Testing different machine learning techniques for runoff routing in a highly glacierized Djankuat river basin (the North Caucasus, Russia).

<p>Distributed, physically based modelling of runoff routing in highly glacierized river basins is an extremely complicated task as glacier drainage systems functioning is very sophisticated, close to karst river systems but also dynamically developing within very short time periods. Accordingly, runoff routing of glacier melt water is most often based on the concept of linear storage. The number of reservoirs generally vary from 1 to 3. For example, one  ‘fast’ reservoir for melt  and  rain  in  glacierized  grid  cells in GERM model, three parallel different linear reservoirs representing snow, firn and ice in GSM-SOCONT model.</p><p>Here we test applicability of different machine learning techniques (gradient boosting, random forest, LSTM) for runoff routing in a highly glacierized river basin. We use the data from Djankuat alpine research catchment located in the North Caucasus (Russia) for the period of 2007­­-2019. The dataset contains different parameters measured with an hourly or sub-daily time step: water runoff, conductivity, turbidity, temperature, 18O, D content at the main gauging station; measurements of precipitation amount, standard meteorological parameters and radiation fluxes. Results of snow and ice melting modelling in the Djankuat river basin over a regular net with an hourly time step using energy-balance distributed A-Melt model are also used as input data.</p><p>Total runoff from the Djankuat river basin (1) and meltwater runoff according to isotopic hydrograph separation (2) were chosen as target functions. Different sets of features to predict the target functions were generated from the original time series using different combinations of the input parameters as well as variable lag times. To score different machine learning techniques and sets of features to predict target function we use correlation coefficient, Nash-Sutcliff efficiency index (NSE), root mean square error (RMSE).</p><p>The study was supported by the Russian Foundation for Basic Research, grant No. 20-35-70024</p>

A New Runoff Routing Scheme for Xin’anjiang Model and Its Routing Parameters Estimation Based on Geographical Information

The Xin’anjiang model is a conceptual hydrological model, which has an essential application in humid and semi-humid regions. In the model, the parameters estimation of runoff routing has always been a significant problem in hydrology. The quantitative relationship between parameters of the lag-and-route method and catchment characteristics has not been well studied. In addition, channels in Muskingum method of the Xin’anjiang model are assumed to be virtual channels. Therefore, its parameters need to be estimated by observed flow data. In this paper, a new routing scheme for the Xin’anjiang model is proposed, adopting isochrones method for overland flow and the grid-to-grid Muskingum–Cunge–Todini (MCT) method for channel routing, so that the routing parameters can be estimated according to the geographic information. For the new routing scheme the average overland flow velocity can be determined through the land cover and overland slope, and the channel routing parameters can be determined through channel geometric characteristic, stream order and channel gradient. The improved model was applied at a 90 m grid scale to a nested watershed located in Anhui province, China. The parent Tunxi watershed, with a drainage area of 2692 km2, contains four internal points with available observed streamflow data, allowing us to evaluate the model’s ability to simulate the hydrologic processes within the watershed. Calibration and verification of the improved model were carried out for hourly time scales using hourly streamflow data from 1982 to 2005. Model performance was assessed by comparing simulated and observed flows at the watershed outlet and interior gauging stations. The performance of both original and new runoff routing schemes were tested and compared at hourly scale. Similar and satisfactory performances were achieved at the outlet both in the new runoff routing scheme using the estimated routing parameters and in the original runoff routing scheme using the calibrated routing parameters, with averaged Nash-Sutcliffe efficiency (NSE) of 0.92 and 0.93, respectively. Moreover, the new runoff routing scheme is also able to reproduce promising hydrographs at internal gauges in study catchment with the mean NSE ranging from 0.84 to 0.88. These results indicate that the parameter estimation approach is efficient and the developed model can satisfactorily simulate not only the streamflow at the parent watershed outlet, but also the flood hydrograph at the interior gauging points without model recalibration. This study can provide some guidance for the application of the Xin’anjiang model in ungauged areas.

