data-raw/import_model_results.R
In WDNR-Water-Use/CSLSdata: WI DNR Central Sands Lakes Study Field Data
#' Import Model Results
#'
#' This function loads model outputs from csv files and saves as combined data
#' frame. Important assumptions:
#'
#' * Start date is "01-01-1981" for long runs
#' * Start date is "01-01-2012" for calibration runs
#' * LAK package information is read out on the first of the next month (e.g.,
#' January rates are read out on 2/1)
#' * Signs are such that 0 = P + E + GWin + GWout + dV
#'
#' @return df, a data frame with the following columns:
#' \item{sim}{type of model simulation, e.g. "irr", or "no_irr"}
#' \item{lake}{name of lake, i.e. "Pleasant", "Long", "Plainfield"}
#' \item{date}{date of output, POSIXct}
#' \item{level_m}{mean lake stage for the month, mamsl}
#' \item{P_m3}{precipitation volume this month (m^3)}
#' \item{E_m3}{evaporation volume this month (m^3)}
#' \item{GWin_m3}{groundwater inflow volume this month (m^3)}
#' \item{GWout_m3}{groundwater outflow volume this month (m^3)}
#' \item{dV_m3}{change in lake volume this month (m^3)}
#' \item{P_m3_d}{precipitation flow rate (m^3/d)}
#' \item{E_m3_d}{evaporation flow rate (m^3/d)}
#' \item{GWin_m3_d}{groundwater inflow rate into lake (m^3/d)}
#' \item{GWout_m3_d}{groundwater outflow rate into lake (m^3/d)}
#' \item{dV_m3_d}{change in lake volume as a flow rate (m^3/d)}
import_model_results <- function(scenarios = c("cal", "cur_irr", "no_irr",
"fut_irr", "wells_off"),
lakes = c("Pleasant", "Long", "Plainfield")) {
library(dplyr)
library(stringr)
library(lubridate)
library(zoo)
library(reshape2)
df <- NULL
for (scenario in scenarios) {
# Set appropriate start_date for the scenario
if (scenario == "cal") {
start_date <- as_datetime(mdy("01-01-2012"))
} else {
start_date <- as_datetime(mdy("01-01-1981"))
}
# 1. Fluxes ----------------------------------------------------------------
# Read in csv for each lake
ls <- list()
for (lake in lakes) {
ls[[lake]] <- read.csv(sprintf("data-raw/MODFLOW/MODFLOW_%s_fluxes_%s.csv",
lake, scenario))
}
# Combine all lakes into one data frame
lk_fluxes <- NULL
for (lake in names(ls)) {
this_df <- ls[[lake]]
this_df$lake <- lake
lk_fluxes <- bind_rows(lk_fluxes, this_df)
}
# Set date to the start_datetime
lk_fluxes <- lk_fluxes %>%
mutate(date = start_date + days(.data$mf_time) -
months(1) - days(1)) %>%
filter(.data$date >= start_date)
lk_fluxes$mf_time <- NULL
# Read in monte carlo simulations
if (!"gw_in" %in% colnames(lk_fluxes)) {
lk_fluxes <- lk_fluxes %>%
melt(id.vars = c("lake", "date")) %>%
mutate(flux = str_replace(.data$variable, "_real.+", ""),
scenario = scenario,
sim = str_replace(.data$variable, ".+_real", ""),
sim = as.numeric(.data$sim) + 1) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$flux, .data$value) %>%
dcast(lake+date+scenario+sim~flux, value.var = "value")
} else if (scenario %in% c("cal")) {
lk_fluxes <- lk_fluxes %>%
mutate(scenario = scenario,
sim = 0) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$gw_in, .data$gw_out, .data$gw_net)
} else {
lk_fluxes <- lk_fluxes %>%
mutate(scenario = scenario,
sim = 1) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$gw_in, .data$gw_out, .data$gw_net)
}
# 2. Stages ----------------------------------------------------------------
# Import CSVs to a list
ls <- list()
for (lake in lakes) {
ls[[lake]] <- read.csv(sprintf("data-raw/MODFLOW/MODFLOW_%s_stages_%s.csv",
lake, scenario))
}
# Combine in one data frame, rename funny column names
lk_pkg <- NULL
for (lake in names(ls)) {
this_df <- ls[[lake]]
