A.3 Public or Tax-Exempt King County Parcels

Description

A dataset of parcels within King County, Washington. Each record in this datatset is not currently paying taxes - a status which may make them a suitable candidate for the development of affordable housing and early learning facilities.

This data reflects the King County Assessor records as of October 13, 2017.

Summary

Links

• Sources
  • King County Assessor Data Download
  • King County GIS Portal
• Datasets
  • Parcel (tabular)
  • Parcel (spatial)
  • Real Property Account
• Metadata
  • Parcel (tabular)
  • Parcel (spatial)
  • Real Property Account

Processing Steps

Summary

1. Create p: a tibble containing parcel data for the entire county.
2. Create acct: a tibble containing real property account data for the entire county.
3. Create p_sf: a simple feature object containing the spatial polygon representations of each parcel in the county-wide dataset.
4. Clean/process the three dataset: prepare the datasets to be joined and filtered.
   • Load tax_status and tax_reason for transforming coded features into descriptive text.
   • Load pub for joining the public parcels.
5. Create pub_or_tax_exempt_sf: Join and filter the datasets.
   • Join p and acct with inner_join() (an implicit filter that returns all records contained in both datasets)
   • Filter by excluding records whose TAX_STATUS is Taxable.
   • Create a subset of pub which includes only those records not found in p_sf
   • Combine the two filtered objects together using bind_rows()
   • Drop any record which has NA in the TAX_STATUS column (these records are mostly empty)
   • Join and filter the combined object to p_sf using inner_join() - this returns a simple feature class spatial object (i.e., it can be shown on a map).
6. Upload to Drive

Step 1

p_fp <- root_file("./1-data/2-external/EXTR_Parcel_20171013.csv")


p_load <- 
  make_or_read(p_fp,
               {
                 zip_dir <- root_file("./1-data/2-external")
                 
                 target_name <- "EXTR_Parcel_20171013.csv"
                 
                 dr_id <- as_id("0B5Pp4V6eCkhraF9jOTl3bURiMkU")
                 
                 drive_read_zip(
                   dr_id = dr_id,
                   .tempdir = FALSE,
                   dir_path = zip_dir,
                   read_fun = read_csv,
                   target_name = target_name
                 ) 
               },
               {
                 read_csv(p_fp)
               }) 


p <- 
  p_load %>% 
  rename_all(to_screaming_snake_case) 

rm(p_load)
gc(verbose = FALSE)

Step 2

acct_fp <- root_file("./1-data/2-external/EXTR_RPAcctl_20171013.csv")

acct_load <- 
  make_or_read(acct_fp,
               {
                 dr_id <- "0B5Pp4V6eCkhreGt2djZCTnFUeFU"
                 
                 zip_dir <- root_file("./1-data/2-external")
                 
                 target_name <- "EXTR_RPAcctl_20171013.csv"
                 
                 drive_read_zip(dr_id = dr_id,
                                .tempdir = FALSE,
                                dir_path = zip_dir,
                                read_fun = read_csv,
                                target_name = target_name)
                 
               },
               {read_csv(acct_fp)}) 

acct <- 
  acct_load %>% 
  rename_all(to_screaming_snake_case)

rm(acct_load)
gc(verbose = FALSE)

Step 3

p_sf_fp <- root_file("./1-data/2-external/parcel")

p_sf_load <- 
  make_or_read(p_sf_fp,
               {
                 dr_id <- "0B5Pp4V6eCkhrRnM4bHFmWTBTQnM"
                 
                 zip_dir <- root_file("./1-data/2-external")
                 
                 target_name <- "parcel"
                 
                 drive_read_zip(dr_id = dr_id,
                                dir_path = zip_dir,
                                read_fun = st_read,
                                target_name = target_name,
                                .tempdir = FALSE,
                                layer = "parcel",
                                stringsAsFactors = FALSE)
               },
               {st_read(p_sf_fp, layer = "parcel", stringsAsFactors = FALSE)})

p_sf <-
  p_sf_load %>%
  rename_if(not_sfc, to_screaming_snake_case)

rm(p_sf_load)
gc(verbose = FALSE)

Step 4

tr_fp <- root_file("./1-data/3-interim/kc-tax-reason.rds")

tax_reason <- 
  make_or_read(tr_fp,
               {
                 dr_id <- as_id("0B5Pp4V6eCkhrQzdjY0hKVUhOR3M")
                 
                 drive_read(dr_id = dr_id,
                            .tempfile = FALSE,
                            path = tr_fp,
                            read_fun = read_rds)
                 
