In [ ]:
%%html
<!--Script block to left align Markdown Tables-->
<style>
  table {margin-left: 0 !important;}
</style>

Download (right-click, save target as ...) this page as a jupyterlab notebook from: Lab24

Laboratory 24: "Predictor-Response Data Models"

LAST NAME, FIRST NAME

R00000000

ENGR 1330 Laboratory 24

Exercise: Watershed Response Metrics

Background

Rainfall-Runoff response prediction is a vital step in engineering design for mitigating flood-induced infrastructure failure. One easy to measure characteristic of a watershed is its drainage area. Harder to quantify are its characteristic response time, and its conversion (of precipitation into runoff) factor.

Study Database

The watersheds.csv dataset contains (measured) drainage area for 92 study watersheds in Texas from Cleveland, et. al., 2006, and the associated data:

Columns Info.
STATION_ID USGS HUC-8 Station ID code
TDA Total drainage area (sq. miles)
RCOEF Runoff Ratio (Runoff Depth/Precipitation Depth)
TPEAK Characteristic Time (minutes)
FPEAK Peaking factor (same as NRCS factor)
QP_OBS Observed peak discharge (measured)
QP_MOD Modeled peak discharge (modeled)

:

Using the following steps, build a predictor-response type data model.

Step 1:

Read the "watersheds.csv" file as a dataframe. Explore the dataframe and in a markdown cell briefly describe the summarize the dataframe.

In [1]:
# import packages
# read data file
# summarize contents + markdown cell as needed

Step 2:

Make a data model using TDA as a predictor of TPEAK ($T_{peak} = \beta_{0}+\beta_{1}*TDA$)
Plot your model and the data on the same plot. Report your values of the parameters.

In [2]:
#

Step 3:

Make a data model using log(TDA) as a predictor of TPEAK ($T_{peak} = \beta_{0}+\beta_{1}*log(TDA)$)

In your opinion which mapping of TDA (arithmetic or logarithmic) produces a more useful graph?

In [3]:
#