Rivers of the Anthropocene?

T he ancient Greek philosopher Heraclitus famously asserted that one can never step into the same river twice, thus providing a physical metaphor for life's constant change. The growth in human population and advances in technology in recent centuries have certainly changed rivers. From a scientific perspective, Heraclitus' dictum has never been more apt than now, at the dawn of the so-called " Anthropocene ". Rivers are exceedingly complex biophysical systems. Observable ecological patterns reflect temporally averaged, spatially distributed, multi-scaled processes arising from watershed controls on precipitation, erosion, and nutrient inputs into river channels. Aquatic and riparian species and communities reflect a long evolutionary history of adaptations to dynamic and heterogeneous environments that are hydrologically connected via channel networks through which movement and gene exchange occur. Ecological complexity in rivers has been conceptualized scientifically by viewing these systems as existing in a kind of dynamic equilibrium, or balance, defined by prevailing hydro-climatic and watershed controls and by evolutionary species pools. Humans act to disrupt components of this equilibrium, causing " impacts " that are quantified as measurable deviations in riverine biophysical processes and patterns from some unperturbed baseline condition. Human impacts on rivers are extensive and pervasive. Perhaps nowhere is this better observed than in the case of land-use change (eg urbanization, deforestation) and channel-spanning water infrastructure. Dams dramatically transform rivers by altering the downstream flux of water, sediment, and nutrients, modifying water temperatures, and blocking species movement. They are ubiquitous; in the US alone, there are 75 000 dams exceeding 2 m in height, one on average for every 50 km along mid-sized rivers. Only 42 rivers in the US have undammed reaches greater than 200 km. On a global scale, over 40 000 large dams (>15 m in height) have been built and thousands more are proposed, chiefly in developing countries, to produce so-called " green energy " via hydropower. There is growing societal interest in " restoring " regulated rivers by deliberately releasing reservoir water from dams to provide more reference-like flow conditions downstream. For example, experimental flow releases in the Colorado River system, such as the recent pulse of water from Arizona's Morelos Dam into the near dry downstream Mexican riverbed (Science 2014; 343: 1301), represent efforts to regain some historical ecosystem function. There are limits to the effectiveness of such efforts, however. The physical–chemical modifications to systems associated with larger dams, combined with the human-assisted …