Estimating the suspended sediment yield in a river network by means of geomorphic parameters and regression relationships

Abstract

Abstract. An application of regression relationships depending on geomorphic parameters is proposed to predict the amount of the average annual suspended sediment yield at different sections of the drainage network. Simple and multiple regression relationships, utilising the drainage density and the hierarchical anomaly index as independent variables, based on data from 20 river basins of different size located in Italy, are here tested. An application is also shown for a small river basin located in central Italy where it is possible to compare the obtained suspended sediment yield estimates with reservoirs siltation data. The results confirm the potential applicability of regression equations for estimating the suspended sediment yield depending on the topological behaviours of the river network. A discussion of the reliability of the method for ungauged basins is also provided, which puts in light the necessity of additional tests to support the application of the approach to small size watersheds.

Principal Component and Multiple Regression Analyses for the Estimation of Suspended Sediment Yield in Ungauged Basins of Northern Thailand

“…It is worth noting that a few sub-basins could not be delineated in areas that are relatively plains, which create difficulties in delineating the river and basin boundary. The river network's properties, namely the hierarchical anomaly index and the hierarchical anomaly density [1,21,30,52,53], were estimated based on the digitized river network derived from 1:50,000 topographical maps obtained from the Royal Thai Survey Department, which was satisfactorily compared to the existing river network [51]. To elaborate on the river network's properties, let us assume G as the number of first order streams necessary to make a drainage network perfectly ordered in a binary tree-shaped structure with streams of order K flowing into streams of order K + 1, and N is the number of first order streams present in the drainage network.…”

“…When the input data is scarce, the large number of involved parameters may cause significant uncertainty in soil erosion estimates [1]. Furthermore, the simulation of sediment transport at the basin scale is still computationally very expensive.…”

“…Bray and Xie [29] identified six categories of variables that can be related to the processes associated with the generation and delivery of suspended sediment to the basin outlet in Canada, which are hydroclimatic conditions, basin topographic features, land surface features, soil characteristics, channel network features and human activities. Ciccacci et al [30] and Grauso et al [1] investigated the correlation between the sediment yield and some geomorphologic, hydrological and climatic parameters in Italy. They found a significant relationship between average yearly sediment yield per unit watershed area and the drainage density.…”

“…Such effects may disturb the sediment balance in the basin. In particular, sediment that is eroded from sloping areas can accumulate in the river's network, thereby affecting channel water conveyance [1]. Moreover, several problems due to soil erosion, such as the loss of fine and nutrient-rich topsoil that reduces land productivity, as well as the pollution of surface water bodies, are evident [2][3][4][5].…”

“…Several authors have shown the effectiveness of statistical relationships, which allow one to estimate river sediment transport depending on easily available geomorphologic, hydrological and climatic parameters [1,[18][19][20][21][22][23][24][25][26][27][28][29][30][31][32][33]. Sediment yield is controlled by factors that control erosion and sediment delivery, including local topography, soil properties, climate, vegetation cover, catchment morphology, drainage network characteristics and land use [26,28].…”