A New Algorithm For The Grid Cell-Based Runoff Routing Model Based on Travel Time Concept

The grid cell-based routing model has recently been used to simulate direct runoff hydrographs at catchment scales. This study develops a flexible event-based runoff routing algorithm to simulate a direct runoff hydrograph (DRH). The experiment was based on the spatiotemporal inputs of a hydrological data set. The flexibility is based on the time step and grid cell size applied in the original STORE-DHM. Rainfall distribution was obtained using radar data adjusted by the measured point ground, while the runoff yield was determined using the NRCS-CN method. The parameter distribution was captured in the GIS environment as raster data formats. Furthermore, it was converted into ASCII data formats for scripting the routing algorithm using Matlab programming codes. The model algorithm was tested for storm events within two small study river systems in Yogyakarta, Indonesia. One event in each catchment was selected and calibrated to the observed hydrograph, treating the Curve Number (CN) and Manning coefficient (n) values as parameter calibrations. In the end, two events were selected for validation. The proposed routing model algorithm simulates DRHs of all selected events in the study areas with excellent performance. The Nash-Sutcliffe coefficient was greater than 0.75 for all DRH during validation, and the volume bias and peak discharge error were less than 25%. Keywords: Algorithm; Cell-based runoff routing; Travel time; GIS; Direct runoff hydrograph. Copyright (c) 2020 Geosfera Indonesia Journal and Department of Geography Education, University of Jember This work is licensed under a Creative Commons Attribution-Share A like 4.0 International License

Future evolution and uncertainty of river flow regime change in a deglaciating river basin

The flow regimes of glacier-fed rivers are sensitive to climate change due to strong climate–cryosphere–hydrosphere interactions. Previous modelling studies have projected changes in annual and seasonal flow magnitude but neglect other changes in river flow regime that also have socio-economic and environmental impacts. This study employs a signature-based analysis of climate change impacts on the river flow regime for the deglaciating Virkisá river basin in southern Iceland. Twenty-five metrics (signatures) are derived from 21st century projections of river flow time series to evaluate changes in different characteristics (magnitude, timing and variability) of river flow regime over sub-daily to decadal timescales. The projections are produced by a model chain that links numerical models of climate and glacio-hydrology. Five components of the model chain are perturbed to represent their uncertainty including the emission scenario, numerical climate model, downscaling procedure, snow/ice melt model and runoff-routing model. The results show that the magnitude, timing and variability of glacier-fed river flows over a range of timescales will change in response to climate change. For most signatures there is high confidence in the direction of change, but the magnitude is uncertain. A decomposition of the projection uncertainties using analysis of variance (ANOVA) shows that all five perturbed model chain components contribute to projection uncertainty, but their relative contributions vary across the signatures of river flow. For example, the numerical climate model is the dominant source of uncertainty for projections of high-magnitude, quick-release flows, while the runoff-routing model is most important for signatures related to low-magnitude, slow-release flows. The emission scenario dominates mean monthly flow projection uncertainty, but during the transition from the cold to melt season (April and May) the snow/ice melt model contributes up to 23 % of projection uncertainty. Signature-based decompositions of projection uncertainty can be used to better design impact studies to provide more robust projections.

Glacio-hydrological melt and runoff modelling: a limits of acceptability framework for model selection

Glacio-hydrological models (GHMs) underpin our understanding how future climate change will affect river flow regimes in glaciated watersheds. A variety of simplified GHM structures and parameterisations exist, yet the performance of these are rarely quantified at the process-level or with metrics beyond global summary statistics. A fuller understanding of the deficiencies in competing model structures and parameterisations and the ability of models to simulate physical processes require performance metrics utilising the full range of uncertainty information within input observations. Here, the glacio-hydrological characteristics of the Virkisá river basin in southern Iceland are characterised using 33 signatures derived from observations of ice melt, snow coverage and river discharge. The uncertainty of each set of observations are harnessed to define limits of acceptability (LOA), a set of criteria used to objectively evaluate the acceptability of different GHM structures and parameterisations. This framework is used to compare and diagnose deficiencies in three melt and three runoff-routing model structures. Increased model complexity is shown to improve acceptability when evaluated against specific signatures, but does not always result in better consistency across all signatures, emphasising the difficulty in appropriate model selection and the need for multi-model prediction approaches to account for model selection uncertainty. Melt and runoff-routing structures demonstrate a hierarchy of influence on river discharge signatures with melt model structure having the most influence on discharge hydrograph seasonality and runoff-routing structure on shorter-timescale discharge events. None of the tested GHM structural configurations returned acceptable simulations across the full population of signatures. The framework outlined here provides a comprehensive and rigorous assessment tool for evaluating the acceptability of different GHM process hypotheses. Future melt and runoff model forecasts should seek to diagnose structural model deficiencies and evaluate diagnostic signatures of system behaviour using the LOA framework.