MESMIS Sustainability Assessment of Plantain and Banana Farms in Quindío: Reproducibility Package

This repository contains the data, analytical code, supplementary material, and reference outputs associated with the study “Sustainability of Plantain and Banana Production Systems under Contrasting Management Profiles in Colombia: An Operational MESMIS Framework.”

The study assessed the sustainability of 36 plantain and/or banana farms located in six municipalities of Quindío, Colombia. The farm was defined as the analytical unit. The seven conceptual MESMIS attributes were organized into four integrated operational constructs: Productivity (P), an operational Stability–Resilience–Reliability proxy (SRR), Economic Self-Reliance (ESR), and the Integrated Social Attribute (ISA).

The reproducibility package includes the anonymized analytical database, data dictionary, standalone R analysis script, supplementary Tables S1A, S1B, and S2, reference tables and Figures 2–6, quality-control outputs, software session information, licenses, and supporting documentation. An archived version of the MESMIS-SSD R Shiny application originally used for data entry and preliminary indicator calculation is also provided for methodological provenance.

The definitive analytical workflow is implemented in code/run_analysis.R. The script reproduces the descriptive statistics, sustainability categories, Spearman correlations, leave-one-out sensitivity analysis, weighting sensitivity analysis, Kruskal–Wallis tests, principal component analysis, tables, and figures reported in the study.

The Stability–Resilience–Reliability construct should be interpreted as an operational biophysical proxy based on soil cover and aboveground biomass rather than as a direct temporal measurement of stability, reliability, or recovery following disturbance. Similarly, the conventional, mixed, and regenerative/transitional categories represent qualitative operational management profiles rather than certified production systems.

Producer names and detailed farm locations were removed to protect participant confidentiality. Users must not attempt to re-identify participating producers or farms. The analytical data are distributed under the Creative Commons Attribution 4.0 International license (CC BY 4.0), while the code is distributed under the MIT License.