Water demand forecasting is a crucial task in the efficient management of the water supply system. This paper compares classical and adapted machine learning algorithms used for water usage predictions including ARIMA, support vector regression, random forests and extremely randomized trees. These models were enriched with human mobility data to improve the predictive power of water demand forecasting. Furthermore, a framework for processing mobility data into time-series correlated with water usage data is proposed. This study uses 51 days of water consumption readings and over 7 million geolocated mobility records from urban areas. Results show that using human mobility data improves water demand prediction. The best forecasting algorithm employing a random forest method achieved 90.4% accuracy (measured by the mean absolute percentage error) and is better by 1% than the same algorithm using only water data, while classic ARIMA approach achieved 90.0%. The Blind (copying) prediction achieved 85.1% of accuracy.

The Sentinel-1 constellation provides an effective new radar instrument with a short revisit time of six days for the monitoring of intensive mining surface deformations. Our goal is to investigate in detail and to bring new comprehension of the mine life cycle. The dynamics of mining, especially in the case of horizontally evolving longwall technology, exhibit rapid surface changes. We use the classical approach of differential radar interferometry (DInSAR) with short temporal baselines (six days), which results in deformation maps with a low decorrelation between the satellite images. For the same time intervals, we compare the radar results with prediction models based on the Knothe–Budryk theory for mining subsidence. The validation of the results with ground levelling measurements reveals a high level of resemblance of the DInSAR subsidence maps (−0.04 m bias with respect to the levelling). On the other hand, aside from the explicable exaggeration, the location of the subsidence trough needs improvement in the forecasted deformations (0.2 km shift in location, a deformation velocity four times higher than in DInSAR). In addition, a time lag between DInSAR (compatible with extraction) and prediction is revealed. The model improvement can be achieved by including the DInSAR results in the elaboration of the model parameters.

Movement analysis has become an integral part of many disciplines, yet with relatively little overlap. A foresight paper in this journal entitled “Towards an integrated science of movement: converging research on animal movement ecology and human mobility science” argued for a better integration of concepts across the divide of animal and human movement, which would lead to the Integrated Science of Movement, but did so from a top-down perspective based on a series of expert workshops. We argue that for a solid establishment of the Integrated Science of Movement, a bottom-up approach is necessary, one based on existing literature which identifies similarities and differences across disciplines. We therefore review, compare, and contrast movement analysis methodologies from GIScience, movement ecology, geography, transportation, public health, computer science, and physics. We structure our review along the dichotomy of individual versus population-based movement or, using terminology from wildlife ecology, between the Lagrangian and Eulerian perspectives. We further introduce a new unifying framework for movement research that is sufficiently general to cover any type of movement study in any discipline and that spans the Lagrangian/Eulerian divide, with the ambitious goal to bridge the gap between disciplines and lay a solid foundation for a new Integrated Science of Movement.

Nowcasting of severe weather events and summer storms, in particular, are intensively studied as they have great potential for large economic and societal losses. Use of Global Navigation Satellite Systems (GNSS) observations for weather nowcasting has been investigated in various regions. However, combining the vertically integrated water vapour (IWV) with vertical profiles of wet refractivity derived from GNSS tomography has not been exploited for short-range forecasts of storms. In this study, we introduce a methodology to use the synergy of IWV and tomography-based vertical profiles to predict 0–2 h of storms using a machine learning approach for Poland. Moreover, we present an analysis of the importance of features that take part in the prediction process. The accuracy of the model reached over 87%, and the precision of prediction was about 30%. The results show that wet refractivity below 6 km and IWV on the west of the storm are among the significant parameters with potential for predicting storm location. The analysis of IWV demonstrated a correlation between IWV changes and storm occurrence.

Predictability of human movement is a theoretical upper bound for the accuracy of movement prediction models, which serves as a reference value showing how regular a dataset is and to what extent mobility can be predicted. Over the years, the predictability of various human mobility datasets was found to vary when estimated for differently processed datasets. Although attempts at the explanation of this variability have been made, the extent of these experiments was limited. In this study, we use high-precision movement trajectories of individuals to analyse how the way we represent the movement impacts its predictability and thus, the outcomes of analyses made on these data. We adopt a number of methods used in the last 11 years of research on human mobility and apply them to a wide range of spatio-temporal data scales, thoroughly analysing changes in predictability and produced data. We find that spatio-temporal resolution and data processing methods have a large impact on the predictability as well as geometrical and numerical properties of human mobility data, and we present their nonlinear dependencies.

Water vapour is a highly variable constituent of the troposphere; thus, its high-resolution measurements are of great importance to weather prediction systems. The Global Navigation Satellite Systems (GNSS) are operationally used in the estimation of the tropospheric state and assimilation of the results into the weather models. One of the GNSS techniques of troposphere sensing is tomography which provides 3-D fields of wet refractivity. The tomographic results have been successfully assimilated into the numerical weather models, showing the great potential of this technique. The GNSS tomography can be based on two different approaches to the parameterisation of the model’s domain, i.e. block (voxel-based) or grid (node-based) approach. Regardless of the parameterisation approach, the tomographic domain should be discretised, which is usually performed in a regular manner, with a grid resolution depending on the mean distance between the GNSS receivers. In this work, we propose a new parameterisation approach based on the optimisation of the tomographic nodes’ location, taking into account the non-uniform distribution of the GNSS information in the troposphere. The experiment was performed using a dense network of 16 low-cost multi-GNSS receivers located in Wrocław and its suburbs, with a mean distance of 3 km. Cross-validation of four different parameterisation approaches is presented. The validation is performed based on the Weather Research and Forecasting model as well as radiosonde observations. The new approach improves the results of wet refractivity estimation by 0.5–2 ppm in terms of RMSE, especially for altitudes of 0.5–2.0 km.