Exploring psychrometer time series

After visually inspecting and cleaning the half-hourly time series of Juniperus osteosperma stem water potentials, we turn to exploratory plots that will help understand how the trees and loggers relate to each other, to meteorological variables, and in diurnal patterns.

First, we bring in the relevant datasets, including manual validation of leaf water potential using a pressure chamber.

# Psychrometer (cleaned)
psy <- read_csv(file = "../data_cleaned/psy_hourly.csv",
                locale = locale(tz = "America/Denver")) %>%
  mutate(Tree = as.factor(Tree),
         Logger = as.factor(Logger))

# Manual validation data
man <- read_csv("../data_raw/Pressure-chamber-data.csv") %>%
  rename(dt = 1,
         WP_bar = 4,
         WP_mpa = 5,
         Psy = 6) %>%
  mutate(Tree = as.factor(Tree),
         Logger = as.factor(Logger),
         dt_orig = as.POSIXct(dt, format = "%m/%d/%Y %H:%M", tz = "America/Denver"),
         dt = round.POSIXct(dt_orig, units = "30 mins")) %>%
  select(Tree, Logger, dt, WP_mpa) %>%
  filter(!is.na(WP_mpa))

# Met data
col_names <- names(read_csv("../data_raw/Other-tower-data.csv",
                skip = 1, n_max = 0))

met <- read_csv("../data_raw/Other-tower-data.csv",
                skip = 4, 
                col_names = col_names, 
                na = c("-9999")) %>%
  mutate(dt = as.POSIXct(TIMESTAMP, format = "%m/%d/%Y %H:%M", 
                         tz = "America/Denver"),
         VPD_Avg = RHtoVPD(RH_Avg, AirTemp_Avg)) %>%
  select(dt, AirTemp_Avg, RH_Avg, VPD_Avg, 
         Precip_Tot, contains("VWC")) %>%
  filter(dt >= min(psy$dt) & dt <= max(psy$dt))

# Combine psy and manual and met, create time indices for plotting
psy_met <- psy %>%
  left_join(man) %>%
  left_join( met) %>%
  mutate(month = factor(month.name[lubridate::month(dt)],
                        levels = month.name),
         day = lubridate::day(dt),
         hour = lubridate::hour(dt),
         mo = sprintf("%02d", lubridate::month(dt)),
         Month = as.POSIXct(paste0("2021-", mo, "-01 00:00:00"), 
                            tz = "America/Denver"),
         month_diff = as.numeric(difftime(dt, Month, "secs")),
         month_ind = as.POSIXct(month_diff, 
                                origin = "1970-01-01",
                                tz = "America/Denver")) %>%
  select(-mo, -Month, -month_diff)

Plot all psychrometers together

After separately plotting each tree, we bring together all loggers together to see patterns across the seven individual trees

Examine relationship between met variables

Examine each tree with met variables

For each tree, plot stem water potential against soil moisture and VPD by month.The color axis indicates day within each month.

The presence of multiple relationships within each panel indicates that the effect of environmental drivers could be interactive and dependent on past conditions.

Plot diurnal patterns

Within each month, plot the diurnal patterns of stem water potential. Plots are separated for each tree to better check for differences between loggers.

Time series by month

Finally, the time series for each tree is broken down by month.