Litter fall in forest, harvest left overs and other plant residues is a source of organic material to HYPE. The plant residues increase the immobile pool of organic carbon (OC) in the form of fastC in top two layers in soil. Thus the OC model does not use the fractionation of plant material (resfast) into fast and humus soil pools as is done for nitrogen and phosphorus in plant residues. The organic carbon addition by plant residues is defined based on crop/vegetation. Input, resc (kg/ha/yr), gives a daily supplement to the pool during the number of days determined by parameter litterdays.

The initial size of organic carbon pools in the soil is dependent on land use and determined by the user. The parameters (humusc1, humusc2, humusc3, fastc1, fastc2, fastc3) give immobile OC content of the three soil layers. The unit for these parameter values is mg/m3. With this information and soil layer thickness, the size of the pools in the layers are calculated. The model works with pools of the unit kg/km2. In addition an initial concentration of dissolved organic carbon in the soil water of different land uses may be specified (occonc0). The amount of DOC in the soil water is below called the OCpool.

Many soil processes depend on temperature and soil moisture. They use the same common functions as nitrogen and phosphorus. The organic carbon soil transformations (both production of humusC from fastC, and turnover of fastC and turnover of humusC) use the soil moisture function with values partly given by the user instead of the given coefficients of the equation as is used used for nitrogen and phosphorus processes. The coefficient is replaced by the land-use dependent parameter ocsoimslp (), and the coefficient satact is replaced by land-use dependent parameter ocsoimsat. The coefficients and keep their values (i.e. 0.12 and 1.0). The percolation reduction of OC uses all the given coefficients, and is not affected by the parameters.

Figure 2: Organic carbon processes in upper soil layer. The transformation modelled are shown by hollow green arrows. The mineralization is a fraction of the organic carbon transformed between the pools, as is shown by the little perpendicular arrow. humusC is here denoted slowC

Some of the fastC of soil is converted into humusC. For HYPE this means that fastC (the pool where plant residues were added) is transformed to humusC in the uppermost soil layer. In Figure 2 the pool of humusC is denoted slowC because of its slower transformation rates.

The other soil layers (k) also have a transition from fastC to humusC. The loss of fastC does not all end up in the humusC pool, but a proportion (parameter minc) is mineralized in the process. The transformation (fasttohumus, mg/m2/d) depends on soil moisture (smfcn) and temperature (tmpfcn), amount of OC in the pool (fastC) and a vegetation dependent parameter klh.

Turnover of fastC is a sink for fastC and a source of dissolved OC in soil water (DOC) in all soil layers (k = 1-3). The loss of fastC does not all go to the OCpool, but a proportion (parameter minc) is mineralized. Turnover (transfC, mg/m2/d) depends on a general parameter (klo), the temperature function (tempfcn), humidity function (smfcn) and the pool of fastC (fastC).

In dry conditions a transfer in the opposite direction can also occur. The transformation of DOC to fastC is a decrease of DOC and a source of fastC in all soil layers (k = 1-3). The loss of DOC is not all ending up in fastC but a proportion (parameter minc) is mineralized. Turnover (doctofast, mg/m2/d) depends on a general parameter (kof) and the pool of DOC (OCpool). The transfer is limited that the soil layer temperature must be less than 5 °C, the soil moisture (sm) must be less than field capacity and moisture function (smfcn) must be less than the parameter koflim.

Turnover of humusC is a sink for humusC and a source of DOC in all soil layers (k = 1.3). The turnover rate of humusC is lower than that of fastC, why it is sometimes called slowC (e.g. in Figure 2). The loss of humusC does not all go to the soil water OC, but a proportion (parameter minc) is mineralized. Turnover (transhC, mg/m2/d) depends on a general parameter (kho), temperature function (tempfcn), humidity function (smfcn) and the pool of humusC (humusC).

Organic carbon is lost from the soil water as it percolates down through the soil layers and where it is dissipated to become a regional groundwater flow. The decrease in concentration of percolating water during transport between the soil layers depends on soil moisture and temperature and a calibration parameter.

The soil moisture function and temperature function are the general functions described for soil processes. Percolation uses the nitrogen and phosphorus coefficients for the soil moisture function, not the parameters that the OC transformations uses. The parameter, par in the equation, is kcgwreg for regional groundwater flow formation and koc for percolation between soil layers. Both are general parameters.

The OC transported by surface and groundwater runoff and tile drainage (runoffC) is collected in a temporary storage pool (relpool (kg/km2)).

From the temporary pool organic carbon is released (release (kg/km2)) and then transported to the local river depending on the size of the total runoff (runoff (mm)). The parameters ocfldelx and ocfldele are general parameters.

The released OC give the current OC concentration of runoff.

Runoff from soil may flows through a riparian zone before it reaches the local river. Surface runoff and drainage water from drainage pipes reaches the local river without passing through the riparian zone. In the riparian zone the levels of OC are affected, while flows remain unchanged. The change in OC depends on soil temperature, class altitude (elev (in masl)), the water table (gwat) and its recent change, season and soil moisture (sm). The runoff concentration (conc(i)) of each soillayer (k) increases with the factor:

The temperature function (tmpfcn) is the usual of soil processes (see above). The following equations describe the other process functions:

The activation of riparian zone processes is based on land use. It is primarily thought to act on forest runoff. The land use dependent parameter ripz determines the overall level of increase in concentration in the riparian zone, and if set to zero no riparian zone processes are used. In addition, two general parameters can influence the effect of the riparian processes; ripe which determines the groundwater level dependence, and rips which determines the seasonal influence. Season division is determined by ten-day and twenty-day averages of air temperature (T10, T20). Autumn is defined as the period when T10 is less than T20. The soil moisture function is different for an increasing (rising) and sinking ground water table (Figure 3). It uses coefficients , and saturation (satact = 0.6). It depends on the soil moisture of all layers together (sm) and the water-holding capacity parameters; wp - wilting border and pw - total pore volume, in fractions of total soil layer thickness (d).

Figure 3: Example of riparian zone soil moisture function, and the dependence on changes in the groundwater levels.

The initial organic carbon concentration in rivers are assumed to be zero, while the lakes' concentration are set by the user. The parameter, tocmean, is lakeregion dependent, but can also be set for each lake separately.

Primary production is a source of organic carbon in rivers and lakes, while mineralization is a sink. Primary production and mineralization are calculated the same way as for nitrogen, but with its own calibration parameter (wprodc). The equations are repeated below. The production/mineralization depend on temperature and total phosphorus and lake area (area). The potential carbon transformation (minprodCpot, kg / day) is proportional to the potential nitrogen transformation (minprodNpot, see also NP section) with a transformation rate that depends on the carbon-nitrogen ratio (NCratio = 5.7). The calculated mineralization of organic carbon is limited to a maximum of 50% of the available OC pool. The phosphorus dependence is based on a half-saturation equation using a long-term average of total phosphorus. If phosphorus is not modelled by HYPE, a lake region dependent parameter (tpmean) is used instead. The long-term average concentration of phosphorus is reduced by the general parameter limsedPP before using it in the concentration function. The water depth (depth) is the lake depth, and for the river the depth calculated above.

Sedimentation in lakes is a sink for OC and works the same way as for organic nitrogen and particulate phosphorus. Thus sedimentation (, kg/day) is calculated as a function of OC concentration in lake water (conc)) and lake area (area). The settling velocity parameter sedoc is general or can be specified for each lake.