,
Rafael Poyatos [aut] ,
Victor Flo [aut] ,
Jacob Nelson [ctb] ,
Sapfluxnet Core Team [cph]| Title: | Working with 'Sapfluxnet' Project Data |
|---|---|
| Description: | Access, modify, aggregate and plot data from the 'Sapfluxnet' project (<http://sapfluxnet.creaf.cat>), the first global database of sap flow measurements. |
| Authors: | Victor Granda [aut, cre] ,
Rafael Poyatos [aut] ,
Victor Flo [aut] ,
Jacob Nelson [ctb] ,
Sapfluxnet Core Team [cph] |
| Maintainer: | Victor Granda <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.1.4 |
| Built: | 2025-03-12 05:33:20 UTC |
| Source: | https://github.com/sapfluxnet/sapfluxnetr |
Imported from magrittr package
# piping sites ARG_TRE %>% daily_metrics()# piping sites ARG_TRE %>% daily_metrics()
Example site for package usage demonstration based on ARG_MAZ
ARG_MAZARG_MAZ
An sfn_data class object with the data and metadata for ARG_MAZ site
data('ARG_MAZ', package = 'sapfluxnetr') ARG_MAZdata('ARG_MAZ', package = 'sapfluxnetr') ARG_MAZ
Example site for package usage demonstration based on ARG_TRE
ARG_TREARG_TRE
An sfn_data class object with the data and metadata for ARG_TRE site
data('ARG_TRE', package = 'sapfluxnetr') ARG_TREdata('ARG_TRE', package = 'sapfluxnetr') ARG_TRE
Example site for package usage demonstration based on AUS_CAN_ST2_MIX
AUS_CAN_ST2_MIXAUS_CAN_ST2_MIX
An sfn_data class object with the data and metadata for AUS_CAN_ST2_MIX site
data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') AUS_CAN_ST2_MIXdata('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') AUS_CAN_ST2_MIX
helper for sfn_metrics
data_coverage(x, timestep, period_minutes)data_coverage(x, timestep, period_minutes)
x |
a vector, usually a variable in the sapflow or environmental data. |
timestep |
numeric value with the timestep in minutes |
period_minutes |
numeric value with the period in minutes |
This helper function calculates the coverage percentage in a vector, and is designed to be used inside a dplyr summarise statement. It calculates the coverage as the percentage of no NAs in the expected length of the summarising period stated by the timestep.
a single value (numeric) with the percentage of coverage for that variable
# data for one day, 60 minutes timestep (24 values) with a 75% of coberture x <- rep(c(1,2,3,NA), 6) data_coverage(x, 60, 1440) # 75# data for one day, 60 minutes timestep (24 values) with a 75% of coberture x <- rep(c(1,2,3,NA), 6) data_coverage(x, 60, 1440) # 75
describe_md_variable prints in console a detailed description for the
requested variable. Useful to know which values to filter or in which units
the variables are.
describe_md_variable(variable)describe_md_variable(variable)
variable |
A character with the name of the variable |
Nothing, prints information to console
# info about the method used to measure sapflow (pl_sens_meth) describe_md_variable('pl_sens_meth')# info about the method used to measure sapflow (pl_sens_meth) describe_md_variable('pl_sens_meth')
Calculate the diurnal centroid for sapflow variables
diurnal_centroid(variable)diurnal_centroid(variable)
variable |
A numeric vector containing the sapflow values for a day at a regular intervals. Missing values are allowed but not recommended |
The code for this function has been kindly provided by Jacob Nelson in python
(see https://github.com/jnelson18/FluxnetTools/blob/master/FileS3.py) and has
been translated to a tidy data philosophy in R to be used inside a
summarise statement.
A numeric value with the diurnal centroid value (0 to 24 h)
Given a continuous subdaily sapflow values at regular intervals
to obtain the diurnal centroid each value is
multiplied by its interval index and summed up and divided
by the sum of the values for the day and finally the value is normalized to
24h:
With even values for all the intervals (i.e. 100 for all), centroid converges to 12h at more than 1000 intervals per day. With only 48 (half hourly measurements) centroid converges to 12.25h and with 24 intervals (hourly measurements) centroid converges to 12.5h. So, using diurnal centroid value in half hourly datasets or above can have a considerable error associated.
Jacob Nelson & Víctor Granda
# dplyr library(dplyr) # check convergence to 12h: diurnal_centroid(rep(1, 1000)) # 12.012 h diurnal_centroid(rep(10000, 1000)) # 12.012 h, variable scale not affects calculation # sapflow diurnal centroid data('ARG_TRE', package = 'sapfluxnetr') sfn_metrics( ARG_TRE, period = '1 day', .funs = list(~ diurnal_centroid(.), ~ data_coverage(., timestep, period_minutes)), solar = FALSE, interval = 'general' )# dplyr library(dplyr) # check convergence to 12h: diurnal_centroid(rep(1, 1000)) # 12.012 h diurnal_centroid(rep(10000, 1000)) # 12.012 h, variable scale not affects calculation # sapflow diurnal centroid data('ARG_TRE', package = 'sapfluxnetr') sfn_metrics( ARG_TRE, period = '1 day', .funs = list(~ diurnal_centroid(.), ~ data_coverage(., timestep, period_minutes)), solar = FALSE, interval = 'general' )
filter_sites_by_md function takes logical expressions for the metadata
variables (i.e. pl_sens_meth == 'HR'), and list the sites that met
the criteria from those supplied
filter_sites_by_md(sites, metadata, ..., .join = c("and", "or"))filter_sites_by_md(sites, metadata, ..., .join = c("and", "or"))
sites |
character vector with the sites codes to filter, generally the
result of |
metadata |
metadata tbl object, usually the result of
|
... |
Logical expressions for the metadata variables, as in
|
.join |
Character indicating how to filter the sites, see details. |
.join argument indicates how sites must be filtered between metadata
classes. 'and' indicates only sites meeting all conditions for all
metadata classes are returned. 'or' indicates all sites meeting any
condition between classes are returned. For two or more filters of the same
metadata class, they are combined as 'and'.
