Package 'portalr'

Add Seasons

Description

Higher-order data summaries, by 6-month seasons, 3-month seasons, or year. Also applies specified functions to the specified summary level.

yearly generates a table of yearly means

Usage

add_seasons(
  data,
  level = "site",
  season_level = 2,
  date_column = "yearmon",
  summary_funs = NA,
  path = get_default_data_path(),
  download_if_missing = TRUE,
  clean = TRUE
)

yearly(...)

Arguments

data	data frame containing columns: date, period, newmoonnumber, or year and month
level	summarize by "Plot", "Treatment", or "Site"
season_level	either year, 2: winter = Oct-March summer = April-Sept 4: winter = Dec-Feb spring = March-May summer = Jun-Aug fall = Sep-Nov
date_column	either "date" (must be in format "y-m-d"), "period", "newmoonnumber", or "yearmon" (data must contain "year" and "month")
summary_funs	A function specified by its name (e.g. "mean"). Default is NA (returned with seasons added but not summarized).
path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
clean	logical, load only QA/QC rodent data (TRUE) or all data (FALSE)
...	arguments passed to add_seasons

Value

a data.frame with additional "season" and "year" column, and other columns summarized as specified. If no summary function is specified, "season" and "year" columns are added to original dataframe, as well as a "seasonyear" column which correctly assigns months to seasons for grouping (eg December 2000 in winter 2001, rather than winter 2000).

---

Ant Bait Presence Absence

Description

Get ant species presence/absence by year/plot/stake from bait census data

Bait census data is more consistent over time than the colony census data. This function assumes that all species present in at least one census were censused in all years.

Usage

bait_presence_absence(
  path = get_default_data_path(),
  level = "Site",
  download_if_missing = TRUE,
  quiet = FALSE
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
level	level at which to summarize data: 'Site', 'Plot', or 'Stake'
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to run without version messages

Value

data frame with year, species, (plot if applicable), and presence [1, 0]

---

Manage the default path for downloading Portal Data into

Description

check_default_data_path checks if a default data path is set, and prompts the user to set it if it is missing.

get_default_data_path gets the value of the data path environmental variable

use_default_data_path has 3 steps. First, it checks for the presence of a pre-existing setting for the environmental variable. Then it checks if the folder exists and creates it, if needed. Then it provides instructions for setting the environmental variable.

Usage

check_default_data_path(
  ENV_VAR = "PORTALR_DATA_PATH",
  MESSAGE_FUN = message,
  DATA_NAME = "Portal data"
)

get_default_data_path(fallback = "~", ENV_VAR = "PORTALR_DATA_PATH")

use_default_data_path(path = NULL, ENV_VAR = "PORTALR_DATA_PATH")

Arguments

ENV_VAR	the environmental variable to check (by default '"PORTALR_DATA_PATH"“)
MESSAGE_FUN	the function to use to output messages
DATA_NAME	the name of the dataset to use in output messages
fallback	the default value to use if the setting is missing
path	character Folder into which data will be downloaded.

Value

FALSE if there is no path set, TRUE otherwise

None

---

Check for latest version of data files

Description

Check the latest version against the data that exists on the GitHub repo

Usage

check_for_newer_data(path = get_default_data_path())

Arguments

path	Folder in which data will be checked

Value

bool TRUE if there is a newer version of the data online

---

Do basic cleaning of Portal plant data

Description

This function does basic quality control of the Portal plant data. It is mainly called from summarize_plant_data, with several arguments passed along.

The specific steps it does are, in order: (1) correct species names according to recent vouchers, if requested (2) restrict species to annuals or non-woody (3) remove records for unidentified species (5) exclude the plots that aren't long-term treatments

Usage

clean_plant_data(
  data_tables,
  type = "All",
  unknowns = FALSE,
  correct_sp = TRUE
)

Arguments

data_tables	the list of data_tables, returned from calling load_plant_data
type	specify subset of species; If type=Annuals, removes all non-annual species. If type=Non-woody, removes shrub and subshrub species If type=Perennials, returns all perennial species (includes shrubs and subshrubs) If type=Shrubs, returns only shrubs and subshrubs If type=Winter-annual, returns all annuals found in winter IF type=Summer-annual, returns all annuals found in summer
unknowns	either removes all individuals not identified to species (unknowns = FALSE) or sums them in an additional column (unknowns = TRUE)
correct_sp	T/F whether or not to use likely corrected plant IDs, passed to rename_species_plants

---

Do basic cleaning of Portal rodent data

Description

This function does basic quality control of the Portal rodent data. It is mainly called from summarize_rodent_data, with several arguments passed along.

