library(tidyverse)
library(naniar)
#install.packages(c('s2', 'units','rJava','raster', 'sf','terra', 'sftime','stars', 'gstat','RNetCDF', 'ncmeta', 'areal', 'ncdfgeom', 'leafem', 'leafgl', 'leaflegend', 'leaflet', 'leafsync', 'maptiles', 'tmaptools', 'tmap', 'lwgeom', 'leafpop','satellite', 'mapview'), repos = "https://cloud.r-project.org", dependencies = TRUE)
library(sf)
#library(s2)
#library(units)

Data Import

BMF file from: https://nccsdata.s3.amazonaws.com/harmonized/bmf/unified/MD_BMF_V1.1.csv Neighborhood csv file from: https://data.baltimorecity.gov/datasets/baltimore::neighborhood-statistical-area-nsa-boundaries/about

BMF <- read_csv("data/MD_BMF_V1.1.csv")

## Rows: 73768 Columns: 49
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (22): EIN2, NTEE_IRS, NTEE_NCCS, NTEEV2, NCCS_LEVEL_1, NCCS_LEVEL_2, NCC...
## dbl (27): EIN, F990_TOTAL_REVENUE_RECENT, F990_TOTAL_INCOME_RECENT, F990_TOT...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

neighborhoods <- read_csv("data/Neighborhood_Statistical_Area_(NSA)_Boundaries.csv")

## Rows: 279 Columns: 60
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr  (1): Name
## dbl (59): OBJECTID, Population, White, Blk_AfAm, AmInd_AkNa, Asian, NatHaw_P...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Get lat and long for shape file

Shape file from: https://data.baltimorecity.gov/datasets/baltimore::neighborhood-statistical-area-nsa-boundaries/about

https://stackoverflow.com/questions/66381795/check-whether-point-coordinate-lies-within-polygon https://www.statsilk.com/maps/convert-esri-shapefile-map-geojson-format

neighborhood_shape <-st_read("data/Neighborhood_Statistical_Area_(NSA)_Boundaries/Neighborhood_Statistical_Area_(NSA)_Boundaries.shp")

## Reading layer `Neighborhood_Statistical_Area_(NSA)_Boundaries' from data source 
##   `/Users/carriewright/Documents/GitHub/Baltimore_nonprofits/Baltimore_nonprofits_assets/data/Neighborhood_Statistical_Area_(NSA)_Boundaries/Neighborhood_Statistical_Area_(NSA)_Boundaries.shp' 
##   using driver `ESRI Shapefile'
## Simple feature collection with 279 features and 60 fields
## Geometry type: MULTIPOLYGON
## Dimension:     XY
## Bounding box:  xmin: 1393931 ymin: 557733.6 xmax: 1445504 ymax: 621406.8
## Projected CRS: NAD83 / Maryland (ftUS)

Check for missing data

BMF_geo <- BMF %>% dplyr::select(EIN, LATITUDE, LONGITUDE)
any(is.na(BMF_geo$LATITUDE)) # no missing location info

## [1] FALSE

any(is.na(BMF_geo$LONGITUDE))

## [1] FALSE

Limit to just Baltimore

Based on coordinates…

#CRS <- st_crs(neighborhood_shape$geometry) # get class sf version of coordinates
BMF_sf <- st_as_sf(BMF, coords = c('LONGITUDE', 'LATITUDE'), crs = st_crs(4326)) %>% st_set_crs(4326) # Use LONGITUDE and LATITUDE columns and convert to sf object (with a geometry column), use coordinate reference system 4326
BMF_trans <- st_transform(BMF_sf, 2163) #EPSG:2163, convert lat and long to NAD27 / US National Atlas Equal Area for the org data

## Warning in CPL_crs_from_input(x): GDAL Message 1: CRS EPSG:2163 is deprecated.
## Its non-deprecated replacement EPSG:9311 will be used instead. To use the
## original CRS, set the OSR_USE_NON_DEPRECATED configuration option to NO.

neighborhood_tt <- st_transform(neighborhood_shape$geometry, 2163)# convert Baltimore  shape file to NAD27 / US National Atlas Equal Area coordinate reference system

Look for the intersection of orgs with locations within Baltimore

# Make a new column for each org row about if the location is located within the Baltimore shape outline
intersection <- BMF_sf %>% mutate(
  intersection = as.integer(st_intersects(BMF_trans, neighborhood_tt )))

# filter for just data within Baltimore dropping other MD orgs
in_balt <- intersection %>% filter(!is.na(intersection)) # just Baltimore locations

Check that it worked

Checking that it worked using this image: https://commons.wikimedia.org/wiki/File:Baltimore_neighborhoods_map.png

in_balt[1,]$intersection # row for intersection

## [1] 108

in_balt[1,]$EIN # org id

## [1] 2190718

in_balt[1,]$ORG_ADDR_FULL # this looks like downtown when you look it up

## [1] "225 N CHARLES ST,BALT,MD,21201"

# how does this look in the neighborhood file?
neighborhood_shape[in_balt[1,]$intersection,]$Name # neighborhood name

## [1] "Downtown"

#Looks like this is in that location

in_balt[43,]$ORG_ADDR_FULL 

## [1] "901 S BOND ST STE 400,BALTIMORE,MD,21231-3340"

neighborhood_shape[in_balt[43,]$intersection,]$Name # neighborhood name

## [1] "Fells Point"

#Looks like this is in that location

in_balt[58,]$ORG_ADDR_FULL 

## [1] "5114 WINDSOR MILL RD,BALTIMORE,MD,21207"

neighborhood_shape[in_balt[58,]$intersection,]$Name # neighborhood name

## [1] "Wakefield"

#Looks like this is in that location

in_balt[65,]$ORG_ADDR_FULL

## [1] "2700 LIGHTHOUSE POINT EAST,BALTIMORE,MD,21224"

neighborhood_shape[in_balt[65,]$intersection,]$Name # neighborhood name

## [1] "Canton"

#Looks like this is in that location

in_balt[100,]$ORG_ADDR_FULL

## [1] "2404 PENNSYLVANIA AVE 2ND FLR,BALTIMORE,MD,21217-1722"

neighborhood_shape[in_balt[100,]$intersection,]$Name # neighborhood name

## [1] "Penn North"

#Looks like this is in that location

Combining it all together

Want the org info, and neighborhood info and shape information together to enable making maps later.

neighborhood_shape <-as_tibble(neighborhood_shape) # convert to tibble
neighborhood_shape<-neighborhood_shape %>%  mutate(id = row_number()) # add row number column
org_data <-left_join(in_balt, neighborhood_shape, by = c("intersection" = "id")) # look for what row numbers align with the shape file

# save neighborhood R file and org data with neighborhood info
write_rds(neighborhood_shape, file = "data/processed/neighborhood_shape_data.rds")
write_rds(org_data, file = "data/processed/org_data.rds")