A character vector with the sites fulfilling the premises
# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) save(AUS_CAN_ST2_MIX, file = file.path(folder, 'AUS_CAN_ST2_MIX.RData')) # we need the metadata and the site names metadata <- read_sfn_metadata(folder = folder, .write_cache = TRUE) sites <- sfn_sites_in_folder(folder) # Filter by Heat Ratio method filter_sites_by_md( pl_sens_meth == 'HR', sites = sites, metadata = metadata ) # Both, Heat Ratio and Heat Dissipation filter_sites_by_md( pl_sens_meth %in% c('HR', 'HD'), sites = sites, metadata = metadata ) # more complex, Heat Ratio method AND Woodland/Shrubland biome filter_sites_by_md( pl_sens_meth == 'HR', si_biome == 'Woodland/Shrubland', sites = sites, metadata = metadata, .join = 'and' # default ) # join = 'or' returns sites that meet any condition filter_sites_by_md( pl_sens_meth == 'HR', si_biome == 'Woodland/Shrubland', sites = sites, metadata = metadata, .join = 'or' )# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) save(AUS_CAN_ST2_MIX, file = file.path(folder, 'AUS_CAN_ST2_MIX.RData')) # we need the metadata and the site names metadata <- read_sfn_metadata(folder = folder, .write_cache = TRUE) sites <- sfn_sites_in_folder(folder) # Filter by Heat Ratio method filter_sites_by_md( pl_sens_meth == 'HR', sites = sites, metadata = metadata ) # Both, Heat Ratio and Heat Dissipation filter_sites_by_md( pl_sens_meth %in% c('HR', 'HD'), sites = sites, metadata = metadata ) # more complex, Heat Ratio method AND Woodland/Shrubland biome filter_sites_by_md( pl_sens_meth == 'HR', si_biome == 'Woodland/Shrubland', sites = sites, metadata = metadata, .join = 'and' # default ) # join = 'or' returns sites that meet any condition filter_sites_by_md( pl_sens_meth == 'HR', si_biome == 'Woodland/Shrubland', sites = sites, metadata = metadata, .join = 'or' )
Obtain the site timezone from a sfn_data/sfn_data_multi object
get_timezone(sfn_data)get_timezone(sfn_data)
sfn_data |
An sfn_data or sfn_data_multi object |
a character with the site timezone
# timezone of ARG_TRE site get_timezone(ARG_TRE)# timezone of ARG_TRE site get_timezone(ARG_TRE)
Initialize an sfn_data_multi object
## S4 method for signature 'sfn_data_multi' initialize(.Object, ...)## S4 method for signature 'sfn_data_multi' initialize(.Object, ...)
.Object |
sfn_data_multi object to create |
... |
sfn_data elements |
Initialize an sfn_data object
## S4 method for signature 'sfn_data' initialize( .Object, sapf_data, env_data, sapf_flags, env_flags, si_code, timestamp, solar_timestamp, site_md, stand_md, species_md, plant_md, env_md )## S4 method for signature 'sfn_data' initialize( .Object, sapf_data, env_data, sapf_flags, env_flags, si_code, timestamp, solar_timestamp, site_md, stand_md, species_md, plant_md, env_md )
.Object |
sfn_data object to create |
sapf_data |
A tibble (or any object coercible to one) with the sapf_data (without the TIMESTAMP variable) |
env_data |
A tibble (or any object coercible to one) with the env_data (without the TIMESTAMP variable) |
sapf_flags |
A tibble (or any object coercible to one) with the same
dimensions of |
env_flags |
A tibble (or any object coercible to one) with the same
dimensions of |
si_code |
A character vector of length one indicating the site code |
timestamp |
A POSIXct vector of length |
solar_timestamp |
A POSIXct vector of length |
site_md |
A tibble (or any object coercible to one) containing the site metadata |
stand_md |
A tibble (or any object coercible to one) containing the stand metadata |
species_md |
A tibble (or any object coercible to one) containing the species metadata |
plant_md |
A tibble (or any object coercible to one) containing the plant metadata |
env_md |
A tibble (or any object coercible to one) containing the env metadata |
This set of functions returns a complete set of statistics for a site (using
sfn_data) or several sites (using sfn_data_multi)
daily_metrics( sfn_data, solar = TRUE, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) monthly_metrics( sfn_data, solar = TRUE, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) nightly_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 20, int_end = 6, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) daylight_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 6, int_end = 20, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) predawn_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 4, int_end = 6, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) midday_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 11, int_end = 13, probs = c(0.95), tidy = FALSE, metadata = NULL, ... )daily_metrics( sfn_data, solar = TRUE, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) monthly_metrics( sfn_data, solar = TRUE, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) nightly_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 20, int_end = 6, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) daylight_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 6, int_end = 20, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) predawn_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 4, int_end = 6, probs = c(0.95), tidy = FALSE, metadata = NULL, ... ) midday_metrics( sfn_data, period = c("1 day", "1 month"), solar = TRUE, int_start = 11, int_end = 13, probs = c(0.95), tidy = FALSE, metadata = NULL, ... )
sfn_data |
|
solar |
Logical indicating if the solarTIMESTAMP must be used instead of the site local TIMESTAMP. Default to TRUE (use solarTIMESTAMP). |
probs |
numeric vector of probabilities for
|
tidy |
Logical indicating if the metrics must be returned in a tidy format (a long tibble, each observation in its own row) |
metadata |
metadata object, usually the result of
|
... |
additional arguments to be passed to |
period |
Time period to aggregate data by. See period section for an explanation about the periods ('3 hours', '1 day', '1 month', '1 year', ...) |
int_start |
Integer value indicating the starting hour of the special interval in 24h format. See Interval section in details. |
int_end |
Integer value indicating the ending hour of the special interval in 24h format. See Interval section in details. |
*_metrics functions are wrappers for sfn_metrics with a
set of fixed arguments.
*_metrics functions return all or some of the following statistics:
mean: mean of variable (tree or environmental variable) for the given period. NAs are removed
sd: standard deviation of the variable for the given period. NAs are removed
coverage: Data coverage percentage (percentage of measures without NAs)
q_XX: 0.XX quantile value for the period
centroid: Diurnal centroid value (hours passed until the half of the summed daily value was reached). Only returned for sapflow measures when period is '1 day'
accumulated: Accumulated values for precipitation only
If tidy is TRUE, a tibble with the metrics for
sapflow and environmental data, with all the metadata included. If
tidy is FALSE (default), a list of tibbles with the calculated
metrics.
daily_metrics summarise daily data for all hours in the day
monthly_metrics summarise monthly data for all hours in the day.
nightly_metrics will return the metrics for night
periods, summarised daily or monthly
Night for daily period starts in DOY x and ends in DOY x+1 (i.e. if
night_start = 20, night_end = 6 values for the night starting at
2018-03-28 20:00:00 and ending at 2018-03-29 06:00:00 are summarised).
Night for monthly period summarises all night periods in the month, that includes from 00:00:00 of the first month night to 23:59:59 of the last month night.
daylight_metrics will return the metrics for daylight
periods, summarised daily or monthly. Daylight interval is selected by start
and end hours.
predawn_metrics will always return the metrics for predawn
period, summarised daily or monthly. Predawn interval is selected by start and
end hours.
Predawn metrics did not return the centroid metric.
midday_metrics will always return the metrics for midday
period, summarised daily or monthly. midday interval is selected by start and
end hours.
Midday metrics did not return the centroid metric.
