Advance preparation: Prepare a bag for each pair (or group) of students that has two different colors of dried beans (any two colors will work). You want to have enough beans in the bag so that students need to take the beans out of the bags and carefully count them. We suggest bags of 40-50 beans per color. Each bag needs to have the same amount of beans- but don’t tell the students this! You will limit the amount of time it takes them to count the beans, and providing a distraction while they are counting can also be helpful. We suggest giving students 1 minute to count if you use 50 beans per color. This reminds students how easy it is to make mistakes when you aren’t paying close attention! For a fun introduction to this idea, consider watching the “Gorilla in the Room” video: http://www.youtube.com/watch?v=m_8nJZ_VUKY . This video shows two teams of students (team white vs team black) passing basketballs. Students are asked to count how many times the white team passes their ball. Most students will completely miss the gorilla that walks into the middle of the basketball court during the 1-minute exercise.
Hand out the worksheet “Salty, salty streams”. Watch the video clip (about 5 minutes long) of Dr. Sujay Kaushal explaining his research on Baltimore streams. This video is available on the Cary Institute Vimeo page (https://vimeo.com/caryinstitute ). Students should answer questions #1-4 as they watch.
After watching the video, review the questions on the worksheet. Students should make a claim about what type of stream they think will have the highest chloride levels: forested, suburban, or urban. In small groups, ask the students to interpret the graph of Baltimore stream data provided in question #5 (also provided as powerpoint slide #1). Students should notice the large difference between the urban and forested streams, and should be able to think about how land use might affect the different streams. The graph of seasonal chloride levels is provided for you in the powerpoint (slide #1), in case you would like to project the image and discuss it as a whole class. These data allow the students to critique the claim they made beforehand.
Next, show students one year of data from the suburban Baltimore stream (in the powerpoint, slide #2). Ask students to explain why the data points vary between winter and summer; students should talk about the increase in road salt application in the winter. Then, focus on the data points in the summer months – ask students to explain why there might be differences between the data points in June or July.
To introduce the concept of variability, give each pair (or group) of students a bag of beans. Give students a defined amount of time (less time if you have fewer beans – I recommend limiting the counting to 1 minute if you have 50 beans per bag) and ask them to count all of the beans in the bags. Tell them that both speed AND accuracy matter. Once students have counted, they should come up to the board and write down their counts on the board. Don’t allow students to recount their beans.
Compare and discuss the students’ results – are they the same? Do some counts differ? Why? What are some reasons for the differences – could it be due to human error, or actual differences between the bags?
Next, provide students with the natural objects – leaves, pine needles, etc. They should have enough of the objects so that when they compare them, they are similar but not exactly the same. Ask students to measure the diameter of the leaves or the lengths of the pine needles, and record the results. Students should share their results. Ask: Why are there differences between the leaves? Are these differences due to the leaves, or to human error? Brainstorm answers to both questions.
Place this chart up on the board, and fill in with student help (students have a copy of this chart as question #6 in their worksheet). Ask students to think about potential error with reference to the Baltimore salt data. Some examples are included below.
|Real/Natural : variability due to the ecosystem||Human/experimental : variability due to human error, sampling effort or design, etc|
- Rainfall events could dilute the chloride concentrations in the summer
- Drought might increase the chloride concentrations in the summer
- There might have been a chemical spill that cause the high spikes of chloride
- Snow or ice events would mean more salt spread on the roads
- The sampling equipment might have been different throughout the year
- Researchers may have collected data in different areas
- Data might not have been collected often enough
- The sampling equipment might have been working incorrectly or someone used the equipment incorrectly
- Sampling during different times of year might provide different results
Students often focus on human error, or experimental error, instead of thinking about the types of changes that might allow the chloride values to change in the ecosystem.
Finally, show students the “Beans Example Template” that gives an example of the bean data – this spreadsheet will show you an example of the activity. (You can also put in your bean count data.) You can see the bean counts individually (as a scatter plot), the averaged bean counts (bar graph), and a bar graph with error bars. Ask students to explain the benefits and drawbacks of the different types of graphs. Students should recognize that error bars allows them to have some information about the range of the data in the sample, similar to a scatter plot, but with the additional benefit of knowing the average of the set of data. This idea will be reviewed in the last lesson.
Explore: Give students data that explore chloride concentrations in area streams; pick the dataset that is closest to your region. Available data sets include:
- New York City: East Branch of the Delaware
- Kingston/Cairo: Esopus Creek
- Rockland County/Pearl River: Sparkill Creek
- Poughkeepsie/Arlington/Chatham: Sawkill Creek
Students should graph these data and explain the trends they notice. Depending on your students, you may want to have them graph these data by hand (using the provided “simplified” data tables in Word), or use Excel. The Excel data sheets provide two tabs - one that has all of the data, and one that is a summary tab. Make sure that you decide ahead of time how much data you want your students to use, and provide them with the appropriate data sets. Graphs are embedded in the Excel sheets. If students are using Excel, have them add a regression line and place the r-squared value on the graph. This will help them begin to understand the idea of significance.
Chloride levels are increasing throughout the northeast. Use the powerpoint to review the content behind this idea. The third powerpoint slide shows the increasing chloride concentrations in Maryland, New York, and New Hampshire. Here is an abstract from a paper in the Proceedings of the National Academies of Sciences:
Chloride concentrations are increasing at a rate that threatens the availability of fresh water in the northeastern United States. Increases in roadways and deicer use are now salinizing fresh waters, degrading habitat for aquatic organisms, and impacting large supplies of drinking water for humans throughout the region. We observed chloride concentrations of up to 25% of the concentration of seawater in streams of Maryland, New York, and New Hampshire during winters, and chloride concentrations remaining up to 100 times greater than unimpacted forest streams during summers. Mean annual chloride concentration increased as a function of impervious surface and exceeded tolerance for freshwater life in suburban and urban watersheds. Our analysis shows that if salinity were to continue to increase at its present rate due to changes in impervious surface coverage and current management practices, many surface waters in the northeastern United States would not be potable for human consumption and would become toxic to freshwater life within the next century. - Kaushal et al, 2005
After slide 3, the powerpoint reviews how sodium chloride molecules accumulate in a stream over time. Students can view these slides to understand the gradual accumulation of this pollutant in our groundwater.
The steady increase of chloride throughout the region is likely due to a lag effect of long-term salt use & subsurface build-up. Chloride levels in a stream should level off if road salt is no longer applied, but currently this is not the case. In Dutchess County, salt has increased steadily in Wappinger Creek since 1986, with some spikes in last 5 years. The primary source is road salt, even though there has not been a significant increase in the use of road salt in the last 9 years. The groundwater is able to store the salt and slowly release it throughout the year. In contrast to Baltimore, streams in Dutchess County are saltier in the summer. This is due to the amount of water that is used by transpiration, which removes water from the streams and thus increases the concentration of salt in the streams. Challenge students to do the math to understand this phenomenon by walking them through slides 20-35.
An optional background reading is provided: "Chloride and Salt".
Students should complete the reading on “Salt makes roads safe but can pollute water” from the Poughkeepsie Journal and answer the four accompanying questions.