The risk assessment of urban trees is a challenge that must be addressed by many cities, in consideration of their climatology, management and spatial heterogeneity. Since these trees are in stages of aging, added to bad practices, inadequate selection and establishment, amongst others, their early elimination, and hence the loss of the ecosystem services they provide, is being promoted. For this reason, two tree risk assessment methods were developed to be applied in a complementary manner in urban parks, one basic visual (BV) and the other detailed visual (DV), both structured with the components of likelihood of failure, likelihood of impact, consequence and risk rating, being analyzed qualitatively and semi quantitatively. The methods were applied in 24 trees of Eucalyptus and Tipuana genus by two groups of assessors, experienced and inexperienced, totaling 192 assessments. Results were analyzed with a linear mixed model (LMM), which showed differences between both methods for the likelihood of failure, where the assessment of branches turned out to be the attribute with the highest impact. The methods proved to be suitable to be used in a complementary manner, since BV turns out to be efficient when it is necessary to assess a large number of trees. DV turned out to be more effective, since it was able to identify more precisely the attributes of the tree that must be considered for adequate risk mitigation and management. Therefore, their combined use is recommended for risk assessment in trees established in parks.