Other metrics:
sfn_metrics()
## daily_metrics # data load data('ARG_TRE', package = 'sapfluxnetr') data('sfn_metadata_ex', package = 'sapfluxnetr') # non tidy raw metrics (default) ARG_TRE_raw_daily <- daily_metrics(ARG_TRE) str(ARG_TRE_raw_daily) # tidy daily metrics ARG_TRE_daily <- daily_metrics( ARG_TRE, tidy = TRUE, metadata = sfn_metadata_ex ) ARG_TRE_daily ## monthly_metrics # data load data('ARG_TRE', package = 'sapfluxnetr') data('sfn_metadata_ex', package = 'sapfluxnetr') # non tidy raw metrics (default) ARG_TRE_raw_monthly <- monthly_metrics(ARG_TRE) str(ARG_TRE_raw_monthly) # tidy monthly metrics ARG_TRE_monthly <- monthly_metrics( ARG_TRE, tidy = TRUE, metadata = sfn_metadata_ex ) ARG_TRE_monthly ## nightly_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily night metrics (default) AUS_CAN_ST2_MIX_night <- nightly_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_night) AUS_CAN_ST2_MIX_night[['sapf']] AUS_CAN_ST2_MIX_night[['env']] # change the night interval AUS_CAN_ST2_MIX_night_short <- nightly_metrics( AUS_CAN_ST2_MIX, int_start = 21, int_end = 4 # night starting and ending hour ) AUS_CAN_ST2_MIX_night_short[['env']] # tidy nightly metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_night_tidy <- nightly_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_night_tidy ## daylight_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily daylight metrics (default) AUS_CAN_ST2_MIX_daylight <- daylight_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_daylight) AUS_CAN_ST2_MIX_daylight[['sapf']] AUS_CAN_ST2_MIX_daylight[['env']] # change the daylight interval AUS_CAN_ST2_MIX_daylight_short <- daylight_metrics( AUS_CAN_ST2_MIX, int_start = 8, int_end = 18 # night starting and ending hour ) AUS_CAN_ST2_MIX_daylight_short[['env']] # tidy daylight metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_daylight_tidy <- daylight_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_daylight_tidy ## predawn_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily predawn metrics (default) AUS_CAN_ST2_MIX_predawn <- predawn_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_predawn) AUS_CAN_ST2_MIX_predawn[['sapf']] AUS_CAN_ST2_MIX_predawn[['env']] # change the predawn interval AUS_CAN_ST2_MIX_predawn_short <- predawn_metrics( AUS_CAN_ST2_MIX, int_start = 8, int_end = 18 # night starting and ending hour ) AUS_CAN_ST2_MIX_predawn_short[['env']] # tidy daylight metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_predawn_tidy <- predawn_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_predawn_tidy ## midday_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily midday metrics (default) AUS_CAN_ST2_MIX_midday <- midday_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_midday) AUS_CAN_ST2_MIX_midday[['sapf']] AUS_CAN_ST2_MIX_midday[['env']] # change the midday interval AUS_CAN_ST2_MIX_midday_short <- midday_metrics( AUS_CAN_ST2_MIX, int_start = 8, int_end = 18 # night starting and ending hour ) AUS_CAN_ST2_MIX_midday_short[['env']] # tidy daylight metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_midday_tidy <- midday_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_midday_tidy## daily_metrics # data load data('ARG_TRE', package = 'sapfluxnetr') data('sfn_metadata_ex', package = 'sapfluxnetr') # non tidy raw metrics (default) ARG_TRE_raw_daily <- daily_metrics(ARG_TRE) str(ARG_TRE_raw_daily) # tidy daily metrics ARG_TRE_daily <- daily_metrics( ARG_TRE, tidy = TRUE, metadata = sfn_metadata_ex ) ARG_TRE_daily ## monthly_metrics # data load data('ARG_TRE', package = 'sapfluxnetr') data('sfn_metadata_ex', package = 'sapfluxnetr') # non tidy raw metrics (default) ARG_TRE_raw_monthly <- monthly_metrics(ARG_TRE) str(ARG_TRE_raw_monthly) # tidy monthly metrics ARG_TRE_monthly <- monthly_metrics( ARG_TRE, tidy = TRUE, metadata = sfn_metadata_ex ) ARG_TRE_monthly ## nightly_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily night metrics (default) AUS_CAN_ST2_MIX_night <- nightly_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_night) AUS_CAN_ST2_MIX_night[['sapf']] AUS_CAN_ST2_MIX_night[['env']] # change the night interval AUS_CAN_ST2_MIX_night_short <- nightly_metrics( AUS_CAN_ST2_MIX, int_start = 21, int_end = 4 # night starting and ending hour ) AUS_CAN_ST2_MIX_night_short[['env']] # tidy nightly metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_night_tidy <- nightly_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_night_tidy ## daylight_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily daylight metrics (default) AUS_CAN_ST2_MIX_daylight <- daylight_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_daylight) AUS_CAN_ST2_MIX_daylight[['sapf']] AUS_CAN_ST2_MIX_daylight[['env']] # change the daylight interval AUS_CAN_ST2_MIX_daylight_short <- daylight_metrics( AUS_CAN_ST2_MIX, int_start = 8, int_end = 18 # night starting and ending hour ) AUS_CAN_ST2_MIX_daylight_short[['env']] # tidy daylight metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_daylight_tidy <- daylight_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_daylight_tidy ## predawn_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily predawn metrics (default) AUS_CAN_ST2_MIX_predawn <- predawn_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_predawn) AUS_CAN_ST2_MIX_predawn[['sapf']] AUS_CAN_ST2_MIX_predawn[['env']] # change the predawn interval AUS_CAN_ST2_MIX_predawn_short <- predawn_metrics( AUS_CAN_ST2_MIX, int_start = 8, int_end = 18 # night starting and ending hour ) AUS_CAN_ST2_MIX_predawn_short[['env']] # tidy daylight metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_predawn_tidy <- predawn_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_predawn_tidy ## midday_metrics # data load data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') # non tidy daily midday metrics (default) AUS_CAN_ST2_MIX_midday <- midday_metrics(AUS_CAN_ST2_MIX) str(AUS_CAN_ST2_MIX_midday) AUS_CAN_ST2_MIX_midday[['sapf']] AUS_CAN_ST2_MIX_midday[['env']] # change the midday interval AUS_CAN_ST2_MIX_midday_short <- midday_metrics( AUS_CAN_ST2_MIX, int_start = 8, int_end = 18 # night starting and ending hour ) AUS_CAN_ST2_MIX_midday_short[['env']] # tidy daylight metrics data('sfn_metadata_ex', package = 'sapfluxnetr') AUS_CAN_ST2_MIX_midday_tidy <- midday_metrics( AUS_CAN_ST2_MIX, tidy = TRUE, metadata = sfn_metadata_ex ) AUS_CAN_ST2_MIX_midday_tidy
Transform the nested list of metrics in a tidy tibble where each observation has its own row
metrics_tidyfier( metrics_res, metadata, interval = c("general", "predawn", "midday", "night", "daylight") )metrics_tidyfier( metrics_res, metadata, interval = c("general", "predawn", "midday", "night", "daylight") )
metrics_res |
Nested list containing the metrics results as obtained
from |
metadata |
List containing the metadata nested list, as obtained from
|
interval |
Interval to return, it depends on the |
a tibble with the following columns:
TIMESTAMP: POSIXct vector with the date-time of the observation
si_code: Character vector with the site codes
pl_code: Character vector with the plant codes
sapflow_*: Variables containing the different metrics for the sapflow measurements (i.e. sapflow_mean, sapflow_q_95)
ta_*; rh_*; vpd_*; ...: Variables containing the different metrics for environmental variables (i.e. ta_mean, ta_q_95)
pl_*: plant metadata variables (i.e. pl_sapw_area, pl_sens_meth)
si_*: site metadata variables (i.e. si_biome, si_contact_firstname)
st_*: stand metadata variables (i.e. st_aspect, st_lai)
sp_*: species metadata variables (i.e. sp_basal_area_perc)
env_*: environmental metadata variables (i.e. env_timezone)
# data multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) data('sfn_metadata_ex', package = 'sapfluxnetr') # metrics multi_metrics <- daily_metrics(multi_sfn) # tidyfing multi_tidy <- metrics_tidyfier( multi_metrics, sfn_metadata_ex, interval = 'general' ) multi_tidy # A really easier way of doing the same multi_tidy_easy <- daily_metrics(multi_sfn, tidy = TRUE, metadata = sfn_metadata_ex)# data multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) data('sfn_metadata_ex', package = 'sapfluxnetr') # metrics multi_metrics <- daily_metrics(multi_sfn) # tidyfing multi_tidy <- metrics_tidyfier( multi_metrics, sfn_metadata_ex, interval = 'general' ) multi_tidy # A really easier way of doing the same multi_tidy_easy <- daily_metrics(multi_sfn, tidy = TRUE, metadata = sfn_metadata_ex)
Calculate the normalized diurnal centroid for sapflow variables
norm_diurnal_centroid(sapf_var, rad_var)norm_diurnal_centroid(sapf_var, rad_var)
sapf_var |
A numeric vector containing the sapflow values for a day at a regular intervals. Missing values are allowed but not recommended. |
rad_var |
A numeric vector containing the incoming radiation for a day at a regular intervals. Missing values are allowed but not recommended. Must be the same length as sapf_var. |
The code for this function has been kindly provided by Jacob Nelson in python
(see https://github.com/jnelson18/FluxnetTools/blob/master/FileS3.py) and has
been translated to a tidy data philosophy in R to be used inside a
summarise statement.
