SALADPLAN: a model of the decision-making process in lettuce and endive croppingMireille Navarretea and Marianne Le Bailb
a INRA Unité d'Écodéveloppement, Site Agroparc, 84914 Avignon Cedex 09, France
b INAPG, UMR SADAPT, 16 rue Claude-Bernard, 75231 Paris Cedex 05, France
(Accepted 22 February 2007; published online 11 July 2007)
Abstract - Market gardeners are facing new environmental, commercial and safety requirements and regulations that weigh heavily on their crop planting and management decisions. Producing high-quality products and simultaneously reducing the use of chemical products needs change in the multi-field planning of crops. In order to identify which cropping practice may be altered depending on each farm's characteristics, we built a rule-based model that represents how market gardeners plan cropping cycles of lettuce and endive, here denoted as salad vegetables, on a farm scale. The model is adapted to the specificities of salad vegetable crops in the Mediterranean region, in particular the combination of several cycles per year and of open field and sheltered crops. The output variables are the developed areas allocated to the different salad types and the range of the harvest dates, which are two key factors for commercialisation. The model is based on identifying the successive decision steps of growers when building their farming systems; descriptive variables are the effective surface area and period for salad cultivation, the number of salad cycles per year, the developed surface area of each salad type and the dates of harvest. The decision-making rules leading to each descriptive variable are based on the agronomic characteristics of soil, climate and crops, farmers' strategic objectives and organisation of farm resources. The cropping decisions were recorded in 38 farms in the large market-gardening region of Languedoc-Roussillon. In the sample, the number of salad types cultivated per farm varied from 1 to 8 and the harvesting period from 3 to 7 months. Our results demonstrate that the model can explain a substantial part of this variability, 26 and 28 farms among 38, respectively, being correctly classified by the model for the two variables. The original result of this work is to enhance the generality of a similar framework built earlier for arable cropping systems. From a practical point of view, this model can be used to determine the leeway available to the farmers for the introduction of technical change in salad cropping. It could therefore be used by agricultural advisors to plan salad crops either for individual farms or on the scale of a marketing structure.
Key words: lettuce / endive / sheltered crop / decision / model / planning
Corresponding author: email@example.com
© INRA, EDP Sciences 2007