This part of the exercise aims to develop two simple models that enable us to predict whether a given gaging station experienced flooding during the Hurricane Floyd period. The first model will use temperature and precipitation data to estimate streamflow. The second model will estimate gage height by using streamflow values estimated by the first model. We can then compare estimated gage heights with the flood stage for this given gage station to predict if this particular gaging station experienced flooding during Hurricane Floyd.

STEP 1: OBTAINING FLOOD STAGES

First, we need to find the flood stages for each station. A flood stage is simply the elevation at which overflow of the natural banks of a stream or body of water begins in the reach or area in which the elevation is measured.

We need to obtain flood stage numbers for the available stations that belong to the following rivers in North Carolina:

• Cape Fear River

• Deep River

• Fishing River

• French Board River

• Haw River

• Neuse River

• Roanake River

• Rocky River

• Tar River

• Yadkin River

The following map shows these rivers and gaging stations.

You can see a more detailed map with gage station IDs by clicking here. 

For simplicity, we have created an Excel spreadsheet with a template showing all of the available stations in these rivers.  You can download this spreadsheet file by clicking here. This spreadsheet, titled flood_stage_template.xls, has the flood stages for 37 stations, but we still don't have actual gage heights pertaining to the Hurricane Floyd period. Unfortunately, no gage height data is available for these stations for that period because the gages were overwhelmed by the event. That is why we will estimate the gage heights by using streamflow data.  To do this, we will create a model of that relationship.  A model is an approximation of the real system being described using mathematical relationships and defining, to the degree possible, the accuracy of our estimates.