In this study, the cumulative distribution function approach (CDFA) and the Monte Carlo approach (MCA) were used to develop two new approaches for model evaluation within an uncertainty framework. These proposed methods could be extended to watershed models to provide a substitution for traditional model evaluations within an uncertainty framework.

An unprecedented flash drought took place across parts of the US Northern Great Plains and Canadian Prairie Provinces during the summer of 2017 that in some areas was the worst in recorded history. We show that this drought was preceded by a breakdown of land–atmosphere coupling, reducing the likelihood of convective precipitation. It may be useful to monitor land–atmosphere coupling to track and potentially forecast drought development.

Water depth observations are essential data to forecast flood hazard, predict sediment transport, or monitor in-stream habitats. We retrieved bathymetry with a sonar wired to a drone. This system can improve the speed and spatial scale at which water depth observations are retrieved. Observations can be retrieved also in unnavigable or inaccessible rivers. Water depth observations showed an accuracy of ca. 2.1 % of actual depth, without being affected by water turbidity or bed material.

Kilometre-scale climate-model data are of great benefit to both hydrologists and end users studying extreme precipitation, though often unavailable due to the computational expense associated with such high-resolution simulations. We develop a method which identifies days with enhanced risk of extreme rainfall over a catchment, so that high-resolution simulations can be performed only when such a risk exists, reducing computational expense by over 90 % while still well capturing the extremes.

Long-range flood forecast is very challenging over monsoonal regions. We investigate the potential of extending seasonal predictability of Yangtze River summer floods by using atmospheric moisture flux prediction with a perfect model assumption. The finding suggests a potential to use atmospheric moisture flux which is more predictable than precipitation at the seasonal timescale, to improve flood forecasting and early warning, especially during post-El Niño summers.

Many chemicals such as pesticides, pharmaceuticals or household chemicals impair water quality in many areas worldwide. Measuring pollution everywhere is too costly. Models can be used instead to predict where high pollution levels are expected. We tested a model that can be used across large river basins. We find that for the selected chemicals predictions are generally within a factor of 2 to 4 from observed concentrations. Often, knowledge about the chemical use limits the predictions.

We present a multivariate application of the ensemble Kalman filter (EnKF) in hydrological modeling of a real-world hillslope test case with dominant unsaturated dynamics and strong nonlinearities. Overall, the EnKF is able to correctly update system state and soil parameters. However, multivariate data assimilation may lead to significant tradeoffs between model predictions of different variables, if the observation data are not high quality or representative.

Groundwater contamination by nitrate is a global problem. Understanding sources and transformations of nitrate in aquifers is key to inform remediation, and isotopes of nitrate (¹⁵N, ¹⁸O) are used for this. Most isotopic investigations are synoptic studies. Here we show that high-frequency nitrate isotope sampling of aquifer recharge better reveals the impact of short-term changes in land use practice or climatic conditions that need to be taken into account when considering long-term trends.

We have developed a new mobile automated water sampling device for environmental research and other applications where waters need to be tested for compliance with environmental/health regulations. It has two main advantages over similar devices: firstly, it injects water samples directly into airtight vials to prevent any change in sample properties through contamination, evaporation and gas exchange. Secondly, it can hold up to 160 sample vials, while other devices only hold up to 24 vials.

The hydrology of high-elevation headwaters in midlatitudes is typically dominated by snow processes, which are very sensitive to changes in energy inputs at the top of the snowpack. We present a data analyses that reveal how snowmelt and transpiration waves induced by the diurnal solar cycle generate water pressure fluctuations that propagate through the snowpack–hillslope–stream system. Changes in diurnal energy inputs alter these pressure cycles with potential ecohydrological consequences.

Winds carry air moisture from one place to another. Thus, land-use change that alters air moisture content can also modify downwind rainfall and distant river flows. This aspect has rarely been taken into account in studies of river flow changes. We show here that remote land-use change effect on rainfall can exceed that of local, and that foreign nation influence on river flows is much more prevalent than previously thought. This has important implications for both land and water governance.

Both radar reflectivity and GTS data are good choices for assimilation in improving high-resolution rainfall of the NWP systems, which always fails in providing satisfactory rainfall products for hydrological use. Simultaneously assimilating GTS and radar data always performs better than assimilating radar data alone. The assimilation efficiency of the GTS data is higher than both radar reflectivity and radial velocity considering the number of data assimilated and its effect.

