Dear Mohana,

Normally Enhanced Vegetation Index (EVI) is computed after Modis goes through atmospheric correction process so the values are between -1 and 1. I do not know how did you get the elevation values added to EVI data? What do you mean normalized data is EVI computed using radiance value of MODIS or reflectance values after Atmo. corr.?  if this is the case then it is probably why you get large values.

Please clarify your question.

Best regards

I have extracted elevation values from the DEM and assign it to the respective farms.

I have been given the EVI values for all these 1000 farm locations(I don't have raw data of MOD13a). The values are ranging between 0 and 10000 (where 0 representing no veg and 10,000 represents100% cover of well growing healthy green vegetation). This is the provided information along with the data by the other team.

Using normalization, I have rescaled the value between 0 and 1. All I want to do is to remove the EVI anomalies as a function of elevation values.

Sorry for the confusion.

Why do you think EVI is driven by elevation, does your data show that as a strong trend. I could think vegetation is linked to elevation somehow, but normally the land use/ human's impact is much stronger.

Example: we could have rice field on the flat land with very high EVI, while hilly land is being deforested to bare land that gives low EVI.