A Little help interpreting spectral data from an article

[Guilherme Franco]

I'm trying to find a good database of absorption or reflection spectra in visible light for pigments.

I've found a wonderful database in this article: http://e-conservation.org/issue-2/36-FORS-spectral-database#CSV

It's almost exactly what I needed

Except I don't understand the data

The graphs in the article show a lot of reflectance values over 100% (and doesn't explain them) and the data in the CSV files have both numbers greater than 100 and lower than 0 (negative numbers), not to mention I don't know if those values are percentages (the same on the graphs of the article) or some signal intensity value.

I can't seem to find explanation for these things on the article.

Can anyone help me?

Thanks!

 

[mjc123]

I assume the numbers in the csv files are the same as the percentages in the graphs. The article explains that reflectance values are not absolute but are with reference to a white reflectance standard. If the sample reflects more strongly than the reference, the reflectance value is > 100%. The few negative numbers look like just the result of noise at low wavelengths.

 

[sophiecentaur]

Seems crazy, on the face of it. There must be something in the 'small print' that allows for >100% reflectance. I have been skipping around a number of Google hits and the reference seems to be a 99% Spectralon diffuse reflectance standard, whatever that is. That makes it even more weird.
On the face of it, you'd need some frequency shifting mechanism to get more out at a given frequency than is put in. Fluorescence happens in some materials but I thought it needed UV energies.

 

[davenn]

this isn't my field of expertise
Am some others will chime in

@Dale @Andy Resnick

I don't know if those values are percentages (the same on the graphs of the article) or some signal intensity value.

this comment in the introduction explains that part of the query

A FORS spectrum shows for each wavelength, the ratio between the intensity of the reflected light and the incident light, measured with respect to a standard white reference. This ratio is called reflectance and is given in percentage (%).

I am just also not sure how more than 100% is achieved ?

I assume the numbers in the csv files are the same as the percentages in the graphs. The article explains that reflectance values are not absolute but are with reference to a white reflectance standard. If the sample reflects more strongly than the reference, the reflectance value is > 100%. The few negative numbers look like just the result of noise at low wavelengths.

yes, exactly

 

[sophiecentaur]

I am just also not sure how more than 100% is achieved ?

I guess this is so familiar to the 'experts' that they don't even feel the need to explain it away. Or perhaps it takes a Physicist to even notice something like this? We have a 'learned intuition'.

 

[Andy Resnick]

I'm trying to find a good database of absorption or reflection spectra in visible light for pigments.

The graphs in the article show a lot of reflectance values over 100% (and doesn't explain them) and the data in the CSV files have both numbers greater than 100 and lower than 0 (negative numbers), not to mention I don't know if those values are percentages (the same on the graphs of the article) or some signal intensity value.

I assume the numbers in the csv files are the same as the percentages in the graphs. The article explains that reflectance values are not absolute but are with reference to a white reflectance standard. If the sample reflects more strongly than the reference, the reflectance value is > 100%. The few negative numbers look like just the result of noise at low wavelengths.

this isn't my field of expertise
Am some others will chime in

@Dale @Andy Resnick

I've done a lot of these kinds of measurements, there could be a number of reasons why reflectance measurements > 100%, and unfortunately, I couldn't get a clear understanding of the measurement method from the article. For example: it's not clear if the reflectance measurement is 'normal' reflection or retroreflection (back reflection). It could be 'normal' (http://oceanoptics.com//wp-content/uploads/example-setup-reflectance-1.jpg), but it's not explicit. For normal incidence this doesn't matter, but it does for 45-degree angle incidence. Also, I didn't see in the article where they discuss a 'reflection standard': not only could certain materials reflect more than a standard material (like spectralon), if the material is more specular than the diffuse standard, the reflectance measurement could >100%.

A more complete reflectance specification is the 'bi-directional reflectance distribution function':

https://en.wikipedia.org/wiki/Bidirectional_reflectance_distribution_function

One other (probably minor) effect: the authors mention some of the pigments fluoresce; if they illuminate the sample with white light some of the short-wavelength energy will convert to longer wavelengths increasing the apparent reflectance.