Keywords: Climate change; Increased climate variability; Vulnerability; Smallholder farmers; Adaptation Climate change and increased climate variability are currently seen as the major constraints to the already stressed smallholder farming livelihood system in southern Africa. The main objectives of this study were first to understand the nature and sources of vulnerability of smallholder farmers to climate variability and change, and second to use this knowledge to evaluate possible farm-level management options that can enhance the adaptive capacity of smallholder farmers in the face of increased climate variability and long-term change in climate. The study was conducted in Makoni and Hwedza districts in eastern Zimbabwe. Local famers’ and expert empirical knowledge were combined using research tools that mainly included detailed field observations and surveys, systems analysis and field experimentation, and simulation modelling (the Agricultural Production Systems Simulator (APSIM)). To understand the nature and sources of vulnerability, long term climate data were analysed and farmers were interviewed individually and in groups. On-farm experimentation and simulation modelling were conducted to evaluate the impacts and interactions of adaptation options namely maize cultivar choice, staggered planting dates, and variable fertilizer rates, on maize yield under both short-term climate variability and long-term climate change. Another on-farm experiment was conducted to assess whether small grains (finger millet and sorghum) perform as well as maize under variable soil and rainfall conditions. The long-term rainfall and temperature analyses closely supports farmers’ perceptions that the total annual rainfall has so far not changed, but variability in the rainfall distribution within seasons has increased. The number of rain days has decreased, and the frequency of dry spells within season increased. The mean daily minimum temperature increased by 0.2°C per decade in Makoni, and by 0.5°C per decade in Hwedza, over the period from 1962 to 2000. The surface air temperature is further projected to increase significantly in Makoni and Hwedza, by 2100. The impacts of rising temperatures and increased rainfall variability among smallholder households were highly differentiated because different households depend on varied farming livelihood sub-systems, which were exposed uniquely to aspects of climatic risk. For example, livestock production was sensitive to drought due to lack of feed, affecting resource-endowed farmers, who often own relatively large herds of cattle. Crop production was more sensitive to increased rainfall variability, affecting especially farmers with intermediate resource endowment. Availability of wild fruits and social safety nets were affected directly and indirectly by extreme temperatures and increased rainfall variability, impacting the livelihoods of poorer farmers. Farmers have also access to different biophysical and socioeconomic resources such as fertilizer and farm labour inputs, and as a result they respond variedly to impacts of a changing climate. Thus, alongside climate variability and change, farmers also faced biophysical and socioeconomic challenges, and these challenges had strong interactions with adaptation options to climate change. Experimentation in this studydemonstrated that the maize cultivars currently on the market in Zimbabwe, and in many parts of southern Africa, exhibit narrow differences in maturity time such that they do not respond differently to prolonged dry spells. The yield performance for all three cultivars is projected to be similar in future change in climates, consistent with results from the experiments.In the current cropping system farmers can select any cultivar available on the market without a yield penalty. However, with climate change none of the available cultivars will be able to compensate for the decline in yield. Greater maize grain yields were obtained with both the early (25 October – 20 November) and normal (21 November – 15 December) plantings, with no significant differences between these planting windows(e.g. on average 5 t ha-1 in Makoni, and 3 t ha-1 in Hwedza for the high fertilization rate).Contrary to previous research findings, there is a reasonably wide planting window in which good yields can be obtained if the rains start on time, but if the start of the rains is delayed until after the beginning of December planting should be done as soon as possible. Regardless of the amount of fertilizer applied, yields were reduced strongly when planting was substantially delayed by four weeks after the start of the rainy season. Maize yielded more than finger millet and sorghum even when rainfall was poor in the 2010/2011 season. For example, maize yielded 2.4 t ha-1 compared with 1.6 t ha-1 for finger millet and 0.4 t ha-1 for sorghum in the 2010/2011 rainfall season in Makoni. Finger millet and sorghum failed to emerge unless fertilizer was applied. Application of manure alone failed to address this challenge of poor emergence until fertilizer was added. Sorghum suffered critical yield losses due to bird damage. The better performance of maize over finger millet and sorghum suggested that the recommendation to substitute small grains for maize as a viable adaptation option to a changing climate, will neither be the best option for robust adaptation nor attractive for farmers in southern Africa. Alternatively spreading crops across the farm and in time can be a viable strategy to spread climatic risk as well as improve human nutrition. Poor soil fertility constrained yield more strongly than rainfall and late planting, as demonstrated by the large yield gap (> 1.2 t ha-1) between the unfertilized and fertilized cultivars even in the poor rainfall season (2010/2011). Fertilization increased yield significantly under both the baseline and future climates particularly when planting before mid-December.The maize response to mineral nitrogen is, however, projected to decline as climate changes, although effects only become substantial towards the end of the 21st Century. Soil fertility management is therefore likely to be a major entry point for increasing the adaptive capacity of smallholder farmers to climate change and increased climate variability. However, management of factors related to both nutrient resource access and farmers decisions to enhance resource use efficiencies are critical if agriculture is to be used as robust adaptation options to climate change by smallholder in Southern Africa.