In this data lesson, we explore and visualize stream discharge time series data collected by the United States Geological Survey (USGS). You will use everything that you learned in the previous lessons to create your plots. You will use these plots in the report that you submit for your homework.
Note: this page just shows you what the plots should look like. You will need to use your programming skills to create the plots!
After completing this tutorial, you will be able to:
- Plot USGS Stream Discharge time series data in
What you need
RStudio to complete this tutorial. Also you should have an
earth-analytics directory setup on your computer with a
/data directory with it.
R libraries to install:
If you haven’t already downloaded this data (from the previous lesson), do so now.
About the data - USGS stream discharge data
The USGS has a distributed network of aquatic sensors located in streams across the United States. This network monitors a suit of variables that are important to stream morphology and health. One of the metrics that this sensor network monitors is Stream Discharge, a metric which quantifies the volume of water moving down a stream. Discharge is an ideal metric to quantify flow, which increases significantly during a flood event.
As defined by USGS: Discharge is the volume of water moving down a stream or river per unit of time, commonly expressed in cubic feet per second or gallons per day. In general, river discharge is computed by multiplying the area of water in a channel cross section by the average velocity of the water in that cross section.
As you can imagine, stream gages can be sensitive to high flows and in the case of an extreme event like a flood are sometimes damaged. However, during the 2013 floods, one stream gage in Boulder, Colorado remained in tact. USGS stream gauge 06730200 located on Boulder Creek at North 75th St. collected data that we will use in the lesson below!
Work with USGS stream gage data
Let’s begin by loading our libraries and setting our working directory.
# set your working directory # setwd("working-dir-path-here") # load packages library(ggplot2) # create efficient, professional plots library(dplyr) # data manipulation # set strings as factors to false options(stringsAsFactors = FALSE)
Import USGS stream discharge data into R
Let’s first import our data using the
discharge <- read.csv("data/week_02/discharge/06730200-discharge-daily-1986-2013.csv", header=TRUE) # view first 6 lines of data head(discharge) ## agency_cd site_no datetime disValue qualCode ## 1 USGS 6730200 10/1/86 30 A ## 2 USGS 6730200 10/2/86 30 A ## 3 USGS 6730200 10/3/86 30 A ## 4 USGS 6730200 10/4/86 30 A ## 5 USGS 6730200 10/5/86 30 A ## 6 USGS 6730200 10/6/86 30 A
Now that the data are imported, plot
disValue (discharge value) over time. To do this, you will need to use everything that you learned in the previous lessons.
Hint: when converting the date, take a close look at the format of the date - is the year 4 digits (including the century) or just 2? Use
?strptime to figure out what format elements you’ll need to include to get the date right.
Your plot should look something like the one below:
Similar to the previous lesson, take the cleaned discharge data that you just plotted and subset it to the time span of 2013-08-15 to 2013-10-15. Use
dplyr pipes and the
filter() function to perform the subset.
Plot the data with
ggplot(). Your plot should look like the one below.
Additional information on USGS streamflow measurements and data:
- Find peak streamflow for other locations
- USGS: How streamflow is measured
- USGS: How streamflow is measured, Part II
- USGS National Streamflow Information Program Fact Sheet
API data access
USGS data can be downloaded via an API using a command line interface. This is particularly useful if you want to request data from multiple sites or build the data request into a script. Read more here about API downloads of USGS data.