The Tibetan Plateau (TP) is experiencing high rates of climatic change. We present a novel combined mechanistic-bioclimatic modeling approach to determine how changes in precipitation and temperature on the TP may impact net primary production (NPP) in four major biomes (forest, shrub, grass, desert) and if there exists a maximum rain use efficiency (RUE(MAX)) that represents Huxman et al.'s "boundary that constrain[s] site-level productivity and efficiency." We used a daily mechanistic ecosystem model to generate 40-yr outputs using observed climatic data for scenarios of decreased precipitation (25-100%); increased air temperature (1 degrees - 6 degrees C); simultaneous changes in both precipitation (+/- 50%, +/- 25%) and air temperature (+1 to +6 degrees C) and increased interannual variability (IAV) of precipitation (+1 sigma to +3 sigma, with fixed means, where sigma is SD). We fitted model output from these scenarios to Huxman et al.'s RUE(MAX) bioclimatic model, NPP = alpha + RUE x PPT (where alpha is the intercept, RUE is rain use efficiency, and PPT is annual precipitation). Based on these analyses, we conclude that there is strong support (when not explicit, then trend-wise) for Huxman et al.'s assertion that biomes converge to a common RUE(MAX) during the driest years at a site, thus representing the boundary for highest rain use efficiency; the interactive effects of simultaneously decreasing precipitation and increasing temperature on NPP for the TP is smaller than might be expected from additive, single-factor changes in these drivers; and that increasing IAV of precipitation may ultimately have a larger impact on biomes of the Tibetan Plateau than changing amounts of rainfall and air temperature alone.