The benefits of a massive harvesting of legacy measured profiles for mapping primary and functional soil properties in the Coastal Plain of Occitanie (Southern France)

International audience

إنجليزي. The coastal plain of Occitanie is facing up to important issues related with the dramatic decrease of the water and soil resources caused respectively by the global climate change and the progresses of urbanization boosted by a large demographic growth. Mitigating such issues requires relevant decision making based on an accurate knowledge of the spatial distribution of soil properties over the region. For the last fifteen years, such a knowledge has been provided to end-users by the successive implementations of new digital soil mapping approaches: Developing GlobalSoilMap products (Vaysse et al, 2015), Regional mapping of Soil Available Water Capacities (Styc and Lagacherie, 2021) and of a Potential Soil multifunctionality index (Angelini et al, 2022). However, the low spatial density of the sets of legacy measured profiles severely limited the accuracy of these mapping products, which in turn significantly hampered the consideration of soils in the land planners' decision making.To face this problem, a massive harvesting of legacy soil profiles was conducted in the region, involving semi-automatic entry procedures based on automatic reading. This allowed to considerably increase the spatial density of legacy measured soil profiles that were used as input of DSM models (from 1 profile/ 9.9 km2 to 1 profile/ 1.3 km2). The new DSM models that were calibrated from this new soil dataset produced more accurate soil mapping products with lower predicted uncertainty. A subsequent effort was also conducted to communicate the remaining uncertainty under an adequate formalism (spatial aggregation, adapted semiology) that enabled a good appropriation of the soil maps by decision-makers.The results obtained in this pilot region illustrated the feasibility of operational DSM approaches to support land planners' decision making at regional and territorial scales.ReferencesAngelini, M.E., Heuvelink, G.B.M., Lagacherie, P., 2023. A multivariate approach for mapping a soil quality index and its uncertainty in southern France. Eur. J. Soil Sci. 74:e13345., 1–17.Styc, Q., Lagacherie, P., 2021. Uncertainty assessment of soil available water capacity using error propagation : A test in Languedoc-Roussillon. Geoderma 391, 114968.Vaysse, K., Lagacherie, P., 2015. Evaluating Digital Soil Mapping approaches for mapping GlobalSoilMap soil properties from legacy data in Languedoc-Roussillon (France). Geoderma Reg. 4, 20–30.