The specific steps it does are, in order: (1) add in missing weight data (2) remove records with "bad" period codes or plot numbers (3) remove records for unidentified species (4) exclude non-granivores (5) exclude incomplete trapping sessions (6) exclude the plots that aren't long-term treatments

Usage

clean_rodent_data(
  rodent_data,
  species_table,
  fillweight = FALSE,
  type = "Rodents",
  unknowns = FALSE
)

Arguments

rodent_data	the raw rodent data table
species_table	the species table
fillweight	specify whether to fill in unknown weights with other records from that individual or species, where possible
type	specify subset of species; either all "Rodents" or only "Granivores"
unknowns	either removes all individuals not identified to species (unknowns = FALSE) or sums them in an additional column (unknowns = TRUE)

---

Ant Colony Presence Absence

Description

Get ant species presence/absence by year/plot/stake from colony census data

Anomalies in ant colony census protocol over the years means that it can be difficult to discern true absences of all species in all years. This function uses information from Portal_ant_species.csv and Portal_ant_dataflags.csv to predict true presence/absence of species per plot per year. If a more conservative estimate is desired, setting the argument 'rare_sp = T' will only include species we are confident were censused regularly. Setting 'rare_sp = F' may include some false absences, since it is unknown if some rare species were censused in all years. Unknowns may also be excluded from output if desired.

Usage

colony_presence_absence(
  path = get_default_data_path(),
  level = "Site",
  rare_sp = FALSE,
  unknowns = FALSE,
  download_if_missing = TRUE,
  quiet = FALSE
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
level	level at which to summarize data: 'Site', 'Plot', or 'Stake'
rare_sp	include rare species (T) or not (F). Rare species may or may not have been censused in all years. Setting 'rare_sp = FALSE' gives a more conservative estimate of presence/absence
unknowns	include unknown species (TRUE) or not (FALSE). Unknowns include those only identified to genus.
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to run without version messages

Value

data frame with year, species, (plot if applicable), and presence [1, 0, NA]

---

Download the PortalData repo

Description

Downloads specified version of the Portal data.

Usage

download_observations(
  path = get_default_data_path(),
  version = "latest",
  source = "github",
  quiet = FALSE,
  verbose = FALSE,
  pause = 30,
  timeout = getOption("timeout"),
  force = FALSE
)

Arguments

path	character Folder into which data will be downloaded.
version	character Version of the data to download (default = "latest"). If NULL, returns.
source	character indicator of the source for the download. Either "github" (default) or "zenodo".
quiet	logical whether to download data silently.
verbose	logical whether to provide details of downloading.
pause	Positive integer or integer numeric seconds for pausing during steps around unzipping that require time delayment.
timeout	Positive integer or integer numeric seconds for timeout on downloads. Temporarily overrides the "timeout" option in options.
force	logical indicator of whether or not existing files or folders (such as the archive) should be over-written if an up-to-date copy exists (most users should leave as FALSE).

Value

NULL invisibly.

---

Forecast ndvi using a seasonal auto ARIMA

Description

Forecast ndvi using a seasonal auto ARIMA

Usage

fcast_ndvi(hist_ndvi, level, lead, moons = NULL)

Arguments

hist_ndvi	historic ndvi data
level	specify "monthly" or "newmoon"
lead	number of steps forward to forecast
moons	moon data (required if level = "newmoon")

Details

ndvi values are forecast using auto.arima with seasonality (using a Fourier transform)

Value

a data.frame with time and ndvi values

---

Fill in historic ndvi data to the complete timeseries being fit

Description

Fill in historic ndvi data to the complete timeseries being fit

Usage

fill_missing_ndvi(ndvi, level, last_time, moons = NULL)

Arguments

ndvi	ndvi data
level	specify "monthly" or "newmoon"
last_time	the last time step to have been completed
moons	moon data (required if level = "newmoons" and forecasts are needed)

Details

missing values during the time series are replaced using na.interp, missing values at the end of the time series are forecast using auto.arima with seasonality (using Fourier transform)

Value

a data.frame with time and ndvi values

---

Period code for incomplete censuses

Description

Determines incomplete censuses by finding dates when some plots were trapped, but others were not.

