A novel method for mapping agricultural intensity reveals its spatial aggregation: Implications for conservation policies
Highlights
► We study the spatial distribution of agricultural intensity in France. ► A continuous, cross production type intensity indicator is developed. ► Its value is estimated at accurate spatial resolution. ► We show aggregation clusters within the two intensity extremes.
Introduction
The lack of spatial targeting has been identified as an important cause of the low effectiveness of agri-environmental schemes (AESs) for promoting biodiversity (Feehan et al., 2005, Whittingham, 2007). AESs are policies designed to encourage farmers of the European Union (EU) member states to protect and enhance the environment on their farmland. Horizontal schemes applied across one or several European countries do not always have similar effects on species among regions (Whittingham et al., 2007). Spatial targeting is expected to improve the cost-effectiveness of AESs, as applying conservation measures on the most suitable areas would provide environmental effects at lower costs than if conducted elsewhere (van der Horst, 2007, Piorr et al., 2009, Uthes et al., 2010). Two contrary types of spatially targeted policies have been suggested. On the one hand, policies could target areas with aggregation of extensively managed farmlands, which have been shown to host higher biodiversity levels (Kleijn and Sutherland, 2003) and provide more resources for the multiple species requirements (Whittingham, 2007). Zonal scheme targeting these regions would thereby reinforce the quality of areas with high biodiversity potential, which could be the most effective option (Feehan et al., 2005). On the other hand, policies could target areas with aggregation of intensively managed farmlands in order to reverse biodiversity decline in regions where it is the most severe (Primdahl et al., 2003). In all cases, however, policy options require data on farming intensity for large gradients and scales and at a resolution relevant for fine policy targeting, and these have been lacking up to now.
Many studies tackling the environmental impacts of agricultural intensity focused on a single component, such as nitrogen input (Billeter et al., 2008, Kleijn et al., 2009, Temme and Verburg, 2011) or pesticides (Boutin and Jobin, 1998). Other studies used indirect indicators of agricultural intensity such as yield (Donald et al., 2001) or the relative amount of arable fields (Ekroos et al., 2010). Few studies integrate the various components of agricultural intensity into a more complete indicator. Assessing several complex intensity variables requires a large amount of data. Farm surveys can be conducted (Herzog et al., 2006) although this would not be feasible on large scales.
Because of the above-mentioned difficulties, a dichotomous view of agricultural intensity prevails in studies addressing intensity distribution on a large spatial scale and at a high resolution. Several studies focus on the distribution of organic versus conventional farming (Gabriel et al., 2009 and Ilbery and Maye, 2011 in the UK, Frederiksen and Langer, 2004 in Denmark, Rundlöf and Smith, 2006 in Sweden). Even though organic farming seems to have a generally positive effect on species richness and abundance (Bengtsson et al., 2005), it is not the only agricultural management option for promoting biodiversity. Other studies focus on extensively managed farmlands, which are crucial for European biodiversity (Bignal and McCracken, 1996, Benton et al., 2002). For instance, the indicator of high nature value (HNV) characterizes and maps such low-input agricultural systems with high environmental qualities (Baldock et al., 1993, Beaufoy et al., 1994, Andersen et al., 2003, Pointereau et al., 2007). As for organic farming, it remains a dichotomous view of agriculture (HNV vs. non-HNV, organic vs. conventional) regarding its effects on biodiversity.
This dichotomous view of agricultural intensity is insufficient because biodiversity can display continuous response to a large intensity gradient (Kleijn et al., 2009, Ekroos et al., 2010). Studying the distribution of agricultural systems belonging to a large, continuous gradient of intensity is thus important. A few studies have addressed this question and partially overcome the issue of scale and resolution. Reidsma et al. (2006) developed an aggregated intensity indicator based on input costs, but due to poor availability of data, it described intensity at a low spatial resolution, unsuitable for the fine-tuning of public policies (i.e. NUTS 2). Temme and Verburg (2011) developed a method to estimate and map agricultural intensity with very high resolution (1 km2 grid) on the scale of Europe. However, their intensity measure was not continuous (two or three intensity classes for livestock and arable farming respectively).
The objective of this study was to map the spatial distribution of agricultural intensity for the whole of France with a spatial resolution that would be adequate for a better targeting and adaptation of conservation policies. Three steps were conducted to fulfill this objective. (i) We produced an intensity indicator relevant for the main agricultural production types in France and studied the distribution of the production types along the intensity gradient. (ii) We developed a method that relied on existing datasets to estimate the value of this indicator at the Small Agricultural Region (SAR) level compatible with the fine-tuning of conservation policies. (iii) We mapped the spatial distribution of our intensity indicator and tested for its spatial aggregation.
Section snippets
Data
Five datasets from year 2006 were combined to estimate an intensity indicator at the SAR level. All data were provided by the INRA service unit managing the French Observatory of Rural Development (ODR, 2011).
The first dataset was the French FADN (Farm Accountancy Data Network), the FADN follows the same methodology in the 27 countries of EU. It contains a very broad set of variables at the individual farm level. It provides a limited sample of farms surveyed on a yearly basis (n = 7361 farms in
Agricultural intensity and input categories distribution across production types
The characteristics of the seven groups computed by the cluster analysis of input cost structure are detailed in Table 2. It shows the distribution of the whole IC/ha intensity gradient across production types. The seven groups were distributed along a broad intensity gradient, group means ranging from 231.5 to 681.6 €/ha. Along this intensity gradient, groups dominated by livestock alternated with groups dominated by arable farming. Groups 1, 3, 5 and 7 were dominated by livestock and mixed
Discussion
This study describes the distribution of farming intensity on the scale of France with spatial resolution relevant for the policy targeting of homogeneous conditions in terms of agricultural systems, soil and climate. The intensity indicator, based on input costs, was relevant for the main production types. We showed strong spatial aggregation of low-input systems and also revealed some aggregation clusters of high-input systems.
