Learn · Investigating the lakes · Lesson 3 of 4
Doing science with the data
Every number on In The Lakes was measured, estimated or modelled by someone, and most of it is open for anyone to use. That makes the lakes a real laboratory. You don't need a research ship to do science on them: you need a good question, some care with the numbers, and the honesty to say what the data can and can't tell you.
See it live
Tap any buoy for its card: it says whether each number is measured, estimated from wind, or an old reading. Open Conditions at the helm to replay a past day or the buoy's storm of record. The Granite Island buoy on Lake Superior holds the biggest storm in the archive: waves averaging almost 9 metres on 24 October 2017.
Not for navigation. In The Lakes is a simulation for learning and curiosity. It's a fine place to practise science, but for a trip on the water, use the official Environment Canada or National Weather Service marine forecast.
For kids
Kindergarten to Grade 6Scientists are people who ask questions about the world and then go and check. You can do that too. The lakes are covered in floating buoys that measure the water every few minutes, and you can read what they say right on the map.
A good question you can test
"Is the lake nice?" is hard to test. "Is the water colder in the morning than in the afternoon?" is a great science question, because you can look at the numbers and find out. A guess about the answer, before you look, is called a hypothesis.
Where do the numbers come from?
- Measured: a sensor on a buoy felt it. On the map, these buoys are filled-in dots.
- Estimated: some buoys only measure wind, so the site works out the waves from how hard the wind is blowing. These buoys are rings.
- Old reading: if a buoy hasn't sent anything for more than three hours, its dot gets a dashed edge. Old news isn't today's lake.
Same thing, different numbers
A wind of 10 knots is the same wind as 18.5 kilometres an hour. Water at 20 °C is the same as 68 °F. Always write down the unit, or your number doesn't mean anything.
buoys and weather stations, Canadian and American, are on the map. Only some are reporting now (sent a reading in the last 3 hours), and fewer still have wave sensors; the line at the bottom of the map gives today's numbers. Many buoys come out of the water from late October to May. The site checks them every 30 minutes.
The average height of the biggest waves at the Granite Island buoy on Lake Superior in a storm on 24 October 2017. That's about as tall as a three-storey building.
Try this
Keep a lake diary. Pick one buoy near you. Every day for a week, at the same time, write down the date, the time, the water temperature, the wind and the waves, with their units. At the end of the week, draw a line graph of the water temperature. Did it go up, go down, or stay the same? What do you think made it change?
For students
Grades 7 to 12Start with a testable question
A good investigation names what you'll change or compare (the independent variable), what you'll measure (the dependent variable), and what you'll keep the same. "Does the water get colder after a strong north wind?" works: wind direction and strength on one side, water temperature on the other, at one buoy so the place stays the same.
Know what kind of number you have
- Measured values come from a sensor at that spot. Wave height, period and direction are measured only where a buoy has a wave sensor.
- Estimated from wind values (marked est.) are worked out from the wind speed and the fetch. They're reasonable, but they're not observations.
- Old readings (over three hours) were true once. Don't use them to describe now.
- Satellite values, like the daily lake surface temperature map, see the whole lake but only the top skin of the water, and not through cloud.
- Model values, like the 48-hour wave forecast or the temperature down to the lake floor, come from a computer model built on physics and real observations. They fill gaps, but they can be wrong.
Units and conversions
- Speed: a knot is one nautical mile (1,852 m) per hour, so 1 kn = 1.852 km/h, about 0.514 m/s. Going the other way, 1 m/s = 3.6 km/h, about 1.94 kn. Many buoys report in m/s; the map shows knots, like marine forecasts.
- Temperature: °F = °C × 9/5 + 32. So 15 °C is 59 °F, and 4 °C is about 39 °F.
- Length: 1 ft = 0.3048 m exactly, so 1 m is about 3.28 ft.
- Mass: a short ton is 2,000 lb (about 907 kg); a tonne is 1,000 kg. Canadian and U.S. shipping figures use different ones, so check before you compare.
Reading a time series
A time series is a list of readings in time order, and its graph puts time along the bottom. Look for the trend (is it rising or falling overall?), cycles (does it go up every afternoon?), sudden jumps, and gaps. Check the scale on the side: a line that looks like a cliff might only change by half a degree.
Averages, and the significant wave height
The mean adds up the readings and divides by how many there are. Waves need a smarter average, because a few big ones matter more than many small ones. The significant wave height is the average of the highest one third of the waves in a sample. Individual waves are bigger and smaller than that: roughly 1 wave in 3,000 is twice as tall, so over a few hours a "1 m" sea can still throw a 2 m wave.
Uncertainty and sensor error
Every measurement has some uncertainty: a thermometer good to ±0.1 °C can't tell 12.61 from 12.65. Sensors also fail. They drift as they age, which is why buoys come ashore each fall for calibration, and they can be fooled: ice on an anemometer, a wave sensor knocked about in a storm. A reading that doesn't fit the rest is an outlier. Don't delete it quietly; ask whether the lake or the instrument made it, and say what you decided.