Usage

find_incomplete_censuses(trapping_table, min_plots, min_traps)

Arguments

trapping_table	Data_table of when plots were censused.
min_plots	minimum number of plots within a period for an observation to be included
min_traps	minimum number of traps for a plot to be included

Value

Data.table of period codes when not all plots were trapped.

---

Return Citation for Portal Data

Description

Return Citation for Portal Data

Usage

get_dataset_citation()

Value

An object of class "citation". For more details, see 'citation()'

---

Get future newmoon dates and numbers

Description

Get next newmoon dates and assign newmoon numbers for forecasting

Usage

get_future_newmoons(newmoons, nfuture_newmoons = NULL)

Arguments

newmoons	current newmoon table
nfuture_newmoons	number of future newmoons to get

Value

expected newmoons table for requested future newmoons

---

read in a raw datafile from the downloaded data or the GitHub repo

Description

does checking for whether a particular datafile exists and then reads it in, using na_strings to determine what gets converted to NA. It can also download the dataset if it's missing locally.

Usage

load_datafile(
  datafile,
  na.strings = "",
  path = get_default_data_path(),
  download_if_missing = TRUE,
  quiet = TRUE,
  ...
)

Arguments

datafile	the path to the datafile within the folder for Portal data
na.strings	a character vector of strings which are to be interpreted as NA values. Blank fields are also considered to be missing values in logical, integer, numeric and complex fields. Note that the test happens after white space is stripped from the input, so na.strings values may need their own white space stripped in advance.
path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to perform operations silently
...	arguments passed to download_observations

---

Read in the Portal data files

Description

Loads Portal data files from either a user-defined path or the online Github repository. If the user-defined path is un- available, the default option is to download to that location.

load_rodent_data loads the rodent data files

load_plant_data loads the plant data files

load_ant_data loads the ant data files

load_trapping_data loads just the rodent trapping files

Usage

load_rodent_data(
  path = get_default_data_path(),
  download_if_missing = TRUE,
  clean = TRUE,
  quiet = FALSE,
  ...
)

load_plant_data(
  path = get_default_data_path(),
  download_if_missing = TRUE,
  quiet = FALSE,
  ...
)

load_ant_data(
  path = get_default_data_path(),
  download_if_missing = TRUE,
  quiet = FALSE,
  ...
)

load_trapping_data(
  path = get_default_data_path(),
  download_if_missing = TRUE,
  clean = TRUE,
  quiet = FALSE,
  ...
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
clean	logical, load only QA/QC rodent data (TRUE) or all data (FALSE)
quiet	logical, whether to run without version messages
...	arguments passed to download_observations

Value

load_rodent_data returns a list of 5 dataframes:

rodent_data	raw data on rodent captures
species_table	species code, names, types
trapping_table	when each plot was trapped
newmoons_table	pairs census periods with newmoons
plots_table	rodent treatment assignments for each plot

load_plant_data returns a list of 7 dataframes:

quadrat_data	raw plant quadrat data
species_table	species code, names, types
census_table	indicates whether each quadrat was counted in each census; area of each quadrat
date_table	start and end date of each plant census
plots_table	rodent treatment assignments for each plot
transect_data	raw plant transect data with length and height (2015-present)
oldtransect_data	raw plant transect data as point counts (1989-2009)

load_ant_data returns a list of 4 dataframes:

bait_data	raw ant bait data
colony_data	raw ant colony data
species_table	species code, names, types
plots_table	treatment assignments for each plot

load_trapping_data returns a list of 2 dataframes:

trapping_table	when each plot was trapped
newmoons_table	pairs census periods with newmoons

---

Conform NA entries to "NA" entries

Description

Given the species abbreviation Neotoma albigula (NA), when data are read in, there can be an NA when it should be an "NA". This function conforms the entries to be proper character values.

Usage

na_conformer(dfv, colname = "species")

Arguments

dfv	Either [1] a data.frame containing colname as a column with NAs that need to be conformed to "NA"s or [2] a vector with NAs that need to be conformed to "NA"s.
colname	character value of the column name in tab to conform the NAs to "NA"s.

Value

x with any NA in colname replaced with "NA".

