Department Systems Analysis, Integrated Assessment and Modelling

Calibration of precipitation-streamflow models to streamflow signatures is a promising approach for streamflow prediction in ungauged basins (PUB). The estimation of parameter and prediction uncertainty in this case is not trivial because: (a) calibration takes place in the signature domain, while predictions are required in the time domain, and (b) streamflow signatures are estimated (e.g., from donor catchments) rather than “observed” (computed from observed streamflow in the target catchment), and therefore particularly uncertain. This study investigates model calibration using estimated signatures in an Approximate Bayesian Computation framework. First, we construct a stochastic signature transfer model, based on seasonal flow duration curves. Second, we calibrate a precipitation-streamflow model to the estimated signatures, accounting for their uncertainty. The proposed method is tested in six catchments of the Thur basin, Switzerland. Three data availability scenarios are considered: (a) concomitant scenario, where signatures are "observed," (b) non-concomitant scenario, where signatures are transferred from a different time period, and (c) regionalization scenario, where signatures are transferred from (neighboring) donor catchments. In this study, the switch from observed to regionalized signatures increases predictive streamflow uncertainty by 38% and worsens the (deterministic) fit to observations by 17% (in terms of Nash-Sutcliffe efficiency). Despite this deterioration, posterior predictive uncertainty remains lower than prior predictive uncertainty generated using uniform priors over representative parameter ranges (“uncalibrated” model), which demonstrates the effectiveness of the proposed signature-based calibration. More importantly, uncertainty is reliably estimated at the ungauged catchments, which represents a key advance in stochastic streamflow PUB.

The traditional description of a hydrological system with a deterministic, conceptual model and a lumped output error term does not explicitly consider the main mechanisms of uncertainty generation due to approximate process representation, unobserved variability in processes and influence factors, and input uncertainty. In this study, we test the description of such intrinsic uncertainty in conceptual models by making process rates stochastic through stochastic, time-dependent rate parameters. We analyze the advantages and challenges of this approach by using Bayesian inference to jointly estimate model parameters, parameters of the stochastic processes, and the time series of the stochastic parameters. Numerically, we use a particle filter to infer the stochastic time series and to approximate the marginal likelihood for Markov chain sampling of the constant parameters. Compared to the lumped error formulation, we achieve a more realistic description of uncertainty, as we obtain larger errors in intrinsic states, larger uncertainty during prediction than calibration periods (a feature missing for simple lumped error models), and autocorrelated model outputs. However, the additional degrees of freedom introduced with stochastic parameters can lead to an unintentional compensation for model deficits or input errors. This problem is symptomatic of model structure inadequacy or poor selection of the parameters that are made stochastic, and can be diagnosed through cross-validation and careful posterior analysis. The proposed approach is computationally more demanding and its implementation more challenging than the traditional description of hydrological systems using a deterministic model with a lumped error term. However, its advantages both in providing a more realistic representation of uncertainties and in diagnosing model deficiencies suggest its adoption and further development in future studies.

Daily streamflow dynamics can be accurately simulated by conceptual models as simple as a single bucket in some catchments, while they require more complex configurations in other catchments. However, without resorting to calibration, anticipating where and why a given model structure may be appropriate remains difficult. In this work, we explored the feasibility of relating suitable model structures to the climate and streamflow characteristics of 508 catchments in Brazil. Specifically, we tested four model structures using up to three reservoirs, where each reservoir is intended to represent a catchment function: the rainfall-runoff threshold, the fast, and the slow hydrograph response. We hypothesized a relationship between suitable model structures and hydrological signatures of aridity (I_A) and baseflow index (I_B). Our results show that different classes of signatures resulted in distinct patterns of model performance. Wet catchments (I_A < 0.9) with low baseflow (I_B < 0.4) were the easiest to model, with a single-reservoir model presenting a relatively good performance. In the case of low baseflow, adding a rainfall-runoff threshold reservoir resulted in better performance than adding a slow response reservoir, whereas in the case of high baseflow (I_B < 0.6) the opposite occurred. In the case of low baseflow, the inclusion of a slow response reservoir helped the simulation of dry catchments (I_A < 1.1), but not of wet ones, which we attributed to the impact of permeability in dry catchments. These results indicate a path toward model structure identification from streamflow signatures and potentially from landscape features.

