Existing empirical evapotranspiration formulas were evaluated using climatic data for two consecutive years (1980 and 1981) from the Lower Colorado River Valley (Blythe, California). It was demonstrated that existing simple empirical models are inapplicable in arid zones where climatic ranges are wide. Extensive measurements of evapotranspiration rates using the Bowen ratio energy budget approach and selected climatological variables were made over well irrigated alfalfa sites in Avra Valley near Tucson, Arizona, for the season beginning in early May 1982 and ending in mid-October 1982. Forty-three days of 12-minute data gathered during the study were used as a basis for developing new models and for calibrating the Penman combination model. Several empirical models estimating evapotranspiration from various numbers of climatic variables were derived using a multiple linear regression technique. Their performance over the wide range of climatic conditions during the study prove the utility of certain empirical models for estimating evapotranspiration. The most promising is a simple solar radiation and wind model PET = a + bK + CU which uses only two climatic variables, incoming solar radiation (I%) and wind (U) which are widely available from routine weather measurements. The Penman combination model, with improvements suggested in literature, underestimated evapotranspiration as measured over irrigated alfalfa. A new arid zone wind function was developed using 12-minute data throughout the entire measurement period. The Penman model with the arid zone wind function performed very well throughout the season.

The accuracy of the geochemical model PHRQPITZ (Plummer et al., 1988), which uses the ion-interaction method (Pitzer, 1973) for the calculation of activity coefficients in predicting mineral solubilities, is evaluated. Furthermore, a method for estimating computational error in saturation indices for minerals in natural systems is developed. The published experimental solubility data for celestite (SrSO₄), barite (BaSO₄), fluorite (CaF₂), and calcite (CaCO₃) are evaluated and compared to calculated solubility data for the minerals in simple binary (mineral-H₂O) and ternary (mineral-NaCl-H₂O) systems. The solubility modeling error determined for minerals in these simple systems is then extrapolated to several examples of natural systems. The error in magnitude and direction gives a good estimate of the computational uncertainty in the mineral saturation indices calculated by PHRQPITZ for the natural systems. The only adjustment to PHRQPITZ is the addition of polynomial-based log K(sp) equations for the minerals being evaluated.

In this study, model parameters were estimated using five different watershed configurations for the SWRRB (A Basin Scale Simulation model for Soil and Water Resources Management) model, based on soil types and vegetation characteristics for watershed 11 in the Walnut Gulch Experimental Watershed in Tombstone, Arizona. Different watershed configurations were used, starting with one subbasin, up to a maximum of five subbasins. GIS was used to delineate the subbasins and to calculate some of the input parameters. Surface runoff was simulated for a period of six years (1982-1987). The simulated runoff values were compared with observed values. In addition, this study evaluated the effects of watershed configuration on runoff prediction by the SWRRB model.

Three electrical and electromagnetic geophysical methods were used to examine the moisture and permeability distribution and identify fractures in a partially saturated, fractured ash-flow tuff near Superior, Arizona. Experiments using ground penetrating radar (GPR) to locate targets with known orientations in the tuff were used to assess the angular limits of fracture detection from radar profiles. Identification of fractures from radar records appears to be limited to those with dips of less than 45°. The application of a single well electrical tracer (SWET) technique proved effective in identifying relative permeability in a borehole drilled in the tuff when compared to air permeability logs of the same borehole. Laboratory resistivity measurements of electrical resistivity were conducted to determine electrical resistivity/moisture saturation relationships. Twenty-seven tuff cores were measured at equilibrated suctions of 0 to 500 kPa. Resistivity values ranged from 50 to 1600 ohm-m.

Resolution of litigation between El Paso, Texas, and New Mexico over the water resources of the Mesilla Basin prompted a ground-water modeling study to address the common concerns of both parties. Participants requested development of a model which eventually could be used to simulate the ground-water and surface-water system's response to lining a portion of the canal system and to changes in pumping stresses within the Mesilla Basin. An existing U.S. Geological Survey ground-water flow simulation (Frenzel and Kaehler, 1990; Frenzel, 1992) was updated and revised through modification and incorporation of a new Streamflow-Routing (Prudic, 1989) package for use with MODFLOW (a three-dimensional finite-difference groundwater flow model; McDonald and Harbaugh, 1988). The modified Streamflow-Routing package was used to simulate the complex Mesilla Valley river, canal, and drain system. Updating also included incorporation of information on pumping stresses; canal, river, and drain parameters; Mesilla Valley boundary fluxes; and stream/aquifer interactions.