A numeric value with the normalized diurnal centroid value
This function calculates the diurnal centroid of sapflow measures relative to the diurnal centroid of incoming radiation (in any units). For that the incoming radiation diurnal centroid is substracted from the sapflow diurnal centroid:
Jacob Nelson & Víctor Granda
Given a site code and a route, read_sfn_data will return the selected
sfn_data object
read_sfn_data(site_codes, folder = ".")read_sfn_data(site_codes, folder = ".")
site_codes |
A character vector with the site code/s |
folder |
Route to the folder containing the |
If site_codes is a vector of length 1, an sfn_data object with
the selected site data. If site_codes is a vector of length > 1, then
a sfn_data_multi object containing all selected sites.
# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) # now we read a single site ARG_TRE_test <- read_sfn_data('ARG_TRE', folder) ARG_TRE_test # or we can read multiple sites at once multi_sfn <- read_sfn_data( c('ARG_TRE', 'ARG_MAZ'), folder ) multi_sfn# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) # now we read a single site ARG_TRE_test <- read_sfn_data('ARG_TRE', folder) ARG_TRE_test # or we can read multiple sites at once multi_sfn <- read_sfn_data( c('ARG_TRE', 'ARG_MAZ'), folder ) multi_sfn
Read metadata from all sites in folder and write it to disk to cache the info for easy and fast access
read_sfn_metadata(folder = ".", .write_cache = FALSE)read_sfn_metadata(folder = ".", .write_cache = FALSE)
folder |
Route to the folder containing the data. Default to working directory |
.write_cache |
Logical indicating if a cached copy of the metadata must
be written in |
Load all data in memory to collect metadata info can be resource limiting. For easy and quick access to metadata, this function stores an .RData file in the specified folder along the data with all the metadata preloaded. Also it return it as an object to use in filtering and selecting sites.
A list of tibbles with the five metadata classes (site, stand, species, plant and environmental)
# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) # create and load the metadata. The first time we use .write_cache = TRUE, # to ensure creating a file containing the metadata for speed the process # for the next times read_sfn_metadata( folder = folder, .write_cache = TRUE ) # a cached copy must have been written to "folder" file.exists(paste0(folder, '.metadata_cache.RData')) # TRUE # after that, we only need to especify the folder sites_metadata <- read_sfn_metadata(folder = folder) # quicker than before sites_metadata# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) # create and load the metadata. The first time we use .write_cache = TRUE, # to ensure creating a file containing the metadata for speed the process # for the next times read_sfn_metadata( folder = folder, .write_cache = TRUE ) # a cached copy must have been written to "folder" file.exists(paste0(folder, '.metadata_cache.RData')) # TRUE # after that, we only need to especify the folder sites_metadata <- read_sfn_metadata(folder = folder) # quicker than before sites_metadata
Validation checks for generating sfn_data_multi class objects
This method is used internally to ensure the correctness of the sfn_data_multi object. Basically ensures that the object returned is a list of sfn_data class objects
Multi sfn data class, derived from list
This class inherits from list, but modified to contain sfn_data objects
as elements. This will allow to work with several sites at the same time
obtaining results for all of them combined or individually as elements of
the resulting list (with lapply or purrr::map)
Validation checks for generating sfn_data class objects
This method is used internally when creating and/or modifying sfn_data class objects to ensure that the object returned is correct in terms of content classes and dimensions (i.e. sapflow data and environmental data has the same length)
Main class for storing sapfluxnet project site data and metadata
This class allows to store all the data and metadata for a sapfluxnet site
in one single object, to easily work with it. See
vignette('sfn-data-classes', package = 'sapfluxnetr') for more info.
sapf_dataA data frame with the sapf data
env_dataA data frame with the env data
sapf_flagsA data frame with the same dimensions of sapf_data
with the flag info for each tree/TIMESTAMP combination
env_flagsA data frame with the same dimensions of env_data with
the flag info for each env_var/TIMESTAMP combination
si_codeA character vector of length one indicating the site code
timestampA POSIXct vector of length nrow(sapf_data) with the
timestamp
solar_timestampA POSIXct vector of length nrow(sapf_data) with
the solar timestamp
site_mdA data frame containing the site metadata
stand_mdA data frame containing the stand metadata
species_mdA data frame containing the species metadata
plant_mdA data frame containing the plant metadata
env_mdA data frame containing the env metadata
Port of filter for sfn_data and
sfn_data_multi objects
sfn_filter(sfn_data, ..., solar = FALSE)sfn_filter(sfn_data, ..., solar = FALSE)
sfn_data |
|
... |
expressions to pass to the |
solar |
Logical indicating if solar timestamp must used to subset |
'sfn_filter' will remove the rows not matching the logical expression/s
provided. So, it will remove cases and will create TIMESTAMP gaps, so its use
is not recommended except in the case of filtering by TIMESTAMP (i.e. to
set several sites (sfn_data_multi) in the same time frame). For other
scenarios (removing extreme environmental conditions values or strange
sapflow measures patterns) see sfn_mutate and
sfn_mutate_at
For sfn_data objects, a filtered sfn_data or NULL if
no data meet the criteria. For sfn_data_multi another
sfn_data_multi with the sites filtered, and an empty
sfn_data_multi if any sites met the criteria
library(dplyr) library(lubridate) # data data('ARG_TRE', package = 'sapfluxnetr') # by timestamp foo_timestamp <- get_timestamp(ARG_TRE) foo_timestamp_trimmed <- foo_timestamp[1:100] sfn_filter( ARG_TRE, TIMESTAMP %in% foo_timestamp_trimmed ) # by wind speed value ws_threshold <- 25 sfn_filter( ARG_TRE, ws <= ws_threshold ) ## multi data('ARG_MAZ', package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ) # by timestamp sfn_filter( multi_sfn, between(day(TIMESTAMP), 18, 22) ) # by wind speed value sfn_filter( multi_sfn, ws <= ws_threshold )library(dplyr) library(lubridate) # data data('ARG_TRE', package = 'sapfluxnetr') # by timestamp foo_timestamp <- get_timestamp(ARG_TRE) foo_timestamp_trimmed <- foo_timestamp[1:100] sfn_filter( ARG_TRE, TIMESTAMP %in% foo_timestamp_trimmed ) # by wind speed value ws_threshold <- 25 sfn_filter( ARG_TRE, ws <= ws_threshold ) ## multi data('ARG_MAZ', package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ) # by timestamp sfn_filter( multi_sfn, between(day(TIMESTAMP), 18, 22) ) # by wind speed value sfn_filter( multi_sfn, ws <= ws_threshold )
Generics for getting the info in the slots of SfnData
get_sapf_data(object, ...) get_env_data(object, ...) get_sapf_flags(object, ...) get_env_flags(object, ...) get_timestamp(object, ...) get_solar_timestamp(object, ...) get_si_code(object, ...) get_site_md(object, ...) get_stand_md(object, ...) get_species_md(object, ...) get_plant_md(object, ...) get_env_md(object, ...)get_sapf_data(object, ...) get_env_data(object, ...) get_sapf_flags(object, ...) get_env_flags(object, ...) get_timestamp(object, ...) get_solar_timestamp(object, ...) get_si_code(object, ...) get_site_md(object, ...) get_stand_md(object, ...) get_species_md(object, ...) get_plant_md(object, ...) get_env_md(object, ...)