Conclusions
Our method combines existing agricultural datasets to accurately estimate the distribution of agricultural intensity at higher resolution than directly available through the FADN data. We only accessed French data but similar datasets are available in other European countries where our method could be used. Moreover, we developed a single, continuous intensity indicator that is appropriate for all the main agricultural production types covering the territory. However, the intensity indicator
Acknowledgements
This work was carried out with the financial support of the “ANR – Agence Nationale de la Recherche – The French National Research Agency” under the “SYSTERRA program – Ecosystems and Sustainable Development”, project “NR-08-STRA-007, FARMBIRD – Coviability models of FARMing and BIRD biodiversity”.
References (55)
- et al.
Accuracy and cost of models predicting bird distribution in agricultural grasslands
Agriculture, Ecosystems & Environment
(2010) - et al.
Effects of an agri-environment scheme on farmland biodiversity in Ireland
Agriculture, Ecosystems & Environment
(2005) - et al.
Assessing the intensity of temperate European agriculture at the landscape scale
European Journal of Agronomy
(2006) - et al.
Integrated assessment of future CAP policies: land use changes, spatial patterns and targeting
Environmental Science & Policy
(2009) - et al.
Environmental effects of agri-environmental schemes in Western Europe
Journal of Environmental Management
(2003) - et al.
Impacts of land-use change on biodiversity: an assessment of agricultural biodiversity in the European Union
Agriculture, Ecosystems & Environment
(2006) Ecological impacts of arable intensification in Europe
Journal of Environmental Management
(2001)- et al.
Mapping and modelling of changes in agricultural intensity in Europe. Agriculture
Ecosystems & Environment
(2011) Assessing the efficiency gains of improved spatial targeting of policy interventions; the example of an agri-environmental scheme
Journal of Environmental Management
(2007)- et al.
The influence of non-governmental organisations on the creation of Natura 2000 during the European Policy process
Forest Policy and Economics
(2002)
Méthodologie Rica France Tableaux standard 2006
Agreste Chiffres et Données Agriculture
Contractualisation et diffusion spatiale des mesures agro-environnementales herbagères
Revue d’Etudes en Agriculture et Environnement
Farming and the environment in the European Union—using agricultural statistics to provide farm management indicators
Local Indicators of Spatial Association – LISA
Geographical Analysis
The effects of organic agriculture on biodiversity and abundance: a meta-analysis
Journal of Applied Ecology
Linking agricultural practice to insect and bird populations: a historical study over three decades
Journal of Applied Ecology
Low-intensity farming systems in the conservation of the countryside
The Journal of Applied Ecology
Indicators for biodiversity in agricultural landscapes: a pan-European study
Journal of Applied Ecology
High nitrogen costs of dairy production in Europe: worsened by intensification
Ambio
Intensity of agricultural practices and effects on adjacent habitats
Ecological Applications
Spatial Processes: Models & Applications
The R Book
Agricultural intensification and the collapse of Europe’s farmland bird populations
Proceedings of the Biological Sciences/The Royal Society
Homogenization of lepidopteran communities in intensively cultivated agricultural landscapes
Journal of Applied Ecology
Cited by (41)
Spatial patterns of county-level arable land productive-capacity and its coordination with land-use intensity in mainland China
2022, Agriculture, Ecosystems and EnvironmentLinking the human appropriation of net primary productivity-based indicators, input cost and high nature value to the dimensions of land-use intensity across French agricultural landscapes
2019, Agriculture, Ecosystems and EnvironmentCitation Excerpt :Studies focusing on inputs either evaluate the fertilizer use in general (Billeter et al., 2008; Dormann et al., 2007), nitrogen inputs (Kleijn et al., 2009), pesticides (Geiger et al., 2010) or more integrative inputs indicator for crops and livestock (Teillard et al., 2012). Among the indicators used, the Input-Cost per hectare (“IC/ha”; Teillard et al., 2012) evaluates the total input cost across French agricultural landscapes as the total expenses in fertilizers, feed products, pesticides, seeds, fuel, veterinary products and irrigation water. Studies focusing on outputs are rather based on food production of crops or livestock (Dross et al., 2018; Teillard et al., 2016); or the yield per time and unit area (Lambin et al., 2000).
Monitoring the Sustainable Intensification of Arable Agriculture: the Potential Role of Earth Observation
2019, International Journal of Applied Earth Observation and GeoinformationTrade-offs and synergies between livestock production and other ecosystem services
2019, Agricultural SystemsCitation Excerpt :Land cover fraction F is composed by five sub-categories, namely Broad Leaved Forest (CLC 311), Coniferous Forest (CLC 312), Mixed Forest (CLC 313), Sclerophyllous Vegetation (CLC 323), and Transitional Woodland-Shrub (CLC 324), respectively. The set of land-use-related production factors, U, is formed by the annual pesticide expense [€ ha−1 yr−1] per unit area of fodderland and cropland, pFOD and pC, representing farm intensity (Teillard et al., 2012), as well as the production factor k [Mcal], which represents the average crop energetic content in the cropland fraction. Such a production factor is a proxy of the crop associations in the different parts of the country.