I'm facing difficulties with area measurements when converting a raster to polygon in ArcGIS Pro and then working with the resulting data in R. My process involves a number of steps:
Data collection: I download land cover rasters for each year since 1985 from the Map Biomas website, an international organization based in Brazil.
Processing in ArcGIS Pro: I use a model in ArcGIS Pro that raises the rasters one by one. Then I convert them into polygons, project them and apply an intersect with various subdivision layers (departments, hydrographic basins, localities, etc.) of the territory of Uruguay.
CSV generation: I add a column with the year and generate a CSV file where I am interested in the following variables:
gridcode: soil category codes, defined based on Map Biomas materials.
shape_area: area of the polygons (here lies the problem).
bc_anio: value of the year of registration.
Categorical variable that assigns names to each subdivision of the territory, whose categories change depending on the layer used for the intersect.
Processing in R: I structure the data in R, eliminating unnecessary variables and grouping them by year, geographic subdivision and gridcode. I then pivot gridcode to have each category as a column. This allows me to know how much area there is for each land cover typology in a specific year and location.
The problem: Although I understand that the values I get in ArcGIS Pro are in square meters, when I review them and compare them with external sources, the extents do not match. As I am not a geographer and lack knowledge of trigonometry behind UTM measurements, I suspect the error could be here, but I have not been able to identify it.
Available resources: I have access to the model code in Python and the R code that structures the data. However, I clarify that the Python code is generated by the program and I do not use it directly. In ArcGIS Pro, I operate manually through the graphical interface.
I would greatly appreciate any help or suggestions to resolve this issue. Thank you!
