Bamboo-dominated forests in Southwestern Amazonia encompass an estimated 180,000 km2 of nearly contiguous primary, tropical lowland forest. This area, largely composed of two bamboo species, Guadua weberbaueri Pilger and G. sarcocarpa Londoo & Peterson, comprises a significant portion of the Amazon Basin and has a potentially important effect on regional carbon storage. Numerous local REDD(+) projects would benefit from the development of allometric models for these species, although there has been just one effort to do so. The aim of this research was to create a set of improved allometric equations relating the above and belowground biomass to the full range of natural size and growth patterns observed. Four variables (DBH, stem length, small branch number and branch number ≥ 2cm diameter) were highly significant predictors of stem biomass (N≤ 278, p< 0.0001 for all predictors, complete model R2=0.93). A secondary field model (containing DBH and branch number > 2cm diameter), proved highly significant as well (N= 278, p< 0.0001 for both predictors, R2=0.84). The belowground biomass was estimated to be 19.2+/-6.2% of the total dry biomass of the bamboo species examined. To demonstrate the utility of these models in the field and derive stand-level estimates of bamboo biomass, ten 0.36-ha plots were analyzed (N= 3,966 culms), yielding above + belowground biomass values ranging from 4.3-14.5 Mg/ha. The results of this research provide novel allometric models and estimates of the contribution of G. weberbaueri and G. sarcocarpa to the total carbon budget of this vast and largely unexplored Amazonian habitat.