this_df$lake <- lake
lk_pkg <- bind_rows(lk_pkg, this_df)
}
colnames(lk_pkg) <- str_to_lower(colnames(lk_pkg))
colnames(lk_pkg) <- str_replace(colnames(lk_pkg), "\\.", "_")
# Convert time to dates, keep only the 1st of the month
lk_pkg$date <- start_date + ddays(lk_pkg$time) - days(1)
lk_pkg <- lk_pkg %>% filter(day(.data$date) == 1)
# Retain only useful columns
col_matches <- "lake|stage|rainfall|lak|storage|evaporation|date|volume"
lk_pkg <- lk_pkg[,str_detect(colnames(lk_pkg), col_matches)]
# Read in monte carlo simulations
if (!"stage" %in% colnames(lk_pkg)) {
lk_pkg <- lk_pkg %>%
melt(id.vars = c("lake", "date")) %>%
mutate(flux = str_replace(.data$variable, "_real.+", ""),
scenario = scenario,
sim = str_replace(.data$variable, ".+_real", ""),
sim = as.numeric(.data$sim) + 1) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$flux, .data$value) %>%
dcast(lake+date+scenario+sim~flux, value.var = "value")
} else if (scenario %in% c("cal")) {
lk_pkg <- lk_pkg %>%
mutate(scenario = str_replace(scenario, "_baseline", ""),
sim = 0)
} else {
lk_pkg <- lk_pkg %>%
mutate(scenario = scenario,
sim = 1)
}
# Calculate lake stage as average between 1st of month & 1st of next month
lk_stages <- NULL
for (lake in lakes) {
for (sim in unique(lk_pkg$sim)) {
this_stage <- lk_pkg %>%
filter(.data$lake == !!lake,
.data$sim == !!sim) %>%
arrange(.data$date) %>%
mutate(stage_next = lead(.data$stage, 1, order_by = .data$date),
stage = 0.5*(.data$stage + .data$stage_next)) %>%
select(.data$lake, .data$date, .data$scenario,
.data$sim, .data$stage)
lk_stages <- bind_rows(lk_stages, this_stage)
}
}
lk_vols <- lk_pkg %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$volume)
lk_stages <- left_join(lk_stages, lk_vols,
by = c("lake", "date", "scenario", "sim"))
# lk_stages <- lk_pkg %>%
# select(.data$lake, .data$date, .data$scenario, .data$sim,
# .data$stage, .data$volume)
# Roll back dates for lake fluxes to first of the previous month
# For example, a LAK values on 2/1 represent the net groundwater flow from
# 1/1 to 2/1, so assign it the date 1/1.
lk_pkg$date <- lk_pkg$date - months(1)
lk_pkg <- lk_pkg %>%
filter(.data$date >= start_date) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$rainfall, .data$lak, .data$evaporation,
.data$storage)
# Merge stages and fluxes back together
lk_stages <- full_join(lk_stages, lk_pkg,
by = c("lake", "date", "scenario", "sim"))
# 3. Combine to single data frame ------------------------------------------
this_sim <- full_join(lk_stages, lk_fluxes,
by = c("lake", "date", "scenario", "sim"))
# Fill in missing gw values
this_sim <- this_sim %>%
mutate(gw_net = ifelse(is.na(.data$gw_net),
.data$lak, .data$gw_net),
gw_out = ifelse(is.na(.data$gw_out),
.data$gw_net - .data$gw_in,
.data$gw_out),
gw_in = ifelse(is.na(.data$gw_in),
.data$gw_net - .data$gw_out,
.data$gw_in))
# Include values as m3/month and m3/day
this_sim <- this_sim %>%
filter(!is.na(.data$stage)) %>%
mutate(days = day(.data$date + months(1) - days(1)),
sim = ifelse(.data$sim == 10000,
0, .data$sim),
P_m3 = .data$rainfall*.data$days,
E_m3 = .data$evaporation*.data$days,
GWin_m3 = .data$gw_in*.data$days,
GWout_m3 = .data$gw_out*.data$days,
dV_m3 = .data$storage*.data$days,
vol_m3 = .data$volume) %>%
select(scenario = .data$scenario,
sim = .data$sim,
lake = .data$lake,
date = .data$date,
level_m = .data$stage,
vol_m3 = .data$vol_m3,
P_m3 = .data$P_m3,
E_m3 = .data$E_m3,
GWin_m3 = .data$GWin_m3,
GWout_m3 = .data$GWout_m3,
dV_m3 = .data$dV_m3,
P_m3_d = .data$rainfall,
E_m3_d = .data$evaporation,
