Rivers are shaped by water and sediment inputs, and changes in either of these (from land use, river regulation, or shifting climate) have the potential to drive changes in how rivers look and behave. Numerical modeling offers one of the best options for understanding how the changes we make to watersheds might affect how river systems operate. Unfortunately, river evolution models are often complex and difficult to run over the timescales of river change: decades to a century at the scale of river reaches several kilometers long. To address this, I developed a simplified river evolution model that can predict how channels are likely to respond to shifting water and sediment supply using easily-measured channel geometry metrics that can be extracted from geospatial datasets.
For further reading: Kasprak A, Brasington J, Hafen K, Williams R, Wheaton JM. 2019. Modelling braided river morphodynamics using a particle travel distance framework. Earth Surface Dynamics (ESurfD) DOI: 10.5194/esurf-2018-17.
Results of a one-year simulation of channel change on a braided, gravel-bed river in Scotland, showing actual field-measured elevation changes (left), simulated changes (center), and inferred processes driving change (right).
In-channel large woody debris promotes quality aquatic habitat through sediment sorting, pool scouring, and in-stream nutrient retention and transport. LWD recruitment can occur by numerous ecological and geomorphic mechanisms including channel migration, mass wasting, and natural tree fall, yet LWD sourcing on the watershed scale remains poorly understood. We have developed a rapid and spatially extensive method for using high-resolution light detection and ranging (lidar) elevation data to (1) constrain tree height throughout a watershed, (2) determine the likelihood for streams to recruit channel-spanning trees at reach scales, (3) establish whether adjacent tree fall, mass wasting, or channel migration may be the dominant mechanism for delivery of LWD, and (4) understand the past and future role of LWD at the watershed scale. We utilized this method on the 78 km long Narraguagus River in coastal Maine.
For more information see: Kasprak A, Magilligan FJ, Nislow KH, Snyder NP. 2012. A lidar-derived evaluation of watershed-scale large woody debris sources and recruitment mechansisms: coastal Maine, USA. River Research and Applications 28: 1462-1476. DOI: 10.1002/rra.1532.
Clipped portion of lidar-derived vegetation height DEM. Vegetation pixels capable of spanning adjacent channel are shown in green. Base is an orthophoto from 2007.
Increased fine (less than 2 mm) sediment in streams has been attributed to numerous factors, including anthropogenic activities. In this case, fine sediment production and delivery is believed to result from a loss of soil cohesion stemming from deforestation for timber harvest or road construction. Increases in fine sediments are particularly detrimental to spawning salmonids, and a great deal of research has been performed in high-gradient mountain streams of the Pacific Northwest. Comparatively less is known about the impact of timber harvest on fine sediment delivery to the lower-gradient streams of coastal Maine. We utilized field-based stream surveys (pebble counts, embeddedness measurements, shelter space surveys) augmented with isotopic analysis (210Pb and 7Be) to measure the quantity and temporal flux of fine sediment at various sites in the Narraguagus River basin in coastal Maine. These measurements were then correlated with the amount of area harvested upstream from each field site to determine the influence of timber harvest on fine sediment delivery.
For more Information see: Kasprak A, Magilligan FJ, Nislow KH, Renshaw CE, Snyder NP, Dade WB. 2013. Differentiating the relative importance of land cover change and geomorphic processes on fine sediment sequestration in a logged watershed. Geomorphology 185: 67-77. DOI: 10.1016/j.geomorph.2012.12.005.
Harvested areas in Sinclair Brook, Maine watershed since 1981. Base is an orthophoto from 2007.
We performed a stratigraphic survey on the inlet delta of Beddington Lake, a mainstem pond on the Narraguagus River in coastal Maine. Traditional soil pits, sediment cores, and ground penetrating radar surveys were utilized. Beddington Lake underwent a water level transgression due to the installation of a dam on the lake in ~1850. When the dam was removed in 1951, a subsequent water level regression occurred. Delta stratigraphy revealed differences in sediment physical characteristics (grain size, bulk density, loss-on-ignition) between pre-dam, dam-influenced, and post-dam sedimentary packages. Sedimentation rates over each historic period were also quantified.
Historic water extents in Beddington Lake, Maine measured from aerial photos. Base is an orthophoto from 1996.