select grid cells based on reference grid

Author: wlangera

data/b3data_package/datapackage.json

@@ -10,7 +10,7 @@
       "mediatype": "text/csv",
       "encoding": "utf-8",
       "title": "Occurrence cube for birds in Belgium (MGRS 10 km)",
-      "description": "Occurrence cube for birds in Belgium between 2000 en 2024. The taxonomical resolution is 'species' and the temporal resolution is 'year' Spatial aggregation is done using the MGRS grid at 10 km scale.",
+      "description": "Occurrence cube for birds in Belgium between 2000 en 2024. The taxonomical resolution is 'species' and the temporal resolution is 'year' Spatial aggregation is done using the MGRS grid at 10 km scale. Only grid cells that fall within the 10 km MGRS reference grid for mainland Belgium (see b3data: `mgrs10_refgrid_belgium.geojson`) are included.",
       "sources": ["GBIF Occurrence Download", "https://doi.org/10.15468/dl.y3wpwk"],
       "licenses": [" CC BY-NC 4.0", "https://creativecommons.org/licenses/by-nc/4.0/", "Creative Commons Attribution-NonCommercial 4.0 International"],
       "schema": {

source/create_b3data_package.Rmd

@@ -21,6 +21,7 @@ knitr::opts_chunk$set(echo = TRUE)
 library(tidyverse)     # Data wrangling and visualisation
 library(frictionless)  # Create frictionless data package
 library(rgbif)         # Create occurrence cubes from GBIF data
+library(sf)            # Spatial objects
 
 # Source functions
 source(here::here("source", "R", "download_occ_cube.R"))
@@ -36,8 +37,8 @@ Create [frictionless](https://docs.ropensci.org/frictionless/) data package with
 
 # Methods
 
-This report focuses on creating the **b3data** Frictionless data package for **tabular datasets**.  
-A separate report adds **spatial resources** (e.g. GeoJSON, raster) to the same data package.
+This report focuses on creating the **b3data** frictionless data package for tabular datasets.  
+A separate report adds spatial resources (e.g. GeoJSON, raster) to the same data package.
 
 ## Data Package Overview
 
@@ -130,14 +131,30 @@ bird_cube_query <- "SELECT
 # nolint end
 
 # Download and load cube
-bird_cube_belgium_mgrs10 <- download_occ_cube(
+bird_cube_belgium_mgrs10_full <- download_occ_cube(
   sql_query = bird_cube_query,
   file = "bird_cube_belgium_mgrs10.csv",
   path = data_path,
   overwrite = FALSE
 )
 ```
 
+We only select the grid cells that belong to the MGRS 10 km reference grid for the mainland of Belgium (see b3data `mgrs10_refgrid_belgium.geojson`).
+
+```{r}
+# Read reference grid
+utm10_bel <- st_read(file.path(data_path, "utm10_bel.shp"))
+
+# Visualise
+ggplot() + geom_sf(data = utm10_bel)
+```
+
+```{r}
+# Only select grid cells from reference grid
+bird_cube_belgium_mgrs10 <- bird_cube_belgium_mgrs10_full %>%
+  filter(substr(mgrscode, 4, nchar(mgrscode)) %in% utm10_bel$TAG)
+```
+
 We create the package and add the dataset.
 
 ```{r}
@@ -150,7 +167,10 @@ b3data_package <- create_package() %>%
       "Occurrence cube for birds in Belgium between 2000 en 2024.",
       "The taxonomical resolution is 'species' and",
       "the temporal resolution is 'year'",
-      "Spatial aggregation is done using the MGRS grid at 10 km scale."),
+      "Spatial aggregation is done using the MGRS grid at 10 km scale.",
+      "Only grid cells that fall within the 10 km MGRS reference grid for",
+      "mainland Belgium (see b3data: `mgrs10_refgrid_belgium.geojson`) are",
+      "included."),
     sources = c(
       title = "GBIF Occurrence Download",
       path = "https://doi.org/10.15468/dl.y3wpwk"