Environmental Engineering Reference
An overview on the components, modules and calibration procedure of the
summary version of FSSIM-MP used for EU25 level analysis is described below:
Components: the selected components are: (i) the farm typology; (ii) the
simple computer-based survey; and (iii) the mathematical programming model
FSSIM-MP modules: the selected modules are the crops, premiums, PMP, trend,
policy and common modules.
Calibration procedure: The calibration procedure used in the detailed version of
FSSIM-MP is also used here with minor adjustments.
Tested Scenario: The model was calibrated for the base year and used to predict
changes in the baseline scenario. Sensitivity of crop product quantities to prices
changes was simulated to assess price supply relationships at higher levels.
The price of each crop product was changed iteratively to 60%, 80%, 120% and
140% of the original price keeping the other product prices constant. The effects
on supply were assessed in each iteration.
Running FSSIM for all farm types of the regions with summary information on
agro-management requires some adjustments of FSSIM to restrict the data requirements
to what is available in FADN and the simple survey. Those adjustments are:
1. In the irst phase of PMP the observed crop levels are used as upper bounds to
the added calibration constraints. In FADN there is no information on single
crop levels, instead there is information on groups of crops (e.g. fresh vegetables
which refers to the area of a number of crops such as area of onion, carrot and
cabbage). In the detailed version of FSSIM, expert knowledge is used to transform
the observed levels of FADN crop groups to observed levels of single crops.
Finding experts in all sampled regions would be a resource demanding process.
To avoid this process it was decided to evaluate and calibrate the reduced version
of FSSIM directly on FADN crop groups.
2. In some cases, the observed cropping pattern of some farm types included crops
that are not part of any rotation identiied in the simple survey. This implies that
it is not possible to simulate such crops. To avoid this problem it wad decided to
treat the area of these crops as ixed land and it was subtracted from the total
available farm land.
3. Finally, in order to ensure that there is at least a linear combination of activities
that reproduces the observed cropping pattern, we decided to include some
mono-crop activities, deined as rotations of a single crop; this is justiied because
this represents rented land (e.g. from dairy farms) on which speciic crops are
grown for 1 year.
An example of the sensitivity analysis of prices performed for one of the farm types
in Flevoland is presented in Table 5.6 . Note, that spring wheat substitutes winter
wheat when prices of spring wheat are high or those of winter wheat are low.
These farm level results are used to estimate price-supply relationships at regional
level and subsequently they will be extrapolated with advanced econometric
procedures (EXPAMOD) to non-sampled regions (Pérez Domínguez et al. 2009) .