Evaporation is the biggest water consumer of the rainfall that falls on the land. Knowing its magnitude will help water resources to develop water use strategies. This study describes a model that can estimate the magnitude of evaporation on a global level. It does not use local information, only information from rainfall and vegetation patterns derived from satellites.

We improved lake mixing process simulations by applying a vertical mixing scheme, K profile parameterization (KPP), in the Community Land Model (CLM) version 4.5, developed by the National Center for Atmospheric Research. The current vertical mixing scheme in CLM requires an arbitrarily enlarged eddy diffusivity to enhance water mixing. The coupled CLM-KPP considers a boundary layer for eddy development. The improved lake model provides an important tool for lake hydrology and ecosystem studies.

We investigate the influence of anticipated climate change on water security in the Jaguari Basin, which is the main source of freshwater for 9 million people in the São Paulo metropolitan region. Our findings indicate an expansion of the basin critical period, and identify October and November as the most vulnerable months. There is an urgent need to implement efficient mitigation and adaptation policies that recognize the annual pattern of variation between insecure and secure periods.

The combined effects of changes in climate, vegetation, and soils on mountain hydrology were modeled in three mountain basins. In the Yukon, an insignificant increasing effect of vegetation change on snow was found to be important enough to offset the climate change effect. In the Canadian Rockies, a combined effect of soil and climate change on runoff became significant, whereas their individual effects were not significant. Only vegetation change decreased runoff in the basin in Idaho.

Over semi-arid regions, it is essential to have a correct representation of the groundwater processes in climate modelling. We present a land surface and groundwater model that incorporates groundwater–soil interactions, groundwater–rivers flow and lateral transport at the subsurface. We study the groundwater influence on soil moisture distribution and memory, and on evapotranspiration in the Iberian Peninsula. Shallow water table regions persist and provide water to the surface during droughts.

Digital and thermal images taken near midday were used to scale daily point observations of key factors driving actual-evaporation estimates across a complex Canadian Prairie landscape. Point estimates of actual evaporation agreed well with observed values via eddy covariance. Impacts of spatial variations on areal estimates were minor, and no covariance was found between model parameters driving the energy term. The methods can be applied further to improve land surface parameterisations.

This study presents an analysis of 135 soil moisture profiles for identification of the spatial and temporal preferential flow occurrence in a complex landscape. Especially dry conditions and high rainfall intensities were found to increase preferential flow occurrence in soils. This results in a seasonal pattern of preferential flow with a higher occurrence in summer. During this time grasslands showed increased flow velocities, whereas forest sites exhibited a higher amount of bypass flow.

While conventional approaches focus on flood designs at individual locations, there are many situations requiring an understanding of spatial dependence of floods at multiple locations. This research describes a new framework for analyzing flood characteristics across civil infrastructure systems, including conditional and joint probabilities of floods. This work leads to a new flood estimation paradigm, which focuses on the risk of the entire system rather than each system element in isolation.

Flowing stream networks extend and retract seasonally and in response to precipitation. This affects the distances and thus the time that it takes a water molecule to reach the flowing stream and the stream outlet. When the network is fully extended, the travel times are short, but when the network retracts, the travel times become longer and more uniform. These dynamics should be included when modeling solute or pollutant transport.

A new stochastic rainfall model evaluation framework is introduced, with three key features: (1) streamflow-based, to directly evaluate modelled streamflow performance, (2) virtual, to avoid confounding errors in hydrological models or data, and (3) targeted, to isolate errors according to specific sites/months. The framework identified the importance of rainfall in the wetting-up months for providing reliable predictions of streamflow over the entire year despite their low flow volumes.

This study evaluated the performance of 36 CMIP5 GCMs in simulating seasonal precipitation and maximum and minimum temperature over Pakistan using spatial metrics (SPAtial EFficiency, fractions skill score, Goodman–Kruskal's lambda, Cramer's V, Mapcurves, and Kling–Gupta efficiency) for the period 1961–2005. NorESM1-M, MIROC5, BCC-CSM1-1, and ACCESS1-3 were identified as the most suitable GCMs for simulating all three climate variables over Pakistan.