               },
               {read_rds(tr_fp)})

ts_fp <- root_file("./1-data/3-interim/kc-tax-status.rds")

tax_status <- 
  make_or_read(ts_fp,
               {
                 dr_id <- as_id("0B5Pp4V6eCkhrWFlYLWM3RC1pRDA")
                 
                 drive_read(dr_id = dr_id,
                            .tempfile = FALSE,
                            path = ts_fp,
                            read_fun = read_rds)
                 
               },
               {read_rds(ts_fp)})


pub_fp <- root_file("./1-data/3-interim/public_parcels_20170727.rds")

pub <- 
  make_or_read(pub_fp,
               {
                 dr_id <- as_id("0B5Pp4V6eCkhrSUhXdHNtdjBfa28")
                 
                 drive_read(dr_id = dr_id,
                            path = pub_fp,
                            read_fun = read_rds,
                            .tempfile = FALSE)
               },
               {read_rds(pub_fp)})

p_ready <- 
  p %>% 
  mutate(PIN = str_c(MAJOR, MINOR),
         DATE = ymd(20171013)
         ) %>% 
  select(PIN,
         DATE,
         PROP_NAME,
         PROP_TYPE,
         DISTRICT_NAME,
         CURRENT_ZONING,
         PRESENT_USE,
         SQ_FT_LOT,
         ACCESS,
         TOPOGRAPHY,
         RESTRICTIVE_SZ_SHAPE,
         PCNT_UNUSABLE,
         CONTAMINATION,
         HISTORIC_SITE,
         EASEMENTS,
         SEISMIC_HAZARD,
         LANDSLIDE_HAZARD,
         STEEP_SLOPE_HAZARD 
  )


acct_frmt <- 
  acct %>% 
  mutate(PIN = str_c(MAJOR,MINOR),
         DATE = ymd(20171013),
         TAX_STATUS = tax_status[TAX_STAT],
         TAX_REASON = tax_reason[TAX_VAL_REASON]) %>% 
  select(PIN, 
         DATE,
         ACCT_NBR,
         BILL_YR,
         TAX_STATUS,
         TAX_REASON,
         APPR_LAND_VAL,
         APPR_IMPR_VAL = APPR_IMPS_VAL,
         TAXABLE_LAND_VAL,
         TAXABLE_IMPR_VAL = TAXABLE_IMPS_VAL)

acct_ready <- 
  acct_frmt %>% 
  miscgis::subset_duplicated("PIN") %>% 
  group_by(PIN) %>% 
  drop_na %>% 
  ungroup %>% 
  bind_rows(subset_duplicated(acct_frmt,"PIN",notin = TRUE)) %>% 
  arrange(PIN)

p_sf_ready <- 
  select(p_sf, PIN)

Step 5

p_join <- 
  list(acct_ready,p_ready) %>% 
  reduce(.f = inner_join, by = c("PIN","DATE")) 

pub_join <- 
  pub %>% 
  select(-matches("APPR")) # remove these columns to avoid duplicates later

# in_pub_not_p <- anti_join(pub_join,p_join, by = "PIN")  # note: the public parcels not found in the master parcel dataset are dropped

not_taxed <- 
  p_join %>% 
  filter(TAX_STATUS %!in% "Taxable") %>% 
  left_join(pub_join, by = c("PIN", "DATE"))

pub_join_other <- 
  anti_join(pub_join, not_taxed, by = "PIN") %>% 
  left_join(p_join, by = c("PIN","DATE"))

pub_or_tax_exempt <- 
  bind_rows(not_taxed, pub_join_other) %>% 
  filter(!is.na(TAX_REASON))

pub_or_tax_exempt_sf <- 
  inner_join(pub_or_tax_exempt, p_sf, by = "PIN") %>% 
  st_as_sf

Step 6

pub_or_tax_exempt_sf_fp <- "./1-data/3-interim/pub_or_tax_exempt_sf_20171013.gpkg"

st_write(pub_or_tax_exempt_sf,dsn = pub_or_tax_exempt_sf_fp, driver = "GPKG")

drive_folder <- as_id("0B5Pp4V6eCkhrZ3NHOEE0Sl9FbWc")

drive_upload(pub_or_tax_exempt_sf_fp, path = drive_folder, name = "pub_or_tax_exempt_sf_20171013.gpkg")