River flow drives the structure and function of aquatic systems on sub-daily to decadal timescales, and sculpts landscapes on geological timescales from months to millennia (Fisher et al., 1998;Pinay et al., 2018;Tucker & Hancock, 2010). For people, variability in river flow regulates access to freshwater, with extreme flow events such as floods and droughts imposing immense personal and societal costs (
River flows change on timescales ranging from minutes to millennia.
These variations influence fundamental functions of ecosystems,
including biogeochemical fluxes, aquatic habitat, and human society.
Efforts to describe temporal variation in river flow—i.e., flow
regime—have resulted in hundreds of unique descriptors, complicating
interpretation and identification of global drivers of flow dynamics.
Here, we used a cross-disciplinary analytical approach to investigate
two related questions: 1. Is there a low-dimensional structure that can
be used to simplify descriptions of streamflow regime? 2. What catchment
characteristics are most associated with that structure? Using a global
database of daily river discharge from 1988-2016 for 3,120 stations, we
calculated 189 traditional flow metrics, which we compared to the
results of a wavelet analysis. Both quantification techniques
independently revealed that streamflow data contain substantial
low-dimensional structure that correlates closely with a small number of
catchment characteristics. This structure provides a framework for
understanding fundamental controls of river flow variability across
multiple timescales. Climate was the most important variable across all
timescales, especially those lasting several weeks, and likely
contributes as much as dams in controlling flow regime. Catchment area
was critical for timescales lasting several days, as was human impact
for timescales lasting several years. In addition, both methods
suggested that streamflow data also contain high-dimensional structure
that is harder to predict from a small number of catchment
characteristics (i.e. is dependent on land use, soil structure, etc.),
and which accounts for the difficulty of producing simple hydrological
models that generalize well.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.