropamide was applied to the entire field surface and a subsequent leaching was observed during two sampling periods. The ratio of leaching depths of the chloride and pesticide defined a third field adsorption coefficient. The data are currently being analyzed using statistical procedures such as autocorrelation and Kriging. In addition, the frequency distribution defined by the set of adsorption
coefficient measurements is being used in a stochastic transfer function model to predict pesticide movement from a knowledge of adsorption coefficients and water flow variability.
Impacts
(N/A)
Publications
|
Progress 01/01/81 to 12/30/81
Outputs
A stochastic transfer function model has been developed and tested on the field experiment at Etiwanda, California with good agreement between predictions and measurements of a single bromide pulse observed between 30 and 360 cm depth over one-half year. The model uses a variable travel time distribution and a piston flow to simulate solute movement past a given depth as a stochastic function of net applied water. Experiments now under way will test the utility of the model to predict absorbed chemical movement. Other experiments currently under way are looking at the comparison between a batch equilibrium and dynamic determination of the partition coefficient for absorption using a devrinol pesticide chemical on undisturbed field cores. This information, taken on 30 field cores, will be used to form a frequency distribution for absorption for use in the transfer function model experiment.
Impacts
(N/A)
Publications
|
Progress 01/01/80 to 12/30/80
Outputs
Experiments and modeling were continued on the field experiment studying the feasibility of reusing power plant cooling water for irrigation. The saline water, varying in concentration from 3000 to 5000 mg/l has been applied to 1.5 ha of sandy loam soil. Sixteen monitoring stations on the field are being used to follow the solute profile with time. Calibration experiments using bromide pulses have been used to characterize the spatial variability of solute movement and has served as the input to a stochastic model for predicting the variation in solute movement in space as well as in time. In separate studies, soil column experimemts have been conducted studying solute movement with and without the presence of roots. A second study, now near completion, is looking at the influence of temperature on phase-partitioning of pesticides.
Impacts
(N/A)
Publications
|
Progress 01/01/79 to 12/30/79
Outputs
The salt transport model developed in this study was used to make predictions ofthe initial and ultimate drainage salt burden of any project irrigating with saline water. The drainage salt burden was a function of irrigation water concentration and composition, irrigation management (leaching fraction) and soil properties. The model was also used to propose a strategy for achieving partial reclamation of a crop root zone by diluting the irrigation water. In a separate study, a new model for water vapor transport in soil was proposed. This model does not require measurement of a transport coefficient.
Impacts
(N/A)
Publications
|
Progress 01/01/78 to 12/30/78
Outputs
A theoretical model of solute transport including chemical reactions was appliedto the saline irrigation experiments conducted over the past two years in the UCR lysimeters. Salt precipitation and cation exchange reactions were satisfactorily represented in the model. Following the conclusion of the saline irrigation lysimeter experiments, seven of the lysimeters were selected for a reclamation experiment. Although initial exchangeable sodium percentages were as high as 60% near the surface, no permeability degradations were observed, even when distilled water was used as the leaching solution. Previously precipitated gypsum from the irrigation experiments maintained soil solution concentration high enough to prevent clay migration or swelling.
Impacts
(N/A)
Publications
|
Progress 01/01/77 to 12/30/77
Outputs
A third (Wheat) and fourth (sorghum) crop were grown in the lysimeters under thethree levels of synthesized saline irrigation water (2.0, 4.1, 7.1 mmho/cm). In the third crop, yields of all treatments were reduced from optimun values, and were below the corresponding yields of the first wheat crop. Correlation of yield with soil water salinity indexes showed only modest negative correlations (minus .5 to minus .7). The fourth crop was substantially reduced from optimum values due to a nitrogen deficiency which was not corrected until the third week after emergence. Salinity-yield correlations were very small. Root water uptake distribution has been moving slowly closer to the surface with succeeding crops, with over 50% of water extraction presently occurring within 5 cm of the surface. This is a consequence both of the high salinity irrigation water and the use of daily irrigation. The zone of maximum salinity is presently between 50 and 70 cm with little or no
water uptake occurring there, making it difficult to avoid leaching nutrients below the zone of extraction. /83 9885.
Impacts
(N/A)
Publications
|
Progress 01/01/76 to 12/30/76
Outputs
Winter wheat followed by grain sorghum was grown in twenty-four lysimeters containing four soil types irrigated with three levels of synthesized saline irrigation water (2.0, 4.1, 7.1 mmho/cm). Complete water and salt balances weretaken, permitting an estimate to be made of evapotranspiration and salt precipitation as a function of irrigation concentration and leaching fraction. There were no significant grain yield or evapotranspiration differences by treatment in the wheat experiment, indicating that the roots were extracting water from the higher quality water initially present in the soil profile. Substantial gypsum precipitation occurred in the upper profile of the high and medium salinity treatments. Sorghum grain yields were reduced 7% and 30% in themedium and high salinity treatments with similar reductions in water use. Gypsum precipitation was observed in all treatments. Leaching fractions in mostlysimeters were below 10% for the two crops, and the
saline profiles have not yet reached the bottom of the lysimeters (150 cm). The ionic composition of theleachate is changing with time, indicating that a quasisteady state has not beenreached for solute movement through the system.
Impacts
(N/A)
Publications
|
Progress 01/01/75 to 12/30/75
Outputs
Nine lysimeters covering four soil types have been equipped with tensiometers, solution extractors, neutron tubes, and a surface drip irrigation network. Capillary conductivity, soil suction, and soil water capacity values were determined in situ as a function of volumetric water content for each lysimeter. Two computer models of water flow, a finite difference solution of the flow equation and a vertical layer system, have been calibrated and tested against lysimeter measurements, with each simulation satisfactorily reproducing deep percolation rates. A model has been constructed to estimate the influence of solute travel time on effluent concentration of tile-drained fields. Variables required for graphical estimation are drain spacing, barrier depth, porosity, and cumulative discharge. A field experiment measuring heat and watermovement under surface rocks on a bare soil has been completed. The rocks were shown to be both a cover against evaporative loss and an
instrument for focusingheat and water vapor toward the soil under the rock. Related laboratory experiments and a computer calculation of heat and vapor movement were consistent with these findings.
Impacts
(N/A)
Publications
|
|