GEOG 4/521:  Advanced Climatology
Climate Anomaly Analysis Project

Overview

This document describes the climate anomaly analysis project.  The aim of the project is twofold:  1) to gain experience in using on-line data sources to learn about recent climate anomalies, and 2) to understand the structure of those anomalies, i.e. how atmospheric circulation controls the nature of the climate anomaly at a particular location or region.

The analysis of climate anomalies is quite straightforward, and consists of three steps: 

1. the identification of the the key years to be composited,
2. the construction of composites (usually expressed as maps) of one or more climate variables for those key years, usually in terms of anomalies or comparison with long-term averages of "normals", and
3. the interpretation of the resulting composites.

The key years or cases are usually identified by examining a time series that describes the variation of a climate variable of interest.  There are two basic kinds of variables used in such an analysis: 

1. variables that describe a particular response of climate, for example, the temperature or precipitation that occurs at a station or within a climate region, and
2. variables that describe state of a particular control (either proximate or ultimate) of regional climate anomalies (e.g. Pacific SSTs).

Other information can also be used to identify the cases used to construct the composites , like a list of the years and months in which, say, large landslides occurred in western Oregon.  Typically, a set of positive cases (e.g. the five warmest years at a specific location) are contrasted with a set of negative (or neutral) cases (e.g. the five coldest years).

The practical aspects of constructing composite anomalies are also straightforward, given the web resources provided by NOAA's Earth System Research Laboratory in Boulder.  It is important to carefully follow the directions and review the selections on the web pages to make sure the appropriate maps and time series are being generated.

Sources of Information

• ERSL =  NOAA Earth System Research Laboratory, Physical Sciences Division, Climate Analysis Branch (formerly the Climate Diagnostics Center, CDC) http://www.esrl.noaa.gov/psd/ 
  ERLS/PSD Interactive Plotting and Analysis web page http://www.esrl.noaa.gov/psd/cgi-bin/data/getpage.pl  
• CPC = NOAA Climate Prediction Center http://www.cpc.ncep.noaa.gov/
  CPC Monitoring and Data Index http://www.cpc.ncep.noaa.gov/products/MD_index.shtml
• NCDC = National Climate Data Center http://www.ncdc.noaa.gov/oa/ncdc.html
  NNDC Climate Data Online http://www.ncdc.noaa.gov/oa/climate/onlineprod/drought/xmgr.html
• WRCC = Western Regional Climate Data Center http://www.wrcc.dri.edu/

Time Series

Time Series of Climate Responses (e.g. Temperature, Precipitation, PDSI)

• Interactive Plotting and Analysis Pages (ERSL)
  • Create a Monthly or Seasonal Time Series of Climate Variables  (ERSL)
• Plot time series history of monthly divisional precipitation or temperature (WRCC)
  • WESTMAP Western U.S. times series
• NOAA National Data Center / National Climate Data Center (NCDC)
  • NNCD Climate Data Online
  • U.S. Divisional Data and Global Summaries

Time Series of Atmospheric circulation and SST Indices (e.g. SOI, PNA, NAO, etc.)

• Monthly atmospheric & SST indices (CPC)
• Northern Hemisphere teleconnection patterns (CPC)
• Climate Indices: Monthly Atmospheric and Ocean Time Series (ERSL)
• U.S. Nation, Region, State, and Division time series (NNDC)
• GCOS -- Global Climate Observing System (NCDC)

Maps

ESRL On-Line Maps

• Interactive Plotting and Analysis Pages (ERSL)
  • Monthly/Seasonal Mean Composites
  • US Climate Division Maps
  • The Global Climate during El Niño and La Niña

Text

Data bases

• NCDC Storm Events Data Base

Example

What kind of upper-level circulation pattern favor dry Januaries  in the Willamette Valley?

To answer this question, we'll examine the precipitation January for the Willamette Valley Climate Division to identify the five driest years in the record, and then produce maps of the 500mb height, vertical velocity and vector-wind anomalies that occur during those dry years.

Before beginning, you might want to start up Word to provide a document that you can paste images into to save them.

Step 1:  Get and plot a time series of precipitation

(It may be helpful to look at these screen shots (or thumbnails) before doing the following.)

To get and plot a time series of January precipitation for the Willamette Valley climate division:

1. Navigate to the ERSL Create a Monthly/Seasonal mean time series web page.
2. Select US Climate Division data, and click on "Go to Selection Options"
3. Create a plot.  On the "Create a monthly/seasonal mean time series from the US Climate Division Dataset" page, make the following selections:

1. click on Precipitation
2. State = OR
3. Division = 2 (Willamette Valley)
4. Set the  "Year range" to 1948 to 2010
5. click on "Anomaly" (differences from 1971-2000 "normals")
6. click on "Seasonal average"
7. Leave months at "Jan to Jan"
8. click on "Plot data: get a plot
9. select "boxes" as plot type
10. get the plot by clicking on "Create Timeseries"

4.  The plot should look something like:

5.  Next, use the browser's "Back"  button to return to the "Create a monthly/seasonal mean time series from the US Climate Division Dataset" page, and repeat the above steps, only this time requesting "Ranked sorted values" instead of "Anomaly" and "Raw data values" instead of "Plot data"  (See the screen shots to verify that you've made the right selections.)

6.  The five driest Januaries in the record are 1985 (driest), 1977, 1949, 1962, and 2001.

Step 2:  Get some maps

Next, get "composite anomaly" maps for the five driest Januaries.  Such maps show the average (mean) anomaly over the individual cases.  Start with the 500mb anomaly map, which shows how circulation in the upper atmosphere differs from the long-term average during the five driest years.

1. Navigate to the Monthly/Seasonal Mean Composites page at the ERSL.
2. Make the following selections:

1. Which variable? = Geopotential Height
2. Analysis level? = 500mb
3. Beginning month of season = Jan; Ending month = Jan
4. Enter years for composites = 2001, 1962, 1949, 1977, 1985
5. Plot type? = Anomaly
6. Map projection = Northern Hemisphere
7. Click on "Create Plot" button

After a few seconds, an image should appear.  You can conveniently save the image by right-clicking on it, and selecting copy.  Then in Microsoft Word, for example, you can paste the image in, and save the document.

Use the "BACK" link to go back to the composites page, and repeat the steps above to obtain "Omega" (vertical velocity) at the 500mb analysis level, and "Vector Winds" (wind arrows and wind-speed shading) at the 500mb analysis level for the same selection of years.  The resulting figures should look like the following:

Step 3:  Interpretation

The three maps of the composite anomalies (or differences from the long-term average) 500mb height, vertical velocity, and vector winds and wind speeds during the five driest Januarys for the Willamette Valley climate division provide a good example of how such anomalies can be interpreted.  In particular, the dry conditions during the five years in the composite can be seen to be the product of a) generally weaker onshore flow of moisture, and b) the absence of precipitation-producing mechanisms, as indicated by large-scale sinking of air over the Pacific Northwest. 

The Assignment

The project should consist of an illustrated discussion of either a) a particular climate anomaly, such as the unusually dry winter of 2000-2001 in western Oregon, or perhaps the floods of 1997, or b) of a systematic response of a region like the Pacific Northwest to variations in a large-scale control of climate as represented by a teleconnection index like the SOI.

The report should include a (double-spaced) page or two of introduction and description of the analysis, a presentation of maps and timeseries (2-4 pages), and a one- or two-page discussion of the results.