River oases at the southern fringe of the Taklamakan desert in NW China are surrounded by belts of spontaneous vegetation that protect the oases from sand drift. As an important source of forage, fuel and construction wood, this foreland vegetation is also a component part of the agricultural system of the oases but has been, and still is, destroyed through overuse. Within a broader study that aimed to provide a basis for a sustainable management of this foreland vegetation, biomass and production were studied in four vegetation types dominated either by Alhagi sparsifolia, Calligonum caput-medusae, Populus euphratica, or Tamarix ramosissima that were thought to occur under different regimes of natural flooding in the foreland of Qira (Cele) oasis, Xinjiang, NW China. Shoot biomass components were closely correlated to basal area (Calligonum, Populus, Tamarix) or shrub volume and projection area (Alhagi), enabling non-destructive estimation of stand biomass from shoot diameters or shrub dimensions with sufficient precision using allometric regression equations. Relationships between shoot basal area and biomass of the woody species (Calligonum, Populus and Tamarix) agreed with predictions by a theoretical model of plant vascular systems, suggesting that they are determined by hydraulic and mechanical requirements for shoot architecture. Average aboveground biomass densities of typical stands in late summer were 2.97 Mg/ha in Alhagi, 3.6 Mg/ha in a row plantation and 10.9 Mg/ha in homogenous stands of Calligonum, 22–29 Mg/ha in 22 year-old Populus forests and 1.9–3.1 Mg/ha in Tamarix-dominated vegetation. Annual aboveground production including wood and assimilation organs ranged from 2.11 to 11.3 Mg/ha in plantations of Calligonum, 3.17 to 6.12 Mg/ha in Populus, and 1.55 to 1.74 Mg/ha (based on total ground area) or 3.10 to 7.15 Mg/ha (in homogenous stands) in Tamarix. Production of Alhagi is equal to peak biomass. A thinning treatment simulating use by the local population enhanced productivity of Calligonum, Populus and Tamarix. A complete harvest of Alhagi in late August decreased production in the following year. An artificial flood irrigation treatment did not sufficiently increase soil water content except in the uppermost layer and had no clear beneficial effect on growth of the four species and even a negative effect on Alhagi, which was due to increased competition from annual species. As biomass and production with or without artificial irrigation were much higher than values expected for rain-fed desert vegetation at a mean annual precipitation of 35 mm, it is concluded that the existence of all vegetation types studied is probably based on permanent access to groundwater and that natural floods or precipitation do not contribute to their water supply. The effects of agricultural groundwater use in the oasis on groundwater in the foreland of the oasis need further study. Sustainable use of this productive vegetation is possible but requires proper management.