Accurate monitoring of surface water extent is essential for hydrological investigation of small lakes (1–10 ha), which supports millions of smallholder farmers. Landsat monitoring of long-term surface water dynamics is shown to be suited to lakes over 3 ha based on extensive hydrometric data from seven field sites over 15 years. MNDWI water classification optimized here for the specificities of small water bodies reduced mean surface area errors by 57 % compared to published global datasets.

We suggested and tested an exploratory approach for the detection of dominant changes in multivariate water quality data sets with irregular sampling in space and time. The approach is especially recommended for the exploratory assessment of existing long-term low-frequency multivariate water quality monitoring data.

We developed 60 modelling chain combinations based on either experimentalists' (bottom-up) or modellers' (top-down) thinking and forced them with data of increasing accuracy. Results showed that the differences in performance arising from the forcing data were due to compensation effects. We also found that modellers' and experimentalists' concept of model realism differs, and the level of detail a model should have to reproduce the processes expected must be agreed in advance.

Using the arithmetic averages of precipitation isotope values, Wu et al. (2017) concluded that the Badain Jaran Desert (BJD) groundwater is recharged by modern local meteoric water. However, based on weighted mean precipitation isotope values, our further analysis shows that modern precipitation on the Qilian Mountains is more likely to be the main source of the groundwater and lake water in the BJD, as found. We believe this comment provides an important improvement for their study.

In this study we investigated whether climate change and wildfires are likely to alter water quality of streams in western boreal Canada, a region that contains large permafrost-affected peatlands. We monitored stream discharge and water quality from early snowmelt to fall in two streams, one of which drained a recently burned landscape. Wildfire increased the stream delivery of phosphorous and possibly increased the release of old natural organic matter previously stored in permafrost soils.

Satellite soil moisture products can provide critical information on incipient droughts and the interplay between vegetation and water availability. However, time-variant systematic errors in the soil moisture products may impede their usefulness. Using a novel statistical approach, we detect such errors (associated with changing vegetation) in the SMAP soil moisture product. The vegetation-associated biases impede drought detection and the quantification of vegetation–water interactions.

Precipitation events associated with rain and snow on the eastern slopes of the Rocky Mountains, Canada, are a critical aspect of the regional water cycle. The goal is to characterize the precipitation and weather conditions in the Kananaskis Valley, Alberta, during a field experiment. Mainly dense solid precipitation reached the surface and occurred during downslope and upslope conditions. The precipitation phase has critical implications on the severity of flooding events in the area.

Observing the amount of water evaporated in nature is not easy, and we need to combine accurate local measurements with estimates from satellites, more uncertain but covering larger areas. This is the main topic of our paper, in which local observations are compared with global land evaporation estimates, followed by a weighting of the global observations based on this comparison to attempt derive a more accurate evaporation product.

Based on warm season data from 259 weather stations across the US, we analyze the correlation between precipitation, potential evaporation, and “apparent” potential evaporation (measured by pan evaporation). Over 93 % of the stations show negative correlation between precipitation and apparent potential evaporation, but no clear relationship is shown between precipitation and potential evaporation. The collected data points follow the trend of the newly derived Bouchet–Budyko curve.

In Africa field-based meteorological data are scarce; therefore global data sources based on remote sensing and climate models are often used as alternatives. To assess their suitability for a large and topographically complex area in East Africa, we evaluated multiple climate data products with available ground station data at multiple timescales over 21 regions. The comprehensive evaluation resulted in identification of preferential data sources to be used for climate and hydrological studies.

By realizing that hydrological models are not one single hypothesis, but an assemblage of many hypotheses, new ways to scrutinize hydrological models are needed. Up until now, studies concentrate on comparing existing models or built models incrementally. This approach here tries to tackle the problem the other way around. We construct a complex model, containing all processes important for the catchment, and deconstruct it step by step to understand the influence of single processes.

The Kling and Gupta efficiency (KGE) is a score used in hydrology to evaluate flow simulation compared to observations. In order to force the evaluation on the low flows, some authors used the log-transformed flow to calculate the KGE. In this technical note, we show that this transformation should be avoided because it produced numerical flaws that lead to difficulties in the score value interpretation.

It is a critical step to gap-fill streamflow data for most hydrological studies, such as streamflow trend, flood, and drought analysis and predictions. However, quantitative evaluation of the gap-filled data accuracy is not available. Here we conducted the first comprehensive study, and found that when the missing data rate is less than 10 %, the gap-filled streamflow data using hydrological models are reliable for annual streamflow and its trend analysis.