Catchment-scale hydrological models are widely used to represent and improve our understanding of hydrological processes and to support operational water resource management. Conceptual models, which approximate catchment dynamics using relatively simple storage and routing elements, offer an attractive compromise in terms of predictive accuracy, computational demands, and amenability to interpretation. This paper introduces SuperflexPy, an open-source Python framework implementing the SUPERFLEX principles (Fenicia et al., 2011) for building conceptual hydrological models from generic components, with a high degree of control over all aspects of model specification. SuperflexPy can be used to build models of a wide range of spatial complexity, ranging from simple lumped models (e.g., a reservoir) to spatially distributed configurations (e.g., nested sub-catchments), with the ability to customize all individual model components. SuperflexPy is a Python package, enabling modelers to exploit the full potential of the framework without the need for separate software installations and making it easier to use and interface with existing Python code for model deployment. This paper presents the general architecture of SuperflexPy, discusses the software design and implementation choices, and illustrates its usage to build conceptual models of varying degrees of complexity. The illustration includes the usage of existing SuperflexPy model elements, as well as their extension to implement new functionality. Comprehensive documentation is available online and provided as a Supplement to this paper. SuperflexPy is available as open-source code and can be used by the hydrological community to investigate improved process representations for model comparison and for operational work.

Models that mimic an original model might have a different model structure than the original model, that affects model output. This study assesses model structure differences and their impact on output by comparing 7 model implementations that carry the name HBV. We explain and quantify output differences with individual model structure components at both the numerical (e.g., explicit/implicit scheme) and mathematical level (e.g., lineair/power outflow). It was found that none of the numerical and mathematical formulations of the mimicking models were (originally) the same as the benchmark, HBV-light. This led to small but distinct output differences in simulated streamflow for different numerical implementations (KGE difference up to 0.15), and major output differences due to mathematical differences (KGE median loss of 0.27). These differences decreased after calibrating the individual models to the simulated streamflow of the benchmark model. We argue that the lack of systematic model naming has led to a diverging concept of the HBV-model, diminishing the concept of model mimicry. Development of a systematic model naming framework, open accessible model code and more elaborate model descriptions are suggested to enhance model mimicry and model development.

River networks are one of the main routes by which the public could be exposed to environmental sources of antibiotic resistance, that may be introduced e.g. via treated wastewater. In this study, we applied a comprehensive integrated analysis encompassing mass-flow concepts, chemistry, bacterial plate counts, resistance gene quantification and shotgun metagenomics to track the fate of the resistome (collective antibiotic resistance genes (ARGs) in a microbial community) of treated wastewater in two Swiss rivers at the kilometer scale. The levels of certain ARGs and the class 1 integron integrase gene (intI1) commonly associated with anthropogenic sources of ARGs decreased quickly over short distances (2-2.5 km) downstream of wastewater discharge points. Mass-flow analysis based on conservative tracers suggested this decrease was attributable mainly to dilution but ARG loadings frequently also decreased (e.g., 55.0-98.5 % for ermB and tetW) over the longest studied distances (6.8 and 13.7 km downstream). Metagenomic analysis confirmed that ARG of wastewater-origin did not persist in rivers after 5 ∼ 6.8 km downstream distance. sul1 and intI1 levels and loadings were more variable and even increased sharply at 5 ∼ 6.8 km downstream distance on one occasion. While input from agriculture and in-situ positive selection pressure for organisms carrying ARGs cannot be excluded, in-system growth of biomass is a more probable explanation. The potential for direct human exposure to the resistome of wastewater-origin thus appeared to typically abate rapidly in the studied rivers. However, the riverine aquatic resistome was also dynamic, as evidenced by the increase of certain gene markers downstream, without obvious sources of anthropogenic contamination. This study provides new insight into drivers of riverine resistomes and pinpoints key monitoring targets indicative of where human sources and exposures are likely to be most acute.

This study documents the development of a semi-distributed hydrological model aimed at reflecting the dominant controls on observed streamflow spatial variability. The process is presented through the case study of the Thur catchment (Switzerland, 1702 km²), an alpine and pre-alpine catchment where streamflow (measured at 10 subcatchments) has different spatial characteristics in terms of amounts, seasonal patterns, and dominance of baseflow. In order to appraise the dominant controls on streamflow spatial variability and build a model that reflects them, we follow a two-stage approach. In a first stage, we identify the main climatic or landscape properties that control the spatial variability of streamflow signatures. This stage is based on correlation analysis, complemented by expert judgement to identify the most plausible cause–effect relationships. In a second stage, the results of the previous analysis are used to develop a set of model experiments aimed at determining an appropriate model representation of the Thur catchment. These experiments confirm that only a hydrological model that accounts for the heterogeneity of precipitation, snow-related processes, and landscape features such as geology produces hydrographs that have signatures similar to the observed ones. This model provides consistent results in space–time validation, which is promising for predictions in ungauged basins. The presented methodology for model building can be transferred to other case studies, since the data used in this work (meteorological variables, streamflow, morphology, and geology maps) are available in numerous regions around the globe.