Data set accompanying the publication "Reconstructing five decades of sediment export from two glaciated high-alpine catchments in Tyrol, Austria, using nonparametric regression".
The data set contains
1. the code of the quantile regression forest model applied in the above mentioned manuscript to reconstruct suspended sediment concentrations and yields for the past five decades at the two nested gauges Vernagt (VF) and Vent (V) in the Upper Ötztal, Austria,
2. the preprocessed input data (discharge, precipitation, air temperature and suspended sediment concentrations derived from turbidity measurements) for both gauges, and
3. the model results.
- The model code is a modified version of this code: https://github.com/TillF/ssc_prediction (Jan 18th 2022)
Modifications:
- extended preprocessing steps in different temporal resolutions
-
includes tuning of aggregation factors, which determine the time span of antecedent hydroclimatic (discharge, temperature, precipitation) conditions included in the calculations
-
The input data are preprocessed versions of the original data which have been provided by the Hydrographic Service of Tyrol, Austria, and the Bavarian Academy of Sciences and Humanity, Munich, Germany, or are already available online (see table A1 in the appendix of the above mentioned manuscript for details):
- gaps have been filled as far as possible, which includes combining and converting data from different stations, and
-
the data have been aggregated or disaggregated to achieve consistent temporal resolutions.
-
The model results can be found in the folder GAUGE_NAME/reconstruction_model/5_model_application/MC_01
- in the temporal resolution of the model (i.e. daily or hourly) as one table per year (e.g. flood_periods_quantForest_GAUGE_NAME_hist1.txt)
- aggregated to annual suspended sediment yields (flood_periods_quantForest_GAUGE_NAME.txt)
