Abstract This paper seeks to identify the leading social, economic, political, and cultural barriers to the adoption of climate-smart agriculture in Kenya. The study is based on primary qualitative data collected from 146 farmers in Southern Kenya, 22 key informants, and a targeted literature review. We employ a five-stage process-tracing technique with Bayesian updating to trace causal pathways between low CSA uptake and possible alternative explanations. We find poverty and access to financial resources are the leading explanations for low CSA uptake. This barrier was raised in all three evidence sources with high frequency. We observe a disconnect between farmers’ experiences and the support provided by public and private actors that tends to focus on education and training. We conclude that farmers in the sampled regions are motivated to take up CSA practices, but most will be unable to adopt CSA without direct financial or in-kind support. We draw recommendations from our analysis to support more effective and efficient support for smallholder farmers in Kenya and in other low-income contexts. Although our results are limited to the contexts we study, the methodological approach we use that gives primacy to farmers’ explanations for CSA uptake is shown to be a useful tool that can be replicated to inform tailored support to CSA in other contexts.

The online version contains supplementary material available at 10.1007/s10113-025-02511-9.

Abstract Mountain ecosystems are highly sensitive to climate variations. At fine spatial scales, high-elevation microclimates play a critical role in shaping biodiversity, hydrological processes, and ecosystem services, while also influencing the occurrence of natural hazards such as landslides, avalanches, and floods. Heat waves, which have been increasing in frequency and intensity due to global climate change, present significant challenges to these vulnerable environments. This study examines the impacts of summer heat waves on mountain microthermal conditions in three compartments: rockwalls, soils, and lakes. We assembled data across a latitudinal and elevational gradient in the French Alps including years of two recent heat wave events (2015 and 2022). We calculated thermal indicators to evaluate the buffering or amplifying effects of the atmospheric signal on the investigated compartments. The average summer temperature and the growing/thawing degree days were more responsive to heat waves than the phenological indicator (e.g., spring mixing date in lakes) and maximum temperature. We found significant anomalies for both the 2015 and 2022 heat waves across almost all compartments and indicators. Lakes tended to amplify atmospheric temperatures (especially in 2022) whereas rockwalls and soils tended to buffer them. However, residues from the relationship between compartment and atmospheric temperatures were large during heat waves, suggesting that these events may reduce the compartments’ buffering capacity. Our study underscores the importance of long-term monitoring of microthermal conditions to provide a more integrative assessment of mountain ecosystem response to extreme meteorological events. Graphical abstract