Rapid changes in environmental conditions cause the water level in various Bays to vary. Such variations can be attributed to several factors. The tide is usually considered the main contributor to water level. However, it is not the only factor. It is well-known that water height is affected by several factors. These factors in-clude, but are not limited to, freshwater discharge, rainfall, wind speed and direction, and groundwater runoff.

Conventional approaches for modeling the water height utilize regression-based methods to show how these different factors contribute to the water height. Time series analysis of water height data presents its own limitations.

This project introduces a novel concept of using wavelets and multiple levels of data decomposition to show, in much greater detail than regression, how these factors contribute to the water height.

Time series of water height data collected once per hour from Dauphin Island and the Mobile State Docks was used for this project. Tests with wavelet analysis demonstrate superior cause:effect correlation, than as compared to analysis on time series data.