In-situ bioremediation is a promising biotechnology for removing aqueous phase contaminants from groundwater. Utilizing indigenous bacteria to degrade organic contaminants into non-toxic components, bioremediation is relatively inexpensive, fast, and complete. Making predictions about its applicability and success is difficult because of the complexity and variability intrinsic to the subsurface environment. Analytical studies of models, independent of this detailed subsurface data, are essential to finding accurate quantitative results, yet few have been obtained. This dissertation is a collection of three mathematical reports on a one-dimensional model for bioremediation. Using degree theory, the elliptic maximum principle, and comparison theorems, existence of traveling wave solutions to the biodegradation model is proved, a formula for the speed of the traveling concentration front is derived, and bounds on the biomass concentration are obtained. In the second section, the model is shown to reduce to a single equation in the relaxation limit by using properties of systems of hyperbolic conservation laws. In the third section, a formula is found for the parameters at which an unstable traveling wave solution bifurcates to a stable limit cycle (oscillatory solution). These results provide practical information about the structure of concentration fronts for the contaminant, nutrient, and biomass. The fronts travel at speeds that are either constant or time-periodic, depending on the kinetic parameters of the bacteria and the sorption properties of the contaminant. When there is little growth in biomass, many critical properties of the concentrations are derived. For aquifers with low permeability, the model is reduced to a much simpler system, also allowing the derivation of many analytical properties. Though comparisons with experimental data have not yet been done, numerical simulations support these results.

Vegetation change, particularly from the grass to shrub lifeform, is a critical issue on the world's rangelands. The plant community present on a site is the primary determinant of the land's value for watershed protection, wildlife habitat, livestock production, and recreation. Studying past vegetation composition can help separate natural from anthropogenic sources of change and guide natural resource conservation and management decisions. Stable carbon isotope (δ¹³C) values and associated radiocarbon ages from soil organic matter (SOM) were used to evaluate vegetation change across five landscape positions at a small enclosed basin in southeastern Arizona. The utility of the carbon isotope method was verified for this site based on the clear and wide separation in δ¹³C values between grasses having the C₄ photosynthetic pathway and shrubs having the C₃ pathway. The direction and timing of vegetation dynamics differed with landscape position along a gentle elevation gradient from the basin outlet to a nearby volcanic ridge top. Warm-season C₄ perennial grasses have dominated the basin outlet, center, and toe slope landscape positions since at least 5000-6000 yr BP, except for a dramatic increase in C₃ plants at the bottom of the outlet excavation around 5000 yr BP. This isotopic change is associated with rounded cobbles that may have been a stream channel, suggesting the presence of C₃ herbaceous or woody riparian vegetation. On mid-slope and ridge top landscape positions, where semidesert shrubs now dominate, warm-season perennial grass, composition decreased from approximately 60% as recently as 400 yr BP to only 1.5% now. SOM density separates were also analyzed. The youngest SOM is represented by the <2 g/cm³ density fraction that turns over in a few years to several decades and has a post-bomb radiocarbon age. For the ridge top landscape position, this fraction yielded 39% C₄ vegetation, suggesting that the conversion from grass to shrub vegetation occurred recently.

Understanding the extent of microbial transport, distribution and activity in the subsurface is paramount for effective in-situ bioremediation. In one study, we investigated the impact a substrate pulse has on the movement of inoculated or indigenous bacteria through saturated porous media. In another study, we developed a method to visualize the distribution of bacteria on soil surfaces. The elution of either inoculated or indigenous bacteria was monitored from model (homogenous) sand or natural (heterogenous) soil column systems. Sand columns receiving salicylate resulted in enhanced elution of inoculated P. putida. However, the salicylate pulse did not result in enhanced elution of P. putida from a natural system. For natural heterogenous systems, the salicylate pulse significantly affected the elution of certain indigenous bacteria. Specifically, more heterotrophs were eluted from soil columns receiving salicylate than from those that did not for both loamy sand soils tested. On the other hand, there were consistently fewer salicylate-degrading cells eluted in the presence of salicylate from one of the two soils tested. These data suggest that bacterial transport is a function of both the porous medium and the microbial population(s) under investigation. In the second study, an agar lift-DNA/DNA hybridization technique was developed to visualize the distribution of eubacteria on soil surfaces. Briefly, a single layer of soil was lifted from the surface of soil microcosms onto agar slabs and allowed to incubate. Bacterial colonies were lifted from the agar slabs onto membranes. The location of individual colonies was detected on the membranes by hybridization with a probe complementary to a conserved region of the eubacterial genome. This method was able to detect active microorganisms on different soil surfaces. The probe signal correlated well with the number of metabolically active microorganisms found in soils amended with a carbon source. This technique also allowed for visualization of localized microbial activity. A combined approach utilizing both soil column studies and the agar-lift technique should allow researchers to better elucidate microbial transport, distribution and activity in subsurface environments.

