Hello. I am new to this community and I am not an expert in coordinate systems or projections, so please pardon my ignorance on these topics. I am, however, a long term ArcGIS user (using Pro now for several years) and I would like to try to get an authoritative answer about something that has been confusing me for years.
I frequently work on projects that cover large spatial extents (e.g., the whole eastern U.S.) but require accurate area measurements for hundreds of thousands of polygons that lack regional bias. I've read for years about how projections affect area calculations and I've experienced these effects enough to know some basic rules of thumb like "never measure area in a Web Mercator projection" or "always measure acreage using the most local equal-area projection you can."
Based on this surface level understanding, I've recently started to do the following, as my best guess as how to calculate area accurately across large spatial extents.
1. I do a spatial join using a feature class of State Plane Zone boundaries to assign each of my polygons with the State Plane Zone it is in.
2. I iterate the same process over and over until I have area calculations for all of my polygons:
a) I create a new field called something like "AcresStatePlane"
b) I select all polygons in one of the State Plane Zones in my study area.
c) I use the Calculate Geometry tool to calculate area for these polygons using the State Plane Zone projection that matches all of my selected polygons.
d) I clear my selection, and then select all polygons in the next State Plane Zone in my study area.
e) I use the Calculate Geometry tool to calculate area for these polygons using the State Plane Zone projection that matches all of the newly selected polygons.
f) Using this method, I eventually end up with accurate (I think) area measurements for all of my polygons.
I don't know if this method makes sense or not. It can be clunky and takes some time, but in the end I feel confident that I've minimized regional bias in my area calculations (should I?).
The reason I go to this trouble is that I've done experiments on my own comparing area calculations of the same sets of polygons using State Plane Zones, UTM Zones, other equal-area projections like Albers Equal Area Conic, and some non-equal-area projections. When I've summarized results from these experiments, I've found relative bias in calculations that varies regionally (as one might expect given the characteristics of these different projections, where it "seems" to me like calculations are most accurate, or is it least regionally biased?, in the following order, from best to worst: a) State Plane Zones, b) UTMs, c) continental equal-area projections, and then d) non-equal area projections, which always perform terribly).
I recently posted about this method to the Natural Resources community board and asked if anyone had a suggestion for alternative ways to achieve this same result in fewer steps. I also asked if there is any way to calculate "true area," since I'm guessing such a thing exists, at least theoretically. Finally, I asked some practical questions related to how the actual software ArcGIS Pro works, which I'll summarize in a minute.
The response I got to the post was simple... "Use calculate geodesic." Reading more about this, I had the following questions...
1) Is it really this simple? 2) If so, why isn't the first thing anyone ever told me as a GIS user, 3) If this method is always best, why would I ever want to use the "Calculate Area" option instead of the "Calculate Area (geodesic) option? Wouldn't this just be inviting inaccuracy? 4) Does the "geodesic" method represent a gold standard that might be interpreted as being as close to "true area" as possible?
The person I asked these questions said he didn't think he was qualified to answer and suggested I create a new post to ask more qualified people. I think this user group is probably the right place for these questions.
Pragmatically, if the answers to these questions turn out to be something like "yes, always use the calculate geodesic option," then I have the following questions related to how the ArcGIS Pro software works... If I use the Calculate Geometry tool and the Calculate Area Geodesic option for all of my area calculations, do any of the following things matter at all? a) the projection of my feature class? b) the projection of the map? c) the projection of other layers in the map? I experimented with this by using the Calculate Area Geodesic option for feature classes in different projections and each time, the area results I got were the same, making me think that this method over-rides/ignores projection information (which would be great as far as I am concerned).
My take home from all of this is that accurate area calculation may be so much easier than I ever knew and may just boil down to the choice to always use the Calculate Area geodesic selection of the Calculate Geometry tool. If this is true, Hallelujah! I think this means I can stop going through the laborious process I describe above to always use State Plane Zone projections to calculate area. Is this true?
Like I said, I'm really not well versed at all in coordinate systems and projections. I've come to all of these conclusions by asking questions of people who know more than me. I'm starting to feel like I might be getting closer to knowing what to do so that I can assure my collaborators/user groups that my area calculations area as accurate as possible. Is this right?
My guess is that it's not as simple as all of this and that I have more to learn. Therefore, I ask those of you that might know how to answer these questions to please weigh in (knowing that the more practical, less theoretical, the answer, the better chance I'll have at understanding it). Thank you all in advance for any light you can shed on this topic.
Batch projection of polygons to correct state lines lane zone? Environment and Natural Resources UC24 Flier