GWin_m3_d = .data$gw_in,
GWout_m3_d = .data$gw_out,
dV_m3_d = .data$storage)
this_sim$lake <- factor(this_sim$lake, levels = lakes)
df <- rbind(df, this_sim)
}
return(df)
# # Code used to explore LAK vs. NET_GW relationship
# NA_values <- this_sim %>% filter(is.na(.data$gw_net))
# ggplot(NA_values) +
# geom_point(aes(x = .data$date,
# y = .data$lak,
# color = "LAK",
# shape = "LAK"),
# size = 3) +
# geom_point(aes(x = .data$date,
# y = .data$gw_in,
# color = "GW_IN",
# shape = "GW_IN"),
# size = 2) +
# geom_point(aes(x = .data$date,
# y = .data$gw_out,
# color = "GW_OUT",
# shape = "GW_OUT"),
# size = 2) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "Flux (m3/d)", color = "") +
# scale_color_manual(name = "",
# breaks = c("LAK", "GW_IN", "GW_OUT"),
# values = c("grey80", "blue", "darkred")) +
# scale_shape_manual(name = "",
# breaks = c("LAK", "GW_IN", "GW_OUT"),
# values = c(15, 16, 17))
#
# ggplot(filter(this_sim, .data$lake == "Pleasant")) +
# geom_line(aes(x = .data$date,
# y = .data$lak,
# color = "LAK")) +
# geom_line(aes(x = .data$date,
# y = .data$gw_net,
# color = "GW_NET")) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "Flux (m3/d)", color = "") +
# scale_color_manual(name = "",
# breaks = c("LAK", "GW_NET"),
# values = c("black", "blue"))
#
# ggplot(this_sim) +
# geom_line(aes(x = .data$date,
# y = .data$lak_balance,
# color = "LAK")) +
# geom_line(aes(x = .data$date,
# y = .data$gw_balance,
# color = "GW_NET")) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "Water Balance (m3/d)", color = "", title = "Water Balance ('irr' scenario) derived via 'LAK' and 'GW_NET'") +
# scale_color_manual(name = "",
# breaks = c("LAK", "GW_NET"),
# values = c("black", "blue"))
#
# ggplot(filter(this_sim, .data$lake == "Pleasant")) +
# geom_line(aes(x = .data$date,
# y = abs(.data$diff))) +
# theme_bw() +
# scale_y_continuous(limits = c(0,200)) +
# labs(x = "", y = "LAK - GW_NET (%)")
#
# ggplot(this_sim) +
# geom_line(aes(x = .data$date,
# y = .data$lak - .data$gw_net),
# color = "black") +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "LAK - GW_NET (m3/d)")
#
# ggplot(df) +
# geom_line(aes(x = .data$date,
# y = .data$GWin_m3,
# color = .data$sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "GW_IN (m3/d)")
# Code used to explore GW vs. stage relationships
# plot_df <- df %>%
# group_by(.data$sim, .data$lake, year(.data$date)) %>%
# summarise(level_m = mean(level_m),
# P_m3 = sum(P_m3),
# E_m3 = sum(E_m3),
# GWin_m3 = sum(GWin_m3),
# GWout_m3 = sum(GWout_m3),
# dV_m3 = sum(dV_m3))%>%
# ungroup()
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTmeterTOft(level_m),
# y = -100*GWin_m3/(-GWin_m3 + P_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Mean Lake Elevation (ft)", y = "Groundwater Inflow (%)")
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTmeterTOft(level_m),
# y = NISTunits::NISTcubMeterTOacreFt(-GWin_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Mean Lake Elevation (ft)", y = "Groundwater Inflow (ac-ft)")
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTmeterTOft(level_m),
# y = NISTunits::NISTcubMeterTOacreFt(GWout_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Mean Lake Elevation (ft)", y = "Groundwater Outflow (ac-ft)")
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTcubMeterTOacreFt(-GWin_m3),
# y = NISTunits::NISTcubMeterTOacreFt(GWout_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Groundwater Inflow (ac-ft)", y = "Groundwater Outflow (ac-ft)")
}
WDNR-Water-Use/CSLSdata documentation built on Nov. 12, 2021, 2:36 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