Water management in mountains is challenging when there are no climate data of good quality. This can be addressed by using statistical methods or by including alternative sources of information. This project tests a relatively complex statistical method and compares it with simpler alternatives while including satellite data. It was found that the simple alternative may behave as well as the complex one, and how good the alternative sources of information are could also be established.

In this paper we address one of the big challenges in snow hydrology, namely the accurate simulation of the seasonal snowpack in ungauged regions. We do this by assimilating satellite observations of snow cover into a modelling framework. Importantly (and a novelty of the paper), we include a clustering approach that permits highly efficient ensemble simulations. Efficiency gains and dependency on purely global datasets, means that this method can be applied over large areas anywhere on Earth.

Dissolved methane in Lake Kivu (East Africa) represents a precious energy deposit, but the high gas loads have also been perceived as a threat by the local population. Our measurements confirm the huge amount of methane and carbon dioxide present, but do not support the current theory of a significant recharge. Direct measurements of gas pressure indicate no imminent danger due to limnic eruptions. A continuous survey is mandatory to support responsible action during industrial exploitation.

We characterised for the first time the rainfall microphysics for Southern Hemisphere temperate latitudes. Co-located instruments were deployed to provide information on the sampling effect and spatio-temporal variabilities at micro scales. Substantial differences were found across the instruments, increasing with increasing values of the rain rate. Specific relations for reflectivity–rainfall are presented together with related uncertainties for drizzle and stratiform and convective rainfall.

A database of soil moisture measurements from the middle Heihe River basin of China was used to test the potential of a soil moisture database in estimating the soil water balance components (SWBCs). We determined SWBCs using a method that combined the soil water balance method and the inverse Richards equation. This work confirmed that relatively reasonable estimations of the SWBCs in coarse-textured sandy soils can be derived using soil moisture measurements.

This paper examines the influence of the subsurface structure on water routing, water residence times, and the hydrologic response of distinct groundwater stores and further investigates their contribution to streamflow. We conclude that deep groundwater from the fractured aquifer system, rather than shallow groundwater, is the dominant source of streamflow, which highlights the need to better characterize the deep subsurface of mountain systems using interdisciplinary studies such as this one.

High-resolution regional climate modeling that resolves convection was conducted over western Canada for the current climate and a high-end greenhouse gas emission scenario by 2100. The simulation demonstrates its good quality in capturing the temporal and spatial variation in the major hydrometeorological variables. The warming is stronger in the northeastern domain in the cold seasons. It also shows a larger increase in high-intensity precipitation events than moderate and light ones by 2100.

This work demonstrates a new approach to obtain flood level trend information from surveillance footage with minimal prior information. A neural network trained to detect flood water is applied to video frames to create a qualitative flooding metric (namely, SOFI). The correlation between the real water trend and SOFI was found to be 75 % on average (based on six videos of flooding under various circumstances). SOFI could be used for flood model calibration, to increase model reliability.

This study explores novel modelling avenues using machine learning in combination with process-based models to predict the shallow water table at high spatial resolution. Due to climate change and anthropogenic impacts, the shallow groundwater is rising in many parts of the world. In order to adapt to risks induced by groundwater flooding, new modelling tools need to emerge. In this study, we found that machine learning is capable of reaching the required accuracy and resolution.

It is important to understand historic droughts in order to plan and prepare for possible future events. In this study we use the standardised streamflow index for 1891–2015 to systematically identify, characterise and rank hydrological drought events for 108 near-natural UK catchments. Results show when and where the most severe events occurred and describe events of the early 20th century, providing catchment-scale detail important for both science and planning applications of the future.

Recently, the first global groundwater models were developed to better understand surface-water–groundwater interactions and human water use impacts. However, the reliability of model outputs is limited by a lack of data as well as model assumptions required due to the necessarily coarse spatial resolution. In this study we present the first global maps of model sensitivity according to their parameterization and build a foundation to improve datasets, model design, and model understanding.