Seasonally snow covered alpine areas play a larger role in the hydrologic cycle than their area would indicate. Their ecosystems may be sensitive indicators of climatic and atmospheric change. Assessing the hydrologic and bio-geochemical responses of these areas to changes in inputs of water, chemicals and energy should be based on a detailed understanding of watershed processes. This dissertation discusses the development and testing of a model capable of predicting watershed hydrologic and hydrochemical responses to these changes. The model computes integrated water and chemical balances for watersheds with unlimited numbers of terrestrial, stream, and lake subunits, each of which may have a unique, variable snow-covered area. Model capabilities include (1) tracking of chemical inputs from precipitation, dry deposition, snowmelt, mineral weathering, baseflow or flows from areas external to the modeled watershed, and user-defined sources and sinks, (2) tracking water and chemical movements in the canopy, snowpack, soil litter, multiple soil layers, streamflow, between terrestrial subunits (surface and subsurface movement), and within lakes (2 layers), (3) chemical speciation, including free and total soluble species, precipitates, exchange complexes, and acid-neutralizing capacity, (4) nitrogen reactions, (5) a snowmelt optimization procedure capable of exactly matching observed watershed outflows, and (6) modeling riparian areas. Two years of data were available for fitting and comparing observed and modeled output. To the extent possible, model parameters are set based on physical or chemical measurements, leaving only a few fitted parameters. Thc effects of snowmelt rate, rate of chemical elution from the snowpack, nitrogen reactions, mineral weathering, and flow routing on modeled outputs are examined.

Water level recovery data collected in wells following controlled pumping tests are affected by both borehole and formation factors. The borehole factors comprise those effects attributed to the presence of the wellbore, such as step-increases in pumping rate, wellbore storage, well efficiency, and skin effects. The formation factors comprise those effects associated with the geologic environment in which an aquifer system occurs, such as variation of the coefficient of storage, and aquifer barrier boundaries. The recovery data should plot as a straight line on a semilogarithmic plot. Step-increases in the discharge rate during the pumping period cause the water level recovery plot to be concave downward. The curvature of the recovery data plot can be eliminated by applying a correction proposed by Harrill in 1970. However, the effect of step-increases in pumping rate on the recovery data is minimal so long as the duration of the pumping steps is less than about one-third of the total duration of pumping. The well efficiency and skin effects cause an additional component of drawdown in a pumped well, which is manifested as an initially rapid recovery rate after pumping stops. The effects of skin and well efficiency are usually dissipated within a few minutes after pumping stops. Wellbore storage effects can be critical in large diameter wells (wellbore radius greater than 0.5 feet) that penetrate aquifers with transmissivities less than about 2,700 feet squared per day. The time required to dissipate wellbore storage effects in the water level recovery data is inversely proportional to the aquifer transmissivity, and directly proportional to the borehole size. Variation of the coefficient of storage during the recovery period results in a semi-logarithmic recovery plot that is concave downward. The curvature of the recovery plot increases as the variation of the coefficient of storage increases. Variation in the coefficient of storage of one order of magnitude during the recovery period introduces an error of more than fifty percent in the transmissivity calculation at late recovery times. The recovery plot of data collected in a well influenced by a barrier boundary defines two straight line segments. The early-time straight line segment has a slope one-half that of the late-time straight line segment. Analysis of the early-time straight line yields the true aquifer transmissivity. Analysis and interpretation of water level recovery data collected in 59 wells following controlled pumping tests in aquifers of various rock types indicate that, in general, the shape of the recovery plot can be used to diagnose the presence of skin effects, low well efficiency, wellbore storage, and variation of the coefficient of storage. Analysis of data from seventeen wells in alluvial aquifers and thirteen wells in sandstone aquifers indicates that the concave downward recovery plot is the most common type of response curve. This shape of recovery curve indicates that the coefficient of storage is commonly smaller during the recovery period than during the drawdown period. Recovery data collected in twenty wells in fractured hard-rock aquifers indicate that the characteristic shape of the recovery plot predicted by Warren and Root in 1963 is generally diagnostic of flow in non-homogeneous, anisotropic, fractured aquifers. When the fracturing approaches being homogeneous and isotropic, the recovery plot can resemble data collected in non-fractured rocks. Recovery data from nine wells in composite limestone-sandstone aquifers indicate that the recovery plot is sometimes similar to the concave downward shape exhibited in sandstone and alluvial aquifers, and sometimes is similar to the shape predicted by Warren and Root for fractured rocks.

Evapotranspiration rates of plots with vegetative cover and evaporation rates from bare soil differed during the active growing season of desert hackberry (Celtis pallida) plants but total water losses from both plots for the year were the same. Thermally induced vapor flux appeared to contribute insignificantly to moisture movement under the desert hackberry plant. The difference in measured available soil moisture was independent of location from the plant center during the growing season. During the winter months, when the plants were semidormant, soil moisture measurements had more variability and measurement locations appeared to be important due to differential rainfall input. The determined soil moisture release curve and soil water conductivity values (using an in situ technique) appeared to be representative of the conditions at the study site. A model using soil and plant parameters predicted evapotranspiration rates during the active growing season of the plants when water was not a limiting factor. Calculated results using the model were unreliable when plants were under stress -- very low soil water content. Monitoring of climatic parameters delineated only major differences in surface albedo and net radiation between plant cover and bare ground. Potential evapotranspiration estimations were high but within acceptable bounds for desert conditions. Plant diffusion resistance for the desert hackberry plant, determined from a climatological model and measured soil moisture changes, appeared to increase linearly with decreasing soil moisture until it reached a critical value, below which it rose sharply.

Infrared (IR) imagery collected by geostationary satellites provides useful information about the dirunal evolution of cloud systems. These JR images can be analyzed to indicate the location of clouds as well as the pattern of cloud top temperatures (Tbs). During the past several decades, a number of different approaches for estimation of rainfall rate (RR) from Tb have been explored and concluded that the Tb-RR relationship is (1) highly nonlinear, and (2) seasonally and regionally dependent. Therefore, to properly model the relationship, the model must be able to: (1) detect and identify a non-linear mapping of the Tb-RR relationship; (2) Incorporate information about various cloud properties extracted from IR image; (3) Use feedback obtained from RR observations to adaptively adjust to seasonal and regional variations; and (4) Effectively and efficiently process large amounts of satellite image data in real-time. In this study, a kind of artificial neural network (ANN), called Modified Counter Propagation Network (MCPN), that incorporates these features, has been developed. The model was calibrated using the data around the Japanese Islands provided by the Global Precipitation Climatology Project (GPCP) First Algorithm Intercompari son Project (AIP-I). Validation results over the Japanese Islands and Florida peninsula show that by providing limited ground-truth observation, the MCPN model is effective in monthly and hourly rainfall estimation. Comparison of results from MCPN model and GOES Precipitation Index (GPI) approach is also provided in the study.