Very few abnormal pixels

Hello INVEST community!
I run NDR with 10m resolution raster data (about 2017713000 pixels in total), and found that there are about 100 pixels(blue in the picture) at the edge of the data ‘moadified_load_n’ is smaller than ‘export_n’, resulting in ‘retain_n’ being a negative value. What is the reason? The proportion of these abnormal pixels is almost zero, they can be ignored, right? This problem does not occur when using 30m resolution raster data.

Yes, if it’s very few you can probably ignore them. But if you want to fix it, first check the values of your input rasters in the blue areas. Are any of them nodata? Make sure that your nodata values are set correctly.

1 Like

I checked one of the pixels, there is no nodata, ‘modified_load_n’ is 0.06…, ‘export_n’ is 0.08…, because the number of abnormal pixels is almost negligible, so I directly converted negative valued pixels in the ‘retention_n’ data (’ modified_load_n’-‘export_n’) to 0, thanks for your answer!

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.