A study was conducted in Magamba Nature Reserve (MNR), Tanzania to determine the influence of human activities on the degradation of Montane forest in order to generate information for guiding forest conservation and management approaches. Field survey coupled with remote sensing and Geographic Information System (GIS) were used to collect data on land use and human activity patterns. Participatory Rural Appraisal (PRA) tools including focused group discussion, questionnaire survey and participant observation were used for human activities data collection. Ninety respondents were randomly selected for the questionnaire interview in order to determine the human activities in MNR. The forest degradation was determined by means of systematic sample plot inventories, where stems density, basal area and stand volume of stand trees including biomass for 2009 and 2013 were measured. Data collected were analysed using GIS and descriptive statistical procedures. Multiple regression analysis was used to establish the relationship between forest degradation and human activities. Results indicate that the major human activities in the study area on average are collection of firewood (23 %) and medicinal plants (19 %), illegal lumbering (10 %), cutting trees for building poles (3 %) and animal grazing (10 %). The results indicate a significant correlation between biomass change due to firewood collection and distance from village centre (R2 = 0.706) and distance from major roads (R2 = 0.650). These results suggest that forest degradation is higher near major roads and village centres. Distance from major roads and village centres can thus be taken as predictors of forest degradation. The findings of this study will enable the understating of the drivers of deforestation and forest degradation and will also be fundamental for the development of policies and measures that aim to alter current trends in forest activities toward a more climate and biodiversity friendly outcome.Climate Change Impact, Adaptation and Mitigation Project (CCIAM)