        Uncertainty in CO2 Data
Every measurement has some uncertainty associated with it, due to the precision of the instrument. Take a ruler for instance. The smallest division on a standard ruler is 1 mm. Let's say we measure a pencil and to the nearest millimeter we find its length to be 21 mm. Now it is pretty unlikely that the pencil is exactly 21 mm long. It is probably a little longer or shorter. If someone else were to measure the pencil, they may state its length to be 22 mm. Who is right?

Within the limits of the precision of the ruler, 1 mm, we both are correct. We should then state the length of the ruler as 21 ± 1 mm, and the other guy should state the length as 22 ± 1 mm. A scientist looking at our data now knows that the precision of our measurement is 1 mm. Some people call the ± 1 mm an error or probable error in our measurement. It is not an error, but an uncertainty. We took the best measurement we could with the equipment available. If we had made an error, we would have taken the measurement again!

To minimize the uncertainty in a quantity, scientists measure it several times. This also helps to eliminate true errors in a measurement, because the value in error will most likely be substantially different from the others. The value stated is the average value from all the measurements. For example, let's say we took five length measurements of an object: The average value would be: The uncertainty is found by first taking the difference between each measurement and the average value. This is called the deviation and is stated as a positive number. Deviation: The mean deviation, dm, is the sum of the deviations divided by the number of measurements: Mean deviation: The probable uncertainty is then found by dividing the mean deviation by the square root of the number of measurements: The measurements of the length would then be reported as Sometimes uncertainties are given in percents. The percent uncertainty is found by dividing the probable uncertainty by the average value and multiplying by 100. What if we add, subtract, multiply, or divide several numbers with uncertainties? Let's say we have three numbers: For addition and subtraction, the numerical uncertainty is equal to the square root of the sum of the squares of the probable uncertainties. If we add our three numbers, we get 24.41 with an uncertainty of So the result would be written as For multiplication and division, the numerical percent uncertainty is equal to the square root of the sum of the squares of the percent uncertainties. If we multiply our three numbers, we get 510 with an percent uncertainty of

First, turn probable uncertainties into percent uncertainties: Then, So the result would be written as If many measurements are taken, the uncertainty is sometimes measured by the standard deviation. One standard deviation means that 68% of all measurements will be within this value. Two standard deviations mean 95% of all measurements will be within this value. Calculation of the standard deviation is beyond the scope of this note.

[ Yellowstone Biomass Burning ] [ Seasonal Vegetation Changes ]
[
Fossil Fuel Burning ] [ Fit CO2 Curve ]
[
Balancing the Carbon Cycle ] [ Uncertainty in CO2 Data ]
[ Glossary] [ Related Links ] [ References ] [ PBL Model ]

[ Home ] [ Teacher Pages ] [ Modules & Activities ] HTML code by Chris Kreger
Maintained by ETE Team
Last updated November 10, 2004

Some images © 2004 www.clipart.com

Privacy Statement and Copyright © 1997-2004 by Wheeling Jesuit University/NASA-supported Classroom of the Future. All rights reserved.

Center for Educational Technologies, Circuit Board/Apple graphic logo, and COTF Classroom of the Future logo are registered trademarks of Wheeling Jesuit University.