Random search methods are becoming more widely used to estimate model parameters. Their ability to globally search a parameter space makes them attractive for solving problems that have multi-local optima, as are non-linear hydrologic models such as Conceptual Rainfall-Runoff (CRR) models. The investigation of this thesis is on the ability of the Adaptive Random Search (ARS) to find the global optimum of the CRR model known as the Soil Moisture Accounting Model of the National Weather Service River Forecast System (SMA-NWSRFS) and compares its performance to that of the Uniform Random Search (URS). Research results indicate that, although the ARS was slightly more efficient than the URS, neither strategy demonstrated an ability to converge to the globally optimum parameter set. Factors which inhibit the convergence include model structure characteristics and an insufficient number of points searched. Ways for random search techniques to identify and address these problems are discussed.

The surface water composition of Pinal Creek has been affected by input of a low-oxygen ground water plume with high metal concentrations, particularly manganese and copper, and low pH. The pH of the stream water increases from approximately 6.0 to 7.8 downstream of the point of entry, which may affect the rate of manganese oxide precipitation. The purpose of this study was to determine the role of gas exchange processes on the observed downstream increase in pH at Pinal Creek. A finite-difference model was developed to calculate downstream changes in concentration of volatile solutes in small streams. Gas exchange rate constants used in the model were estimated using propane tracer techniques. Modeled simulations of dissolved oxygen and dissolved inorganic carbon transport in a 2.6 km reach at Pinal Creek, Arizona closely matched field observations. Based on the carbon balance and alkalinity balances, it can be concluded that downstream changes in pH can be sufficiently explained by acid/base chemistry of carbonate and CO2 degassing. Laboratory batch studies conducted in the laboratory using Pinal Creek water indicated that biological activity may be an important control on the precipitation of the black manganese oxides that were observed to coat the sediments in Pinal Creek.

Hell Canyon and Sycamore Canyon are major ungaged tributaries in the upper Verde River basin of central Arizona. Gage data implies that the record discharge of 1507 cms occurred on February 20, 1993 measured at the Verde River gage near Clarkdale, Arizona was derived primarily from these tributaries. 1993 flow reconstructions measure 800-900 cms in Sycamore Canyon and 600-700 cms in Hell Canyon. Historic and pre-historic units were exposed in various stratigraphic exposures in these canyons; as many as 11 floods are recorded at any one site. The 1993 floodwaters typically overtop all prior stratigraphy; however, dendrochronology suggests that similar floods occurred prior to the gage record. These results confirm that Hell Canyon and Sycamore Canyon are major contributors to floods on the Verde River in both the historic and paleoflood record.

