Humanity is exerting unprecedented pressure on the natural environment, threatening tens of thousands of species with extinction, and degrading the condition of ecosystems worldwide. The extraordinary value of the last ecologically intact landscapes, free from industrial level degradation (which I call wilderness), is becoming increasingly recognised (chapter 1). Wilderness areas support many of the evolutionary and ecological processes that underpin all life on Earth, are critical for biodiversity conservation, and support the cultural integrity of many indigenous communities. The overarching aim of this thesis is to address key questions relevant to conserving wilderness areas and their values, with a focus on biodiversity conservation.Despite the exceptional value of wilderness areas, information on their location, condition, and threat status globally has been limited. In chapter 2, I utilise a high-resolution (1km 2 ) global dataset of human pressure on the terrestrial environment for the years 1993-2009 (appendix 1) to develop the first temporally inter-comparable global maps of terrestrial wilderness areas. I define wilderness as places free from human pressures such as land clearing, dense human settlements, agriculture, and infrastructure developments, which significantly damage the environment. A spatial analysis of changes in wilderness extent between 1993 and 2009 showed catastrophic declines amounting to 3.3 million km 2 , with the greatest losses occurring in the Amazon and Central Africa (appendix 2). Only 30 million km 2 of wilderness remains (23% of terrestrial areas).Considering rates and extent of wilderness loss varies throughout the world, it is important to identify where this loss is impacting species. In chapter 3, I present a global analysis of cumulative human impacts on threatened species. I develop a novel spatial framework that jointly considers the co-occurrence of threats and the distribution of 5,457 vertebrates. I discover that human impacts extend across 84% of Earth's terrestrial surface, and identify 'hotspots' of impacted species richness. One quarter (n=1237) of species are impacted by threats across >90% of their distribution, and 395 species are impacted across their entire range. The methodology represents a conceptual advance for analysing threats to biodiversity, moving beyond analysing human pressures, which are agnostic to species type and their individual sensitivities to threats, to analysing realised impacts on individual species.