The resulting databases serve as the new soil information for the Mars Crop Yield Forecasting System. Due to the new SGDBE version and required better geographical coverage, soil information used within the CGMS had to be updated. The work consisted in:
- generation of soil parameters with FAO soil map;
- rewriting the Pedotransfer Rules for salinity, alkalinity, chemical toxicity and drainage according to the new nomenclature;
- redefinition of new Soil Physical Groups and their parameters;
- redefinition of new Rooting Depth classes;
- re-evaluation of CGMS Suitability Criteria;
- new evaluation of Soil Suitability Criteria.
New soil parameters with the FAO soil map have been generated for specific countries. This was necessary as not all countries are covered by the SGDBE, but information is needed in the frame of the MCYFS. The attributes SOIL, SLOPE, TEXT and PHASE/AGLIM were generated using additional databases such as the 1:5 mil Soil Map of World.
The pedotransfer rules for salinity, alkalinity, chemical toxicity and drainage have been adapted to the WRB standard and were applied to the SMU database to retrieve the information. These attributes are needed amongst others to determine the soil suitability for the crop growth simulation. Also a new rule for estimating rooting depth has been established by using the values taken by the different attributes and the correlations established within the soil name and the ROO attribute.
The redefinition of new soil physical groups and their parameters was performed as well. In the CGMS, soil physical groups were introduced per STU to determine the available water capacity (AWC) of a soil. Adaptation and extension of the existing rules was necessary because of the other soil classifications in the database. Rules have been defined now in a more general way so that they can handle all possible values that occur for the different attributes. Adaptation of the rules led to different available water capacity values than previously used within the CGMS and to a larger spatial differentiation. Overall the AWC value has increased. The largest differences are due to the recognition that organic soils have very high AWC values. The second largest effect is that in subsoils of secondary chalk and loess, the sponge effect of the subsoil, leading to capillary rise of moisture during the growing season, has been translated into a much higher AWC value (an additional amount of about 100 mm of water). The soil physical group is defined by the volumetric soil moisture content at wilting point, field capacity and saturation. To link the AWC values produced by the CERU32 program to soil physical groups the volumetric soil moisture content of wilting point given by the HYPRES rules (see 5.2.3) was used. The volumetric soil moisture content at field capacity was calculated by adding up the AWC value. This procedure led to 743 unique soil physical groups instead of the former 7 groups.
Based on the re-written PTRs and the application of the former suitability criteria to the SGDBE v.4 the suitability criteria have been evaluated and were re-defined. The assessment of yield level at national and sub national scale with CGMS requires information on locations and soils where the major crops are grown. This information is especially important for the spatial aggregation from simulated crop indicators from simulation units to administrative regions. Most logically this information could be obtained by combining a European land use map and the European soil map. However the main constraint for application of this procedure is formed by the lack of a harmonized land use map at the level of major crops. The solution found to overcome these constraints was to derive from the soil map a new map of plausible land use based on soil suitability rules. A rule is defined by listing the unsuited classes of soil parameters, or by specifying a threshold value for a given soil parameter. The parameters used to assess the soil suitability were rooting depth, slope, and volume of stones, texture, drainage, salinity and sodicity (alkalinity). The selection criteria are defined for each parameter as the range of unsuitable classes.
It was the aim to define a set of soil suitability rules for each major crop group, but the analysis revealed that this is not realistic when using the SGDBE v4. The conclusion is that the application of uniform soil unsuitability criteria across Europe is possible only for the broad selection of the least productive soils on the basis of extreme adverse soil conditions. Any refinement in criteria in order to focus on the soils of the most important centres of agricultural production leads to inconsistent selections of soils across Europe, due to a lack of uniformity in the soil map of Europe between the countries. It is in particular not justified to apply crop group specific or even crop specific sets of unsuitability criteria, as the differences in criteria are rather small, while the precision of the soil indicators derived from the soil map is rather low. It is concluded that the use of European land cover data like CLC2000 and/or GLC2000 will be a more precise source of information than the results of soil suitability rules within the aggregation from EMU to NUTS level 2.