Vitali

A national pedoclimatic characterization of cropping schemes based on FADN Italian data

Giuliano VITALI 1± – Davide RIZZO 2* – Guido BALDONI 1 – Guido Maria BAZZANI 3– Concetta CARDILLO 4 – Maurizio CANAVARI 1

1 DipSA, University of Bologna (ITA)
2 Unité INTERACT, UniLaSalle, Campus de Beauvais (FRA)
3 CNR-IBIMET, Bologna (ITA)
4 CREA, Roma(ITA)

* Speaker: Davide Rizzo
± Corresponding author: Giuliano Vitali, giuliano.vitali@unibo.it

Introduction

Agriculture is the human activity that uses and transforms most of the Earth’s surface. The intensification of farming practices and the reduction of species grown occurred in the past decades is compromising sustainability in all of its meanings. In this regard, it is crucial to scale up the agronomic research from the plot/farm level to the landscape level (Benoit, Rizzo et al., 2012). This challenge is taking benefits from the recent advances in remote sensing that are achieving increasingly detailed crop maps. However, the effects of farming on the environment may take several years to be detectable, so research is needed to analyse the temporal evolution of crop patterns. Methods to map crop sequences can significantly improve the spatially explicit understanding of the interactions between farming practices and natural resources. Another contribution to understanding could come from analysing data collected for different purposes, but rich of information, as the Farm Accountancy Data Network (FADN) a source that could provide relevant insights on the farm management of yearly crop choices.

Materials and methods

The FADN was established by the European Community in 1965 (Regulation EC No 79/65). The purpose of FADN database is monitoring and evaluating the effects of the Common Agricultural Policies (CAP) on agricultural dynamics. It is fed by an annual survey of a farm sample representative of national agricultural situations. Altogether, FADN is delivering a detailed snapshot of farm structure, collecting yearly economic and technical information. An additional advantage is its availability for whole Europe. In each Country, a national agency is mandate to perform the survey that is composed by two parts: (i) a general set of variables that remains the same across Europe; (ii) a national specific part. The National Institute of Agricultural Economics (INEA) is the Italian agency that feeds the FADN database (DPR n. 1708/65. In Italy, FADN sample can be connected with the agricultural general census universe, performed every 10 years by ISTAT, so to improve its representativeness.

We faced three major constraints of FADN. (1) First, the yearly sample is random within the predefined statistical layers, because all farms belonging to the same layer are assumed to be equivalent. Hence, the list of sampled farms changes every year. Therefore, farms are sampled just once or more times (even not sequential) with no strictly plan. (2) Second, climate characterization is very poor in FADN, as it is limited to the altitude and few geomorphological classes (plain, slope, and mountain). To explore the relevant role of climatic features as driving factor of local cropping schemes, a fit of Pavari’s phyto-climatic zoning and Tomaselli’s biomes has been developed and included to RICA that is based on five climate classes (Vitali et al., 2012). (3) Third, RICA provides only partial data about crop sequences (i.e., timely ordered set of crops). A rational way to conceive a crop sequence is to define a pattern to be repeated over a given number of years (i.e., crop rotation) then to divide the arable surface into integer ratios, where integers reflects the number of time steps (years) the crop is grown during the cycle (cf. Dury et al., 2012). As the FADN dataset reports the surface allocated to the various crops within each farm, our analysis was oriented to understand if such partitioning, though with no details about the location of the fields, could reflect the presence of rotations. To the scope, we introduced the concept of “cropping scheme” to deal with FADN data that do not deliver neither with a spatially explicit distribution (crop pattern), nor the sequence adopted to grow the crops (cropping plan). To make the analysis feasible, we regrouped the initial 132 FADN crops into 5 classes. Altogether, the cropping schemes are described by a code based on the crop groups and their proportion on the total surface.

Early results and perspectives

Earlier results of the study are illustrated in Table 1. Agronomist can build upon these patterns with fieldwork on farming practices. Beyond the agronomic domain, the spatial segmentation based on cropping systems can enhance some existing agroecosystem zoning, so far implemented on single-year databases. Schemes emerging from the following study are already included in the framework of a model, based on mathematical programming, designed to optimise a holding structure under varying exogenous conditions: optimisation mainly focuses on arable land, looking for the rotation more suitable for the existing combinations of climate and terrain (Vitali et al., 2015).

Table 1. Quantitative description of the Italian FADN dataset (=RICA) subsample on the total 2008-2014 population per classification criteria. The last column shows an example of analysis of the associated crop schemes.

Classification criteria Consecutive years of observation Total  36_fig1
3 4 5 6 7
Climate
Zone 1 270 298 410 186 217 1381
Zone 2 134 173 107 132 39 585
Zone 3 940 898 841 861 1645 5185
Zone 4 47 45 45 29 118 284
Zone 5 4 4 7 15
Zone 6 5 8 8 8 15 44
Geomorphology
Plain 803 758 874 644 1350 4429
Gentle slope 547 595 476 499 637 2754
Slope 49 73 61 73 53 309
Steep slope 1 1 2
Italy (total) 1400 1426 1411 1216 2041 7494

References

Benoît M, Rizzo D, Marraccini E, et al. (2012) Landscape agronomy: a new field for addressing agricultural landscape dynamics. Landscape Ecology 10:1385–1394. doi: 10.1007/s10980-012-9802-8

Dury J, Schaller N, Garcia F, et al. (2011) Models to support cropping plan and crop rotation decisions. A review. Agronomy Sust Developm. doi: 10.1007/s13593-011-0037-x

Vitali G, Cardillo C, Albertazzi S, et al. (2012) Classification of Italian Farms in the FADN Database Combining Climate and Structural Information. Cartographica: The International Journal for Geographic Information and Geovisualization 47:228–236. doi: 10.3138/carto.47.4.1478

Advertisement