Correlation is not causation
If water temperature falls whenever the wind blows from the north, the two are correlated. That alone doesn't prove the wind caused it. Maybe both follow a third thing, like a cold front bringing cold air and north wind together. Look for a mechanism (a reason it would happen) and test it more than once. One storm is a sample size of one: interesting, never proof.
Data challenge
Does the water get colder after a north wind? Open the Granite Island storm of record and step through the day. The National Weather Service says strong north winds drove the storm. Write down, every few hours, the wave height, the wave period and the water temperature. In this record the water starts near 12.6 °C and ends the day near 11.3 °C. Now be a sceptic: the air also got colder, and storm waves stir up cooler water from below. Which explanation fits better? Then look at the wind column: some hours in the middle of the gale read almost calm. Is that the lake, or the sensor? To test your idea properly, pick a live buoy, wait for the next strong north wind, and use the date picker to compare the days before and after. How many storms would you need before you'd trust the answer?
For adults
Going deeperA worked mini-investigation
Question: after a strong north wind, does the water at a Lake Superior buoy get colder? Data: the hourly storm-of-record rows for the Granite Island buoy, 24 October 2017, from GLOS Seagull. What it shows: the significant wave height climbs from 0.2 m overnight to a peak of 8.8 m (28.8 ft) in the afternoon, the wave period stretches from about 2 seconds to 11, and the water temperature slips about 1.3 °C over the day. Caveats: it's one storm at one buoy; the air cooled from about 9 °C to 4 or 5 °C at the same time; and the buoy's wind readings drop to near zero in the middle of the gale, which is hard to believe when the National Weather Service reported hurricane-force gusts at Granite Island that day. A careful conclusion is "consistent with storm mixing and cooling, worth testing on more storms", not "north wind cools the lake".
Keep a data log
Record the station name and id, the date and time (and time zone), each value with its unit, whether it was measured, estimated or modelled, and anything odd you noticed. Note the metadata too: the sensor depth, when the buoy went in, and who runs it. A log someone else can follow is the difference between a hunch and a finding.
Open data, and citing it
Much of the Great Lakes data is open data. GLOS gathers buoys and stations run by many different owners on its Seagull platform; Great Lakes DataStream shares water-quality monitoring by community groups, First Nations, governments and researchers; NOAA and Environment and Climate Change Canada publish satellite products and forecasts. Open isn't the same as anonymous. When you use a dataset, credit who collected it and where you got it, name the station and dates, give the date you downloaded it, and follow the licence. In The Lakes is an independent project, and its own Data & Science page lists every source it uses.
Citizen science: adding to the record
Citizen science is research where members of the public collect or analyze real data. Around the lakes:
- Adopt-a-Beach, from the Alliance for the Great Lakes: volunteers clean shorelines and record every piece of litter they find, data that has been used in plastics policy.
- DataStream: community monitoring groups share their water-quality results openly.
- eBird: birders submit checklists of what they see; in Canada it's run by Birds Canada with the Cornell Lab of Ornithology.
- CoCoRaHS: volunteers measure rain and snow at home every morning, in Canada and the U.S.
- Crowdsourced depths: GLOS collects depth soundings from volunteer boats, helping map a lake floor that is mostly unmapped (see Lakebed 2030).
For how the instruments themselves work, see Buoys and sensors; for the physics behind the wave numbers, Waves and wind.
Words to know: Hypothesis · Variable · Time series · Mean · Uncertainty · Calibration · Outlier · Correlation · Sample size · Metadata · Open data · Citizen science · Significant wave height · Model · Sensor · Knot · Short ton and tonne
See the data
Look for patterns. Which lake runs warmest? Shallow Lake Erie warms fastest in summer and cools fastest in fall. Is one buoy far colder than the rest of its lake? That can be an upwelling, when wind pushes the warm surface water away and cold water rises from below. Come back tomorrow and see what has changed.
Show the numbers
| Lake | Buoys | Coldest (°C) | Middle (°C) | Warmest (°C) |
|---|---|---|---|---|
| Loading the latest readings… | ||||
Source: the latest readings on the In The Lakes map, from the Great Lakes Observing System's Seagull network of Canadian and US buoys, refreshed every 30 minutes.
For teachers
Teacher notes
- Big idea
- Real data lets anyone test an idea about the lakes, if they know where each number came from and how far to trust it.
- Great Lakes Literacy
- 6 · The lakes and the people of their watershed are bound together
- 7 · Much remains to be learned about the lakes
- Curriculum
Ontario
- Mathematics, Data (D1): every grade, data drawn from real life; broken-line graphs in Grade 6, scatter plots in Grade 8
- Grade 9 Science (SNC1W): A1, investigation skills; A2.4, local environmental issues
- Grade 11 Environmental Science (SVN3M): C2.2, longitudinal data; C2.3, field study of water samples
- Grade 11 Introduction to Spatial Technologies (CGT3O): C2.3, GIS and GPS analysis of local water resources
Ontario links give the expectation numbers, checked in October 2026 against the curriculum in force on the Ontario Curriculum and Resources site; "example" means the lakes appear in the Ministry's examples or prompts, not as a required expectation. They're a planning aid, not an official alignment.