#' Import Model Results
#'
#' This function loads model outputs from csv files and saves as combined data
#' frame. Important assumptions:
#'
#' * Start date is "01-01-1981" for long runs
#' * Start date is "01-01-2012" for calibration runs
#' * LAK package information is read out on the first of the next month (e.g.,
#' January rates are read out on 2/1)
#' * Signs are such that 0 = P + E + GWin + GWout + dV
#'
#' @return df, a data frame with the following columns:
#' \item{sim}{type of model simulation, e.g. "irr", or "no_irr"}
#' \item{lake}{name of lake, i.e. "Pleasant", "Long", "Plainfield"}
#' \item{date}{date of output, POSIXct}
#' \item{level_m}{mean lake stage for the month, mamsl}
#' \item{P_m3}{precipitation volume this month (m^3)}
#' \item{E_m3}{evaporation volume this month (m^3)}
#' \item{GWin_m3}{groundwater inflow volume this month (m^3)}
#' \item{GWout_m3}{groundwater outflow volume this month (m^3)}
#' \item{dV_m3}{change in lake volume this month (m^3)}
#' \item{P_m3_d}{precipitation flow rate (m^3/d)}
#' \item{E_m3_d}{evaporation flow rate (m^3/d)}
#' \item{GWin_m3_d}{groundwater inflow rate into lake (m^3/d)}
#' \item{GWout_m3_d}{groundwater outflow rate into lake (m^3/d)}
#' \item{dV_m3_d}{change in lake volume as a flow rate (m^3/d)}
import_model_results <- function(scenarios = c("cal", "cur_irr", "no_irr",
"fut_irr", "wells_off"),
lakes = c("Pleasant", "Long", "Plainfield")) {
library(dplyr)
library(stringr)
library(lubridate)
library(zoo)
library(reshape2)
df <- NULL
for (scenario in scenarios) {
# Set appropriate start_date for the scenario
if (scenario == "cal") {
start_date <- as_datetime(mdy("01-01-2012"))
} else {
start_date <- as_datetime(mdy("01-01-1981"))
}
# 1. Fluxes ----------------------------------------------------------------
# Read in csv for each lake
ls <- list()
for (lake in lakes) {
ls[[lake]] <- read.csv(sprintf("data-raw/MODFLOW/MODFLOW_%s_fluxes_%s.csv",
lake, scenario))
}
# Combine all lakes into one data frame
lk_fluxes <- NULL
for (lake in names(ls)) {
this_df <- ls[[lake]]
this_df$lake <- lake
lk_fluxes <- bind_rows(lk_fluxes, this_df)
}
# Set date to the start_datetime
lk_fluxes <- lk_fluxes %>%
mutate(date = start_date + days(.data$mf_time) -
months(1) - days(1)) %>%
filter(.data$date >= start_date)
lk_fluxes$mf_time <- NULL
# Read in monte carlo simulations
if (!"gw_in" %in% colnames(lk_fluxes)) {
lk_fluxes <- lk_fluxes %>%
melt(id.vars = c("lake", "date")) %>%
mutate(flux = str_replace(.data$variable, "_real.+", ""),
scenario = scenario,
sim = str_replace(.data$variable, ".+_real", ""),
sim = as.numeric(.data$sim) + 1) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$flux, .data$value) %>%
dcast(lake+date+scenario+sim~flux, value.var = "value")
} else if (scenario %in% c("cal")) {
lk_fluxes <- lk_fluxes %>%
mutate(scenario = scenario,
sim = 0) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$gw_in, .data$gw_out, .data$gw_net)
} else {
lk_fluxes <- lk_fluxes %>%
mutate(scenario = scenario,
sim = 1) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$gw_in, .data$gw_out, .data$gw_net)
}
# 2. Stages ----------------------------------------------------------------
# Import CSVs to a list
ls <- list()
for (lake in lakes) {
ls[[lake]] <- read.csv(sprintf("data-raw/MODFLOW/MODFLOW_%s_stages_%s.csv",
lake, scenario))
}
# Combine in one data frame, rename funny column names
lk_pkg <- NULL
for (lake in names(ls)) {
this_df <- ls[[lake]]
this_df$lake <- lake
lk_pkg <- bind_rows(lk_pkg, this_df)
