Thank you for taking this test as part of the BE Group recruitment process. Please read this content carefully. If you find any issue with the data or critical question please contact with [email protected] and [email protected].

We strongly suggest you to use the latest version of R and possibly RStudio. Please let us know as soon as possible if this may be an issue for you. You can download and install R from CRAN and Rstudio from Posit website.

We’ve prepared two small tests for our recruitment process. The idea is that you will build the best approach to the problem and answer to the specific questions they propose. You will be asked to send us your approach and the main outputs in a short data report by the end of the date agreed by email. Remember that it will require a suite of packages: tidyverse, sf, data.table here, scales, raster, ggplot, and its dependencies.

Your report must include answers to the questions with a a short description of your outputs (maps, figures or tables) as well as the code used.

You will prepare and present a 5 minutes talk with your achievements for each test, focusing on the critical aspects of the proposed questions. Add the details you may think will help to understand it. Be clear and concise. You can use a R Markdown presentation, a Power Point presentation or any other resource that best fits your needs.

• Map the wind energy facilities in mainland Portugal regarding the country and district borders. Apply the appropriate legend and choose one aggregation level (legend) to wind energy facilities.
• Map the wind facilities and and its distribution regarding elevation and protected areas.
• Find the the appropriate theme for the digital elevation model.
• What was the total national installed capacity (in GW) in 2019?
• How facilities are distributed in relation to altitude? plot the attitudinal distribution and describe.
• How it has grown (installed GW) since the first installed facility?
• Are protected areas affected? Where and to which extent (number of Sites)?

Build an interactive map where one can query for each facility its name, year of commissioning and installed capacity (using provided e2p data).

You will find the data to work in the ~.data/ folder.

• Describe the number of bird movements/hour.
• Describe mean flight height/hour.
• Are there any differences between mean number of bird movements/hour if you distinguish day and night periods?
• Are there any differences between mean flight altitude/hour if you distinguish day and night periods?

• Describe the pattern of bird movement in relation to tide height and distance to next high tide.

Suggestion: retain the first observation of each bird movement to map the distance for next high tide.

You will find the required data to work in the ~.data/ folder.

This dataset is GPS-PPT like. Every line is a obtained position. Individual movements are identified by a single unique ID [track_id]. A few variables are available for every position but they can result a NULL/NA (fail to record).

• Take care with projections. Data are provided in geographic (WGS84 EPSG 4326) and projected reference systems (PT-TM06 EPSG 3763).

• Remember to deal with NA values when summarizing.

• Use a raster elevation model to get elevation data from. We are not providing any DEM raster for this test. For 30m SRTM DEM, the tiles for Portugal can be found here. A 25m resampled SRTM can be found here.

• Remember that some questions will be easier to approach using data.table package (specially directional roll joins).

• You may find it useful to programmatically download and read this 25m SRTM raster tiles directly from this temporary source.

# Install and load packages
library(googledrive)
library(raster)

temp <- tempfile(fileext = ".zip")

# For tile srtm_35_05 the link is
# https://drive.google.com/file/d/1D8QNshjsY7X700X8MfXFqO3RoXpKGQ2x/view?usp=sharing
# Retain tile_ID from the link above

# Download a ~20Mb zip file. 
# In the process a wizard will guide you to work with the API (use your Google account)
dl <- drive_download(
  as_id("1D8QNshjsY7X700X8MfXFqO3RoXpKGQ2x"), path = temp, overwrite = TRUE)

# Unzip to a temp folder (~170Mb asc file)
out <- unzip(temp, exdir = tempdir())

# Read asc raster from temp folder
tile1 <- raster(out[2])

# Plot to check
plot(tile1)

• If running your analysis from Windows OS, you can import data directly from github repo with:

dat_url <- 'https://raw.githubusercontent.com/pecard/testeR/master/data/e2p.rda?raw=true'
download.file(dat_url,"e2p", method="curl")
load("e2p")
e2p

Instead, you can simply download the data from the repository into a local folder.