Reconstructing five decades of sediment export from two glacierized high-alpine catchments in Tyrol, Austria, using nonparametric regression

Abstract

Abstract. Knowledge on the response of sediment export to recent climate change in glacierized areas in the European Alps is limited, primarily because long-term records of suspended sediment concentrations (SSCs) are scarce. Here we tested the estimation of sediment export of the past five decades using quantile regression forest (QRF), a nonparametric, multivariate regression based on random forest. The regression builds on short-term records of SSCs and long records of the most important hydroclimatic drivers (discharge, precipitation and air temperature – QPT). We trained independent models for two nested and partially glacier-covered catchments, Vent (98 km2) and Vernagt (11.4 km2), in the upper Ötztal in Tyrol, Austria (1891 to 3772 m a.s.l.), where available QPT records start in 1967 and 1975. To assess temporal extrapolation ability, we used two 2-year SSC datasets at gauge Vernagt, which are almost 20 years apart, for a validation. For Vent, we performed a five-fold cross-validation on the 15 years of SSC measurements. Further, we quantified the number of days where predictors exceeded the range represented in the training dataset, as the inability to extrapolate beyond this range is a known limitation of QRF. Finally, we compared QRF performance to sediment rating curves (SRCs). We analyzed the modeled sediment export time series, the predictors and glacier mass balance data for trends (Mann–Kendall test and Sen's slope estimator) and step-like changes (using the widely applied Pettitt test and a complementary Bayesian approach). Our validation at gauge Vernagt demonstrated that QRF performs well in estimating past daily sediment export (Nash–Sutcliffe efficiency (NSE) of 0.73) and satisfactorily for SSCs (NSE of 0.51), despite the small training dataset. The temporal extrapolation ability of QRF was superior to SRCs, especially in periods with high-SSC events, which demonstrated the ability of QRF to model threshold effects. Days with high SSCs tended to be underestimated, but the effect on annual yields was small. Days with predictor exceedances were rare, indicating a good representativity of the training dataset. Finally, the QRF reconstruction models outperformed SRCs by about 20 percent points of the explained variance. Significant positive trends in the reconstructed annual suspended sediment yields were found at both gauges, with distinct step-like increases around 1981. This was linked to increased glacier melt, which became apparent through step-like increases in discharge at both gauges as well as change points in mass balances of the two largest glaciers in the Vent catchment. We identified exceptionally high July temperatures in 1982 and 1983 as a likely cause. In contrast, we did not find coinciding change points in precipitation. Opposing trends at the two gauges after 1981 suggest different timings of “peak sediment”. We conclude that, given large-enough training datasets, the presented QRF approach is a promising tool with the ability to deepen our understanding of the response of high-alpine areas to decadal climate change.

Projecting sediment export from two highly glacierized alpine catchments under climate change: Exploring non-parametric regression as an analysis tool

“…In a previous study, we trained and validated quantile regression forest models to retrospectively estimate SSY at two gauges for the past 5 decades, using the available records of turbidity-derived suspended sediment concentrations (four and 15 years) and long-term records of the predictors, i.e. discharge, precipitation and temperature (Schmidt et al, 2023) (Figure 2, dashed-line box). In the present study, we use these models and apply 120 them to downscaled and bias-corrected EURO-CORDEX temperature and precipitation projections that were used as input data for the glacio-hydrological model AMUNDSEN as well as the discharge projections of AMUNDSEN (Hanzer et al, 2018)(Figure 1).…”

“…As a promising alternative, geoscientific machine-learning approaches have emerged, and have recently been acknowledged for their potential in applications to Earth System Science (Reichstein et al, 2019). Indeed, first studies showed that machine-learning approaches can easily outperform well-known existing models for sediment 100 yield (Gupta et al, 2021;Rahman et al, 2022;Jimeno-Sáez et al, 2022;Schmidt et al, 2023). In a previous study, we have developed and validated a Quantile Regression Forest (QRF) approach to model SSY in two nested highalpine catchments and estimate yields for the past five decades (Schmidt et al, 2023).…”

“…Indeed, first studies showed that machine-learning approaches can easily outperform well-known existing models for sediment 100 yield (Gupta et al, 2021;Rahman et al, 2022;Jimeno-Sáez et al, 2022;Schmidt et al, 2023). In a previous study, we have developed and validated a Quantile Regression Forest (QRF) approach to model SSY in two nested highalpine catchments and estimate yields for the past five decades (Schmidt et al, 2023). This showed that the QRF model outperformed commonly applied sediment rating curves by about 20 % of explained variance, and other studies found that regression trees and Random Forest models (which QRF is based on) even outperformed other 105 machine learning approaches in modeling sediment dynamics (Talebi et al, 2017;Al-Mukhtar, 2019).…”

“…Thus, the present study is motivated to explore QRF to model future SSY based on measurement data, emission scenarios and subsequent hydrological model results. We test the approach in two glacierized high-alpine catchments in the Ötztal in Austria, where projections of future climatological and glacio-hydrological conditions from the AMUNDSEN model are available (Hanzer et al, 2018), and where we have successfully trained and 110 applied QRF models to reconstruct past sediment export, using records of discharge, precipitation and air temperature (Schmidt et al, 2023). The goals of the present study are (i) to derive estimates of future changes in sediment export with respect to trends in annual yields, shifts in the seasonal distribution and the timing of 'peak sediment' and (ii) to assess uncertainties of the model due to known limitations of the QRF method in order to identify the limitations of the approach.…”

Controls on Sediment Transport From a Glacierized Catchment in the Swiss Alps Established Through Inverse Modeling of Geomorphic Processes

“…Furthermore, individual glaciers respond differently to variations in water discharge within the same season (Delaney et al, 2018). In turn, the brevity of the data collection periods and variability in the records pose challenges when establishing the geomorphic processes and effects of climate with available observations (Schmidt et al, 2022(Schmidt et al, , 2023.…”