plot_scenarios: Plot historic and future scenarios for climate-related...

In chillR: Statistical Methods for Phenology Analysis in Temperate Fruit Trees

Plot historic and future scenarios for climate-related metrics (ggplot2 version)

Description

Visualize outputs from the temperature_generation function used in climate-related assessments. These outputs are usually compiled with the make_climate_scenario function.

Usage

plot_scenarios(
  scenario_list,
  metric,
  add_historic = TRUE,
  ...,
  outlier_shape = 19,
  historic_color = "white",
  group_by = c("Scenario", "Year"),
  y_axis_name = paste("Cumulative response in", metric),
  x_axis_name = "Year",
  legend_title = "Climate model",
  legend_labels = NULL,
  panel_labels = NULL,
  base_size = 11
)

Arguments

scenario_list	is a list of lists containing information and data about the scenarios to be plotted. These lists must have: an element named data, which should be a list containing one or more named dataframes with a column named the same as the metric argument. This column must contain (numeric) information to be plotted. Dataframes of climate-related metrics can be obtained with the tempResponse_daily_list function. For past scenarios, the names of the dataframes can be the reference years used to generate the scenarios. These names will be recycled and used in the x-axis of the historic panel. For future scenarios, the names of the dataframes can be the models used in the projections. These names will appear in the legend for future panels. an element named caption containing information about the scenario which the list is related to. an element named historic_data which represents a data frame for actual observations in past scenarios. This element can be optional but is mandatory if add_historic = TRUE time_series is an optional argument that defines whether the scenario contains a time series. labels is an optional vector that usually contains the names of the elements used for metric_summary in make_climate_scenario.
metric	is a character string corresponding to the name of the column that contains the data of interest in the dataframe of the scenario_list (and, if applicable, in the historic_data).
add_historic	is a boolean parameter to define whether the plot should include the actual observations of historic climate-related metrics.
...	accepts arguments that can be passed to layer and are commonly used outside the aesthetic function for different geoms. In this case, ... is passed to the geom_point function in the case that actual observations of chill or heat are displayed. Options are size, color, among others.
outlier_shape	is the optional shape to replace the outliers in the boxplots. To show no oultiers use NA. See shape for shape options.
historic_color	is a character string corresponding to the color used to fill the boxplots in simulated historic scenarios. Supported options are those provided by colors.
group_by	is a vector of character strings indicating how the plots should be grouped. I.e. by Scenario and then Year or viceversa.
y_axis_name	is a character string representing the title of the y axis in the final plot. Default is set to paste('Cumulative response in', metric) to let the function obtain the name based on the metric argument.
x_axis_name	is a character string representing the title of the x axis in the 'Historic' panel. Default is set to Year.
legend_title	is a character string representing the title of the legend showing the climate models used in the assessment.
legend_labels	is a vector of character strings that allows the user to modify the names of the climate models used in the projections. The length of the vector must coincide with the number of climate models. Default is set to NULL to let the function use the labels generated with the make_climate_scenario function.
panel_labels	is a list of 3 named objects that allows the user to customize the text in the upper part of the plot. Default is set to NULL to let the function use the labels generated with the make_climate_scenario function. If provided, the objects of the list must be: an element named Historic containing the name to be used in the 'Historic' panel. an element named Scenario containing the names of the scenarios used for the projections. If group_by = c("Year", "Scenario") is used, Scenario must be a list of named objects according to the labels used in the Year object. See examples. an element named Year containing the labels to be used for the time horizons used in the assessment. If group_by = c("Scenario", "Year") is used, Year must be a list of named objects according to the labels used in the Scenario object. See examples.
base_size	is an integer to define the relative size of the text in the final plot. This argument is passed to ggpplot2::theme_bw. Default is set to 11.

Details

plot_scenarios uses the ggplot2 syntax for producing separated plots for historic and future scenarios. Later, the plots are merged into one final figure by using the patchwork library.

