The plight of forests and the restoration of forests worldwide has gained recent traction, with emphasis being drawn on forests at the 2021 COP26 summit in Glasgow where the United Nations coined this the “Decade of Restoration”. With forests holding much of the world’s biodiversity, they are of key conservation concern, along with playing an important role in climate regulation. Therefore, monitoring the relationships between the forests themselves and the fauna within are key for successful conservation. Within forests, large herbivores have a crucial role and help to maintain ecosystem functioning through seed dispersal, nutrient and carbon cycling, creation of microhabitats and opening previously inaccessible areas for smaller animals. Of the large herbivores, the megaherbivores (weighing in excess of 1000 kg) have the greatest impact, and of particular interest is the African elephant. From conducting a review of previous research from Kibale National Park, Uganda, three questions were developed: (1) How does disturbance history and vegetation cover affect relative elephant abundance? (2) How does rainfall variation across years affect relative elephant abundance in a forested environment? And (3) How has cumulative diameter at breast height (DBH) of trees in Kibale changed and has there been any influence by elephants? These questions were addressed using a combination of long-term elephant abundance data and forest structure with remotely sensed land cover classification. The results showed that relative elephant abundance in Kibale has increased but has been higher than expected from natural recruitment. This has been attributed to elephants migrating from the Democratic Republic of Congo into Kibale. As a result, the ecological predictors used in the mixed model (disturbance history and rainfall) showed minimal slight significant effects on relative elephant abundance. Land classification did not work for mapping earlier years (1996-2008) and therefore was dropped from the models. We suggest that more information is required for land classification of previous years along with substantial knowledge of the study area to interpret such longitudinal data.