Life history strategies drive size-dependent biomass allocation patterns of dryland ephemerals and shrubs

Plant biomass allocation patterns are important to understanding and predicting ecosystem carbon cycles and other important ecological processes. Consequently, many attempts have been made to study these patterns. However, most studies focus on data from species in temperate forests to the neglect of data from desert species adapted to arid conditions. We hypothesize that different life history strategies drive or at least participate in different plant biomass allocation patterns, as, for example, the life history differences between desert ephemeral and shrub species. We tested this hypothesis using field data gathered directly from the entire desert vegetation in Northwestern China. When the data from each of the two species groups are pooled, ephemeral and shrub species manifest different scaling relationships between above- and belowground biomass, unlike the isometric scaling relationships typically reported for forest tree species. The observed scaling relationships are sensitive to water stress and temperature gradients. The scaling exponents numerically decrease with increasing drought stress, and relatively more biomass is allocated to shoot growth in ephemeral species (presumably to rapidly complete their life history). In the case of shrubs, the numerical value of the scaling exponent is insensitive to rainfall, presumably because these species allocate more biomass to root growth to access belowground water. However, the scaling exponent is significantly sensitive to temperature, which also regulates root growth. Moreover, for the two species groups, root-shoot ratios are jointly regulated by precipitation and temperature. The different biomass allocation patterns appear to result from different life histories that maximize either competition among neighboring plants, or escaping damage from drought and low temperatures. These findings show how overall body size and life histories jointly regulate biomass allocation patterns under different extreme conditions and provide insights into estimating the dry carbon content in dryland ecosystems.