Policy & Practice for composting on Saharan farms in El Oued.

Abstract

Background

Agricultural soils in the El Oued wilaya of the Algerian Sahara suffer from severe organic-matter depletion, secondary salinisation, and high mineral fertiliser costs. Composting local organic feedstocks offers a sustainable soil-restoration solution. However, extreme heterogeneity in farm endowments means uniform extension strategies lead to resource misallocation, requiring an evidence-based tool for targeting interventions. Objectives: This thesis develops and validates a multi-criteria decision analysis (MCDA) pipeline integrating the Analytic Hierarchy Process (AHP) and the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) to rank composting intervention priority for 30 smallholder farms across 15 criteria grouped into five dimensions. Methods: Primary data were collected via surveys and soil sampling (SOM and ECe) in 2023–2024. A five-expert panel constructed 15×15 pairwise comparison matrices using Saaty’s 1–9 scale. Individual matrices were solved for eigenvectors, tested for consistency, and synthesised using the Aggregation of Individual Judgements (AIJ) geometric-mean method. The AIJ weights were implemented in the TOPSIS pipeline. Ranking robustness was validated using four sensitivity procedures: Monte Carlo perturbation, Kendall’s Tau, Full-Range Weight Sensitivity Analysis (FRWSA), and One-at-a-Time (OAT) testing, alongside PCA, clustering, and OLS regression. Principal Results: All expert AHP matrices achieved high internal consistency ( CR < 0.10 ). The AIJ-aggregated matrix yielded CR = 0.021 , identifying Manure (16.07%), SOM (12.61%), and Willingness (12.35%) as the top criteria. TOPSIS closeness coefficients ( C_i ) ranged from 0.149 to 0.771, stratifying farms into five classes (2 Very High, 4 High, 6 Moderate, 12 Low, 6 Very Low). Only 20% of farms crossed the high-priority threshold. Monte Carlo analysis confirmed exceptional ranking stability (mean Spearman rho = 0.976 ). Conclusions and Implications: Feedstock availability, rather than willingness, is the primary constraint on composting adoption in El Oued. This AHP–TOPSIS framework provides the first validated, farm-level composting priority map for the Algerian Sahara, supported by fully reproducible Python code in the annexes.