Examples

na_conformer(c("a", "b", NA, "c"))

---

NDVI by calendar month or lunar month

Description

Summarize NDVI data to monthly or lunar monthly level

Usage

ndvi(
  level = "monthly",
  sensor = "landsat",
  fill = FALSE,
  path = get_default_data_path(),
  download_if_missing = TRUE
)

Arguments

level	specify "monthly" or "newmoon"
sensor	specify "landsat", "modis", "gimms", or "all"
fill	specify if missing data should be filled, passed to fill_missing_ndvi
path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it

---

Phenocam data products by day, calendar month, or lunar month

Description

Summarize phenocam data products to either daily, monthly, or lunar monthly level.

Usage

phenocam(level = "daily", path = get_default_data_path())

Arguments

level	specify 'monthly', 'daily', or 'newmoon'
path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository

---

Prints a portal_data_list object

Description

Prints a portal_data_list object

Usage

## S3 method for class 'portal_data_list'
print(x, ...)

Arguments

x	A portal_data_list object.
...	arguments passed to print

---

If a Value is NULL, Trigger the Parent Function's Return

Description

If the focal input is NULL, return value from the parent function. Should only be used within a function.

Usage

return_if_null(x, value = NULL)

Arguments

x	Focal input.
value	If x is NULL, return this input from the parent function.

Value

If x is not NULL, NULL is returned. If x is NULL, the result of return with value as its input evaluated within the parent function's environment is returned.

Examples

ff <- function(x = 1, null_return = "hello"){
   return_if_null(x, null_return)
   x
 }
 ff()
 ff(NULL)

---

Rodent species abbreviations

Description

Creates a simple character vector of abbreviations for the Portal Rodents.

Usage

rodent_species(
  path = get_default_data_path(),
  type = "code",
  set = "all",
  total = FALSE
)

forecasting_species(
  path = get_default_data_path(),
  total = FALSE,
  type = "abbreviation"
)

Arguments

path	character Folder into which data will be downloaded.
type	character value indicating the output type. Current options include 'abbreviation' or 'code' (default, two-letter abbreviation), 'g_species' (abbreviated genus and species), 'Latin' (full scientific names), 'common' (common names), and 'table' (a data.frame of all the options).
set	character input of a specified set of species. Options include "all" (default, all species included) and "forecasting" (the species used in forecating pipelines).
total	logical value indicating if "total" should be added or not.

Value

character vector of species abbreviations.

---

Generate percent cover from Portal plant transect data

Description

This function calculates percent cover from transect data. It handles the pre-2015 data differently from the current transects, becase they are collected differently. But it returns a single time-series with all years of transect data available. It also returns mean height beginning in 2015.

Usage

shrub_cover(
  path = get_default_data_path(),
  type = "Shrubs",
  plots = "all",
  unknowns = FALSE,
  correct_sp = TRUE,
  download_if_missing = TRUE,
  quiet = FALSE
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
type	specify subset of species; If type=Annuals, removes all non-annual species. If type=Summer Annuals, returns all annual species that can be found in the summer If type=Winter Annuals, returns all annual species that can be found in the winter If type=Non-woody, removes shrub and subshrub species If type=Perennials, returns all perennial species (includes shrubs and subshrubs) If type=Shrubs, returns only shrubs and subshrubs
plots	specify subset of plots; can be a vector of plots, or specific sets: "all" plots or "Longterm" plots (plots that have had the same treatment for the entire time series)
unknowns	either removes all individuals not identified to species (unknowns = FALSE) or sums them in an additional column (unknowns = TRUE)
correct_sp	correct species names suspected to be incorrect in early data (T/F)
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to run without version messages

Value

a data.frame of percent cover and mean height

---

Return cleaned Portal rodent individual data

Description

This function cleans and subsets the data based on a number of arguments. It returns stake number and individual level data.

