Agricultural Horizons: Research, Technology and Sustainable Development Vol. 3

Climate extremes are reshaping the risk profile of agricultural landscapes, and the deliberate retention or introduction of woody perennials within farmland is now promoted widely as an adaptation response. This critical narrative review examines how far that promotion is matched by evidence. The review synthesises research on the capacity of agroforestry to moderate exposure to heat extremes, meteorological and agricultural drought, extreme rainfall and flooding, and on the risks that agroforestry systems themselves face as climatic variability increases. Literature was identified through structured searching of open scholarly indexes and citation registries, supplemented by backward and forward citation tracking and by targeted searching of authoritative institutional sources, with evidence appraised for design adequacy, measurement validity, comparator quality, scale of inference and geographical representativeness. Four conclusions emerge. Thermal buffering is the best-supported function, with sub-canopy moderation of maximum temperatures repeatedly demonstrated and with the buffering effect increasing as ambient conditions become more extreme, although the translation of that microclimatic effect into yield protection is demonstrated far less consistently than the microclimatic effect itself. Hydrological buffering is conditional rather than general: improvements in infiltration and soil structure are robust at plot scale, but net water balance outcomes depend on tree density, rooting depth, species water use and rainfall regime, and catchment-scale flood attenuation weakens sharply for the largest events. Drought represents the sharpest internal contradiction in the field, since the same canopy that reduces evaporative demand also increases whole-system transpiration, and at least one carefully instrumented study records higher crop mortality under shade during extreme drought. Emerging risks, including compound hot and dry events, rising atmospheric dryness, shade-mediated pathogen dynamics, fire and tree mortality, are inadequately represented in the adaptation literature. The evidence supports a conditional, design-dependent and context-specific reading of agroforestry resilience rather than a generalised protective claim, and priority research needs centre on extreme-event-resolved measurement, long observational series and transparent reporting of negative results.