Michigan
Science
- 3-5-ETS1-3 (Grades 3–5, Engineering Design) plan and carry out fair tests
- MS-ESS2-5 (Grades 6–8, Earth and Space Sciences) data on air masses and changing weather
- HS-ESS3-5 (High school, Earth and Space Sciences) analyze geoscience data to forecast
Social studies
- 6-G1.1.1 (Grade 6, Geography) Use a variety of geographic tools (maps, globes, and web-based geography technology) to analyze the world at global, regional, and local scales.
- 7-G1.2.1 (Grade 7, Geography) Use a variety of geographical tools (maps, globes, geographic information systems [GIS], and web-based geography technology) to analyze what is happening at different times in different locations.
Michigan science links are the NGSS performance expectations, which Michigan adopted as its K–12 Science Standards; social studies codes are the content expectations in the Michigan K–12 Social Studies Standards, checked in November 2026 against MDE's published standards. They're a planning aid, not an official alignment.
Wisconsin
Science
- SCI.ETS1.B.3-5 (Grades 3–5, Engineering, Technology and the Application of Science) Research on a problem should be carried out before beginning to design a solution. Testing a solution involves investigating how well it performs under a range of likely conditions.
- SCI.ETS1.C.3-5 (Grades 3–5, Engineering, Technology and the Application of Science) Different solutions need to be tested in order to determine which of them best solves the problem, given the criteria and the constraints.
- SCI.ESS2.C.m (Grades 6–8, Earth and Space Sciences) Water cycles among land, ocean, and atmosphere, and is propelled by sunlight and gravity. Density variations of sea water drive interconnected ocean currents. Water movement causes weathering and erosion, changing landscape features.
- SCI.ESS2.D.m (Grades 6–8, Earth and Space Sciences) Complex interactions determine local weather patterns and influence climate, including the role of the ocean.
- SCI.ESS3.D.h (Grades 9–12, Earth and Space Sciences) Global climate models used to predict changes continue to be improved, although discoveries about the global climate system are ongoing and continually needed.
Social studies
- SS.Geog1.b.m (Grades 6–8, Geography) Interpret patterns in a variety of maps, charts, and graphs to display geographic information (contour, cartogram, population, natural resource, historical maps) and explain relationships among them.
- SS.Geog1.b.h (Grades 9–12, Geography) Interpret maps and images (e.g., political, physical, relief, thematic, virtual or electronic) to analyze geographic problems and changes over time.
Codes are the performance indicators in the Wisconsin Standards for Science (adopted 2017, kept after DPI's 2024 review) and the Wisconsin Standards for Social Studies (2018), checked in November 2026 against DPI's published standards; "example" means the lakes appear only as an example. They're a planning aid, not an official alignment.
- In class
- Have each group keep a week-long data log for a different buoy, then compare graphs and argue about what caused the differences.
- Convert a buoy card's wind, temperature and wave height into every unit you know, and check each other's work.
- Replay the Granite Island storm and plot wave height, period and water temperature on one graph. Discuss: is the falling temperature caused by the wind, the air, the waves, or all three?
- Join a citizen-science program as a class, such as an Adopt-a-Beach cleanup or a schoolyard eBird count, and look up your own data afterward.
Pick Ontario, Michigan or Wisconsin above; the curriculum map lists every lesson by expectation or standard. The Great Lakes Literacy Principles are from the Great Lakes Sea Grant Network; the labels here are our short paraphrases.
Keep exploring
Free games, simulations, virtual dives and hands-on activities from other educators. Each one opens on another website.
Sources
- Great Lakes Observing System (GLOS) and its Seagull data platform (open data from many owners; cite the owner of each station)
- Great Lakes DataStream: open water-quality data (data under open licences set by each contributor)
- NOAA National Data Buoy Center: how wave statistics are calculated (public domain)
- NOAA GLERL CoastWatch: satellite lake surface temperature (GLSEA) (public domain)
- National Weather Service Marquette: the strong fall storm of 24 October 2017 (public domain; north winds, gusts and the 20 to 30 ft waves at Granite Island and Munising)
- Michigan Public: buoy logs highest wave ever recorded in Lake Superior (25 October 2017) (news report on the 28.8 ft significant wave height)
- NIST: unit conversion (SI and U.S. customary units) (public domain, U.S. government)
- Alliance for the Great Lakes: Adopt-a-Beach (volunteer beach cleanups that record what they find)
- Birds Canada: eBird Canada (run with the Cornell Lab of Ornithology)
- CoCoRaHS Canada: volunteer rain, hail and snow observers (open data)
- Great Lakes Literacy Principles (Great Lakes Sea Grant Network) (the framework used in the teacher notes)
Written for In The Lakes in our own words, from public-domain and openly licensed government sources, with others cited for further reading. Figures are rounded. Spotted a mistake? Write to admin@inthelakes.ca.
A–Z · LOOK IT UP Glossary 370 lake words in plain language, from AIS to zooplankton. OPEN THE GLOSSARY →