Model attribute areal coverage discrepancy

The model-attributes layer contains fewer divide_id values than the corresponding divides layer.

The below example shows that for comid 6716129, there are 18 unique divide ids in the divides layer, but only 13 in the model-attributes layer. This amounts to 220.5 km^2 represented in the divides layer, whereas the model-attributes layer accounts for 181.4 km^2.

library(hfsubsetR)
library(dplyr)
library(glue)
attr_div <- hfsubsetR::get_subset(comid = "6716129", 
                             lyrs = c("model-attributes","divides"),
                             source   ="s3://lynker-spatial/hydrofabric",
                             hf_version =  '2.1.1',
                             type    = "nextgen",
                             domain  = "conus")

# Combine model attributes with divides
attrs_area <- dplyr::left_join(x=attr_div$`model-attributes`, attr_div$divides, by="divide_id")

# Note that there are additional divide ids inside the divides layer, meaning the model-attributes layer is missing some divide ids
print(glue::glue("Total divide ids inside divides layer: {length(unique(attr_div$divides$divide_id))}"))
print(glue::glue("Total divide ids inside model-attributes layer: {length(unique(attr_div$`model-attributes`$divide_id))}"))

# Total area accounted for in each layer:
print(glue::glue("Total catchment area inside the divides layer: {sum(attr_div$divides$areasqkm)} km^2"))

print(glue::glue("Total catchment area inside the model-attributes layer: {sum(attrs_area$areasqkm)} km^2"))

NOAA-OWP / hydrofabric

Model attribute areal coverage discrepancy #254