Operationalizing Knowledge Co-production: The Hybrid Epistemic Adaptive Resilience (HEAR) Framework and Methodological Lessons from Malawi

Abstract

Indigenous Knowledge Systems (IKS) and real-time digital technologies offer complementary resources for strengthening climate resilience in smallholder agriculture, yet their integration remains methodologically difficult. This paper develops the Hybrid Epistemic Adaptive Resilience (HEAR) framework to address the epistemic, institutional, and technical tensions that arise when Complex Adaptive Systems theory and Participatory Action Research are applied in stratified agricultural extension settings. The framework is grounded in a longitudinal, convergent-parallel mixed-methods case study being conducted across Nsanje, Dedza, and Karonga districts in Malawi from September 2024 to June 2026. HEAR organises knowledge co-production through four linked layers: technological and indigenous data inputs, a Translation Engine, adaptive feedback loops, and socio-ecological outcomes. The Translation Engine comprises knowledge extraction, semantic mapping, evidence weighting, and recommendation generation. It is designed to combine sensor, satellite, meteorological, and crop-modelling data with locally validated phenological, edaphic, celestial, meteorological, ethno-pedological, and ethno-botanical knowledge. The framework also incorporates power-sensitive facilitation, multimodal advisory delivery, longitudinal adoption monitoring, yield-stability assessment, farmer-satisfaction measures, and an Epistemic Justice Index. By specifying measurable constructs, testable pathways, and safeguards for community authority and data governance, HEAR provides a structured methodology for examining whether hybrid advisories can become scientifically credible, culturally legitimate, accessible, and responsive to local agricultural conditions.