Year of Publication

2015

Season of Publication

Summer

Paper Type

Master's Thesis

College

College of Computing, Engineering & Construction

Degree Name

Master of Science in Civil Engineering (MSCE)

Department

Engineering

NACO controlled Corporate Body

University of North Florida. School of Engineering

First Advisor

Dr. Christopher J. Brown

Second Advisor

Dr. Donald T. Resio

Third Advisor

Dr. Peter Bacopoulos

Department Chair

Dr. Murat Tiryakioglu

College Dean

Dr. Mark A. Tumeo

Abstract

As hydrological computer modeling software continues to increase in complexity, the need for further understanding of the value of different model input datasets becomes apparent. Frequently used precipitation model input include rain gauge data and next-generation radar–based (NEXRAD) rainfall data. Rain gauge data are usually interpolated across a model domain using various methods including the Thiessen Polygon methodology, which may be data-sparse in some areas and overly data-dense in others. However, rain gauge data are generally very easy to use in hydrologic model development, often requiring little to no data processing. NEXRAD data have the potential to improve hydrologic runoff estimates due to the increased spatial resolution of the data: but has its own issues regarding accuracy, false precipitation indications, and difficulties due to data processing. Previous studies have investigated the value of NEXRAD input versus traditional rain gauge data inputs for hydrologic studies; however, results are inconclusive as to which precipitation source provides more accurate results. Limited work has been done to compare the value of these datasets at multiple spatial scales, especially in Florida, a study area dominated by low topographic drive and sub-tropical weather. In addition, little to no research has been done regarding the value of NEXRAD versus rain gauge data inputs at different rainfall return frequencies. The proposed research will utilize a hydrological rain-runoff model (HEC-HMS) of the Upper St. Johns River Basin, Florida to compare the performance of the two precipitation data input types at various watershed spatial scales and rainfall return frequencies. Statistical analysis of the hydrological model “goodness-of-fit” results will be utilized to assess the watershed scaling and rainfall frequency requirements to xii which NEXRAD data provide little to no advantage over standard rain gauges using the Thiessen Polygon method for estimating rainfall totals across a model domain.

Share

COinS