Soil venting, which includes gas injection as well as gas extraction in subsurface media, has become the primary method used in the United States to remove volatile organic compounds (VOCs) from unsaturated subsurface media. The popularity and widespread use of venting is due to its simplicity of operation and proven ability to remove contaminant mass inexpensively compared to competing technologies. Despite the common use of venting in the Superfund program, there is little consistency in approach to assessment of performance and closure. Assessment of the technology's performance and eventual decisions on closure are based primarily on negotiations between responsible parties and regulators. In this process there is widespread use and reliance on empirical methods as opposed to an emphasis on understanding fundamental physical, chemical, and biological processes controlling mass removal during the venting operation. This results in the technology not being used to its fullest potential, nor its limitations being well understood. The overall purpose of the work described in this dissertation was to improve the "state of the art" and "state of the science" of soil venting application. This purpose was accomplished by attainment of three specific objectives. The first objective was to develop an overall regulatory approach to assess venting performance and closure including measures to ensure consistency in ground-water and vadose zone remediation. The second objective was to provide comprehensive and detailed literature reviews on gas flow and vapor transport. These reviews formed the basis of recommendations and methods to improve venting design and monitoring. The third objective was to perform research to improve various aspects of venting application. This research consisted of: (1) analysis of linearization of the gas flow equation, (2) one-dimensional steady-state analysis of gas slippage, (3) two-dimensional steady-state analysis of gas flow and permeability estimation in a domain open to the atmosphere, (4) two-dimensional steady-state analysis of gas flow and permeability estimation in a semi-confined domain, (5) two-dimensional transient gas flow analysis and permeability estimation, (6) analysis and comparison of radius of influence versus critical pore-gas velocity based venting design, (7) modification of a gas extraction well to minimize water-table upwelling, (8) simulation of rate-limited vapor transport with diffusion modeling, (9) assessment of respiration testing, (10) development of a one-dimensional, analytical, vadose zone transport code to simulate mass flux to and from the capillary fringe, and (11) analysis of water-table mounding during sparging.

Three goals of this research were: (1) to study the movement of a non-reactive tracer (bromide) and of water through non-aggregated fine sand, with low clay (1.2%) and organic matter content (<1%); 2) to develop an inverse method for estimating the hydraulic properties of unsaturated soil from intermediate- and field-scale infiltration data; and (3) to develop a transient in-situ method for calibrating the neutron probe. All three goals of this research are crucial for understanding and determining water and solute transport at the intermediate to field scale. The field research was conducted in a field-scale research facility--the University of Arizona Superfund lysimeter (400-cm deep and 250-cm in diameter). We found equilibrium conditions, as evidenced by symmetrical bromide breakthrough curves (BTCs), from data collected during an unsaturated infiltration experiment in the lysimeter. Breakthrough of bromide, however, occurred sooner than was expected based on water arrival, and this observation is inconsistent with previous observations of other investigators. About 21% of the pore water (corresponding to approximately 0.03 cm³ cm⁻³) was found to be isolated from the bromide transport. We postulate that this inaccessible water partly existed as very thin films, adsorbed onto soil particle surfaces, and did not participate in anion transport. The combined effect of these films and of anion exclusion caused the bromide tracer to travel faster than the wetting front in this initially dry soil, because the excluded water fraction was larger than the initial water content. The soil hydraulic properties were estimated by an inverse method using in-situ data collected from this deep infiltration experiment. Soil hydraulic properties determined from laboratory experiments often are non-representative of field conditions. The inverse method developed in this study uses transient tension data during wetting of the profile, and the steady state water content found behind the wetting front. The results indicate that the method is fast and yields a unique estimate of the in-situ hydraulic properties at the field scale, without the need to collect excessive amounts of data. The neutron moisture meter used to determine the soil water content was calibrated using a newly developed transient mass balance method. The method was tested in the field scale lysimeter, and at a field site at the Agriculture Center, Maricopa, AZ. Water content errors using this method were less than 0.01 cm³ cm⁻³ for both sites. Application of the method to the lysimeter data showed excellent agreement between the soil water storage obtained using the calibration curve, and the actual volume of soil water added into the system.