Usage

summarize_individual_rodents(
  path = get_default_data_path(),
  clean = TRUE,
  type = "Rodents",
  length = "all",
  unknowns = FALSE,
  time = "period",
  fillweight = FALSE,
  min_plots = 1,
  min_traps = 1,
  download_if_missing = TRUE,
  quiet = FALSE
)

summarise_individual_rodents(
  path = get_default_data_path(),
  clean = TRUE,
  type = "Rodents",
  length = "all",
  unknowns = FALSE,
  time = "period",
  fillweight = FALSE,
  min_plots = 1,
  min_traps = 1,
  download_if_missing = TRUE,
  quiet = FALSE
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
clean	logical, load only QA/QC rodent data (TRUE) or all data (FALSE)
type	specify subset of species; either all "Rodents" or only "Granivores"
length	specify subset of plots; use "All" plots or only "Longterm" plots (to be deprecated)
unknowns	either removes all individuals not identified to species (unknowns = FALSE) or sums them in an additional column (unknowns = TRUE)
time	specify the format of the time index in the output, either "period" (sequential Portal surveys), "newmoon" (lunar cycle numbering), "date" (calendar date), or "all" (for all time indices)
fillweight	specify whether to fill in unknown weights with other records from that individual or species, where possible
min_plots	minimum number of plots within a period for an observation to be included
min_traps	minimum number of traps for a plot to be included
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to run without producing messages

Value

a data.frame

---

Generate summaries of Portal plant data

Description

This function is a generic interface into creating summaries of the Portal plant species data. It contains a number of arguments to specify both the kind of data to summarize, at what level of aggregation, various choices for dealing with data quality, and output format.

plant_abundance generates a table of plant abundance

Usage

summarize_plant_data(
  path = get_default_data_path(),
  level = "Site",
  type = "All",
  length = "all",
  plots = length,
  unknowns = FALSE,
  correct_sp = TRUE,
  shape = "flat",
  output = "abundance",
  na_drop = switch(tolower(level), quadrat = FALSE, plot = FALSE, treatment = TRUE, site
    = TRUE, TRUE),
  zero_drop = switch(tolower(level), quadrat = TRUE, plot = FALSE, treatment = TRUE, site
    = TRUE, TRUE),
  min_quads = 1,
  effort = TRUE,
  download_if_missing = TRUE,
  quiet = FALSE
)

plant_abundance(..., shape = "flat")

summarise_plant_data(
  path = get_default_data_path(),
  level = "Site",
  type = "All",
  length = "all",
  plots = length,
  unknowns = FALSE,
  correct_sp = TRUE,
  shape = "flat",
  output = "abundance",
  na_drop = switch(tolower(level), quadrat = FALSE, plot = FALSE, treatment = TRUE, site
    = TRUE, TRUE),
  zero_drop = switch(tolower(level), quadrat = TRUE, plot = FALSE, treatment = TRUE, site
    = TRUE, TRUE),
  min_quads = 1,
  effort = TRUE,
  download_if_missing = TRUE,
  quiet = FALSE
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
level	summarize by "Plot", "Treatment", "Site", or "Quadrat"
type	specify subset of species; If type=Annuals, removes all non-annual species. If type=Summer Annuals, returns all annual species that can be found in the summer If type=Winter Annuals, returns all annual species that can be found in the winter If type=Non-woody, removes shrub and subshrub species If type=Perennials, returns all perennial species (includes shrubs and subshrubs) If type=Shrubs, returns only shrubs and subshrubs
length	specify subset of plots; use "All" plots or only "Longterm" plots (to be deprecated)
plots	specify subset of plots; can be a vector of plots, or specific sets: "all" plots or "Longterm" plots (plots that have had the same treatment for the entire time series)
unknowns	either removes all individuals not identified to species (unknowns = FALSE) or sums them in an additional column (unknowns = TRUE)
correct_sp	correct species names suspected to be incorrect in early data (T/F)
shape	return data as a "crosstab" or "flat" list
output	specify whether to return "abundance", or "cover" [cover data starts in summer 2015]
na_drop	logical, drop NA values (representing insufficient sampling) filling missing combinations of year-month-treatment/plot-species with NA could represent one of a few slightly different meanings: 1) that combo doesn't exist 2) that combo was skipped that month, or 3) that combo was trapped, but is unusable (a negative period code))
zero_drop	logical, drop 0s (representing sufficient sampling, but no detection)
min_quads	numeric [1:16], minimum number of quadrats (out of 16) for a plot to be included
effort	logical as to whether or not the effort columns should be included in the output
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to run without version messages
...	arguments passed to summarize_plant_data

Value

a data.frame in either "long" or "wide" format, depending on the value of 'shape'

