PubAg

Main content area

Dynamics of erosion and suspended sediment transport from drained peatland forestry

Author:
Marttila, Hannu, Kløve, Bjørn
Source:
Journal of hydrology 2010 v.388 no.3-4 pp. 414-425
ISSN:
0022-1694
Subject:
drainage channels, forestry, forests, hysteresis, mechanics, models, prediction, rain, regression analysis, runoff, sediment transport, sediment yield, sediments, snowmelt, soil erosion, spring, summer, transposons, watersheds
Abstract:
Erosion and suspended sediment transport dynamics were studied in the open water season (April-November) of three consecutive years in a drained peatland forest. Discharge and suspended solids were monitored continuously and sediment properties surveyed with individual samples. Sediment transport dynamics were studied using statistical methods and plotted as hysteresis, duration and effective discharge plots. Sediment rating curves for averaged daily data for months and total season were established to seek empirical sediment yield predictive model(s). The discharge patterns in the study catchment were dominated by peak runoff events resulting from snowmelt and intensive rainfall. Fluctuations in suspended sediment concentrations (SSC) were associated with discharge fluctuations. Sediment transport varied markedly during the study months and years, with organic sediment playing an important role. The present study demonstrates sediment transport processes and reveals the underlying mechanics operating in forested peatland areas. These include inter-storm sediment storage, bank collapse and the role of peak runoff rate on erosion in peak flow events. Sediment availability in drainage network channels also played a major role in erosion and SSC. The results indicate that summer runoff peaks can be a dominant element controlling annual sediment yield, partly contradicting previous studies in drained peatland areas identifying spring snowmelt as the dominant element. Sediment yield prediction with the empirical sediment-transport rating curve gave good regressions for some data, but simple regression models were not adequate to provide accurate annual predictions.
Agid:
417686