The role of chronic and episodic disturbances on channel–hillslope coupling: the persistence and legacy of extreme floods

Abstract

Landscape form represents the cumulative effects of de-stabilizing events relative to recovery processes. Most geomorphic research has focused on the role of episodic rare events on landscape form with less attention paid to the role and persistence of chronic inputs. To better establish the interplay between chronic and episodic extreme events at regional scales, we used aerial photography and post-flood sediment sampling to assess stream and hillslope response and recovery to a 100-300 yr. flood caused by Tropical Storm Irene in New England. Within a 14 000 km 2 study area, analysis of aerial photographs indicated that the storm initiated (n = 534) and reactivated (n = 460) a large number of landslides. These landslides dramatically increased overall estimates of regional erosion rates (from 0.0023 mm/yr. without Irene to 0.0072 mm/yr. with Irene). Similarly, Irene-generated LWD inputs of 0.25-0.5 trees/km exceeded annual background rates in a single event, and these concentrated inputs (10 1 -10 2 of trees/landslide) are likely to result in large jams and snags that are particularly persistent and geomorphically effective. Finally, we found that landslide scars continue to provide elevated sediment inputs years after the event, as evidenced by sustained higher suspended sediment concentrations in streams with Irene-generated landslides. Overall, our results indicate that infrequent, high-magnitude events have a more important geomorphic role in tectonically stable, more moderate-relief systems than has been previously recognized. Understanding the role of these events has particular relevance in regions such as New England, where the frequency and magnitude of extreme storms is expected to increase. Further, these effects may force reconsideration of conservation and restoration targets (for example in channel form and large wood loading and distribution) in fluvial systems.

Dynamic linear models to explore time‐varying suspended sediment‐discharge rating curves

“…These results suggest a degree of flood event‐driven positive hysteresis in turbidity loads. As discussed earlier, this has been observed elsewhere [ Yellen et al ., ; Dethier et al ., ] and is thought to be linked to stream bank erosion and mass wasting that persist for prolonged periods after the flood. These effects are simulated by the dynamic intercept.…”

“…[] noted that SSC remained elevated in the lower Connecticut River for years following the flood of Hurricane Irene, requiring recalibration of the rating curve to maintain model accuracy. Similar flood event‐driven positive hysteresis has been noted elsewhere [ Megnounif et al ., ; Dethier et al ., ], including in the Catskill Range in New York [ Mukundan et al ., ], which is the study region for this work. Figure shows observed turbidity load, a proxy for sediment load equal to the product of turbidity (T n ) and Q, versus streamflow in the Esopus Creek in the Catskills before and after Hurricane Irene and a major 1996 winter storm.…”

Historically unprecedented erosion from Tropical Storm Irene due to high antecedent precipitation

“…These observations both downstream at Keeney Cove, and in Amherst Lake's densely‐forested and undammed watershed, imply a geographically broad signal of extreme erosion from Irene. Extensive documented cutbank‐initiated landslides into these sediments during Irene corroborate this interpretation (Buraas et al , ; Magilligan et al , ; Dethier et al , ). At Amherst Lake, similarity in peak runoff during Irene and the 1927 storm (Figure e) contrasts with stark differences in resultant sedimentation.…”

Impact of an Extreme Storm Event on River Corridor Bank Erosion and Phosphorus Mobilization in a Mountainous Watershed in the Northeastern United States

“…We estimated that the load of sediment and P from <1 km of stream bank during this one event was similar in magnitude to the long‐term average annual export. High post‐Irene rates of stream bank erosion and sediment transport are likely due, in part, to instability caused by this event (Dethier et al, ). Continued high rates of erosion are probable given the projected increased likelihood of recurring extreme precipitation events (Hayhoe et al, ).…”