cultivar_22_Final_EN

18 ANALYSIS AND PROSPECTIVE STUDIES CULTIVAR Issue 22 APRIL 2021 to buy the new inputs. In this framework, farmers – until then the main agents of local knowledge on which their production systems had been based – were made deeply dependent on global scientific knowledge. This initially belonged to the state and its research and rural extension apparatus, and later to the commercial suppliers of the new inputs. The dual replacement of the chemical-mechanical model allowed higher food production per agricul- tural worker and therefore the transfer of many farm labourers to the emerging industrial and service sectors. This therefore allowed greater occupational freedom of choice which is highly valued today. Fur- thermore, it reduced the global risk of food insuffi- ciency, which stems today from income inequality rather than insufficient tech- nological potential in food production. Agroecosystems influenced by the chemical-mechanical model are highly modified today. They are more pro- ductive, in terms of food production per hectare, and more dependent on outside energy subsidies to ensure their own functioning and stability. Artificialising agroecosystems allowed agricultural production to rise during the second half of the 20th century chiefly by raising production per hectare (intensification) rather than by expanding the area of cultivated land. This has had evident benefits in terms of lower pressure to convert natural habitat into farmland. The inefficient use of chemical inputs has led, however, to serious pollution problems that are far from being localised. The use of nitrogen fertilisers has doubled the overall nitrogen cycle (Vitousek et al. 1997) and bioaccumulative pesticides are today detectable in relatively remote areas where they have never been used, such as Antarctica. Globally, the mainstreaming of the chemical-me- chanical model, even in developing countries (the so-called green revolution), has multiplied global cereals output threefold since 1950 based on the adoption of high-yield varieties of wheat, rice and maize, multiplied irrigated land threefold and multi- plied the global use of industrial fertilisers elevenfold (Brown, 2004). A new technological model: sustainable intensification As we have seen, the need to increase labour pro- ductivity lies behind the chemical-mechanical model in agriculture. The development of this model has led to higher land productivity (agricultural intensification) by raising the use of industrial inputs but this has generally been accompanied by lowering the efficiency with which they are used. Today, overcoming the dilemma of intensification implies raising land produc- tivity (the good part of inten- sification) without increasing inputs per hectare (the bad part), which requires very significant gains in the efficient use of these inputs (“more crop per drop”). In fact, defined as the level of production per hectare and not as the level of inputs per hectare, intensification may, as we have seen, be the key to satisfying the growing demand for food, bioenergy and biomaterials, avoiding the mass conversion of natural habitats into farmland, which would have an unsustainable environmental cost. Under the chemical-mechanical model, rising yields per hectare were generally achieved by increasing inputs per hectare. Therefore, agricultural use of fer- tilisers, pesticides, water and energy have multiplied globally by several factors over the last decades. This growth in the use of inputs has led to lower efficiency in agricultural production and a necessary increase in doses in order to obtain successive identical increases in output (the law of diminishing returns). This drop in efficiency combined with the general increase in inputs has resulted in a range of environmental prob- lems, such as eutrophication of aquatic ecosystems, Agroecosystems influenced by the chemical-mechanical model are highly modified today. They are more productive, in terms of food production per hectare, and more dependent on outside energy subsidies to ensure their own functioning and stability.