I have a broad range of polygons (in vector format and in SpatialPolygonsDataFrame class), some going east-west, others going south-north, while even others span polar regions or very large ranges as mid-to-high/low-to-mid latitudes/longitudes. In addition, as a single polygon can consist of several merged sub-polygons, they do not necessarily have to be aggregated or continuous, but can instead be chopped-up across large ranges. I have until now calculated the area without projecting the data using gArea {rgeos}
as follows
gArea(SpatialPolygons(spdf@polygons))
. Since I am not interested in visualization of any maps but only comparing polygon area and range sizes, I reckoned that the unit (square degrees vs. square meters) did not matter... but now I am not so sure.
So my questions are;
(i)
Calculating polygon areas, would I need to project my data using an e.g. equal area projection
?
If yes,
(ii)
Since I do not care about any map visualizations and my polygons can span very broad ranges making it hard to find a single best projection, can I use the same projection for all polygons e.g. Lambert azimuthal equal-area projection
when calculating the area?