A method to identify and map the land use patterns relevant for agro-environmental policies in agricultural landscapes

The association of natural and farmed areas is of primary importance for ecological networks in agricultural landscapes. The main goal of this work is to present a method to rapidly quantify and map the spatial configurations (SC) of the main landscape elements in farmed landscapes. Such land use patterns are relevant for agro-environmental policies since a better knowledge of main CS in landscapes is needed (Forman and Wilson, 1995; Benoit et al., 2012). Such CS  can support  an evaluation of the ecosystems services provided by agricultural landscapes in view of the “greening” measures of the next CAP (Lefevre et al., 2012) and locally the integration of ecological networks in planning of agricultural landscapes (Pinto-Correia et al., 2006).

The method we developed was in four steps: (1) To identify the main objectives of agro-environmental policies and of their relationship with local land uses; (2) to qualify the CS relevant for policies through field surveys, aerial photos observation, qualitative spatial representations, also in order to describe each CS by spatial indicators; (3) to test the relevance of the indicators describing CS using Pearson Chi-squared and Wilcoxon signed rank tests; (4) to map the spatial distribution of the identified CS in the analyzed agricultural landscape. We applied this method to a case study in Southern Tuscany, a Mediterranean inland hilly landscape of 1500 ha (Figure 1a).

Relevant agro-environmental policy objectives were the protection of surface water from nitrates, the protection of soil quality and the conservation of the cultural landscape diversity (Marraccini et al., 2009). With respect to these objectives, several land elements and uses (patch and linear networks) were described to have an influence on the identified objectives (Lefevre et al., 2012): agricultural land uses (arable crops 83% of the landscape surface, permanent crops 10%, fallows 2%), natural vegetation (woods, hedgerows, isolated trees), water bodies, human settlements. Starting from this preliminary information, we identified four CS in the studied area, differing for the different composition and arrangement of land uses. Arable crops (mainly winter wheat) were part of each CS, e.g. as main land use nearby rivers in CS1 or in a mosaic of human settlements and permanent crops in CS2. The description of the five CS and of the indicators identified is in Table 1. Selected indicators were proper to distinguish CS and were all statistically significant. The mapping of the CS showed a total coverage of the analyzed landscape, with some overlapping among different CS (Figure 1b shows an examples for CS1 and CS2).

The method reliability has been tested on another grassland oriented landscape in Central France showing a higher predictive value of agro-environmental services for Mediterranean landscapes characterized by an higher land use diversity (Marraccini, 2010).