Hurricanes > Lab 3: Putting Hurricanes on the Calendar > 3C: HURDAT Exploration

Putting Hurricanes on the Calendar

Part C: Further Exploration

What Else Do the Data Say? Launching Points for Inquiry

Choose one of the following questions, or a question of your own that you are wondering about after working with the HURDAT data. Use your spreadsheet manipulation and analysis skills to answer it. You can copy the HURDAT data and paste it into a new spreadsheet as many times as you like to do your analysis. Use spreadsheet commands such as Text to Columns, Sort, and/or COUNTIF to see what else you can learn from the data.

HURDAT data are also available in an alternate, easy-to-read version that includes additional labels. A guide for interpreting the alternate format is also available. Consider how you could sort and graph data from this version of the dataset to answer questions about hurricanes.

How well does HURDAT represent reality?

Atlantic Named Storms, 1900-2005. The black line represents the long-term average annual number of reported hurricanes. HURDAT data graphed by Chris Landsea, used with permission.

If you worked with the HURDAT data to examine the number of storms reported per year and came up with a graph that increased over time, could you be sure that the number of tropical storms was actually increasing? Whenever you use a database of information to help answer a question, it's important to consider what biases may be present in the data…

Recall that when the National Weather Service began keeping HURDAT records in 1851, the only tropical storms that were entered into the database were the ones that had been detected by people on land or by ships at sea. Today, a broad array of sensors on satellites, aircraft, and buoys continually monitor the entire Atlantic basin to detect tropical storms.

Before satellite observations began in 1966, some stormsthose that stayed far out in the open ocean, for instancelikely went uncounted. Therefore, comparing annual counts of tropical storms detected before satellites to the number of storms detected each year since satellite monitoring began may not be a fair comparison.

Dr. Christopher Landsea, Science and Operations Officer at the National Hurricane Center, published an article in 2007 detailing his analysis of the biases in the HURDAT database. Counting Atlantic Tropical Cyclones Back to 1900 provides several graphs and examples to compare hurricane counts from different time periods. From his analyses, he has deduced specific values by which storms may have been undercounted during different periods. He argues that the increase in the annual number of storms shown in HURDAT reflects increased monitoring capabilities rather than a change in the climate in which they develop.

Some scientists have used HURDAT data to link Atlantic hurricane trends to climate change. They disagree with Dr. Landsea's assumption that the long-term annual number of tropical cyclones has remained stable over time.

Optional Extension

Follow the real-life scientific drama over the accuracy of HURDAT by downloading and reading papers by Mann and Emanuel (2006), Landsea (2007), and Mann and Emanuel (2007). Use a spreadsheet program to perform HURDAT analyses that the papers use to illustrate their points. Follow up on more recent work by these and other meteorologists to decide if the frequency of tropical storms has changed over time.

Why is HURDAT so important?

Hurricane-related damage in Florida. Photo courtesy of NOAA.

Communities use information from the HURDAT database in setting building codes, assessing risk for emergency managers, and calculating potential losses for insurance and business interests. Additonally, scientists use the database to check their forecasting techniques, verify their predictions, and to study changes that may be attributed to climate change.

The accuracy of the data is important enough that NOAA makes continual efforts to assess and improve the database. For example, some scientists have contributed to the effort by investigating leads in historical documents to help them identify storms that may have been missed. Other efforts focus on the accuracy of storms' paths. Read more about the HURDAT Re-Analysis project to learn how changes are decided upon and implemented to this important collection of data.


Stop and Think

4. Describe an additional data exploration you made using the HURDAT data. Include a graph or chart and a written interpretation of your results.
5. Describe the importance of understanding bias within a dataset. Suggest at least one solution that could remove sampling bias from the data, and describe what would be lost in implementing that solution.

« Previous Page      Next Page »