We use circulation classifications and spatiotemporal deep neural networks to correct raw daily forecast precipitation by combining large-scale circulation patterns with local spatiotemporal information. We find that the method not only captures the westward and northward movement of the western Pacific subtropical high but also shows substantially higher bias-correction capabilities than existing standard methods in terms of spatial scale, timescale, and intensity.

Numerous modelling approaches can be used for studying karst water resources, which can make it difficult for a stakeholder or researcher to choose the appropriate method. We conduct a comparison of two widely used karst modelling approaches: artificial neural networks (ANNs) and reservoir models. Results show that ANN models are very flexible and seem great for reproducing high flows. Reservoir models can work with relatively short time series and seem to accurately reproduce low flows.

This study demonstrates the fact that the large river flows forecasted by the models show an underestimation that is inversely related to the number of locations where precipitation is recorded, which is independent of the model. The higher the number of points where the amount of precipitation is recorded, the better the estimate of the river flows.

The construction of two dams in the Lower Volta River, Ghana, adversely affected downstream riverine ecosystems and communities. In contrast, Ghana has enjoyed vast economic benefits from the dams. Herein lies the challenge; there exists a trade-off between water for river ecosystems and water for anthropogenic water demands such hydropower. In this study, we quantify these trade-offs and show that there is room for providing environmental flows under current and future climatic conditions.

In seasonal climates with a warm and a cold season, low flows are generated by different processes so that return periods used as a measure of event severity will be inaccurate. We propose a novel mixed copula estimator that is shown to outperform previous calculation methods. The new method is highly relevant for a wide range of European river flow regimes and should be used by default.

To provide a possibility for early warning and forecasting of ponding in the urban drainage system, an optimized long short-term memory (LSTM)-based model is proposed in this paper. It has a remarkable improvement compared to the models based on LSTM and convolutional neural network (CNN) structures. The performance of the corrected model is reliable if the number of monitoring sites is over one per hectare. Increasing the number of monitoring points further has little impact on the performance.

Measuring flows in streams allows us to manage crucial water resources. This work shows the automated application of a dual camera computer vision stream gauging (CVSG) system for measuring streams. Comparing between state-of-the-art technologies demonstrated that camera-based methods were capable of performing within the best available error margins. CVSG offers significant benefits towards improving stream data and providing a safe way for measuring floods while adapting to changes over time.

Design rainfall volumes at different duration and frequencies are necessary for the planning of water-related systems and facilities. As the procedure for deriving these values is subjected to different sources of uncertainty, here we explore different methods to estimate how precise these values are for different duration, locations and frequencies in Germany. Combining local and spatial simulations, we estimate tolerance ranges from approx. 10–60% for design rainfall volumes in Germany.