object |
Object to get data from |
... |
Further arguments to pass to the corresponding get method |
see sfn_get_methods for detailed info about using the get
methods in sfn_data class objects and
sfn_multi_get_methods for detailed info about using the get
methods in sfn_data_multi class objects.
Methods to get the data and metadata from the sfn_data class slots
## S4 method for signature 'sfn_data' get_sapf_data(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_env_data(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_sapf_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_env_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_timestamp(object) ## S4 method for signature 'sfn_data' get_solar_timestamp(object) ## S4 method for signature 'sfn_data' get_si_code(object) ## S4 method for signature 'sfn_data' get_site_md(object) ## S4 method for signature 'sfn_data' get_stand_md(object) ## S4 method for signature 'sfn_data' get_species_md(object) ## S4 method for signature 'sfn_data' get_plant_md(object) ## S4 method for signature 'sfn_data' get_env_md(object)## S4 method for signature 'sfn_data' get_sapf_data(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_env_data(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_sapf_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_env_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data' get_timestamp(object) ## S4 method for signature 'sfn_data' get_solar_timestamp(object) ## S4 method for signature 'sfn_data' get_si_code(object) ## S4 method for signature 'sfn_data' get_site_md(object) ## S4 method for signature 'sfn_data' get_stand_md(object) ## S4 method for signature 'sfn_data' get_species_md(object) ## S4 method for signature 'sfn_data' get_plant_md(object) ## S4 method for signature 'sfn_data' get_env_md(object)
object |
Object of class sfn_data from which data is retrieved |
solar |
Logical indicating if the timestamp to return in the |
get_sapf_data and get_env_data methods retrieve sapflow or environmental
tibbles to create a functional dataset to work with.
get_sapf_flags and get_env_flags methods retrieve sapflow or
environmental flags as tibbles.
get_timestamp and get_solar_timestamp methods retrieve only the
timestamp as POSIXct vector.
get_si_code method retrieve a character vector with length(timestamp)
containing the site code.
get_site_md, get_stand_md, get_species_md,
get_plant_md and get_env_md methods retrieve the corresponding
metadata.
library(dplyr) data('ARG_TRE', package = 'sapfluxnetr') sapf_data <- get_sapf_data(ARG_TRE, solar = TRUE) env_data_no_solar <- get_env_data(ARG_TRE, solar = FALSE) plant_md <- get_plant_md(ARG_TRE) # dplyr pipe to get the mean dbh for a site ARG_TRE %>% get_plant_md() %>% summarise(dbh_mean = mean(pl_dbh, na.rm = TRUE)) %>% pull(dbh_mean)library(dplyr) data('ARG_TRE', package = 'sapfluxnetr') sapf_data <- get_sapf_data(ARG_TRE, solar = TRUE) env_data_no_solar <- get_env_data(ARG_TRE, solar = FALSE) plant_md <- get_plant_md(ARG_TRE) # dplyr pipe to get the mean dbh for a site ARG_TRE %>% get_plant_md() %>% summarise(dbh_mean = mean(pl_dbh, na.rm = TRUE)) %>% pull(dbh_mean)
Example metadata cache file content for package usage demonstration
sfn_metadata_exsfn_metadata_ex
A list with five elements, each of one a metadata type.
data('sfn_metadata_ex', package = 'sapfluxnetr') sfn_metadata_exdata('sfn_metadata_ex', package = 'sapfluxnetr') sfn_metadata_ex
Generate metrics from a site/s data for the period indicated
sfn_metrics( sfn_data, period, .funs, solar, interval = c("general", "predawn", "midday", "night", "daylight"), int_start = NULL, int_end = NULL, ... )sfn_metrics( sfn_data, period, .funs, solar, interval = c("general", "predawn", "midday", "night", "daylight"), int_start = NULL, int_end = NULL, ... )
sfn_data |
|
period |
Time period to aggregate data by. See period section for an explanation about the periods ('3 hours', '1 day', '1 month', '1 year', ...) |
.funs |
List of function calls to summarise the data by, see .funs section for more details. |
solar |
Logical indicating if the solarTIMESTAMP must be used instead of the site local TIMESTAMP. Default to TRUE (use solarTIMESTAMP). |
interval |
Character vector indicating if the metrics must be filtered by an special hour interval. See Interval section in details. |
int_start |
Integer value indicating the starting hour of the special interval in 24h format. See Interval section in details. |
int_end |
Integer value indicating the ending hour of the special interval in 24h format. See Interval section in details. |
... |
optional arguments to pass to methods used (i.e. .collapse_timestamp or summarise funs extra arguments) |
For sfn_data objects, a list of tbl_df objects
with the following structure:
$sapf: metrics for the sapflow data
$env: metrics for the environmental data
For sfn_data_multi objects, a list of lists of tbl_df objects
with the metrics for each site:
$SITE_CODE
$sapf: metrics for the sapflow data
$env: metrics for the environmental data
$NEXT_SITE_CODE...
period argument is used by internal function
.collapse_timestamp and it can be stated in two ways:
frequency period format: "3 hours", "1 day", "7 days", "1 month"
As a custom function. This will be the name of a function,
without quotes, that accepts as the first argument the timestamp to collapse.
The result of the function must be a vector of collapsed TIMESTAMPs of the
same length than the original TIMESTAMP which will be used to group by and
summarise the data. Additional arguments to this function, if needed, can
be passed in the ... argument.
.collapse_timestamp also accepts the side argument to
collapse by the starting timestamp or the ending timestamp of each group. This
can be supplied in the ... argument.
