Here, we present a series of 124 simulations with the LPJ–GUESS–SIMFIREglobal dynamic vegetation–wildfire model, including a semi-empiricalformulation for the prediction of burned area based on fire weather, fuelcontinuity and human population density. The simulations use ClimateModel Intercomparison Project 5 (CMIP5) climate predictions from eight Earthsystem models. These were combined with two Representative Concentration Pathways (RCPs) and fivescenarios of future human population density based on the series of SharedSocioeconomic Pathways (SSPs) to assess the sensitivity of emissions to the effect of climate, CO2 and humans. In addition, two alternativeparameterisations of the semi-empirical burned-area model were applied. Contrary toprevious work, we find no clear future trend of global wildfire emissions forthe moderate emissions and climate change scenario based on theRCP 4.5. Only historical populationchange introduces a decline by around 15 % since 1900. Future emissionscould either increase for low population growth and fast urbanisation, orcontinue to decline for high population growth and slow urbanisation. Onlyfor high future climate change (RCP8.5), wildfire emissions start to riseagain after ca. 2020 but are unlikely to reach the levels of 1900 by the endof the 21st century. We find that climate warming will generally increase therisk of fire, but that this is only one of several equally important factorsdriving future levels of wildfire emissions, which include population change,CO2 fertilisation causing woody thickening, increased productivity andfuel load and faster litter turnover in a warmer climate.