Skip to content

Commit

Permalink
update pcakge vignette rmd after package style
Browse files Browse the repository at this point in the history
  • Loading branch information
@dsheng1026
dsheng1026 committed Jan 21, 2025
1 parent 5e0a7c3 commit d5886e6
Showing 1 changed file with 31 additions and 31 deletions.
62 changes: 31 additions & 31 deletions vignettes/HELPS-vignette.Rmd
View file
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@ HELPS has been tested with both ISIMIP data and outputs from BASD.

Besides the climate projections, `HELPS` also takes GGCMI Phase 3 crop calendar (Jägermeyr et al. 2021) (https://zenodo.org/records/5062513) and SPAM harvested area (IFPRI, 2024) (https://doi.org/10.7910/DVN/SWPENT) data as input for temporal and spatial aggregation in the package.

`HELPS` provides measurement of crop specific labor heat stress exposure and heat-induced physical work capacity loss for 93 sectors, consisting of 46 SPAM crops by 2 irrigation practices (rain-fed and irrigated) and 1 noncrop sector. Below is the fUll list of sectors supported in `HELPS`, and refer to Table 1 for crop mapping:
`HELPS` provides measurement of crop specific labor heat stress exposure and heat-induced physical work capacity loss for 93 sectors, consisting of 46 SPAM crops by 2 irrigation practices (rain-fed and irrigated) and 1 noncrop sector. Below is the fUll list of sectors supported in `HELPS`, and refer to Table 1 for crop mapping. Note that for SPAM crops without corresponding ggcmi crop calendar, the crop calendar is assumed to be the whole year.
```
SECTOR_ALL <- c("WHEA_I", "RICE_I", "MAIZ_I", "SOYB_I", "BARL_I", "MILL_I", "PMIL_I", "SORG_I", "OCER_I",
"POTA_I", "SWPO_I", "YAMS_I", "CASS_I", "BEAN_I", "CHIC_I", "COWP_I", "PIGE_I", "LENT_I",
Expand All @@ -54,13 +54,13 @@ SECTOR_ALL <- c("WHEA_I", "RICE_I", "MAIZ_I", "SOYB_I", "BARL_I", "MILL_I", "PMI
"REST_I",
"WHEA_R", "RICE_R", "MAIZ_R", "SOYB_R", "BARL_R", "MILL_R", "PMIL_R", "SORG_R", "OCER_R",
"POTA_R", "SWPO_R", "YAMS_R", "CASS_R", "BEAN_R", "CHIC_R", "COWP_R", "PIGE_R", "LENT_R",
"GROU_R", "SUNF_R", "RAPE_R", "SESA_R", "SUGC_R", "SUGB_R", "COTT_R", "OFIB_R", "BANA_R",
"GROU_R", "MAIZ_R", "RAPE_R", "SESA_R", "SUGC_R", "SUGB_R", "COTT_R", "OFIB_R", "BANA_R",
"PLNT_R", "CITR_R", "TROF_R", "TEMF_R", "TOMA_R", "ONIO_R", "VEGE_R", "ORTS_R", "OPUL_R",
"CNUT_R", "OILP_R", "OOIL_R", "COFF_R", "RCOF_R", "COCO_R", "RUBB_R", "TEAS_R", "TOBA_R",
"REST_R",
"NONCROP")
```
### Table 1. Crop mapping between data sources.
### Table 1. Mapping ggcmi crop calendar to SPAM crops.
| SPAM | SPAM Crop Name | ggcmi | ggcmi Crop Name|
|------|-------------------|-------|----------------|
| WHEA | Wheat | wh | wheat |
Expand Down Expand Up @@ -116,7 +116,7 @@ SECTOR_ALL <- c("WHEA_I", "RICE_I", "MAIZ_I", "SOYB_I", "BARL_I", "MILL_I", "PMI