.funs argument uses the same method as the .funs argument in the
summarise_all function of dplyr package. Basically
it accepts a list of function calls generated by list(). If you want to pass
on a custom function you can specify it here. See details in
summarise_by_period for more complex summarising functions
declaration.
Previously to the metrics summary, data can be filtered by an special
interval (i.e. predawn or nightly). This filtering can be specified with the
interval argument as this:
"general" (default). No special interval is used, and metrics
are performed with all the data.
"predawn". Data is filtered for predawn interval. In this case
int_start and int_end must be specified as 24h value
"midday". Data is filtered for midday interval. In this case
int_start and int_end must be specified as 24h value
"night". Data is filtered for night interval. In this case
int_start and int_end must be specified as 24h value
"daylight". Data is filtered for daylight interval. In this case
int_start and int_end must be specified as 24h value
Other metrics:
metrics
library(dplyr) ### general metrics ## sfn_data data('ARG_TRE', package = 'sapfluxnetr') ARG_TRE_metrics <- sfn_metrics( ARG_TRE, period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()), solar = FALSE, interval = 'general' ) str(ARG_TRE_metrics) ARG_TRE_metrics[['sapf']] ARG_TRE_metrics[['env']] ## sfn_data_multi data('ARG_MAZ', package = 'sapfluxnetr') data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) multi_metrics <- sfn_metrics( multi_sfn, period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()), solar = FALSE, interval = 'general' ) str(multi_metrics) multi_metrics[['ARG_TRE']][['sapf']] ### midday metrics ARG_TRE_midday <- sfn_metrics( ARG_TRE, period = '1 day', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()), solar = TRUE, interval = 'midday', int_start = 11, int_end = 13 ) str(ARG_TRE_midday) ARG_TRE_midday[['sapf']]library(dplyr) ### general metrics ## sfn_data data('ARG_TRE', package = 'sapfluxnetr') ARG_TRE_metrics <- sfn_metrics( ARG_TRE, period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()), solar = FALSE, interval = 'general' ) str(ARG_TRE_metrics) ARG_TRE_metrics[['sapf']] ARG_TRE_metrics[['env']] ## sfn_data_multi data('ARG_MAZ', package = 'sapfluxnetr') data('AUS_CAN_ST2_MIX', package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) multi_metrics <- sfn_metrics( multi_sfn, period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()), solar = FALSE, interval = 'general' ) str(multi_metrics) multi_metrics[['ARG_TRE']][['sapf']] ### midday metrics ARG_TRE_midday <- sfn_metrics( ARG_TRE, period = '1 day', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()), solar = TRUE, interval = 'midday', int_start = 11, int_end = 13 ) str(ARG_TRE_midday) ARG_TRE_midday[['sapf']]
Methods to get the data and metadata from the sfn_data class slots
## S4 method for signature 'sfn_data_multi' get_sapf_data(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_env_data(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_sapf_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_env_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_timestamp(object) ## S4 method for signature 'sfn_data_multi' get_solar_timestamp(object) ## S4 method for signature 'sfn_data_multi' get_si_code(object) ## S4 method for signature 'sfn_data_multi' get_site_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_stand_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_species_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_plant_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_env_md(object, collapse = FALSE)## S4 method for signature 'sfn_data_multi' get_sapf_data(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_env_data(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_sapf_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_env_flags(object, solar = FALSE) ## S4 method for signature 'sfn_data_multi' get_timestamp(object) ## S4 method for signature 'sfn_data_multi' get_solar_timestamp(object) ## S4 method for signature 'sfn_data_multi' get_si_code(object) ## S4 method for signature 'sfn_data_multi' get_site_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_stand_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_species_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_plant_md(object, collapse = FALSE) ## S4 method for signature 'sfn_data_multi' get_env_md(object, collapse = FALSE)
object |
Object of class sfn_data_multi from which data or metadata is retrieved |
solar |
Logical indicating if the timestamp to return in the |
collapse |
Logical indicating if the metadata get methods must collapse the returning list to a data frame with all sites |
get_sapf_data and get_env_data methods retrieve sapflow or
environmental tibbles from the sfn_data objects contained in the
sfn_data_multi and return them in a list.
get_sapf_flags and get_env_flags methods retrieve sapflow or
environmental flags tibbles from the sfn_data objects contained in the
sfn_data_multi and return them in a list.
get_timestamp and get_solar_timestamp methods retrieve only the
timestamps as POSIXct vectors and return them as a list (each element
corresponding to a site in the sfn_data_multi object).
get_si_code method retrieve a character vector with length(timestamp)
containing the site code for each site, returning them as a list.
get_site_md, get_stand_md, get_species_md,
get_plant_md and get_env_md methods retrieve the corresponding
metadata objects for each site returning them as a list, unless collapse is
TRUE, then the list collapses to a tibble.
library(dplyr)library(dplyr)
Port of mutate for sfn_data and
sfn_data_multi objects
sfn_mutate(sfn_data, ..., solar = FALSE)sfn_mutate(sfn_data, ..., solar = FALSE)
sfn_data |
|
... |
Name-value pairs of expressions to pass to the
|
solar |
Logical indicating if solar timestamp must used to subset |
'sfn_mutate' function will maintain the same number of rows before and after
the modification, so it is well suited to modify variables without creating
TIMESTAMP gaps (i.e. to change variable units). For mutating groups of
variables at the same time see sfn_mutate_at.
For sfn_data objects, a mutated sfn_data. For
sfn_data_multi another sfn_data_multi with the sites mutated
'sfn_mutate' internally joins the sapflow and environmental datasets by the TIMESTAMP, so it is possible to mutate variables conditionally between sapflow and environmental measures (i.e. mutate sapflow when wind is high or radiation is zero). Due to this, at the moment any new variable is dropped when building the final results, so this is ONLY intended to mutate existing variables without changing the names.
library(dplyr) library(lubridate) # data data('ARG_TRE', package = 'sapfluxnetr') # transform to NAs any wind value above 25 ws_threshold <- 25 sfn_mutate(ARG_TRE, ws = if_else(ws > 25, NA_real_, ws)) ## multi data(ARG_MAZ, package = 'sapfluxnetr') data(AUS_CAN_ST2_MIX, package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) multi_sfn_mutated <- sfn_mutate( multi_sfn, ws = if_else(ws > 25, NA_real_, ws) ) multi_sfn_mutated[['ARG_TRE']]library(dplyr) library(lubridate) # data data('ARG_TRE', package = 'sapfluxnetr') # transform to NAs any wind value above 25 ws_threshold <- 25 sfn_mutate(ARG_TRE, ws = if_else(ws > 25, NA_real_, ws)) ## multi data(ARG_MAZ, package = 'sapfluxnetr') data(AUS_CAN_ST2_MIX, package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) multi_sfn_mutated <- sfn_mutate( multi_sfn, ws = if_else(ws > 25, NA_real_, ws) ) multi_sfn_mutated[['ARG_TRE']]
Port of mutate_at for sfn_data and
sfn_data_multi objects
sfn_mutate_at(sfn_data, .vars, .funs, ..., solar = FALSE)sfn_mutate_at(sfn_data, .vars, .funs, ..., solar = FALSE)
sfn_data |
|
.vars |
Variables to mutate. Passed to |
.funs |
Function/s for mutate, passed to |
... |
Extra arguments to pass to the functions in |
solar |
Logical indicating if solar timestamp must used to subset |
'sfn_mutate_at' function will maintain the same number of rows before and
after the modification, so it is well suited to modify variables without
creating TIMESTAMP gaps (i.e. to change variable units). For mutating
individual variables see sfn_mutate.
