Monitoring and modeling human interactions with ecosystems

With rapidly increasing consumption rates and global population, there is a growing interest in understanding how to balance human activities with the other components of the Earth system. Humans alter ecosystem functioning with land cover changes, greenhouse gas emissions and overexploitation of natural resources. On the other side, climate and its inherent interannual variability drive global Net Primary Productivity (NPP), the base of energy for all trophic levels, shaping humans' distribution on the land surface and their sensitivity to natural and accelerated patterns of variation in ecosystem processes.; In this thesis, I analyzed anthropogenic influences on ecosystems and ecosystems impacts on humans through a multi-scale approach. Anthropogenic influences were analyzed with a special focus on urban ecosystems, the living environment of nearly half of the global population and almost 90% of the population in the industrialized countries. A poorly quantified aspect of urban ecosystems is the biogeochemistry of urban vegetation, intensively managed through fertilization and irrigation. In chapter 1, adapting the ecosystem model Biome-BGC, I simulated the growth of turf grasses across the United States, and estimated their potential impact on the continental water and carbon budget. Using a remote sensing-based approach, I also developed a methodology to estimate the impact of land cover changes due to urbanization on the regional photosynthetic capacity (chapter 2), finding that low-density urbanization can retain high levels of net primary productivity, although at the expense of inefficient sprawl. One of the feedbacks of urbanization is the urban heat island effect, which I analyzed in conjunction with a remote sensing based estimate of fractional impervious surface area, showing how this is related to increases in land surface temperatures, independently from geographic location and population density (chapter 3). Finally, in chapter 4, I described the distribution of the global human population as a function of terrestrial NPP and its variability, showing that the median person lives at above average levels of NPP but that almost half of the total population lives in regions significantly affected by climatically induced variability in NPP.