r-spatialecology/landscapemetrics

Percentage_inside column shows some values ​far from 100%

Closed this issue · 2 comments

I am trying to get the landscape metrics corresponding to a set of firescars, using a landcover raster and a firescar shape. The total wildfire data set encompasses around 1,000 fire scars, represented by polygons obtained through Google Earth Engine. When calculating the metrics, several polygons with values ​​far from 100% appear in the percentage_inside column. Searching I found this:

The size of the actual sampled landscape in %. It is not always 100% due to two reasons: (i) Clipping raster cells using a buffer not directly intersecting a raster cell center lead to inaccuracies and (ii) Sample plots can exceed the landscape boundary

In this case I think it is something related to the first reason. Is there any way I can fix this in R or Arcgis? or should I just skip these polygons and just keep those with values ​​closer to 100%?
Thanks in advance!

(in the example a firescar is shown with 75%, but I have others with even lower values)
ejemploforo

I think there is not really a way we can change that unfortunately. So I would suggest to skip the ones that are too far off from 100%

@mhesselbarth I think I'll do that u.u, thanks for answering Max