Introduction to geoidep

Antony Barja

22 Setiembre, 2024

1. Introduction

This package aims to provide R users with a new way of accessing official Peruvian cartographic data on various topics that are managed by the country’s Spatial Data Infrastructure.

By offering a new approach to accessing this official data, both from technical-scientific entities and from regional and local governments, it facilitates the automation of processes, thereby optimizing the analysis and use of geospatial information across various fields.

However, this project is still under construction, for more information you can visit the GitHub official repository https://github.com/ambarja/geoidep.

MIDAGRIMINAMMINCUMINEMMINDEFMIMPMIDISMINEDUMTCADSgeoidep(R package)

If you want to support this project, you can support me with a coffee for my programming moments.

2. Package installation

install.packages("geoidep")

Also, you can install the development version as follows:

install.packages('pak')
pak::pkg_install('ambarja/geoidep')
library(geoidep)

3. Basic usage

providers <- get_data_sources()
providers
#> # A tibble: 75 × 7
#>    provider  category      layer       layer_can_be_actived admin_en year  link 
#>    <chr>     <chr>         <chr>       <lgl>                <chr>    <chr> <chr>
#>  1 INEI      General       departamen… TRUE                 Nationa… 2019  http…
#>  2 INEI      General       provincia   TRUE                 Nationa… 2019  http…
#>  3 INEI      General       distritos   TRUE                 Nationa… 2019  http…
#>  4 Midagri   Agriculture   vegetation… TRUE                 Ministr… 2018  http…
#>  5 Midagri   Agriculture   agricultur… TRUE                 Ministr… 2024  http…
#>  6 Midagri   Agriculture   oil_palm_a… TRUE                 Ministr… 2016… http…
#>  7 Midagri   Institucional experiment… TRUE                 Ministr… 2024  http…
#>  8 Geobosque Forest        stock_bosq… FALSE                Ministr… 2001… http…
#>  9 Geobosque Forest        stock_bosq… TRUE                 Ministr… 2001… http…
#> 10 Geobosque Forest        stock_bosq… TRUE                 Ministr… 2001… http…
#> # ℹ 65 more rows
layers_available <- get_providers()
layers_available
#> # A tibble: 5 × 2
#>   provider  layer_count
#>   <fct>           <int>
#> 1 Geobosque           5
#> 2 INEI                3
#> 3 Midagri             4
#> 4 Serfor              2
#> 5 Sernanp            61

4. Download Official Administrative Boundaries by INEI

# Region boundaries download 
loreto_prov <- get_provinces(show_progress = FALSE) |> 
  subset(NOMBDEP == 'LORETO')
library(leaflet)
library(sf)
loreto_prov |> 
  leaflet() |> 
  addTiles() |> 
  addPolygons()
# Defined Ubigeo 
loreto_prov[["ubigeo"]] <- paste0(loreto_prov[["CCDD"]],loreto_prov[["CCPP"]])
# The first five rows
head(loreto_prov)
#> Simple feature collection with 6 features and 9 fields
#> Geometry type: MULTIPOLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -77.82596 ymin: -6.14773 xmax: -69.94904 ymax: -0.03860597
#> Geodetic CRS:  WGS 84
#>      id OBJECTID CCDD CCPP NOMBDEP                NOMBPROV SHAPE_Length
#> 136 136      136   16   02  LORETO           ALTO AMAZONAS     9.955533
#> 137 137      137   16   07  LORETO       DATEM DEL MARAÑON    11.493679
#> 138 138      138   16   03  LORETO                  LORETO    13.602250
#> 139 139      139   16   04  LORETO MARISCAL RAMON CASTILLA    15.113580
#> 140 140      140   16   01  LORETO                  MAYNAS    14.501777
#> 141 141      141   16   08  LORETO                PUTUMAYO    20.443604
#>     SHAPE_Area                           geom ubigeo
#> 136   1.569481 MULTIPOLYGON (((-76.30752 -...   1602
#> 137   3.881998 MULTIPOLYGON (((-76.49832 -...   1607
#> 138   5.574816 MULTIPOLYGON (((-75.74592 -...   1603
#> 139   3.054652 MULTIPOLYGON (((-72.08996 -...   1604
#> 140   6.197942 MULTIPOLYGON (((-75.24086 -...   1601
#> 141   3.725588 MULTIPOLYGON (((-75.10004 -...   1608

5. Working with Geobosque data

my_fun <- function(x){
  data <- get_forest_loss_data(
    layer = 'stock_bosque_perdida_provincia',
    ubigeo = loreto_prov[["ubigeo"]][x],
    show_progress = FALSE )
  return(data)
}
historico_list <- lapply(X = 1:nrow(loreto_prov),FUN = my_fun)
historico_df <- do.call(rbind.data.frame,historico_list)
# The first five rows
head(historico_df)
#>   anio perdida rango1 rango2 rango3  rango4  rango5 tipobosque ubigeo
#> 1 2001 2081.97      0   0.00  40.05  519.21 1522.71          1   1602
#> 2 2002 3923.46      0   0.00 136.71 1254.51 2532.24          1   1602
#> 3 2003 1843.92      0   0.00  23.94  584.73 1235.25          1   1602
#> 4 2004 4269.33      0   0.00 351.36 1818.00 2099.97          1   1602
#> 5 2005 4948.20      0   0.00 351.09 2020.50 2576.61          1   1602
#> 6 2006 2274.48      0 209.07 152.64  689.22 1223.55          1   1602

6. Simple visualization with ggplot

library(ggplot2)
library(dplyr)
historico_df |> 
  inner_join(y = loreto_prov,by = "ubigeo") |> 
  ggplot(aes(x = anio,y = perdida)) + 
  geom_point(size = 1) + 
  geom_line() + 
  facet_wrap(NOMBPROV~.,ncol = 3) + 
  theme_minimal(base_size = 5) + 
  labs(
    title = "Pérdida de bosque histórico del 2001-2023 \npara las provincias de Loreto",
    caption = "Fuente: Geobosque",
    x = "",
    y = "")