Probabilistic soil mapping by Bayesian inference to assess suitability for derocking in northwest Syria

Abstract

Land reclamation by rock removal has wide potential in areas with shallow or rocky soils. In northwest Syria, this practice is hindered in its implementation by a lack of physical soil suitability data, principally soil rockiness and soil depth to hard rock. These soil properties were surveyed in a limited study area, resulting in a hard-boundary thematic soil map (64-94% accuracy per property). Bayesian inference is proposed as a low-cost upscaling method that yields a set of pixel-based probability maps, providing improved input for spatial decision support models. Whereas the achieved spatial upscaling (from 2510 to 19 100 ha) outweighed the decrease in overall accuracy (down to 26-57%), probability maps require dedicated validation and manipulation procedures. This research contributes to methods for the creation, validation and interpretation of probabilistic soil property maps for quantitative land evaluation. First, we evaluated three postprocessing and validation methods for probabilistic soil property maps, identifying the use of 'prediction rates' as the best approach in a spatial planning context. Next, we demonstrated how the maps can support decision-making for land management activities by simulating the expected losses and gains from interventions, in a decision-theoretic approach. Based on the simulations, the investments in largescale derocking projects in northwest Syria will pay off in terms of increased agricultural productivity in less than 10 years.

Numerical modelling of soil–landscape relationships using diversity indices and conditional probability: a case study from an Iranian arid region

“…Conditional probability can be applied as an excellent tool to investigate the soil-landscape ٢٠١ relationship (Bagheri Bodaghabadi and Toomanian, 2019;Hawinkel et al, 2016, Taalab et al, ٢٠٢ 2015. Presume that two events, A and B, are given in a similar sample space, and the event ٢٠٣ probability of B is greater than zero, i ٢٠٤ has occurred (i.e., P (A|B)) is equal to the ٢٠٥ and B (i.e., P (A∩B)), to the unconditional probability B (i.e., ٢٠٦ ٢٠٧ Now, if it is supposed that each soil ( ٢٠٩ landscape, based on conditional probability, the probability ٢١٠ of landform unit (Gj) is observed, can be shown as follows: ٢١١ ٢١٢ It is also possible to show the probability of ٢١٤ observed as follows: ٢١٥ ٢١٦ ٢١٧ When the probability of the existence of soil ( ٢١٨ (Gj) is observed, this means that the soil ( ٢١٩ the closer the P(Si|Gj) average of all map units ٢٢٠ to delineate units or soil types.…”

“…٨٩ From this perspective, there is a need for an approach that can not only determine the role of ٩٠ soils in detecting landforms but also measure the soil-geomorphology relationship and its ٩١ quantitative interpretation. Statistical and mathematical tools are increasingly used to quantify ٩٢ and model soil variability (Hawinkel et al, 2016;Havaee et al, 2019;Mikhailova et al, 2021). ٩٣…”

Numerical Modeling of soil-landscape relationships using diversity indices and conditional probability: A case study from an Iranian arid region

“…From this perspective, there is a need for an approach that can not only determine the role of soils in the detection of landforms but also to quantitatively measure the soil-geomorphology relationship and its interpretation. In order to quantify and model soil variability, the use of statistical and mathematical tools are growing [19,21]. In this regard, the conditional probability approach can be helpful, because the conditional probability is calculated for a set of values that can cover a wide range of soil variables (quantitative and qualitative).…”