Hindcasting and Forecasting Total Suspended Sediment Concentrations Using a NARX Neural Network

Vladimir J Alarcon

doi:10.3390/su13010363

Abstract

Estimating and forecasting suspended sediments concentrations in streams constitutes a valuable asset for sustainable land management. This research presents the development of a non-linear autoregressive exogenous neural network (NARX) for forecasting sediment concentrations at the exit of Francia Creek watershed (Valparaiso, Chile). Details are presented on input data selection, data splitting, selection of model architecture, determination of model structure, NARX training (optimization of model parameters), and model validation (hindcasting and forecasting). The study explored if the developed artificial neural network model is valid for forecasting daily suspended sediment concentrations for a complete year, capturing seasonal trends, and maximum and baseflow concentrations. Francia Creek watershed covers approximately 3.24 km2. Land cover within the catchment consists mainly of native and exotic vegetation, eroded soil, and urban areas. Input data consisting of precipitation and stream flow time-series were fed to a NARX network for forecasting daily suspended sediments (SST) concentrations for years 2013–2014, and hindcasting for years 2008–2010. Training of the network was performed with daily SST, precipitation, and flow data from years 2012 and 2013. The resulting NARX net consisted of an open-loop, 12-node hidden layer, 100 iterations, using Bayesian regularization backpropagation. Hindcasting of daily and monthly SST concentrations for years 2008 through 2010 was successful. Daily SST concentrations for years 2013 and 2014 were forecasted successfully for baseflow conditions (R2 = 0.73, NS = 0.71, and Kling-Gupta efficiency index (K-G) = 0.84). Forecasting daily SST concentrations for year 2014 was within acceptable statistical fit and error margins (R2 = 0.53, NS = 0.47, K-G = 0.60, d = 0.82). Forecasting of monthly maximum SST concentrations for the two-year period (2013 and 2014) was also successful (R2 = 0.69, NS = 0.60, K-G = 0.54, d = 0.84).

Highlights

Healthy soils are substrates for the habitats of innumerable living organisms
The actual test of a neural network is during the validation phase, in which the NARX predictions are compared to independent datasets of observed suspended sediment (SST) concentrations
The modeling strategy and the resulting neural network can be performed using a non-linear autoregressive artificial neural network with exogewere applied to estimate SST concentrations occurring at Francia Creek (Valparaiso, nous inputs (NARX)

Summary

Introduction

Healthy soils are substrates for the habitats of innumerable living organisms. Besides supplying basic resources for survival (food, water, nutrients, and raw materials), soils provide several ecosystem services. Soil erosion and soil loss affect soil health by substantially modifying soil structure and causing significant damage to natural habitats and biodiversity [1]. Urbanization has a potential to disturb soil in large areas, which may cause substantial erosion if the principles of sustainable land use and land cover management are not applied [2]. Soil loss (under any land management scenario) occurs by weathering of soil during or after rain events. Run-off transports the eroded soil to nearby streams increasing the amount of soil in the water column (suspended sediment concentration). Total suspended sediment (SST) concentrations in the streams that transport the eroded soil, reflect how a particular land management scenario affects soil loss [3]

Methods

Results

Conclusion