For sfn_data objects, a mutated sfn_data. For
sfn_data_multi another sfn_data_multi with the sites mutated
library(dplyr) library(lubridate) # data data('ARG_TRE', package = 'sapfluxnetr') # transform to NAs any sapflow value occured with wind speed above 25 ws_threshold <- 25 # get the names of the variables to mutate (tree names) vars_to_mutate <- names(get_sapf_data(ARG_TRE)[,-1]) # no TIMESTAMP sfn_mutate_at( ARG_TRE, .vars = vars(one_of(vars_to_mutate)), .funs = list( ~ case_when( ws > ws_threshold ~ NA_real_, TRUE ~ . ) ) ) ## multi data(ARG_MAZ, package = 'sapfluxnetr') data(AUS_CAN_ST2_MIX, package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) ## in multi it's better to discard the variables to not mutate: vars_to_not_mutate <- names(get_env_data(ARG_TRE)) # all the environmental multi_sfn_mutated <- sfn_mutate_at( multi_sfn, .vars = vars(-one_of(vars_to_not_mutate)), # we use - .funs = list( ~ case_when( ws > ws_threshold ~ NA_real_, TRUE ~ . ) ) ) multi_sfn_mutated[['ARG_TRE']]library(dplyr) library(lubridate) # data data('ARG_TRE', package = 'sapfluxnetr') # transform to NAs any sapflow value occured with wind speed above 25 ws_threshold <- 25 # get the names of the variables to mutate (tree names) vars_to_mutate <- names(get_sapf_data(ARG_TRE)[,-1]) # no TIMESTAMP sfn_mutate_at( ARG_TRE, .vars = vars(one_of(vars_to_mutate)), .funs = list( ~ case_when( ws > ws_threshold ~ NA_real_, TRUE ~ . ) ) ) ## multi data(ARG_MAZ, package = 'sapfluxnetr') data(AUS_CAN_ST2_MIX, package = 'sapfluxnetr') multi_sfn <- sfn_data_multi(ARG_TRE, ARG_MAZ, AUS_CAN_ST2_MIX) ## in multi it's better to discard the variables to not mutate: vars_to_not_mutate <- names(get_env_data(ARG_TRE)) # all the environmental multi_sfn_mutated <- sfn_mutate_at( multi_sfn, .vars = vars(-one_of(vars_to_not_mutate)), # we use - .funs = list( ~ case_when( ws > ws_threshold ~ NA_real_, TRUE ~ . ) ) ) multi_sfn_mutated[['ARG_TRE']]
Plot the desired data from a site object
sfn_plot( sfn_data, type = c("sapf", "env", "ta", "rh", "vpd", "ppfd_in", "netrad", "sw_in", "ext_rad", "ws", "precip", "swc_shallow", "swc_deep"), formula_env = NULL, solar = TRUE, ... )sfn_plot( sfn_data, type = c("sapf", "env", "ta", "rh", "vpd", "ppfd_in", "netrad", "sw_in", "ext_rad", "ws", "precip", "swc_shallow", "swc_deep"), formula_env = NULL, solar = TRUE, ... )
sfn_data |
sfn_data object to plot. It can be also an sfn_data_multi object. |
type |
Character indicating which data to plot. See Type section for detailed information about the available values. Ignored if formula is provided |
formula_env |
Right side formula indicating an environmental variable to
plot vs. the sapflow values. If NULL (default), |
solar |
Logical indicating if the solar timestamp must be used instead of the site timestamp |
... |
Further arguments to be passed on |
A ggplot object that can be called to see the plot. If input is an sfn_data_multi object, a list with the plots
plot is a base R function which uses the base R plotting system
to show the plot. We prefer the ggplot plotting system, which allow for
storing the plots in objects and can be subjected to further modifications.
This allow the package users to generate rather simple plots that can be
fine tuned afterwards to the user taste. Generating a plot
method for the sfn_data class returning a ggplot object is not
desired (it change the way plot works and can be misleading about the plot
general usage). So, instead, we offer this function, sfn_plot.
type argument controls what is going to be plotted. It accepts
the following:
"sapf": It will plot sapflow data vs. TIMESTAMP
"env": It will plot environmental variables vs. TIMESTMAP
"ta", "rh", "vpd", "ppfd_in", "netrad", "sw_in", "ext_rad", "ws", "precip", "swc_shallow" and "swc_deep": They will plot the corresponding variable vs. TIMESTAMP
formula argument can be used to select an environmental variable to
plot versus all the sapflow measurements. Any environmental variable is
allowed, if it exist in the site provided.
By default sfn_plot generates plots using geom_point
geometry, except in the case of type = "ws" and
type = "precip" where geom_col is used. These
geometries can be modified with the ... argument.
