Primeros pasos en manejo de datos climatológicos, usando R

Main ideas / concepts

• Goal: Learn the basic workflow in R to manage and visualize monthly climatological precipitation data over time.
• Core tasks:
  • Set up the working environment and connect to the folder containing the data.
  • Load a CSV into an R object.
  • Handle the dataset as a time series-style table (using an R package designed for structured time-series data).
  • Extract specific month columns (e.g., January, February).
  • Create plots with customized axes/labels/titles.
  • Compute accumulated precipitation across months (cumulative totals).
  • Derive summary statistics like the mean and add reference lines to plots.
  • Compute and visualize climatology (average precipitation by month, January–December).
  • Export a plot as a PNG file with specified dimensions and resolution.

---

Methodology / step-by-step instructions

1) Prepare the script and connect to the data directory

• Add an initial comment in the script (using #).
• Use a function to set the working directory to where the CSV file is stored:
  • Copy/paste the path to the folder.
  • If the path causes an error due to quoting/slashes or unexpected characters, adjust the way the path is entered (the subtitles mention issues caused by path formatting).

---

2) Load the CSV file into an R object

• Load the CSV using a read function (e.g., read.csv(...)).
• Assign the loaded dataset to an object (e.g., a name like dts.data.csv appears in the subtitles; exact naming may be distorted).
• After running the code, verify:
  • Observations (rows): 39
  • Variables (columns): 13

---

3) Install and load required libraries (package management)

• Install a package that provides the intended data structure / time-series handling (subtitles mention a commonly used package, but the exact name is unclear).
• Use an installation command with options like “download dependencies” (e.g., a boolean such as TRUE).
• Run installation only once unless you later update versions.
• Load the library (via a library(...) step).
• Watch for function masking warnings (e.g., another package might also define filter, causing R to warn that functions are masked).

---

4) Convert/handle the data as a structured time-series table

• After loading the relevant library, convert the dataset into an object suited for time series / “ts” style tabular access.
• The dataset includes:
  • A year column
  • Monthly precipitation columns (January through December)

---

5) Extract precipitation for specific months

January

• Select all rows and the January column by index (using bracket indexing like object[rows, columns]).
• Create a new object (e.g., “chosen” / “january object”) containing:
  • Years
  • January precipitation values

February

• Similarly select:
  • Rows = all observations
  • Column = February
  • The subtitles mention indexing patterns such as using a vector of indices (e.g., c(1, 3)) when selecting specific columns.

---

6) Plot monthly precipitation (basic time series plot + styling)

• Use a plotting function to visualize:
  • X-axis: years
  • Y-axis: precipitation
• Add customization examples:
  • Use points/lines instead of defaults (via plotting arguments controlling “points” vs “lines”).
  • Provide axis labels, such as:
    • X label: years
    • Y label: precipitation (with units like millimeters)
  • Provide a plot title indicating the time span and station/context.
• Note common formatting mistakes:
  • For example, forgetting a comma (or removing punctuation) can cause an “unexpected error.”

---

7) Compute accumulated precipitation (cumulative totals)

• Create a new column (e.g., total) for accumulated precipitation.
• Compute the cumulative total per year by summing across the correct monthly columns.
• The subtitles emphasize correcting an off-by-column issue:
  • Ensure the summation includes the intended month columns (they mention ranges like “columns 2 to 13,” then later clarify the intended January-to-December mapping via column indices).
• Confirm that the dataset now has one additional column (e.g., from 13 to 14 columns).

---

8) Plot accumulated precipitation over time

• Plot total/cumulative precipitation versus years.
• Ensure the mapping is correct:
  • X = year values
  • Y = cumulative totals
• Customize:
  • Labels (years and precipitation units)
  • Title referencing the interannual period (e.g., 1975–2013)

---

9) Compute mean precipitation and add a horizontal reference line

• Compute the mean of the interannual total precipitation series.
• Add a horizontal reference line to the plot using a “lines/abline”-style function:
  • Horizontal line (h) at the mean value
  • Choose color, line type, and thickness
• The subtitles stress using the computed mean from the data so the line updates automatically if the dataset changes.

---

10) Compute climatology (average precipitation by month)

• Compute monthly averages for each month (January–December):
  • Take column means from the time-series table for each month column.
• Plot climatology:
  • Use a bar plot (bars stick together; adjust spacing as needed).
  • Customize:
    • fill color
    • border color
    • axis labels
    • title
  • Adjust y-axis limits if needed.
  • Add a horizontal baseline (e.g., at 0) if desired.
• Compute the climatology mean and add a reference line at that value (similar to the interannual mean step).

---

11) Export the climatology plot to a PNG

• Use an export workflow (e.g., a png(...)-style workflow).
• Specify parameters:
  • Filename (e.g., "climatology")
  • Extension: .png
  • Units (e.g., centimeters)
  • Width (e.g., 15 cm), height (e.g., 10 cm)
  • Resolution (e.g., 300 dpi)
• Save to the working directory:
  • If a graphics folder exists, the subtitles mention it can be used.

---

Speakers / sources featured

• No specific named speakers are clearly identified.
• Main references are:
  • R base functions (CSV reading, plotting, mean, sum, etc.)
  • A time-series/table-handling library (exact package name is unclear due to transcription errors)
  • Base R graphics plotting functions