The APSIM SoilP Module Description

SoilP Module Scope

SoilP is a representation of the availability of phosphorus in soil that provides an index for the soil's ability to supply P to crops that can be used in crop modules to modify growth processes under P limiting conditions.

The module is designed to handle inputs of fertiliser that are either immediately available or slow acting (e.g. rock phosphate), and the improved effectiveness of fertiliser due to placement effects.

To link SoilP with the MANURE model, the MANURE module handles the release of P from manure on the soil surface as a function of time and moisture content. This released P then becomes an input to the SoilP modules in the same manner that fertiliser P is an input. Similarly the corresponding release of N from manure is transferred to the appropriate pools in SoilN.

The dominant processes considered by SoilP are:

• loss of availability through time
• removal by crops
• addition by crop residues
• mineralisation / immobilisation of soil organic P.

SoilP is a multi-layer module that uses the same soil layer structure as SoilN. This approach has been adopted in order that there can be a full accounting of the organic P components in the system.

The availability of P in soil is dominantly influenced by the near-surface layers, and crop response to fertiliser is often adequately described by soil tests, P sorption etc of surface soil samples, i.e. without consideration of subsoils. In most situations, extractable P decreases with depth, P sorption increases, and root density decreases; all three factors contribute to the relative unimportance of the subsoil for P nutrition.

The multi-layer nature of the module requires assumptions to be made as to how P uptake by crops is to be partitioned between layers in the soil profile.

SoilP Module History

Within the CARMASAT Project, there is a need to develop a capability for modelling the nutritional effects of manure, and in particular to accommodate the effects of both nitrogen and phosphorus in manure on crop growth. In order that this can be achieved, an essential stepping stone is the incorporation into crop modules of routines that constrain growth under conditions of limiting P supply. This, in turn, requires a module that specifies the P supply from the soil, namely SoilP.

Citations

Barrow, N.J. (1974). The slow reactions between soil and anions. 1. Effects of time, temperature, and water content of a soil on the decrease in effectiveness of phosphate for plant growth. Soil Science 118, 380-386.

Probert, M.E. (1985). A conceptual model for initial and residual responses to phosphorus fertilizers. Fertilizer Research 6 , 131-8. [This paper contains the broad principles of how the availability of P is conceptualised in SOILP]

Probert, M.E. and Okalebo, J.R. (1992). Effects of phosphorus on the growth and development of maize. In “A search for strategies for sustainable dryland cropping in semi-arid eastern Kenya”. (Ed M.E. Probert). ACIAR Proceedings No 41. pp 55-62.

SoilP Module Structure

Figure 1. Schematic representation of the SoilP module showing the principal processes considered.

SoilP Module Components

Initialising SoilP

Inputs required by the SoilP module are:

Labile_p in each layer (ppm)

Sorption in each layer. These values are the total P sorbed at an equilibrium concentration of 1 ppm in solution, that is the value of a when the Freudlich equation is used to describe P sorption ( Psorbed = ac b ).

Residue_cp which is the C:P ratio of the initial above ground crop residues.

Root_cp which is the C:P ratio of the initial roots.

Rate_dissol_rock_p which is the rate of dissolution of any rock phosphate present at initialisation.

The unavailable_p, banded_p and rock_p pools on a layer basis are optional inputs. In the absence of any of values, banded_p and rock_p will be initialised to zero and the unavailable_p pool will be set such that the labile_p « unavailable_p system is at a steady state.

Currently labile_p is the only permitted input. In the future, input of extractable soil P (with options for different extractants) could be input and labile_p calculated.

Adding P as Fertiliser

The nature of the P source and how it is placed determines how it modelled:

    water soluble and broadcast P is added into labile_p,

    water soluble and placed P is added into banded_p,

    any non water soluble P (eg rock phosphate) is added into rock_p

Applied P fertiliser can be any combination of these three forms.

Processes of P in Soil

• Loss of availability with time . The rate for labile_p à unavail_P (and banded_p à unavail_p) is dependent on temperature and moisture (Barrow). The reverse process, unavail_p à labile_p will have a rate constant commensurate with the relative sizes of the two pools under steady state conditions.
• Dissolution of rock_p à labile_p. The rate coefficient depends on the source and the soil type (soil pH, calcium status, etc.). It is an input for any source/soil combination. A typical value is about 0.2 y -1 .
• Decrease in effect of placement with time , ie banded_p à labile_p. The effectiveness of placed P is assumed to decrease with time. Typically 50% of effect is lost in one year. A tillage event results in total transfer of banded_p to labile_p.
• Mineralisation of organic P sources à labile P. This is linked to decomposition of carbon from HUM, BIOM and FOM pools of SoilN module and surface residues in Residue module. SoilP assumes constant C:P ratios in BIOM and HUM but tracks C:P ratios of FOM and surface residues as crops add residues of varying P concentration.
• Removal of P by crop uptake . P uptake is taken from labile_p and banded_p in proportion to their contributions to effective_p (see below).
• Addition of P by crop death/senescence . P in above and below ground parts of the plant are returned to the Residue and FOM pools respectively.

Effective P in Soil

When P is banded, especially on soils of high sorptivity, it is more effective in increasing P uptake than a corresponding broadcast application. This effect is simulated in SoilP by placing a premium on banded-p.

Hence effective_p = labile_p + eff_band * banded_p

where eff_band is the relative effectiveness of banded_p (>1) and would depend on the band spacing and the sorptivity of the soil. Typically it is expected that eff_band has value of 2-3.

Module Interaction

An index of soil P status, as a function of effective_p, sorption, soil water, and rooting depth is made available to crop modules to determine whether the daily demand for P uptake can be met. SoilP receives back the actual P uptake for the day.

SoilP Module Parameterisation

SoilP Module Dependencies

SOILP uses an APSIM water balance module (eg SOILWAT) to determine the soil water status. The dynamics of P in organic matter provided by SOILP requires the soil nitrogen (SOILN) and residue (RESIDUE) modules for carbon cycling.

SoilP Module Outputs

SoilP Module Special Considerations

Simulation of crop growth under conditions of limiting P supply requires crop modules to provide the following:

•  inclusion of P stress factors (derived from P content of the crop or part of crop, optimum and minimum P concentrations which are functions of stage of growth) to restrict growth rates and modify phenological development and partitioning of biomass between tops and roots

•  calculation of daily P uptake by crop which is passed to SoilP so that P can be taken up from appropriate pools/layers

•  when crop residues (tops or roots) cease to be part of the plant and are added to the soil or surface residues (e.g. at harvest or due to senescence), their P content has to be returned, along with mass and N content, to maintain the P balance for the crop/soil system.

SoilP Module Validation

The qualitative response of SoilP in terms of decreasing availability with time, more rapid loss of availability at higher temperature, effect of P sorption on crop uptake (by maize) has been verified.

The only data set where SoilP and a modified Maize module (which includes routines to enable the crop to respond to P limitations) have been tested against observed data is for an experiment in Kenya described by Probert and Okalebo (1992).

SoilP Module Working Group

Dr Merv Probert is the leader.

Configuration Details

Owner Ivan Hills Modified by

Date created 24-Mar-97

Current version 1.2

Version history