Skip to contents

Sequoia

11_sequoia.Rmd
library(Rsequoia2)
#> Attempting to load the package cache ... No cache found.
#> Will try to reconfigure qgisprocess and build new cache ...
library(tmap)
library(openxlsx2)
library(sf)
#> Linking to GEOS 3.12.1, GDAL 3.8.4, PROJ 9.4.0; sf_use_s2() is TRUE

What is Sequoia

Sequoia is an opiniated way of dealing with forest management. It’s goal is to :

  • Facilitate downloading of data for forest management ;
  • Normalize data (name and structure) ;
  • Facilitate exploration of those data

What is a Sequoia process

Sequoia process always have those step :

  • Setting a directory ;
  • Creating parca matrice (excel with cadastrals parcels info);
  • Downloading parca geometry;
  • Downloading data based on parca;
  • GIS part to create analysis units ;
  • Generating final layer from analysis units

Setting a directory

All seq_* function have dirname = "." argument, meaning they use the current working directory by default. Best practice is to create a dedicated directory and set it as the working directory for the R session.

For the example, a forest named "ECKMUHL" is used:

sequoia_dir <- "ECKMUHL"
dir.create(sequoia_dir)

# setting current dir to our sequoia dir
setwd(sequoia_dir)

Creating parca matrice

Rsequoia2 provides several ways to generate this fundamental file:

  • A manually filled Excel file (see Cadastral Parcels article)
  • Using legal-entity open data (see Cadastral Parcels - Legal Entity article)
  • Generating the file from DGFiP PDF matrices

Downloading parca geometry

All data downloaded by Rsequoia2 are based on cadastral parcels (parca). The first mandatory step is therefore to download them using seq_parca().

Downloading data

Rsequoia2 offer many datasets, which can be downloaded using seq_* functions:

  • Environnemental data from MNHN ;
  • Elevation data from IGN ;
  • Orthophoto data from IGN ;
  • Geological data from BRGM ;
  • Hydrological data from BD Topo IGN ;
  • Patrimony data from Atlas du Patrimoine ;
  • …

Analyst unist

The next step after downloading parca and the additional datasets is to create the analysis units.

This layer is generated using seq_parca_to_ua(). Your task is to segment the layer so that it reflects the actual forest structure.

Additional layers can help with this step, but the real diagnostic comes from field data collection.

For more details, see the Analysis units article.

Finalizing the layer

The final step, after manually completing the UA layer, is to make sure everything is correct (and, more importantly, that nothing was accidently corrupted). Some additional processing is also required to properly finalize this layer.

seq_ua() performs a series of checks and corrections to validate and finalize the UA layer:

  • Checks cadastral IDU consistency between UA and PARCA
  • Ensures UA cadastral areas match PARCA by correcting any detected differences
  • Generates a standardized management unit identifier (UG) in the UA object
  • Computes corrected cadastral areas for analysis units based on cartographic area
  • Checks management unit (UG) consistency within the UA sf object
  • Detects and corrects minor inconsistencies within management units (UG)
  • Cleans geometry topology

Summary

Last but not least, the final step is to generate tables that help users better understand and write their forest management document.

seq_summary() produces a set of opinionated summary tables derived from the UA layer.

How it works

Internal configuration

Rsequoia2 relies on an internal configuration that maps keys to data layers and their metadata. For example, the key "r.sol.geol" refers to the geological raster layer named GEOL.tif.

All available keys can be listed with seq_key(). To access metadata for a specific layer, use seq_layer("r.sol.geol").

Note that full keys are not always required: seq_layer("geol") also works.

seq_* functions

All main functions in Rsequoia2 start with seq_* (e.g. seq_geol(), seq_prsf(), …).

These functions: - Read the project cadastral parcels to defined area of interest ; - Download the required data ; - Save them according to the configuration (name and extension) ; - Return path(s) where the data are stored.

Note: for advanced users, each seq_* function has an equivalent get_* function (e.g. get_geol(), get_prsf(), …). These can be used independently of the Sequoia workflow.

Cheat code !

For those who made it to the end, Rsequoia2 provides a wrapper function to guide you through the entire workflow: sequoia().

Have fun !