The Impacts of Armoring Our Deltas: Mapping and Modeling Large-Scale Deltaplain Aggradation

Tuesday, 16 December 2014: 1:55 PM
Irina Overeem1, Stephanie Higgins2, James P Syvitski1, Albert J Kettner3 and Robert Brakenridge4, (1)University of Colorado at Boulder, CSDMS/INSTAAR, Boulder, CO, United States, (2)University of Colorado at Boulder, Boulder, CO, United States, (3)University of Colorado, Boulder, CO, United States, (4)University of Colorado, CSDMS, Boulder, CO, United States
Humans have hardened land-water boundaries in almost every delta they live on. Engineering includes stabilizing and embanking channels to protect from river floods, building dikes around islands and emerging bars to reclaim land, and putting up sea walls to protect from waves and storm surges. These measures aim to reduce the exchange of water and sediment between the distributary delta channel network and the adjacent deltaplain. To first order, armoring of deltas results in net elevation loss of the floodplain, due to subsidence, compaction and reduced aggradation. Here, we ask what are the mechanisms of aggradation in ‘armored’ deltas? How do aggradation patterns compare to more natural depositional patterns?

We analyze 2-week period aggregates of MODIS satellite data from 2000 onwards to map inundation patterns due to irrigation, river floods and storm surges for selected deltas. Using a MODIS band-ratio, we assess relative concentrations of suspended sediment in stagnant water on the floodplains.

In addition, we use a simple approach to route sediment through the delta distributary network based on the relative channel geometries. A depositional process model then calculates cross-channel sediment flux as an exponential decay function, and determines sediment deposition over inundated areas.

Stacked inundation maps show vast areas of deltaplains have flooded between 2000-2014, despite armoring channels with dikes, and coastlines with seawalls. Flooding is caused by overtopping of levees and more rarely by breaching and in those latter cases the flooded areas are often locally constrained. In Asian deltas, rice paddy irrigation with floodwater can be mapped even in the more distal floodplain. Our model predicts that inundated areas still receive significant amounts of fresh sediment, but that the pattern is more variable than in natural systems. Sparse in-situ observations of floodplain aggradation rates and storm surge deposits corroborate high, but localized sedimentation rates.