}
colnames(lk_pkg) <- str_to_lower(colnames(lk_pkg))
colnames(lk_pkg) <- str_replace(colnames(lk_pkg), "\\.", "_")
# Convert time to dates, keep only the 1st of the month
lk_pkg$date <- start_date + ddays(lk_pkg$time) - days(1)
lk_pkg <- lk_pkg %>% filter(day(.data$date) == 1)
# Retain only useful columns
col_matches <- "lake|stage|rainfall|lak|storage|evaporation|date|volume"
lk_pkg <- lk_pkg[,str_detect(colnames(lk_pkg), col_matches)]
# Read in monte carlo simulations
if (!"stage" %in% colnames(lk_pkg)) {
lk_pkg <- lk_pkg %>%
melt(id.vars = c("lake", "date")) %>%
mutate(flux = str_replace(.data$variable, "_real.+", ""),
scenario = scenario,
sim = str_replace(.data$variable, ".+_real", ""),
sim = as.numeric(.data$sim) + 1) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$flux, .data$value) %>%
dcast(lake+date+scenario+sim~flux, value.var = "value")
} else if (scenario %in% c("cal")) {
lk_pkg <- lk_pkg %>%
mutate(scenario = str_replace(scenario, "_baseline", ""),
sim = 0)
} else {
lk_pkg <- lk_pkg %>%
mutate(scenario = scenario,
sim = 1)
}
# Calculate lake stage as average between 1st of month & 1st of next month
lk_stages <- NULL
for (lake in lakes) {
for (sim in unique(lk_pkg$sim)) {
this_stage <- lk_pkg %>%
filter(.data$lake == !!lake,
.data$sim == !!sim) %>%
arrange(.data$date) %>%
mutate(stage_next = lead(.data$stage, 1, order_by = .data$date),
stage = 0.5*(.data$stage + .data$stage_next)) %>%
select(.data$lake, .data$date, .data$scenario,
.data$sim, .data$stage)
lk_stages <- bind_rows(lk_stages, this_stage)
}
}
lk_vols <- lk_pkg %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$volume)
lk_stages <- left_join(lk_stages, lk_vols,
by = c("lake", "date", "scenario", "sim"))
# lk_stages <- lk_pkg %>%
# select(.data$lake, .data$date, .data$scenario, .data$sim,
# .data$stage, .data$volume)
# Roll back dates for lake fluxes to first of the previous month
# For example, a LAK values on 2/1 represent the net groundwater flow from
# 1/1 to 2/1, so assign it the date 1/1.
lk_pkg$date <- lk_pkg$date - months(1)
lk_pkg <- lk_pkg %>%
filter(.data$date >= start_date) %>%
select(.data$lake, .data$date, .data$scenario, .data$sim,
.data$rainfall, .data$lak, .data$evaporation,
.data$storage)
# Merge stages and fluxes back together
lk_stages <- full_join(lk_stages, lk_pkg,
by = c("lake", "date", "scenario", "sim"))
# 3. Combine to single data frame ------------------------------------------
this_sim <- full_join(lk_stages, lk_fluxes,
by = c("lake", "date", "scenario", "sim"))
# Fill in missing gw values
this_sim <- this_sim %>%
mutate(gw_net = ifelse(is.na(.data$gw_net),
.data$lak, .data$gw_net),
gw_out = ifelse(is.na(.data$gw_out),
.data$gw_net - .data$gw_in,
.data$gw_out),
gw_in = ifelse(is.na(.data$gw_in),
.data$gw_net - .data$gw_out,
.data$gw_in))
# Include values as m3/month and m3/day
this_sim <- this_sim %>%
filter(!is.na(.data$stage)) %>%
mutate(days = day(.data$date + months(1) - days(1)),
sim = ifelse(.data$sim == 10000,
0, .data$sim),
P_m3 = .data$rainfall*.data$days,
E_m3 = .data$evaporation*.data$days,
GWin_m3 = .data$gw_in*.data$days,
GWout_m3 = .data$gw_out*.data$days,
dV_m3 = .data$storage*.data$days,
vol_m3 = .data$volume) %>%
select(scenario = .data$scenario,
sim = .data$sim,
lake = .data$lake,
date = .data$date,
level_m = .data$stage,
vol_m3 = .data$vol_m3,
P_m3 = .data$P_m3,
E_m3 = .data$E_m3,
GWin_m3 = .data$GWin_m3,
GWout_m3 = .data$GWout_m3,
dV_m3 = .data$dV_m3,
P_m3_d = .data$rainfall,
E_m3_d = .data$evaporation,
GWin_m3_d = .data$gw_in,