The objective of this research was to evaluate the impact of bioaugmentation of soil with microorganisms harboring plasmid pJP4 on remediation, plasmid transfer, and plasmid dispersal. Divided into three sections, this research showed that use of microbial inocula harboring self-transmissible plasmids holds promise as an applicable bioremediation approach. In the first study, a pJP4 donor that could readily be counter-selected due to a lack of chromosomal genes necessary for 2,4-dichlorophenoxyacetic acid (2,4-D) mineralization was generated to allow detection of transconjugants in soil. Plasmid pJP4 was introduced into Escherichia coli (ATCC 15224), via plate mating with Ralstonia eutropha JMP134 to create such a donor (E. coli D11). Transfer of Plasmid pJP4 to diverse indigenous populations was detected in soils, and under conditions, where it had not been observed previously. Plexiglass columns were used in the second study to evaluate dissemination of plasmid pJP4 under unsaturated or saturated flow conditions in a 2,4-D contaminated soil. In unsaturated soil, pJP4 was detected in both culturable donor and transconjugant cells extending to 10.5 cm from the inoculated layer. In soil subjected to saturated flow conditions, no transconjugants were detected; however, donors were found throughout the entire length of the column (30.5 cm). Thus, donor transport in conjunction with plasmid transfer to indigenous recipients allowed for significant dissemination of introduced genes through contaminated soil. The last study was conducted using soil contaminated with 2,4-D alone or co-contaminated with 2,4-D and cadmium (Cd). This study assessed the impact of introduction of the pJP4 genes via cell bioaugmentation (R. eutropha JMP134 donor), or via gene augmentation (E. coli D11 donor). Both introduced donors remained culturable and transferred plasmid pJP4 to diverse indigenous recipients. Cell bioaugmentation resulted in the most rapid 2,4-D degradation; however, upon a second exposure to 2,4-D, gene augmentation of indigenous populations was more successful. The presence of Cd (100 μg g dry soil⁻¹) had a minimal impact on 2,4-D degradation and transconjugant formation. The establishment of an array of stable indigenous plasmid hosts may be particularly useful in sites with potential for re-exposure or extensive, and thus, long term contamination.

Forty percent of hazardous waste sites on the U.S. Environmental Protection Agency's National Priority List (NPL) are cocontaminated with organic and metal pollutants. Conventional approaches to remediating these sites are costly and often ineffective. Bioremediation is a promising, cost-effective alternative but metal toxicity at cocontaminated sites may limit its efficacy. The research described in this dissertation provides two new possible approaches to enhance the bioremediation of cocontaminated environments and sheds light on the relationship between metal concentration and inhibition of organic pollutant biodegradation. In Objective 1, a rhamnolipid biosurfactant was employed to increase naphthalene biodegradation in the presence of cadmium. The biosurfactant reduced bioavailable cadmium concentrations and increased naphthalene bioavailability. Neither of these phenomena, however, fully accounted for the ability of rhamnolipid to reduce cadmium toxicity. The ability of rhamnolipid to alter the cell surface appeared critical to its ability to mitigate toxicity. In Objective 2, pH was lowered to increase naphthalene biodegradation in the presence of cadmium. Reductions in pH had previously been reported to mitigate metal toxicity, but the mechanism of such reductions warranted elucidation. Previous studies implicated the formation of monovalent hydroxylated metal in the mechanism by which pH mediates toxicity. Results of this study, however, suggest that the importance of such species in determining toxicity may be much less than that of the increased competition between hydrogen and metal ions for binding sites on the cell surface at reduced pH. An indirect relationship between metal concentration and inhibition of organic biodegradation was revealed in Objective 3. Naphthalene biodegradation was more sensitive to cadmium concentrations of 10 and 37.5 mg/L than 100 mg/L. For this reason, we investigated whether naphthalene biodegradation could be increased in the presence of a toxic concentration of cadmium by raising the total metal concentration to a higher, but relatively less toxic concentration. Only elevated concentrations of zinc reduced cadmium toxicity. High but less toxic levels of metal may more rapidly induce the transcription of a gene(s) important in metal efflux than lower more toxic concentrations.