Value

A plot of classes 'patchwork', 'gg', and 'ggplot'. This allows the user to continue editing some features of the plots through the syntax (i.e. '&', and '+') from both libraries (see examples).

Author(s)

Eduardo Fernandez and Eike Luedeling

Examples

# Make 3 identical objects as scenarios; let's assume these represent the
# years 2000, 2005 and 2010.

library(chillR)

# Compute chill responses for KA_weather data

chill <- tempResponse(stack_hourly_temps(
                      fix_weather(KA_weather[which(KA_weather$Year > 2006), ]),
                      latitude = 50.4), Start_JDay = 305, End_JDay = 60)


# Simulated scenarios labels

past_labels <- c(2000, 2005, 2010)

# Models labels

models_labels <- c("Climate model 1", "Climate model 2",
                   "Climate model 3")

# Add named elements to past and future scenarios

scenario_results_past <- list(`2000` = chill,
                              `2005` = chill,
                              `2010` = chill)
                              
scenario_results_future <- list(`Climate model 1` = chill,
                                `Climate model 2` = chill,
                                `Climate model 3` = chill)

# Define the climate scenario  

climate_scenario_list <- list(list(data = scenario_results_past,
                                   caption = c("Historic", "data"),
                                   time_series = TRUE,
                                   labels = past_labels,
                                   historic_data = chill),
                              list(data = scenario_results_future,
                                   caption = c("Scenario 1", "2050"),
                                   labels = models_labels),
                              list(data = scenario_results_future,
                                   caption = c("Scenario 1", "2075"),
                                   labels = models_labels),
                              list(data = scenario_results_future,
                                   caption=c("Scenario 1", "2100"),
                                   labels = models_labels),
                              list(data = scenario_results_future,
                                   caption=c("Scenario 2", "2040"),
                                   labels = models_labels),
                              list(data = scenario_results_future,
                                   caption=c("Scenario 2", "2080"),
                                   labels = models_labels))
                                   
# Plot the climate scenarios

plot_scenarios(climate_scenario_list, metric = 'Chill_Portions',
               add_historic = TRUE, size = 2, shape = 3, color = 'blue',
               outlier_shape = 12, historic_color = 'skyblue',
               group_by = c("Year", "Scenario"))
               
## Plot scenarios modifying the whole text in the plot
## We will comment the next examples to reduce the running time in CRAN
## submissions...
# plot_scenarios(scenario_list = climate_scenario_list, metric = 'Chill_Portions',
#                add_historic = TRUE, size = 2, shape = 3, color = 'blue',
#                outlier_shape = 12, historic_color = 'skyblue',
#                group_by = c("Scenario", "Year"),
#                y_axis_name = "Acumulacion de frio en CP",
#                x_axis_name = "Tiempo",
#                legend_title = "Modelo climatico",
#                legend_labels = c("Modelo 1", "Modelo 2", "Modelo 3"),
#                panel_labels = list(Historic = "Historico",
#                                    Scenario = c("Escenario 1",
#                                                 "Escenario 2"),
#                                    Year = list(`Escenario 1` = c("Futuro cercano",
#                                                                  "Futuro medio",
#                                                                  "Future lejano"),
#                                                `Escenario 2` = c("Futuro cercano",
#                                                                  "Futuro medio"))))

## Since the output is a ggplot object, it is possible to continue
## modifying some general aspects of the plot

## Define the basic plot 
# plot <- plot_scenarios(climate_scenario_list, metric = 'Chill_Portions',
#                        add_historic = TRUE, size = 2, shape = 3, color = 'blue',
#                        outlier_shape = 12, historic_color = 'skyblue')


## Example to change the color of the climate model scale

# plot & ggplot2::scale_fill_brewer(type = 'qual')

## Modify the format of axis title and axis text

# plot & ggplot2::theme(axis.title = ggplot2::element_text(size = 14,
#                                                          family = 'serif'),
#                       axis.text = ggplot2::element_text(face = 'bold',
#                                                         color = 'blue'))

chillR documentation built on Nov. 28, 2023, 1:09 a.m.