Hydrologic models

Three major teleconnections, Atlantic Multidecadal Oscillation (AMO), North Atlantic
Oscillation (NAO), and the Pacific Decadal Oscillation (PDO), in warm and cool phases,
effect precipitation in Florida. The effects of the oscillation phases on the precipitation
characteristics are analyzed by using long-term daily precipitation data, on different
temporal (annual, monthly, and daily) and spatial scales, utilizing numerous indices, and
techniques. Long-term extreme precipitation data for 9 different durations is used to
examine the effects of the oscillation phases on the rainfall extremes, by employing
different parametric and non-parametric statistical tests, along with Depth-Duration-
Frequency analysis. Results show that Florida will experience higher rainfall when AMO
is in the warm phase, except in the panhandle and south Florida, while PDO cool phase is
positively correlated with precipitation, except for the southern part of the peninsula.

Climate models are common tools for developing design standards in the hydrologic field; however,
these models contain uncertainties in multi-model and scenario selections. Along with these uncertainties,
biases can be attached to the models. Such biases and uncertainties can present difficulties in predicting
future extremes. These hydrologic extremes are believed to be non-stationary in character. Only in the
recent past have model users come to terms that the current hydrologic designs are no longer relevant due
to their assumption of stationarity. This study describes a systematic method of selecting a best fit model in
relationship to location and time, along with the use of that best fit model for evaluation of future extremes.
Rain gage stations throughout Florida are used to collect daily precipitation data used in extreme precipitation and quantitative indices. Through these indices conclusions are made on model selection and
future extremes, as they relate to hydrologic designs.

Environmental simulations are computationally very demanding. South Florida Water Management District has implemented the Everglades Landscape Model (ELM) to simulate the ecosystem in South Florida. The current implementation parallelizes all of the model except the canal system. This thesis describes the parallelization of the canal system. We study the canal system and its interaction with the rest of the ELM, and created three distinct parallel implementations. Two of the methods, one-do-all and all-do-all, provide parallelism via task replication while the third method, task-parallel, decomposes the canal system into tasks and uses a locality-based heuristic algorithm to schedule the tasks. We analyze the performance of three methods and discuss future directions for parallelization of the ELM and other environmental models.

Natural and anthropogenic processes have altered wetland habitats. The simulation of surface water movement and its interaction with groundwater and slough channels as it relates to wetlands is very important for many projects. Currently, most groundwater flow models incorporate the wetland system as general head boundary nodes. The purpose of this research was to develop a computer package for the widely used MODFLOW code to simulate three-dimensional wetland flow hydroperiods interacting with aquifers and slough channels. The groundwater flow model was used to reproduce surface water flow process through wetlands, estimating new flow rates and values using a Manning type of equation. This package represents flow routing, the export/import of water, and the evapotranspiration from wetlands during different hydroperiods. The verification procedure for the numerical solution was based on a test-case that was solved using a two-dimensional surface water model. This test-case example is a transient solution to the diffusion equation starting with initial conditions depicted by a sinusoidal water surface profile and a flat bottom.

The South Florida Water Management Model was developed to evaluate proposed alternatives for the south Florida regional hydrologic system. The degree of certainty of the computed system performance measures is required to correctly apply these measures for evaluation and selection of appropriate water resources policies and investments. Initially, a sensitivity matrix is defined which summarizes the model output sensitivity to incremental changes of key parameters. The method of singular value decomposition is applied to the sensitivity matrix to better understand relations between parameters and output variables. Finally, parameter uncertainty is compared to that of total predictive uncertainty of the system performance measures.