A series of three experiments were conducted to determine the diversity of indigenous chlorobenzoate (CB) degraders in soil and to investigate the use of different methods of bioaugmentation for remediation of contaminated soil. In the first study, soil was amended with either 500 or 1000 μg of 3-CB g⁻¹ and was either uninoculated or inoculated with the 3-CB degrader Comamonas testosteroni BR60. Bioaugmentation with C. testosteroni BR60 increased 3-CB degradation at both contaminant levels, and the increase was more pronounced at the higher level due to contaminant inhibition of indigenous 3-CB degraders. Bioaugmentation also appeared to reduce the deleterious effects that 3-CB contamination had on indigenous soil microbial populations as evidenced by changes in culturable heterotrophic bacterial populations. In the second study, two similar pristine soils were contaminated with 500 μg of 2-, 3-, or 4-CB g⁻¹ . The two soils differed in their ability to degrade the compounds with one degrading 2- and 4-CB and the other degrading 3- and 4-CB. Several hundred degraders were isolated, grouped according to DNA fingerprints, and selected degraders were identified by 16S rDNA sequences. The identity of the CB degraders differed between the two soils. The results indicated that the development of 2-, 3-, and 4-CB degrader populations was site-specific even for the soils that developed under similar soil-forming conditions. The third study also used the two soils from the second study. This project investigated the potential for use of activated soil, which contained an indigenous degrader population, as a bioaugmentation inoculant. An aliquot of a given soil that contained an indigenous 2-, 3-, or 4-CB degrader population was added to a soil that did not have an indigenous degrader population for the same contaminant. The study found that bioaugmentation with activated soil increased degradation of each 2-, 3-, and 4-CB but only if the activated soil was pre-exposed to the contaminant prior to use for bioaugmentation. The results from these three studies indicate that CB degrader populations are diverse and variable in pristine soils and, if not present in contaminated soils, appropriate degrader populations may be established via different bioaugmentation strategies.

The toppling of trees forms mounds of disturbed sediment and pits from which the mound removes sediment, rocks, and organic matter. Sites of uprooted trees in Puerto Rico and Colorado were examined (1) to compare areas and volumes of mounds and pits relative to tree size, (2) to compare areas and volumes of mounds and pits formed during catastrophic events at the landscape scale, and (3) to consider decay of mounds and pits after formation. For a given basal area, the analyses found no difference among sites in area and volume of freshly formed individual mounds and pits. For landscape-level catastrophic uprooting, the percent of toppled trees in a plot can explain 85% and 87% of the areas and volumes, respectively, of the quantity of soil uplifted. Exponential decay coefficients developed by monitoring mound/pit complexes indicate that mounds and pits at the humid tropical site in Puerto Rico decay in about 74% and 57% of the time, respectively, of mounds and pits at a temperate Colorado site. Decay coefficients developed for the Colorado site indicate that mounds and pits are reduced to 10% of their original volume within 30 and 78 years, respectively. Coefficients for Puerto Rico suggest that a similar reduction in volume requires 17 years, whereas pits generally fill within a decade.

The encroachment of deeply rooted woody plants into grasslands throughout the world has the potential to alter local, regional, and global water balance. The consequence of encroachment by woody plants on ecosystem water balance is, in part, related to the sensitivity of these plants to summer and winter precipitation pulses. This dissertation addresses the primary question: does pulse sensitivity of a dominant warm-desert woody plant, velvet mesquite (Prosopis velutina Woot.) vary across soil texture and water availability gradients? To address this question, sap flow and xylem anatomy and function were evaluated in mature velvet mesquite trees at two upland sites varying in soil texture at the Santa Rita Experimental Range (SRER), and one floodplain site along the San Pedro River National Conservation Area (SPRNCA). Experimental irrigation was used to assess the sensitivity of mesquite plants to small and large precipitation pulses. There was a moderate response to both small (10 mm) and large (35 mm) irrigation inputs by trees on sandy-loam soil, while trees on loamy-clay soil were only responsive to the large pulse. The differential response between sites was associated with differences in infiltration of the experimental pulses between the two soil types. Model predictions of the critical transpiration rate (Ecrit )--above which hydraulic conductivity through the soil-plant continuum falls to zero--showed that trees at the sandy-loam site operated well below their maximum transpiration rate before the onset of the monsoon. Conversely, plants on loamy-clay soils likely operate closer to their maximum permissible transpiration rates throughout the growing season. Hydraulic redistribution was observed and rates were tightly coupled to growing season and dormant season precipitation inputs. Hydraulic redistribution could enhance pulse sensitivity by transferring soil water to regions of the root zone that are otherwise dry, thereby allowing a greater proportion of the root system to participate in the extraction of pulse water during transpiration. Results from this research suggest that patterns of mesquite water relations are strongly mediated by soil texture. Nevertheless, once established, mesquite plants substantially modify ecosystem water balance, due to their responsiveness to growing season precipitation pulses, and their ability to withstand severe water deficits between precipitation pulses.