### Step1.1: calculate daily gridded heat stress

This step employs `HeatStress()` function, which takes multiple arguments. `SECTOR` denotes which sector the labor is involved in. The current version of **HELPS** supports 93 sectors, consists of 92 crop sectors (46 SPAM crops by 2 irrigation practices) and 1 non-crop sector.
This step employs `cal_heat_stress()` function, which takes multiple arguments. `SECTOR` denotes which sector the labor is involved in. The current version of **HELPS** supports 93 sectors, consists of 92 crop sectors (46 SPAM crops by 2 irrigation practices) and 1 non-crop sector.
`HS` denotes the heat stress function that takes atmospheric variables as input and outputs heat stress level. The current version of **HELPS** covers 3 `HS()`, `WBGT_ESI()`, `WBGT_shade()`, and `WBGT_sun()`.
`WBGT_ESI`, Environmental Stress Index, takes hurs, tas, and rsds as inputs to approximate WBGT.
`WBGT_shade`, from Dunne et al. (2013), takes hurs, tas, and ps as inputs to approximate WBGT. This method assumes full shade and night adaptation and optimization of structures, clothing, activity scheduling.
Expand All @@ -127,7 +127,7 @@ This step employs `HeatStress()` function, which takes multiple arguments. `SECT
Example below takes monthly *hurs* (relative humidity), *tas* (near-surface air temperature), and *ps* (surface air pressure) and *WBGT_ESI* to calculate heat stress levels.

```{r Step1.1, warning=FALSE}
wbgt.sun.day <- HeatStress(TempRes = "day", SECTOR = "SUNF_R", HS = WBGT_sun, YEAR_INPUT = 2024,
wbgt.sun.day <- cal_heat_stress(TempRes = "day", SECTOR = "MAIZ_R", HS = WBGT_sun, YEAR_INPUT = 2024,
"../HELPS_Example_Data/hurs_example_day.nc",
"../HELPS_Example_Data/tas_example_day.nc",
"../HELPS_Example_Data/ps_example_day.nc")
Expand Down Expand Up @@ -160,7 +160,7 @@ wbgt.day.plot %>%
scale_fill_gradient2(low = "blue", mid = "grey95", high = "red", midpoint = 25,
na.value = "white")+
labs(x = "longitude", y = "latitude", fill = "WBGT",
title = "Heat stress for labor in rain-fed sunflower sector: 2024-07-01") +
title = "Heat stress for labor in rain-fed maize sector: 2024-07-01") +
scale_x_continuous(
breaks = seq(-180, 180, by = 60), # Longitude breaks
Expand All @@ -176,8 +176,8 @@ wbgt.day.plot %>%

### Step1.2: calculate daily gridded physical work capacity

This step employs `PWC()` function, which takes multiple arguments.
`WBGT` denotes the output of `HeatStress()` from last step.
This step employs `cal_pwc()` function, which takes multiple arguments.
`WBGT` denotes the output of `cal_heat_stress()` from last step.
`LHR` denotes the labor-heat response function that takes heat stress level and workload intensity as input and outputs physical work capacity (PWC). `workload` varies across alternative industries, we suggest 'high' for construction or agricultural work, 'moderate' for average manufacturing industry work, and 'low' for office desk work and service industries.
There are 4 built-in `LHR` functions in `HELPS`.
`LHR_Hothaps` is based on the *high occupational temperature health and productivity suppression (Hothaps)* programme, derived by Kjellstrom et al. (2018).
Expand All @@ -188,7 +188,7 @@ Note that `LHR_Foster` does not need workload as input, while the rest three LHR


```{r Step1.2, warning=FALSE}
pwc.hothaps.day <- PWC(WBGT = wbgt.sun.day, LHR = LHR_Hothaps, workload = "high")
pwc.hothaps.day <- cal_pwc(WBGT = wbgt.sun.day, LHR = LHR_Hothaps, workload = "high")
pwc.hothaps.day
```
```{r Step1.2.1, warning=FALSE, fig.width=10, fig.height=6}
Expand Down Expand Up @@ -216,7 +216,7 @@ pwc.day.plot %>%
scale_fill_gradient(low = "grey95", high = "red",
na.value = "white")+
labs(x = "longitude", y = "latitude", fill = "PWC \nloss(%)",
title = "Heat-induced PWC loss for labor in rain-fed sunflower sector: 2024-07-01") +
title = "Heat-induced PWC loss for labor in rain-fed maize sector: 2024-07-01") +
scale_x_continuous(
breaks = seq(-180, 180, by = 60), # Longitude breaks
Expand All @@ -231,12 +231,12 @@ pwc.day.plot %>%