library(ggplot2) # data data('ARG_TRE', package = 'sapfluxnetr') # plotting directly sfn_plot(ARG_TRE, type = 'sapf') # this could be noisy, you can facet by "Tree" (for sapflow) or by # "Variable" (for environmental data): sfn_plot(ARG_TRE, type = 'sapf') + facet_wrap(~ Tree) sfn_plot(ARG_TRE, type = 'env') + facet_wrap(~ Variable, scales = 'free_y') # saving and modifying: env_plot <- sfn_plot(ARG_TRE, type = 'env', solar = FALSE) + facet_wrap(~ Variable, scales = 'free_y') env_plot + labs(title = 'Environmental variables facet plot') # formula sfn_plot(ARG_TRE, formula_env = ~ vpd)library(ggplot2) # data data('ARG_TRE', package = 'sapfluxnetr') # plotting directly sfn_plot(ARG_TRE, type = 'sapf') # this could be noisy, you can facet by "Tree" (for sapflow) or by # "Variable" (for environmental data): sfn_plot(ARG_TRE, type = 'sapf') + facet_wrap(~ Tree) sfn_plot(ARG_TRE, type = 'env') + facet_wrap(~ Variable, scales = 'free_y') # saving and modifying: env_plot <- sfn_plot(ARG_TRE, type = 'env', solar = FALSE) + facet_wrap(~ Variable, scales = 'free_y') env_plot + labs(title = 'Environmental variables facet plot') # formula sfn_plot(ARG_TRE, formula_env = ~ vpd)
Generic functions for replacement functions for sfn_data
get_sapf_data(object) <- value get_env_data(object) <- value get_sapf_flags(object) <- value get_env_flags(object) <- value get_timestamp(object) <- value get_solar_timestamp(object) <- value get_si_code(object) <- value get_site_md(object) <- value get_stand_md(object) <- value get_species_md(object) <- value get_plant_md(object) <- value get_env_md(object) <- valueget_sapf_data(object) <- value get_env_data(object) <- value get_sapf_flags(object) <- value get_env_flags(object) <- value get_timestamp(object) <- value get_solar_timestamp(object) <- value get_si_code(object) <- value get_site_md(object) <- value get_stand_md(object) <- value get_species_md(object) <- value get_plant_md(object) <- value get_env_md(object) <- value
object |
Object to replace |
value |
Object to replace with |
see sfn_replacement_methods for more info about using the
replacement methods in sfn_data objects
Methods to replace the data and metadata from the sfn_data class slots
## S4 replacement method for signature 'sfn_data' get_sapf_data(object) <- value ## S4 replacement method for signature 'sfn_data' get_env_data(object) <- value ## S4 replacement method for signature 'sfn_data' get_sapf_flags(object) <- value ## S4 replacement method for signature 'sfn_data' get_env_flags(object) <- value ## S4 replacement method for signature 'sfn_data' get_timestamp(object) <- value ## S4 replacement method for signature 'sfn_data' get_solar_timestamp(object) <- value ## S4 replacement method for signature 'sfn_data' get_si_code(object) <- value ## S4 replacement method for signature 'sfn_data' get_site_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_stand_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_species_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_plant_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_env_md(object) <- value## S4 replacement method for signature 'sfn_data' get_sapf_data(object) <- value ## S4 replacement method for signature 'sfn_data' get_env_data(object) <- value ## S4 replacement method for signature 'sfn_data' get_sapf_flags(object) <- value ## S4 replacement method for signature 'sfn_data' get_env_flags(object) <- value ## S4 replacement method for signature 'sfn_data' get_timestamp(object) <- value ## S4 replacement method for signature 'sfn_data' get_solar_timestamp(object) <- value ## S4 replacement method for signature 'sfn_data' get_si_code(object) <- value ## S4 replacement method for signature 'sfn_data' get_site_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_stand_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_species_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_plant_md(object) <- value ## S4 replacement method for signature 'sfn_data' get_env_md(object) <- value
object |
sfn_data containing the slot to replace |
value |
object with the data to replace sfn_data slot with |
The replacement object must be a valid object for that slot:
For get_sapf_data, get_env_data, get_sapf_flags and
get_env_flags a data.frame or tibble without the TIMESTAMP
variable
For get_*_md a data.frame or tibble
For get_timestamp and get_solar_timestamp a POSIXct
vector of length == nrow(sapf/env_data)
For get_si_code a character vector
Validity is automatically checked before modifying the sfn_data object, and an error is raised if not valid
# preparation data('ARG_TRE', package = 'sapfluxnetr') sapf_data <- get_sapf_data(ARG_TRE, solar = TRUE) # modifying the slot data sapf_data[1:10, 2] <- NA # replacement. Remember, the sfn_data slot does not contain a TIMESTAMP # variable, it must be removed get_sapf_data(ARG_TRE) <- sapf_data[,-1]# preparation data('ARG_TRE', package = 'sapfluxnetr') sapf_data <- get_sapf_data(ARG_TRE, solar = TRUE) # modifying the slot data sapf_data[1:10, 2] <- NA # replacement. Remember, the sfn_data slot does not contain a TIMESTAMP # variable, it must be removed get_sapf_data(ARG_TRE) <- sapf_data[,-1]
Retrieves the site codes in the specified folder
sfn_sites_in_folder(folder = ".")sfn_sites_in_folder(folder = ".")
folder |
Character vector of length 1 indicating the route to the db folder |
If folder
A character vector with the site codes present in the folder, an error if the folder is not valid or does not contain any site data file.
# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) save(AUS_CAN_ST2_MIX, file = file.path(folder, 'AUS_EUC_ST2_MIX.RData')) # lets see the sites sites <- sfn_sites_in_folder(folder)# Let's access the data in "folder". This typically is the folder where the # sapflow data at the desired unit level is (i.e. "RData/plant"), but in this # example we will create a temporal folder with some data to test the function folder <- tempdir() save(ARG_TRE, file = file.path(folder, 'ARG_TRE.RData')) save(ARG_MAZ, file = file.path(folder, 'ARG_MAZ.RData')) save(AUS_CAN_ST2_MIX, file = file.path(folder, 'AUS_EUC_ST2_MIX.RData')) # lets see the sites sites <- sfn_sites_in_folder(folder)
sfn_vars_to_filter() returns a list with the variables for each
kind of metadata that can be used to select and filter sites
sfn_vars_to_filter()sfn_vars_to_filter()
A list with five elements, site_md, stand_md,
species_md, plant_md and env_md
# all variables sfn_vars_to_filter() # by some metadata sfn_vars_to_filter()$site_md# all variables sfn_vars_to_filter() # by some metadata sfn_vars_to_filter()$site_md
print a summary for sfn_data_multi objects
## S4 method for signature 'sfn_data_multi' show(object)## S4 method for signature 'sfn_data_multi' show(object)
object |
sfn_data_multi object to show |
print a summary for sfn_data objects
## S4 method for signature 'sfn_data' show(object)## S4 method for signature 'sfn_data' show(object)
object |
sfn_data object to show |
This function collapse the TIMESTAMP to the desired period (day, month...) by setting the same value to all timestamps within the period. This modified TIMESTAMP is used to group by and summarise the data.
summarise_by_period(data, period, .funs, ...)summarise_by_period(data, period, .funs, ...)
data |
sapflow or environmental data as obtained by |
period |
period to collapse by. See |
.funs |
funs to summarise the data. See details. |
... |
optional arguments. See details |
This function uses internally .collapse_timestamp and
summarise_all. Arguments to control these functions
can be passed as '...'. Arguments for each function are spliced and applied
when needed. Be advised that all arguments passed to the summarise_all function
will be applied to all the summarising functions used, so it will fail if any
of that functions does not accept that argument. To complex function-argument
relationships, indicate each summary function call within the .funs
argument as explained here summarise_all:
# This will fail beacuse na.rm argument will be also passed to the n function, # which does not accept any argument: summarise_by_period( data = get_sapf_data(ARG_TRE), period = '7 days', .funs = list(mean, sd, n()), na.rm = TRUE ) # to solve this is better to use the .funs argument: summarise_by_period( data = get_sapf_data(ARG_TRE), period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()) )
A 'tbl_df' object with the metrics results. The names of the columns indicate the original variable (tree or environmental variable) and the metric calculated (i.e. 'vpd_mean'), separated by underscore
Previously to the collapsing step, a temporal variable called
TIMESTAMP_coll is created to be able to catch the real timestamp when
some events happens, for example to use the min_time function. If
your custom summarise function needs to get the time at which some event
happens, use TIMESTAMP_coll instead of TIMESTAMP for that:
min_time <- function(x, time) {
time[which.min(x)]
}
summarise_by_period(
data = get_sapf_data(ARG_TRE),
period = '1 day',
.funs = list(~ min_time(., time = TIMESTAMP_coll)) # Not TIMESTAMP
)
library(dplyr) # data data('ARG_TRE', package = 'sapfluxnetr') # simple summary summarise_by_period( data = get_sapf_data(ARG_TRE), period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()) )library(dplyr) # data data('ARG_TRE', package = 'sapfluxnetr') # simple summary summarise_by_period( data = get_sapf_data(ARG_TRE), period = '7 days', .funs = list(~ mean(., na.rm = TRUE), ~ sd(., na.rm = TRUE), ~ n()) )