Channel-floodplain sediment interactions along large rivers: hydrological connectivity and sediment budgets

Abstract:

Understanding the patterns of sediment delivery processes and their budgets between channel and floodplains of large rivers is important because both hydrogeomorphic and biogeochemical alterations in floodplains take place through these interactions. The Amazon River has continuous exchange of sediment with floodplains, which may exceed over 3500 Mt/yr in both directions. However, characterizing the sediment transport and deposition patterns in floodplains and quantifying their budgets still remains a challenge.

In this study, geomorphic units in floodplains are digitized and their hydrological connectivity are assessed by identifying recharge thresholds from the main channel. Historical floodplain recharge records are examined from daily water level data measured at nearby gauge stations by calculating number of days falling in between the connection and disconnection thresholds within a hydrological cycle. Historical recharge patterns of each unit is assessed using Mann-Kendall test. Intensity of hydrological connectivity is further investigated for by building power spectrum of over 15 years water extent time series data through fast Fourier transform, which the power spectral density indicates the intensity of flooding pulses from the main channel.

To quantify the sediment budget stored in floodplains, PALSAR DEM acquired during the lowest water level season is used with the MODIS 8-day composite data. First, shoreline grids derived from MODIS-MNDWI is overlaid on PALSAR image to identify the water level at each floodplain lake unit (h). Total imported Sediment Fluxes (TiSF) entering each floodplain lake during a given period will be calculated as sum of (h_t1-h_t2) x (SSC_(x,y)x1000) x 250², where h_tn is the water level in floodplain lake at time tn; SSC_{(x, y)} denotes sediment concentration at x, y coordinate; 1000 is a scale factor; and 250² is the area of MODIS pixel (m²). Successively summing up TiSF derived from each period will retrieve the amount of total sediment influx from the main channel stored in floodplains.

Reference: Park, E., and E. M. Latrubesse (2014), Modeling suspended sediment distribution patterns of the Amazon River using MODIS data, Remote Sens Environ, 147, 232-242.