Species-specific root-shoot ratios in a diverse grassland community

Existing studies of root/shoot ratios (R/S) in the field show large
differences among environments, from high values in unproductive
environments to low values in productive environments. However, virtually
all such studies are community-wide, i.e. compare biomasses of roots and
shoots summed over all coexisting species. Therefore we do not know to
what extent this is due to innate interspecific differences in the R/S
ratios or to plastic response to nutrient limitations. This is due to
methodological difficulties in assigning roots into species, which
preclude the contribution of species-specific differences to the
community-wide R/S ratio. These limitations can be overcome by DNA
analyses. We used qPCR to determine species-specific R/S ratios in a
montane grassland and compared these values to R/S ratios from a
two-season long single-species pot experiment. To translate the qPCR data
to root biomass units, we extracted DNA from the experimental plants to
perform calibration for each species and to take into account potential
change during the ontogeny of root systems. We used these calibration
relationships to recalculate gene copies estimated in a set of field
samples to species-specific root biomass values. All species except one
showed fairly high correlation between aboveground biomass (determined by
weighing) and belowground biomass (determined by qPCR) across all studied
plots in the field. Root/shoot ratios determined from the field were for
most species slightly lower compared to the aggregate values of the whole
community determined by weighing both roots and aboveground biomass. They
showed significant correlation with the R/S ratios from the pot
experiment, indicating innate differences among species that persist
across two very different sets of conditions (multispecies community in
the field, single species culture in pots). The field R/S ratios based on
qPCR root estimation are informative on the true field patterns. This is
supported by their correlation with the pot-derived values. While the
values seem to be slightly underestimated, they are well in the reasonable
range of values. This opens a new field of study, namely examination how
R/Sratios in the field respond to changed nutrient availability and
diversity and species composition of neighbouring species.