GWout_m3_d = .data$gw_out,
dV_m3_d = .data$storage)
this_sim$lake <- factor(this_sim$lake, levels = lakes)
df <- rbind(df, this_sim)
}
return(df)
# # Code used to explore LAK vs. NET_GW relationship
# NA_values <- this_sim %>% filter(is.na(.data$gw_net))
# ggplot(NA_values) +
# geom_point(aes(x = .data$date,
# y = .data$lak,
# color = "LAK",
# shape = "LAK"),
# size = 3) +
# geom_point(aes(x = .data$date,
# y = .data$gw_in,
# color = "GW_IN",
# shape = "GW_IN"),
# size = 2) +
# geom_point(aes(x = .data$date,
# y = .data$gw_out,
# color = "GW_OUT",
# shape = "GW_OUT"),
# size = 2) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "Flux (m3/d)", color = "") +
# scale_color_manual(name = "",
# breaks = c("LAK", "GW_IN", "GW_OUT"),
# values = c("grey80", "blue", "darkred")) +
# scale_shape_manual(name = "",
# breaks = c("LAK", "GW_IN", "GW_OUT"),
# values = c(15, 16, 17))
#
# ggplot(filter(this_sim, .data$lake == "Pleasant")) +
# geom_line(aes(x = .data$date,
# y = .data$lak,
# color = "LAK")) +
# geom_line(aes(x = .data$date,
# y = .data$gw_net,
# color = "GW_NET")) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "Flux (m3/d)", color = "") +
# scale_color_manual(name = "",
# breaks = c("LAK", "GW_NET"),
# values = c("black", "blue"))
#
# ggplot(this_sim) +
# geom_line(aes(x = .data$date,
# y = .data$lak_balance,
# color = "LAK")) +
# geom_line(aes(x = .data$date,
# y = .data$gw_balance,
# color = "GW_NET")) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "Water Balance (m3/d)", color = "", title = "Water Balance ('irr' scenario) derived via 'LAK' and 'GW_NET'") +
# scale_color_manual(name = "",
# breaks = c("LAK", "GW_NET"),
# values = c("black", "blue"))
#
# ggplot(filter(this_sim, .data$lake == "Pleasant")) +
# geom_line(aes(x = .data$date,
# y = abs(.data$diff))) +
# theme_bw() +
# scale_y_continuous(limits = c(0,200)) +
# labs(x = "", y = "LAK - GW_NET (%)")
#
# ggplot(this_sim) +
# geom_line(aes(x = .data$date,
# y = .data$lak - .data$gw_net),
# color = "black") +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "LAK - GW_NET (m3/d)")
#
# ggplot(df) +
# geom_line(aes(x = .data$date,
# y = .data$GWin_m3,
# color = .data$sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "", y = "GW_IN (m3/d)")
# Code used to explore GW vs. stage relationships
# plot_df <- df %>%
# group_by(.data$sim, .data$lake, year(.data$date)) %>%
# summarise(level_m = mean(level_m),
# P_m3 = sum(P_m3),
# E_m3 = sum(E_m3),
# GWin_m3 = sum(GWin_m3),
# GWout_m3 = sum(GWout_m3),
# dV_m3 = sum(dV_m3))%>%
# ungroup()
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTmeterTOft(level_m),
# y = -100*GWin_m3/(-GWin_m3 + P_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Mean Lake Elevation (ft)", y = "Groundwater Inflow (%)")
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTmeterTOft(level_m),
# y = NISTunits::NISTcubMeterTOacreFt(-GWin_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Mean Lake Elevation (ft)", y = "Groundwater Inflow (ac-ft)")
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTmeterTOft(level_m),
# y = NISTunits::NISTcubMeterTOacreFt(GWout_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Mean Lake Elevation (ft)", y = "Groundwater Outflow (ac-ft)")
#
# ggplot(plot_df) +
# geom_point(aes(x = NISTunits::NISTcubMeterTOacreFt(-GWin_m3),
# y = NISTunits::NISTcubMeterTOacreFt(GWout_m3),
# color = sim)) +
# facet_wrap(~lake, scales = "free") +
# theme_bw() +
# labs(x = "Groundwater Inflow (ac-ft)", y = "Groundwater Outflow (ac-ft)")
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.