TY - JOUR A1 - Ayzel, Georgy A1 - Heistermann, Maik T1 - The effect of calibration data length on the performance of a conceptual hydrological model versus LSTM and GRU BT - a case study for six basins from the CAMELS dataset JF - Computers & geosciences : an international journal devoted to the publication of papers on all aspects of geocomputation and to the distribution of computer programs and test data sets ; an official journal of the International Association for Mathematical Geology N2 - We systematically explore the effect of calibration data length on the performance of a conceptual hydrological model, GR4H, in comparison to two Artificial Neural Network (ANN) architectures: Long Short-Term Memory Networks (LSTM) and Gated Recurrent Units (GRU), which have just recently been introduced to the field of hydrology. We implemented a case study for six river basins across the contiguous United States, with 25 years of meteorological and discharge data. Nine years were reserved for independent validation; two years were used as a warm-up period, one year for each of the calibration and validation periods, respectively; from the remaining 14 years, we sampled increasing amounts of data for model calibration, and found pronounced differences in model performance. While GR4H required less data to converge, LSTM and GRU caught up at a remarkable rate, considering their number of parameters. Also, LSTM and GRU exhibited the higher calibration instability in comparison to GR4H. These findings confirm the potential of modern deep-learning architectures in rainfall runoff modelling, but also highlight the noticeable differences between them in regard to the effect of calibration data length. KW - Artificial neural networks KW - Calibration KW - Deep learning KW - Rainfall-runoff KW - modelling Y1 - 2021 U6 - https://doi.org/10.1016/j.cageo.2021.104708 SN - 0098-3004 SN - 1873-7803 VL - 149 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Cao, Xianyong A1 - Herzschuh, Ulrike A1 - Telford, Richard J. A1 - Ni, Jian T1 - A modern pollen-climate dataset from China and Mongolia: assessing its potential for climate reconstruction JF - Review of palaeobotany and palynology : an international journal N2 - A modern pollen dataset from China and Mongolia (18-52 degrees N, 74-132 degrees E) is investigated for its potential use in climate reconstructions. The dataset includes 2559 samples, 229 terrestrial pollen taxa and four climatic variables - mean annual precipitation (P-ann): 35-2091 mm, mean annual temperature (T-ann): -12.1-25.8 degrees C, mean temperature in the coldest month (Mt(co).): -33.8-21.7 degrees C, and mean temperature in the warmest month (Mt(wa)): 03-29.8 degrees C. Modern pollen-climate relationships are assessed using canonical correspondence analysis (CCA), Huisman-Olff-Fresco (HOF) models, the modern analogue technique (MAT), and weighted averaging partial least squares (WA-PLS). Results indicate that P-ann is the most important climatic determinant of pollen distribution and the most promising climate variable for reconstructions, as assessed by the coefficient of determination between observed and predicted environmental values (r(2)) and root mean square error of prediction (RMSEP). Mt(co) and Mt(wa) may be reconstructed too, but with caution. Samples from different depositional environments influence the performance of cross-validation differently, with samples from lake sediment-surfaces and moss polsters having the best fit with the lowest RMSEP. The better model performances of MAT are most probably caused by spatial autocorrelation. Accordingly, the WA-PLS models of this dataset are deemed most suitable for reconstructing past climate quantitatively because of their more reliable predictive power. (C) 2014 Elsevier B.V. All rights reserved. KW - Calibration KW - Huisman-Olff-Fresco models KW - MAT KW - Pollen-climate transfer function KW - Spatial autocorrelation KW - WA-PLS Y1 - 2014 U6 - https://doi.org/10.1016/j.revpalbo.2014.08.007 SN - 0034-6667 SN - 1879-0615 VL - 211 SP - 87 EP - 96 PB - Elsevier CY - Amsterdam ER -