Example for scenario making
example-scenario.RmdMeteo data
Here we provide an example for a weather data set:
-
TEMP_MAXis the maximum daily temperature in C
-
TEMP_MINis the minimum daily temperature in C
-
RELATIVE_HUMIDITYis the average relative humidity in %
-
WIND_SPEEDis the average daily wind speed in m/s
-
PRECIP_QUANTITYis the total daily precipitation in mm
-
GLOBAL_RADIATIONis expressed in kWh/m².
Meteo_data <- read_delim(paste0(datapath, "data.csv"), delim = ",", show_col_types = F)
head(Meteo_data)
#> # A tibble: 6 × 9
#> YEAR MONTH DAY TEMP_MAX TEMP_MIN RELATIVE_HUMIDITY WIND_SPEED
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 2020 1 1 2.9 1.8 95.7 3.1
#> 2 2020 1 2 9.5 1.5 92.8 4
#> 3 2020 1 3 10.4 3 88.9 5.4
#> 4 2020 1 4 8.8 3.3 85.8 3.5
#> 5 2020 1 5 7.5 3.5 87 2.9
#> 6 2020 1 6 8.6 5.2 81.2 3.8
#> # ℹ 2 more variables: PRECIP_QUANTITY <dbl>, GLOBAL_RADIATION <dbl>This data set has some missing data. We solve this here by using the
approxfun function to iterpolate missing values.
Meteo_data <- Meteo_data %>%
mutate(DATE = as_date(paste(YEAR, MONTH, DAY, sep = "-")))
START <- "2020-01-01"
STOP <- "2020-12-31"
Tmax_fun <- approxfun(Meteo_data$DATE, Meteo_data$TEMP_MAX)
Tmin_fun <- approxfun(Meteo_data$DATE, Meteo_data$TEMP_MIN)
RH_fun <- approxfun(Meteo_data$DATE, Meteo_data$RELATIVE_HUMIDITY)
Wind_fun <- approxfun(Meteo_data$DATE, Meteo_data$WIND_SPEED)
Precip_fun <- approxfun(Meteo_data$DATE, Meteo_data$PRECIP_QUANTITY)
Radiation_fun <- approxfun(Meteo_data$DATE, Meteo_data$GLOBAL_RADIATION)
Alldays <- seq.Date(from = as_date(START), to = as_date(STOP), by = 1)
Meteo_data <- tibble(YEAR = lubridate::year(Alldays),
MONTH = lubridate::month(Alldays),
DAY = lubridate::day(Alldays),
TEMP_MAX = Tmax_fun(Alldays),
TEMP_MIN = Tmin_fun(Alldays),
RELATIVE_HUMIDITY = RH_fun(Alldays),
WIND_SPEED = Wind_fun(Alldays),
PRECIPITATION = Precip_fun(Alldays),
GLOBAL_RADIATION = Radiation_fun(Alldays),
DATE = Alldays)As AquaCrop requires reference evapotranspiration as an input, we
calculate this using the ETo_calc() function available
through the MeteoTools package, based on the
Evapotranspiration package.
