Skip to contents

Read LPJmL irrigated and rainfed crop yields per cell.

Source: R/lpjml_regime_yield.R
read_lpjml_regime_yield.Rd

Derives the harvest per unit crop area of the rainfed and the irrigated stand of each crop in each 0.5-degree cell and year from a finished LPJmL run, and maps the LPJmL crop functional types (CFTs) onto WHEP production items. It is the yield ratio that weights the rainfed/irrigated split of a crop row's synthetic nitrogen, production nitrogen and residue removals toward the more productive hectares.

The harvest is pft_harvestc (harvested carbon excluding residues), which LPJmL already writes per square metre of the band's own stand; it is not divided by the stand fraction again. The evidence is in the header of R/lpjml_regime_yield.R: weighted by cftfrac, the bands reproduce the cell total in harvestc.nc exactly on grassland-only cells, while their plain sum overshoots it 1.5 to 26 times.

Usage

read_lpjml_regime_yield(
  years = NULL,
  run_dir = NULL,
  data = NULL,
  include_others = FALSE,
  example = FALSE
)

Source

LPJmL 6.1.1 (Potsdam Institute for Climate Impact Research), fork lbm364dl/LPJmL at git hash e6e6c42b88354f4a0df945e55f2ccf3c4a3cb605 (as stamped in the output metadata), run global_1750-2023_spinup_300_our_inputs_lpjml611_preindustrial_v2, outputs pft_harvestc.nc and cftfrac.nc.

Arguments

years

Optional integer vector of calendar years. NULL reads every year the run's pft_harvestc.nc carries. A year the run does not cover aborts, naming the coverage it has.

run_dir

Path to a finished LPJmL run output directory holding pft_harvestc.nc and cftfrac.nc. NULL (default) uses WHEP_LPJML_RUN_DIR, and with neither set reads the pinned lpjml-crop-regime-yield layer, built from one LPJmL run together with the other LPJmL-derived pins. A run that is named but lacks either file aborts rather than falling back to the pin.

data

Optional list used in place of reading a run, for testing: harvestc (as read_lpjml_npp() returns it for "harvestc") and stand_frac (as read_lpjml_hydrology() returns it for "stand_frac").

include_others

If TRUE, also return LPJmL's "others" catch-all stand, expanded to the items cft_mapping puts on it. Defaults to FALSE, the crop-specific stands only.

example

If TRUE, return a small fixture instead of reading a run. Defaults to FALSE.

Value

A tibble with one row per cell, production item and year:

  • lon, lat: cell centre, 0.5-degree grid.

  • year: calendar year.

  • item_prod_code, item_cbs_code: the WHEP production item and its commodity balance item.

  • lpjml_crop: the LPJmL CFT the item's yield comes from.

  • yield_rainfed, yield_irrigated: harvested carbon, excluding residues, in grams of carbon per square metre of that regime's own stand per year; NA where the regime has no stand in the cell.

  • method_regime_yield: how the yields were obtained (see Years before 1901).

Which crops carry a yield

Only production items mapped in cft_mapping to one of LPJmL's twelve crop-specific CFTs (temperate and tropical cereals, rice, maize, pulses, temperate and tropical roots, sugarcane, and the soybean, groundnut, sunflower and rapeseed oil crops) get rows: 40 items. The other 113 mapped items sit on LPJmL's "others" catch-all stand (fruits, vegetables, nuts, fibres, stimulants, oil palm, cotton...), whose yield is that of a composite rather than of the crop, and items absent from cft_mapping (fodder crops among them) have no band at all. By default neither gets a row, so a join against this layer leaves them without a yield, and the caller has to decide what to use instead.

include_others = TRUE adds the "others" stand as a crop of its own, expanded to the items cft_mapping puts on it. Its yield is the composite stand's, not the item's. The regime yield ratio uses it as the year-to-year anomaly source for crops without a crop-specific CFT, where only the ratio of the two regimes' yields matters; see build_regime_yield_ratio().

The natural key is item_prod_code: each production item maps to one CFT, while four item_cbs_codes mix CFTs across their production items (2520 "Cereals, Other", temperate and tropical cereals; 2534 "Roots, Other", tropical roots and an unmapped item; 2537 "Sugar beet", temperate roots and an unmapped item; 2605 "Vegetables, Other", where only green maize has a band). item_cbs_code is carried for joining, not as a key.

Zero-area stands

A regime whose stand fraction is zero in a cell has no harvest to divide, so its yield is NA, not zero. A cell-crop-year gets a row when at least one of the two stands has area. A stand with area and no harvest (a crop that failed or never matured) keeps its real zero.

Years before 1901

The run's climate forcing starts in 1901, and LPJmL recycles the 1901-1930 window for 1750-1900 while land use and CO2 stay historical. Those years are stamped "lpjml_band_harvest_recycled_climate" in method_regime_yield; later years "lpjml_band_harvest".

Rainfed rice

LPJmL waters paddy rice on the rainfed stand too (see read_lpjml_hydrology()), so rainfed rice yields are paddy yields and the rice irrigated:rainfed ratio sits closer to one than for an unwatered upland stand.

Examples

read_lpjml_regime_yield(example = TRUE)
#> # A tibble: 6 × 9
#>      lon   lat  year item_prod_code item_cbs_code lpjml_crop       yield_rainfed
#>    <dbl> <dbl> <int>          <int>         <int> <chr>                    <dbl>
#> 1   80.8  21.8  2010             56          2514 maize                     85.5
#> 2  -93.8  33.2  2010             15          2511 temperate cerea…         283. 
#> 3  113.   35.2  2010            236          2555 oil crops soybe…          11.2
#> 4  107.   39.2  2010             27          2807 rice                      70.8
#> 5 -100.   36.2  2010             15          2511 temperate cerea…         148. 
#> 6   69.2  23.8  2010             15          2511 temperate cerea…          15.5
#> # ℹ 2 more variables: yield_irrigated <dbl>, method_regime_yield <chr>