Skip to contents

Annual HSOC trajectory (as in an earlier regional historical reconstruction): a fresh and a humus decomposing pool plus an inert organic matter pool. The inert pool is the Falloon (1998) function of initial carbon and, as in that paper, is a component of the measured stock rather than an addition to it, so the two decomposing pools open on the remainder initial_soc_mgc_ha - iom, split between them in the proportion of their steady states input_pool / k_pool. Each year a pool stock loses first-order decomposition and gains its carbon input, so the trajectory relaxes from the supplied stock toward that steady state. Land-use-change carbon transfer is deferred to a later phase (single land use here).

Usage

calculate_soc_hsoc(
  initial_soc_mgc_ha,
  c_input_mgc_ha_yr,
  years,
  clay_pct = NA,
  climate_modifier = 1,
  humification_fraction = 0.3
)

Source

Coleman, K. & Jenkinson, D. S. (1996). doi:10.1007/978-3-642-61094-3_17 ; inert organic matter: Falloon, P. et al. (1998). doi:10.1016/S0038-0717(97)00256-3 .

Arguments

initial_soc_mgc_ha

Initial soil organic carbon stock (Mg C per ha). The trajectory starts here: year 0 of the returned tibble reports this stock, as it does for the four sibling models.

c_input_mgc_ha_yr

Annual carbon input (Mg C per ha per year).

years

Number of years to simulate.

clay_pct

Soil clay content (percent). Scales humification_fraction by the Aguilera et al. (2018) Eq. 5-6 texture modifier, which runs 0.72 at 5% clay to 1.13 at 60% and is 1 at RothC's Rothamsted reference of 23.4%: coarse soils stabilise less of the same carbon input. NA, the default, means no texture information was supplied and applies no texture adjustment. The gridded balance always supplies clay, so this function and build_carbon_balance(model = "hsoc") now return the same stock for the same inputs.

climate_modifier

Annual climate rate modifier (dimensionless).

humification_fraction

Fraction of carbon input humified into the humus pool (the remainder feeds the fresh pool).

Value

A tibble with one row per year: year, fresh, humus, iom and soc_total.

Examples

calculate_soc_hsoc(
  initial_soc_mgc_ha = 50,
  c_input_mgc_ha_yr = 2,
  years = 5
)
#> # A tibble: 6 × 5
#>    year fresh humus   iom soc_total
#>   <int> <dbl> <dbl> <dbl>     <dbl>
#> 1     0  4.06  41.7  4.22      50  
#> 2     1  3.51  41.5  4.22      49.2
#> 3     2  3.22  41.3  4.22      48.7
#> 4     3  3.08  41.0  4.22      48.3
#> 5     4  3.00  40.8  4.22      48.0
#> 6     5  2.96  40.6  4.22      47.8