Hello

I’m a beginner in GIS and have a few questions about calculating vegetation indices (VIs). Here are the details of my process:

I have PlanetScope images (4 bands, 3B_AnalyticMS_SR). I successfully calculated NDVI, and the result was as expected, ranging from -1 to +1.

However, when I tried to calculate EVI using this formula (2.5 * (NIR - Red) / ((NIR + 6 * Red - 7.5 * Blue) + 1), the results were strange, ranging from about -8000 to 6000.

To address this, I normalized the band values (since some values exceeded 10,000) by dividing by 10,000. But even after that, some values were still greater than 1, which didn’t help.

Then, I found a suggestion from somewhere in this community that I should convert the bands to TOA reflectance using coefficients from the metadata because the initial values are in Digital Numbers (DNs). After doing this, I got EVI values within the expected range of -1 to +1. But I’m not sure if this solution is correct.

So, I have a few questions:

1. Is the solution I used to calculate EVI (converting to TOA reflectance) correct?
2. For calculating EVI (and other VIs with using some constant values), should I use DNs or TOA reflectance?
3. For NDVI (which is a ratio and constant value is no needed), can I use DNs for calculation?

Thanks for your help.