If we acquire a tomography dataset, we can extract alot of physical properties from it, including porosity and permeability. These properties are not directly measured using conventional experiment. Instead, they are calculated using different image processing algorithms. To this end, is there any guideline on how to report such results in terms of significant digits?
Thanks.
When reporting physical properties derived from tomography datasets, it is important to consider the accuracy and precision of the measurements. The number of significant digits used in reporting these properties should reflect the level of uncertainty in the measurements.
In general, it is recommended to report physical properties derived from tomography datasets with the same number of significant digits as the measurement uncertainty. The measurement uncertainty can be estimated based on the method used to calculate the property, as well as the resolution and noise level of the tomography dataset.
For example, if the calculated porosity value is 0.25 with a measurement uncertainty of 0.02, then it is appropriate to report the porosity value as 0.25 ± 0.02. This indicates that the true value of porosity falls within the range of 0.23 to 0.27 with 95% confidence.
Similarly, if the calculated permeability value is 10^-15 m^2 with a measurement uncertainty of 10^-16 m^2, then it is appropriate to report the permeability value as (1.0 ± 0.1) × 10^-15 m^2.
It is important to be consistent in the number of significant digits used when reporting physical properties derived from tomography datasets, as this can affect the comparability and reproducibility of results.
Febin Cherian John Thank you for your response. The issue that I face when dealing with tomography dataset is that the data are usually supplied with resolution in unit length. Consequently, identifying the measurement uncertainty in other complicated properties such as permeability or porosity is not very clear for me. Should it have the same significant digits as the resolution limit? if so, this might result in reporting values with too many digits.
Thanks again
You bring up a good point. In the case where the tomography dataset is provided with a resolution in unit length, it may not be straightforward to estimate the measurement uncertainty of more complex properties such as permeability or porosity.
In this case, one approach is to use the resolution as a guide and estimate the measurement uncertainty based on the expected level of variation in the property. For example, if the resolution of the tomography dataset is 1 micrometer and the expected level of variation in the permeability or porosity is on the order of 10%, then a reasonable estimate for the measurement uncertainty might be on the order of 0.1 times the average value.
When reporting physical properties derived from tomography datasets, it is important to balance the need for accuracy and precision with the practical limitations of the measurement and the significance of the results. In general, it is recommended to report physical properties with the appropriate number of significant digits to convey the level of uncertainty and enable meaningful comparison with other results, but not to report more digits than necessary.
Ultimately, the appropriate number of significant digits to report will depend on the specific context and level of uncertainty associated with the measurement. If there is uncertainty about the appropriate number of significant digits to use, it may be helpful to consult with a subject matter expert or refer to relevant standards or guidelines in the field.
I hope this helps you.
Thank you
Dear Febin Cherian John
This is very lucid explanation. Many thanks.
Faisal
- Badges
- Science topic
- Similar topics
- Phytochemistry
- Extracts