### Step1.3: aggregate daily value to monthly value

This step employs `DAY2MON()` function, which takes the daily output (a raster rack with 365/366 layers) from `HeatStress()` or `PWC()` as input, and outputs a raster rack of 12 layers with monthly mean.
This step employs `daily_to_monthly()` function, which takes the daily output (a raster rack with 365/366 layers) from `cal_heat_stress()` or `cal_pwc()` as input, and outputs a raster rack of 12 layers with monthly mean.

```{r Step1.3, warning=FALSE}
wbgt.sun.mon <- DAY2MON(input_rack = wbgt.sun.day)
wbgt.sun.mon <- daily_to_monthly(input_rack = wbgt.sun.day)
# rm(wbgt.sun.day)
pwc.hothaps.mon <- DAY2MON(input_rack = pwc.hothaps.day)
pwc.hothaps.mon <- daily_to_monthly(input_rack = pwc.hothaps.day)
# rm(pwc.hothaps.day)
wbgt.sun.mon
pwc.hothaps.mon
Expand Down Expand Up @@ -267,7 +267,7 @@ pwc.mon.plot %>%
scale_fill_gradient(low = "grey95", high = "red",
na.value = "white")+
labs(x = "longitude", y = "latitude", fill = "PWC \nloss(%)",
title = "Monthly heat-induced PWC loss for labor in rain-fed sunflower sector: 2024-07") +
title = "Monthly heat-induced PWC loss for labor in rain-fed maize sector: 2024-07") +
scale_x_continuous(
breaks = seq(-180, 180, by = 60), # Longitude breaks
labels = format_longitude # Apply custom labels
Expand All @@ -281,12 +281,12 @@ pwc.mon.plot %>%

### Step1.4: aggregate monthly value to annual value

This step employs `MON2ANN()` function, which takes a raster rack with 12 layers of monthly mean as input, and outputs a data frame with grid-level annual mean. In this step, make sure the `SECTOR` input in `MON2ANN()` is identical to the `SECTOR` input in `HeatStress()`.
This step employs `monthly_to_annual()` function, which takes a raster rack with 12 layers of monthly mean as input, and outputs a data frame with grid-level annual mean. In this step, make sure the `SECTOR` input in `monthly_to_annual()` is identical to the `SECTOR` input in `cal_heat_stress()`.


```{r Step1.4, warning=FALSE}
wbgt.sun.ann <- MON2ANN(input_rack = wbgt.sun.mon, SECTOR = "SUNF_R")
pwc.hothaps.ann <- MON2ANN(input_rack = pwc.hothaps.mon, SECTOR = "SUNF_R")
wbgt.sun.ann <- monthly_to_annual(input_rack = wbgt.sun.mon, SECTOR = "MAIZ_R")
pwc.hothaps.ann <- monthly_to_annual(input_rack = pwc.hothaps.mon, SECTOR = "MAIZ_R")
summary(wbgt.sun.ann)
summary(pwc.hothaps.ann)
Expand Down Expand Up @@ -315,7 +315,7 @@ pwc.hothaps.ann %>%
scale_fill_gradient(low = "grey95", high = "red",
na.value = "white")+
labs(x = "longitude", y = "latitude", fill = "PWC \nloss(%)",
title = "Annual heat-induced PWC loss for labor in rain-fed sunflower sector: 2024") +
title = "Annual heat-induced PWC loss for labor in rain-fed maize sector: 2024") +
scale_x_continuous(
breaks = seq(-180, 180, by = 60), # Longitude breaks
labels = format_longitude # Apply custom labels
Expand All @@ -331,10 +331,10 @@ pwc.hothaps.ann %>%

### Step2.1: calculate monthly gridded heat stress

Similar to *Step1.1*, this steps employs`HeatStress()` function. Example below takes monthly *hurs* (relative humidity), *tas* (near-surface air temperature), and *rsds* (surface downwelling shortwave radiation) and *WBGT_ESI* to calculate heat stress levels.
Similar to *Step1.1*, this steps employs`cal_heat_stress()` function. Example below takes monthly *hurs* (relative humidity), *tas* (near-surface air temperature), and *rsds* (surface downwelling shortwave radiation) and *WBGT_ESI* to calculate heat stress levels.

```{r Step2.1, warning=FALSE}
esi.mon <- HeatStress(TempRes = "month", SECTOR = "SUNF_R", HS = WBGT_ESI, YEAR_INPUT = 2024,
esi.mon <- cal_heat_stress(TempRes = "month", SECTOR = "MAIZ_R", HS = WBGT_ESI, YEAR_INPUT = 2024,
"../HELPS_Example_Data/hurs_example_month.nc",
"../HELPS_Example_Data/tas_example_month.nc",
"../HELPS_Example_Data/rsds_example_month.nc")
Expand All @@ -343,54 +343,54 @@ esi.mon

### Step2.2: calculate daily gridded physical work capacity

Similar to *Step1.2*, this step employs `PWC()` function to translate heat stress to PWC. Here we demonstrate the uncertainty in PWC introduced by alternative choices of `LHR()`.
Similar to *Step1.2*, this step employs `cal_pwc()` function to translate heat stress to PWC. Here we demonstrate the uncertainty in PWC introduced by alternative choices of `LHR()`.

