Grasslands are found on every continent in the world except Antarctica. They benefit humankind, support livestock and provide feedstock for biofuel production. Many modern crops were domesticated from grasslands. More than 800 million people worldwide depend on grasslands and their products. Chapter I provides background information about grasslands, the major threats they face today, and an introduction to the research conducted for this thesis.Despite their global presence, grasslands are critically endangered due to land-use change, agricultural expansion, environmental changes such as elevation in atmospheric CO2 levels, and nitrogen (N) deposition. Previous studies have estimated that 85% of terrestrial carbon sink changes are due to elevated atmospheric CO2 and 10% – 20% are due to increasing atmospheric N deposition. Although it is known that N addition is linked strongly with loss of biodiversity, little is understood about the effect on N addition on the biomass responses of grasslands. Chapter 2 explored the effect of continuous long-term N addition on the aboveground biomass (AGB) of grasslands. Data were provided by the Nutrient Network, a global coordinated research effort of grassland sites. The study explores biomass responses of different plant functional types like grass and forbs, origin characteristics of sites, namely if they are introduced species or native species dominated, mean annual temperature (MAT) and precipitation (MAP) and soil characteristics like soil N content, soil carbon (C) content and soil pH to N addition. Findings revealed that different plant functional types had divergent responses with long-term N addition treatments. Although terrestrial biosphere models (TBMs) are also used to estimate the effect of increasing N deposition on biomass, there are large variations among the estimates. In Chapter 3, observation data from grasslands were used to benchmark the estimates from an ensemble of six TBMs. Findings revealed that all models underestimated the nitrogen use efficiency (NUE) of grasslands in an increasing N deposition scenario. The N saturation threshold of models are not as constrained and estimations of the decline in NUE are unreliable. These findings revealed that TBMs could benefit from improving the vegetation structure module and N saturation threshold.