Dear all,

I need to make a correlation between two raster datasets. I want to use Raster Calculator, but I'm still finding a bit difficult to work it out. I have the formula from statistics books, and I know that I'm required to multiply, and add the grids, right? But it's not so straight forward.

How am I supposed to find N, for instance? Should I just multiply the Columns and Rows from my grid to see how many pixels I have on the dataset?

The formula for the Pearson correlation coefficient is quite long and I don't think I should attempt to do everything in one go. I reckon I should do it by different stages.

Any ideas would be very welcome!

Just a couple more things about the two datasets: one of them is a NDVI that I acquired from LANDSAT 5TM, and the other raster has been originated from an interpolation of points (rainfall gauges). I have used Mask on both grids, therefore they have the same number or rows and columns (same number of pixels?).

Thanks for your time.

Alan.

I need to make a correlation between two raster datasets. I want to use Raster Calculator, but I'm still finding a bit difficult to work it out. I have the formula from statistics books, and I know that I'm required to multiply, and add the grids, right? But it's not so straight forward.

How am I supposed to find N, for instance? Should I just multiply the Columns and Rows from my grid to see how many pixels I have on the dataset?

The formula for the Pearson correlation coefficient is quite long and I don't think I should attempt to do everything in one go. I reckon I should do it by different stages.

Any ideas would be very welcome!

Just a couple more things about the two datasets: one of them is a NDVI that I acquired from LANDSAT 5TM, and the other raster has been originated from an interpolation of points (rainfall gauges). I have used Mask on both grids, therefore they have the same number or rows and columns (same number of pixels?).

Thanks for your time.

Alan.

Thank you very much for your reply. I will try what you suggested and I should be coming back here with either some results for discussions (or more questions!) if that's okay.

Just a couple of things regarding the interpolation. My area of study is data scarce and I didn't have a choice. I would have only three points of NDVI where the data was actually collected. I think I will take the risk and still use the interpolation/NDVI analysis - or do you think the results would be worthless?

Regards,

Alan.