---

Generate summaries of Portal rodent data

Description

This function is a generic interface into creating summaries of the Portal rodent species data. It contains a number of arguments to specify the kind of data to summarize (at what level of aggregation) and various choices for dealing with data quality, and output format.

abundance generates a table of rodent abundance

* biomass() generates a table of rodent biomass

* energy() generates a table of rodent energy (computed as 5.69 * (biomass ^ 0.75) after White et al 2004)

* rates() generates a table of rodent growth rates (computed as r=log(N[t+1]/N[t])

Usage

summarize_rodent_data(
  path = get_default_data_path(),
  clean = TRUE,
  level = "Site",
  type = "Rodents",
  length = "all",
  plots = length,
  unknowns = FALSE,
  shape = "crosstab",
  time = "period",
  output = "abundance",
  fillweight = (output != "abundance"),
  na_drop = TRUE,
  zero_drop = switch(tolower(level), plot = FALSE, treatment = TRUE, site = TRUE),
  min_traps = 1,
  min_plots = 24,
  effort = FALSE,
  download_if_missing = TRUE,
  quiet = FALSE,
  include_unsampled = FALSE
)

abundance(...)

biomass(...)

energy(...)

rates(...)

summarise_rodent_data(
  path = get_default_data_path(),
  clean = TRUE,
  level = "Site",
  type = "Rodents",
  length = "all",
  plots = length,
  unknowns = FALSE,
  shape = "crosstab",
  time = "period",
  output = "abundance",
  fillweight = (output != "abundance"),
  na_drop = TRUE,
  zero_drop = switch(tolower(level), plot = FALSE, treatment = TRUE, site = TRUE),
  min_traps = 1,
  min_plots = 24,
  effort = FALSE,
  download_if_missing = TRUE,
  quiet = FALSE,
  include_unsampled = FALSE
)

Arguments

path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository
clean	logical, load only QA/QC rodent data (TRUE) or all data (FALSE)
level	summarize by "Plot", "Treatment", or "Site"
type	specify subset of species; either all "Rodents" or only "Granivores"
length	specify subset of plots; use "All" plots or only "Longterm" plots (to be deprecated)
plots	specify subset of plots; can be a vector of plots, or specific sets: "all" plots or "Longterm" plots (plots that have had the same treatment for the entire time series)
unknowns	either removes all individuals not identified to species (unknowns = FALSE) or sums them in an additional column (unknowns = TRUE)
shape	return data as a "crosstab" or "flat" list
time	specify the format of the time index in the output, either "period" (sequential Portal surveys), "newmoon" (lunar cycle numbering), "date" (calendar date), or "all" (for all time indices)
output	specify whether to return "abundance", or "biomass", or "energy", or "rates"
fillweight	specify whether to fill in unknown weights with other records from that individual or species, where possible
na_drop	logical, drop NA values (representing insufficient sampling) filling missing combinations of year-month-treatment/plot-species with NA could represent one of a few slightly different meanings: 1) that combo doesn't exist 2) that combo was skipped that month, or 3) that combo was trapped, but is unusable (a negative period code))
zero_drop	logical, drop 0s (representing sufficient sampling, but no detection)
min_traps	minimum number of traps for a plot to be included
min_plots	minimum number of plots within a period for an observation to be included
effort	logical as to whether or not the effort columns should be included in the output
download_if_missing	if the specified file path doesn't have the PortalData folder, then download it
quiet	logical, whether to run without producing messages
include_unsampled	logical, overrides settings for 'na_drop' and 'zero_drop', setting both to FALSE
...	arguments passed to summarize_rodent_data

Value

a data.frame in either "long" or "wide" format, depending on the value of 'shape'

---

Weather by day, calendar month, or lunar month

Description

Summarize hourly weather data to either daily, monthly, or lunar monthly level.

Usage

weather(
  level = "daily",
  fill = FALSE,
  horizon = 365,
  temperature_limit = 4,
  path = get_default_data_path()
)

Arguments

level	specify 'monthly', 'daily', or 'newmoon'
fill	specify if missing data should be filled, passed to fill_missing_weather
horizon	Horizon (number of days) to use when calculating cumulative values (eg warm weather precip)
temperature_limit	Temperature limit (in C) to use when calculating cumulative values (eg warm weather precip)
path	either the file path that contains the PortalData folder or "repo", which then pulls data from the PortalData GitHub repository

---