```{r Step2.2.1, warning=FALSE}
start_t = Sys.time()
pwc.mon.foster <- PWC(WBGT = esi.mon, LHR = LHR_Foster, workload = "high")
pwc.mon.foster <- cal_pwc(WBGT = esi.mon, LHR = LHR_Foster, workload = "high")
end_t = Sys.time()
end_t - start_t
pwc.mon.foster
```

```{r Step2.2.2, warning=FALSE}
start_t = Sys.time()
pwc.mon.hothaps <- PWC(WBGT = esi.mon, LHR = LHR_Hothaps, workload = "high")
pwc.mon.hothaps <- cal_pwc(WBGT = esi.mon, LHR = LHR_Hothaps, workload = "high")
end_t = Sys.time()
end_t - start_t
pwc.mon.hothaps
```
```{r Step2.2.3, warning=FALSE}
start_t = Sys.time()
pwc.mon.niosh <- PWC(WBGT = esi.mon, LHR = LHR_NIOSH, workload = "high")
pwc.mon.niosh <- cal_pwc(WBGT = esi.mon, LHR = LHR_NIOSH, workload = "high")
end_t = Sys.time()
end_t - start_t
pwc.mon.niosh
```

```{r Step2.2.4, warning=FALSE}
start_t = Sys.time()
pwc.mon.iso <- PWC(WBGT = esi.mon, LHR = LHR_ISO, workload = "high")
pwc.mon.iso <- cal_pwc(WBGT = esi.mon, LHR = LHR_ISO, workload = "high")
end_t = Sys.time()
end_t - start_t
pwc.mon.iso
```
### Step2.3: aggregate monthly value to annual value

Same as *Step1.4*, make sure the `SECTOR` input in `MON2ANN()` is identical to the `SECTOR` input in `HeatStress()`.
Same as *Step1.4*, make sure the `SECTOR` input in `monthly_to_annual()` is identical to the `SECTOR` input in `cal_heat_stress()`.

```{r Step2.3, warning=FALSE}
pwc.hothaps.ann2 <- MON2ANN(input_rack = pwc.mon.hothaps, SECTOR = "SUNF_R")
pwc.hothaps.ann2 <- monthly_to_annual(input_rack = pwc.mon.hothaps, SECTOR = "MAIZ_R")
summary(pwc.hothaps.ann2)
```

## 3. Spatial aggregation: grid to region

```{r, warning=FALSE}
# aggregate grid level PWC to country level
ctry_pwc <- G2R(grid_annual_value = pwc.hothaps.ann, SECTOR = "SUNF_R", rast_boundary = country_raster)
ctry_pwc <- grid_to_region(grid_annual_value = pwc.hothaps.ann, SECTOR = "MAIZ_R", rast_boundary = country_raster)
# aggregate grid level PWC to GCAM water basin level
glu_pwc <- G2R(grid_annual_value = pwc.hothaps.ann, SECTOR = "SUNF_R", rast_boundary = reg_WB_raster)
glu_pwc <- grid_to_region(grid_annual_value = pwc.hothaps.ann, SECTOR = "MAIZ_R", rast_boundary = reg_WB_raster)
```

Expand All @@ -408,7 +408,7 @@ df.plot %>% ggplot() +
pattern_spacing = 0.02) +
scale_fill_gradient(low = "grey95", high = "red")+
labs(x = "longitude", y = "latitude", fill = "PWC \nloss(%)",
title = "Country-level heat-induced PWC loss for labor in rain-fed sunflower sector: 2024") +
title = "Country-level heat-induced PWC loss for labor in rain-fed maize sector: 2024") +
theme_bw()
```

Expand All @@ -425,6 +425,6 @@ df.plot %>%
pattern_spacing = 0.02) +
scale_fill_gradient(low = "grey95", high = "red")+
labs(x = "longitude", y = "latitude", fill = "PWC \nloss(%)",
title = "Water basin-level heat-induced PWC loss for labor in rain-fed sunflower sector: 2024") +
title = "Water basin-level heat-induced PWC loss for labor in rain-fed maize sector: 2024") +
theme_bw()
```

0 comments on commit d5886e6

Please sign in to comment.