library(MeteoTools)
ETo_data <- ETo_calc(DATE = Meteo_data$DATE,
T_max = Meteo_data$TEMP_MAX,
T_min = Meteo_data$TEMP_MIN,
RH_max = pmin(Meteo_data$RELATIVE_HUMIDITY, 100),
RH_min = pmin(Meteo_data$RELATIVE_HUMIDITY, 100),
Rs_tot = Meteo_data$GLOBAL_RADIATION*3.6, # from kWh m^-2 to MJ m^-2
uz_mean = Meteo_data$WIND_SPEED)
head(ETo_data)
#> # A tibble: 6 × 2
#> DATE ETo
#> <date> <dbl>
#> 1 2020-01-01 0.226
#> 2 2020-01-02 0.418
#> 3 2020-01-03 0.651
#> 4 2020-01-04 0.407
#> 5 2020-01-05 0.561
#> 6 2020-01-06 0.518Now we put everything in the AquacropOnR format:
Plu_example <- Meteo_data %>%
dplyr::mutate(DAY = lubridate::yday(DATE), PLU = PRECIPITATION) %>%
dplyr::select(DAY, PLU)
Tnx_example <- Meteo_data %>%
dplyr::mutate(DAY = lubridate::yday(DATE), TMAX = TEMP_MAX, TMIN = TEMP_MIN) %>%
dplyr::select(DAY, TMAX, TMIN)
ETo_example <- ETo_data %>%
dplyr::mutate(DAY = lubridate::yday(DATE)) %>%
dplyr::select(DAY, ETo)Soil data
SOL_s <- tribble(~ID, ~Horizon, ~Thickness, ~SAT, ~FC, ~WP, ~Ksat, ~Penetrability, ~Gravel,
"soil_1", 1, 0.1, 40.8, 29.2, 12.5, 120, 100, 0,
"soil_1", 2, 0.2, 44.3, 32.2, 10.5, 96, 100, 0,
"soil_1", 3, 0.6, 32.2, 21.4, 11.5, 52.5, 100, 0,
"soil_1", 4, 1.1, 30.1, 20.2, 10.5, 55, 100, 0,
"soil_2", 1, 0.1, 35.8, 29.2, 12.5, 1200, 100, 0,
"soil_2", 2, 0.5, 31.8, 24.5, 10.5, 1200, 100, 0,
"soil_2", 3, 2.2, 31.2, 22.2, 9.5, 950, 100, 0)
head(SOL_s)
#> # A tibble: 6 × 9
#> ID Horizon Thickness SAT FC WP Ksat Penetrability Gravel
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 soil_1 1 0.1 40.8 29.2 12.5 120 100 0
#> 2 soil_1 2 0.2 44.3 32.2 10.5 96 100 0
#> 3 soil_1 3 0.6 32.2 21.4 11.5 52.5 100 0
#> 4 soil_1 4 1.1 30.1 20.2 10.5 55 100 0
#> 5 soil_2 1 0.1 35.8 29.2 12.5 1200 100 0
#> 6 soil_2 2 0.5 31.8 24.5 10.5 1200 100 0Initial conditions
If you do not provide initial conditions in the
design_scenario() function, it will take the default
setting “FC”, which results in a soil profile water content at field
capacity and absence of salts at the start of the simulation.
The initial conditions need to have the same Horizon and Thickness
variables as the soil input tibble SOL_s. here we provide
an example where the initial water content at all layers lies in the
middle between wilting point (WP) and field capacity (FC).
sw0_factor <- 0.5
SW0_s <- SOL_s %>% mutate(WC = WP + (FC-WP)*sw0_factor, ECe = 0.0) %>%
select(ID, Horizon, Thickness, WC, ECe)
head(SW0_s)
#> # A tibble: 6 × 5
#> ID Horizon Thickness WC ECe
#> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 soil_1 1 0.1 20.8 0
#> 2 soil_1 2 0.2 21.4 0
#> 3 soil_1 3 0.6 16.4 0
#> 4 soil_1 4 1.1 15.4 0
#> 5 soil_2 1 0.1 20.8 0
#> 6 soil_2 2 0.5 17.5 0Management
Through the .MAN file, AquaCrop can take into account several field management settings, such as:
- the effect of mulching, using the percentage of soil covered by
mulch (
mulch_perc), and the effect on reducing soil evaporation (mulch_effbetween 0 (no effect) and 100 (all evaporation inhibited)).
- the fertility stress (
fert_stress, between 0 (no stress) and 100 (complete stress)).
- the presence of soil bunds (
soil_bunds).
- the use of run-off affecting measures (
runoff_aff, 0 or 1 (for complete prevention of run-off)) and their effect on the curve number (CN_eff).
- presence of weeds (
weed_clo,weed_mid,weed_cc).
For an effect of erosion preventing measures, the management file can be defined as follows:
FMAN_s <- tribble(~ID, ~mulch_perc, ~mulch_eff, ~fert_stress, ~soil_bunds, ~runoff_aff, ~CN_eff, ~weed_clo, ~weed_mid, ~weed_cc,
"default", 0, 50, 0, 0, 0, 0, 0, 0, -0.01,
"reduction_1", 0, 50, 0, 0, 0, -25, 0, 0, -0.01)where the default treatment has no measures and the reduction_1 treatment ahas measures that reduce the curve number by 25.