Change Propagation

One objective of this project is to understand how change propagates through systems.

We have quantified the detailed nature of hydrologic change due to agricultural practices and climate change in the MRB. Our TileDrainCopulaanalysis has revealed a changed storage-discharge relationship during the growing season, sharper rising limbs of daily streamflow hydrographs and stronger dependence on previous-day precipitation.  Considering the combined climate-drainage effects, we have shown that artificial drainage has reduced the inherent nonlinearity of daily streamflow dynamics, perhaps reflecting the reduced complexity of the engineered hydrologic system.  Right: Copula inter-quartile analysis for positive slopes of daily streamflow data versus previous day precipitation for the Redwood Basin. The black line represents data before the assumed land use change in 1976 and the red line represents after the land use change. The plot indicates a strengthened dependence of streamflow increase to previous day rain.

We have found that the detrimental response to change within the Minnesota River Basin is not evenly distributed across the landscape but concentrated in select times and locations. For example, the main sediment source to the fluvial network is not agricultural fields or run-off, despite the intensive land management practices of industrial scale row-crop agriculture. Most sediment derives from near channel sources such as bank and bluff erosion that are mobilized under high streamflow conditions.   Left: Eroding bluffs contribute much more sediment than runoff within the Minnesota River Basin. A canoe at the base of the bluff provides a sense of scale.

We have also made advances in linking the river meander process to the sinuous channel planform pattern.  We found that the physically-based yet simplified model dynamics inherently produce a prototypical bend shape, dubbed simple. Cutoffs act as perturbations to the channel planform, resulting in a spectrum of bend geometries including two other archetypal geometries: round and long. Using three measures of bend migration, we showed that bends with similar cutoff geometries shared similar dynamic histories, enabling the inference of historic dynamics from static cutoff shapes. Specifically, simples migrated fastestmeandering and longs slowest with rounds showing significant variability in their dynamic trajectories.  Right: A long-term simulated meandering river. (a) 30,000 years of modeled centerline realizations. Older centerlines are darker; the blue centerline shows t = 30,000 years. The upstream boundary condition fixes the first centerline node in place, leading to the formation of the spiral pattern at the upstream boundary. No restrictions are placed on the downstream node so the river may migrate freely. (b) A reach of simulated centerline selected shows the growth and cutoff of three prototypical bend types that emerge from the physically-based deterministic model dynamics. Realizations are 300 years apart. Note the complex multilobe meander that starts as double lobed but develops a third lobe before cutting off between 900 and 1200 years.