Initial Publication Date: April 28, 2017

Part 3 - Gathering Historical Drought Data

Download a Student Activity Worksheet here. (Microsoft Word 2007 (.docx) 24kB Feb24 22)

Although it may seem a simple matter to measure and analyze tree rings, there are several natural conditions and factors that make analysis a more complex process. You may have noticed that the pattern of tree rings–the alternate banding of early wood and late wood–remains virtually the same as a tree grows but the width of the individual rings changes based on climate conditions. Another pattern that scientists often see in a tree core happens because of how a tree grows over time. This pattern is called an age trenddiffering tree ring widths that result from early tree rings tending to be wider than later growth tree rings. Trees appear to grow faster when they are younger and slower as they get older, but what is actually happening is that the growth ring of a very young tree is spread over the small diameter of the tree and is proportionally larger than the ring you observe in trees that are much older that have the growth ring spread over a larger diameter tree. In the figure below, showing tree-ring measurements over time for a tree core, you can see that the early rings are significantly wider that later ones.

Because of the site conditions, and the consistent patterns of wide and narrow rings among cores at a site like Humpty Dumpty talus slope, scientists know that a natural process is the main driving force (or what scientists call the limiting factor) behind the growth of trees. In this case, it's climate. The amount of precipitation directly determines the width of a tree ring each year it lives. If you analyzed 30 or 40 more core samples from the Humpty Dumpty talus slope area, you would observe tree-ring patterns and marker years similar to those you found in Part B. But scientists have to be aware that some trees are not good candidates for study; for example, ones that have "disturbance patterns."

A disturbance pattern results from some kind of interaction between the ecosystem and the tree. For example, while coring in Wrangell Saint Elias, Alaska, dendrochronologist Gordon Jacoby noticed that a moose came to scratch its head on the same tree every day. As a result, that tree was damaged and its bark was rubbed off. This tree would not be a good candidate for study because the resulting ring patterns could be influenced by the damage caused to the tree.

A disturbance pattern can be created when trees are struck by lightning, or if a nearby tree fell and allowed a neighboring tree to get more sunlight and grow faster. Scientists take note of any evidence of disturbance in the field, and they take samples from as many trees as possible to minimize any disturbance signals. Look at the sample above. The abrupt decline in tree growth rates, marked by the narrower rings in the center of the image, suggests some event such as storm damage due to high winds occurred during the life of the tree that killed or damaged many of the pine needles, reducing its ability to grow.

In this activity, you will see what records from meteorological stations can show us about drought in the Hudson Valley. Does the long term climate data reaffirm that the trees growing at the Humpty Dumpty talus slope capture wet and dry years? To do this we are going to use an online tool called Climate Explorer, developed for the scientific community to analyze climate data.

Instructions

1. Analyze Time Series Data of the Hudson Valley Region

Right click on the Climate Explorer link and open the program in a new tab (note that you must register first for an account in order to use this tool).

2. Click on "select a field" on the right side of the screen, then click on "Monthly Observation."

3. Now scroll all the way down under "Monthly Observations" until you find "Drought Index" and then click the 1901-2017 0.5° Global 3.26 link next to the CRU self-calibrating Palmer Drought Severity Index dataset.

4. You will focus on the trees and region of the Hudson River valley in New York around the Humpty Dumpty talus slope area. Enter these coordinates: 41 N to 43N and -75E to -74E.

5. Once you have input your coordinates in the "get grid points, average area, or generate subset" box, click "Make time series" to see what the drought data looks like in the region that you chose. The graph below is an example from Durham, ME.

6. Note: The y-axis is a measure of the Palmer Drought Severity Indexuses readily available temperature and precipitation data to estimate relative dryness. It is a standardized index that spans -10 (dry) to +10 (wet) and measures drought in terms of both precipitation and temperature. It is commonly used because it doesn't matter where you are in the world--it gives you a measure of drought based on the climate of the area and you can compare drought across large spatial scales. Note: Negative values indicate drought, zero is normal, positive values indicate wetter conditions. Look for years with very low (i.e. negative) PDSI.

7. Copy and paste your time series graph into your worksheet.

Stop and Think

3.1 What trends do you observe in the data?

3.2 Describe any similarities you see between the data you graphed in Climate Explorer and the "marker years" you identified in Part 2 of this lab.

8. Evaluating Seasonal Data

To make the data clearer to see, you can view the seasonal averages. In the "Investigate this time Series" section to the right click on the "season" link under "View per month, season, half year." Now you can see the same drought dataset averaged for four seasons; Dec-Feb, Mar-May, Jun-Aug, and Sep-Nov.

9. Drag and drop this seasonal data graph into your worksheet.

Stop and Think

3.3 What trends do you observe in each seasonal representation?

3.4. How do these observations compare with your marker year results from Part 2?

10. What is the spatial reach of drought?

We can also use Climate Explorer to look at how much of the region experiences drought or wet conditions for each year. In Climate Explorer, click on "Select a Field", "Monthly Observations", scroll down to "Drought Index" and click the 1901-2016 0.5° Global 3.25 link next to the CRU self-calibrating PDSI dataset again.

11. Now we are going to expand the PDSI field that we are looking at. In the first box that reads "Get grid points, average and area or generate subset" change the latitude and longitude to read: 25°N to 50°N and -90°E to -60°E.

12. Then click "Plot this field" on the right hand side under "Investigate this field." A new menu will come up and in the first box, which reads "Lat-long plot".

13. Change the year to any of the marker years that you found when evaluating the tree ring cores in Part 2. Change the month to "June" and average it over 3 months. That means you will be looking at the PDSI field for the summer (averaged June, July and August). This is when most of the rain in the Northeast occurs and is the season that trees are most sensitive to climate.

14. Enter the the latitude/longitude coordinates again and hit "plot." You will see a plot of the spatial characteristics of the dry/wet year. Note the PDSI scale on the right. Red = wetter, Blue = drier conditions on a graded scale (you can change this in the previous screen, under the "colors" drop down screen you can change it to "red-grey-blue".

15. Use the same method to investigate 3 other years (both narrow (dry) and wide (wet)) that you discovered when you were evaluating the tree-ring data in Excel in Part 2.

16. Copy and paste the maps that you created of both very dry and very wet years into your activity sheet (the marker years that you determined from your Excel spreadsheet).

Stop and Think

3.5 Describe the spatial patterns you observe in the data.

3.6 How do these patterns compare with your results from Part 2 (your Excel spreadsheet)?