Lane (2001) proposed that the flow from Lost Creek contributed to the discharge of Spring Creek Springs via an unverified conduit system similar to the verified conduit system that contributes flow to Wakulla Spring. In order to investigate this hypothesis two separate dye trace studies were conducted in 2008 and 2009 at Lost Creek swallet. Fluorescent dye was pumped by a peristaltic pump through tubing anchored below the surface of the water (15 meters in 2008 and 27 meters in 2009) directly into the throat of the Lost Creek swallet. Dye tracer sampling was conducted within the study area at select karst windows that had confirmed conduit access 38 meters to 68 meters below land surface. Sampling was also conducted at Wakulla Spring and Spring Creek Springs. Sampling at the karst windows was accomplished via pumps, timers and tubing to ensure that water samples were acquired from water within the conduits. In preparation for the 2009 Lost Creek swallet dye trace an array of water level elevation stations were established at seven selected karst windows along a north-south transect through the study area. The 2008 dye trace confirmed that flow from Lost Creek swallet contributed to discharge at Spring Creek Springs and Wakulla Spring. The velocities for first detection at Spring Creek Springs and Wakulla Spring were 2,235 m/d and 255 m/d respectively. These linear velocities confirmed a conduit connection between Lost Creek swallet, Spring Creek Springs and Wakulla Spring. The 2008 trace also confirmed detections at Revell Sink, a karst window, equidistant between Lost Creek swallet, Wakulla Spring and Spring Creek Springs. The 2009 dye trace confirmed that flow from Lost Creek swallet contributed to the discharge at three separate spring vents in the Spring Creek Springs with linear velocities of 248 m/d to 313 m/d. In 2009 tracer detections confirmed conduit connections at Revell Sink and Punchbowl Sink (two karst windows) along the water level elevation monitoring transect. The combined results of the 2008 and 2009 tracer tests confirmed that Lost Creek swallet and Spring Creek Springs are an extension of the Leon Sinks – Wakulla Spring cave system. The water level elevation data demonstrated reversals of the hydraulic gradient and reversals of groundwater flow between Wakulla Spring and Spring Creek Springs. Statistical analysis of the water level elevation data at the karst window stations revealed two primary groups of behavior among the karst window stations. The analysis also revealed behavior that supports the conclusion that during negative discharge conditions at Spring Creek Springs the karst windows stations responded as if they were connected to a unified conduit system. Whereas, during positive flow conditions at Spring Creek Springs the karst windows stations revealed water level elevation behavior that indicated there may be two sets of conduits that together form a basin wide unified conduit system. / A Thesis submitted to the Department of Earth, Ocean, and Atmospheric Science in partial fulfillment of the Master of Geology. / Fall Semester 2015. / November 5, 2015. / Conduits, Dye Tracing, Karst, Spring Creek Springs, Wakulla Springs, Woodville Karst Plain / Includes bibliographical references. / Steven Kish, Professor Directing Thesis; William Parker, Committee Member; Ming Ye, Committee Member.

To better understand the hydrochemical interaction between groundwater and lakewater in a mantled karst setting, a comprehensive study was conducted to determine the processes that control the patterns and rates of chemical evolution of groundwater near Lake Barco, a sinkhole lake in northern Florida. Environmental isotopes, chlorofluorocarbons (CFCs), solute tracers, and geochemical modeling techniques were used to determine flow patterns, origin and age of waters, and to quantify the net sources and sinks of major ions, dissolved silica, and dissolved carbon species. The stable isotopic composition $(\delta\sp2$H and $\delta\sp{18}$O) of rainfall and groundwater upgradient from the lake (sampled near the water table and at several depths below the water table) plot together along the global meteoric water line. In contrast, the $\delta\sp2$H and $\delta\sp{18}$O composition of lakewater and groundwater downgradient from the lake were enriched relative to meteoric water as a result of evaporation and mixing. The relation between $\delta\sp2$H and $\delta\sp{18}$O in groundwater downgradient from the lake can be described by the expression $\delta\sp2$H(per mil) = 4.76$\delta\sp{18}$O(per mil) $-$ 0.41 (r$\sp2$ = 0.992). A two end-member mixing model, developed to account for the enriched isotopic composition of groundwater, indicated that the amount of lakewater leakage that mixed with infiltrating meteoric water ranged from 11 to 67%, with a limit of detection of lakewater in groundwater of 4.3%. / Groundwater downgradient of Lake Barco was anoxic, with elevated concentrations of hydrogen sulfide, ferrous iron, and methane, resulting from leakage of lakewater through reducing, organic-rich sediments at the bottom of the lake. Based on the $\delta\sp2$H and $\delta\sp{13}$C content of methane, the dominant process for methane generation was by the carbon dioxide reduction pathway. Dissolved inorganic carbon species are being produced by reactions involving microbially mediated oxidation of organic matter along with three possible terminal electron accepting processes: ferric iron reduction, sulfate reduction, and methanogenesis. Rates of carbon dioxide production, determined from computed mass transfer for microbially mediated oxidation of organic matter and ages based on CFC-12 modeled recharge dates, were consistent with rates of microbial activity in other deeper aquifer systems. / This research study has provided a framework for a better understanding of recharge processes and the importance of both abiotic (mineral dissolution and precipitation) and biotic (microbially mediated degradation of organic matter) processes in controlling the cycling of solutes in a dilute groundwater-lakewater system. These results indicate that in areas where leakage of lakewater recharges shallow aquifer systems, flow patterns can be understood using changes in the concentration of major and minor chemical species as well as stable isotopes. / Source: Dissertation Abstracts International, Volume: 55-01, Section: B, page: 0063. / Major Professor: James B. Cowart. / Thesis (Ph.D.)--The Florida State University, 1993.