Environmental Modelling & Software最新文献

The rise in urban flooding events poses a threat to public safety, property, and economic stability. To prevent urban flooding and manage stormwater effectively, relying solely on engineering solutions is insufficient. Therefore, it is critical to implement non-engineering measures such as urban flood warnings and forecasting. This article reviews the characteristics of different urban flood models based on different hydrological and hydrodynamic principles and deep learning (DL). It highlights the limitations of coupled hydrological-hydrodynamic models in terms of timeliness. Additionally, it discusses research on the use of Numerical Simulation in hydrological early warning and forecasting. Compared to traditional hydrodynamic models that rely on physical mechanisms, models driven by DL methods can effectively and adaptively extract input-output relationships of complex systems. Subsequently, a summary of the current flood models is presented, followed by a discussion of future development trends and challenges.

Weather monitoring and forecasting plays a vital role in a great variety of human activities such as agriculture, transportation, and extreme weather phenomena. This study presents the first outcomes of the development of a fully automated system regarding the real-time recording of basic meteorological parameters and their short-range forecasting (nowcasting). The system itself is divided into five core components: a hardware system for monitoring atmospheric conditions (Commercial Off-The-Shelf structures), a system for storing and managing data, a module for distributing data to support applications, a machine learning algorithm for nowcasting, and a user-friendly interface, all made by modern tools and methods, described analytically. Finally, the nowcasting procedure along with the relative accuracy results, is presented. The nowcasting procedure is based on a Long Short-Term Memory (LSTM) model scheme which is parametrized in such a way that reliable forecasts, up to 2 h ahead of time, can be provided.

Air quality forecasting requires appropriate models and data sources. The Fogees system presented in this paper enables mapping, evaluation and forecasting of the level of PM_x pollution in the air, for urban and suburban areas. Input data were downloaded from the Meteoblue service, the GIS database and – in the case of integration with existing measurement systems – also from local air quality monitoring stations. The system uses a diagnostic model to determine the velocity field and a Lagrange model to describe pollution dispersion. The system has an autocalibration model and uses proprietary algorithms to estimate emissions from domestic heating, transport and background pollution levels. Modular design and parallel computation facilitate simultaneous calculation of forecasts for existing emission conditions and calculations for alternative emission conditions. The system permits forecasting for the next hour and for several consecutive hours. The validation results confirm that the system reliable forecasting of PM concentrations.

The increasing availability of high-resolution climate data has greatly expanded the study of how the climate impacts humans and society. However, the processing of these multi-dimensional datasets poses significant challenges for researchers in this growing field, most of whom are social scientists. This paper introduces stagg, or “space-time aggregator”, a new R package that streamlines three critical components of climate data processing for impacts analysis: nonlinear transformation, spatial and temporal aggregation, and spatial weighting by social or economic variables. The package consolidates the data processing pipeline into a few lines of code, lowering barriers to entry for researchers and facilitating a larger and more diverse research community. The paper provides an overview of stagg's functions, followed by an applied example demonstrating the package's utility in climate impacts research. stagg has the potential to be a valuable tool in generating evidence-based estimates of the likely impacts of future climate change.

Temporal rainfall distribution facilitates the understanding of rainfall patterns at various time scales, extreme events, and corresponding water resources implications. Researchers have developed various metrics of temporal rainfall distribution but there exist no easy-to-use software packages for calculating these metrics. To address this gap, we developed the PyMTRD package, which can be conveniently used to calculate the metrics of temporal rainfall distribution and conduct rainfall pattern analysis. The metrics calculated in the package included rainfall intensity, rainfall frequency, consecutive dry days, Gini index, unranked Gini index, wet-day Gini index, precipitation concentration index, dimensionless seasonality index, and seasonality index. This paper documented our Python software development, which included the architecture design, the Application Programming Interfaces design and algorithms for calculating each metric, and also the point and global scale applications.

This study employed machine learning to comprehensively analyze glacier melting in Tibet Autonomous Region (TAR) and its vital influencing factors. Existing machine learning research often lacks detailed explanations, leading to generalized predictions without considering essential driving factors necessary for yielding an insightful understanding of glacier melting dynamics. To overcome these limitations and fulfill multi-level analysis requirements for comprehending glacier melting, this study identifies factors contributing to glacier melting heterogeneity and assesses distinct melting causes in three spatial melted glacier clusters. We utilized K-means unsupervised classification to cluster Tibet melted glaciers into three categories based on temperature, sunshine hours, evapotranspiration, precipitation, normalized vegetation index, and slope. XGBoost algorithm explores the nonlinear relationships of glacier melting with these features and Shapley values were used for model transparency, quantifying feature's influence on the melting process. Investigating geographical heterogeneity among clusters enhanced our understanding of the observed changes. High fitting accuracy (>0.98) enhanced the result reliability, as well. The results show that Tibetan glaciers melt significantly from 2010 to 2020, and the cluster analysis reveals its unique melting characteristics. Melting glaciers in the same cluster are not only similar in characteristics, but also in spatial and geographical distribution, with two of the clusters concentrating in the eastern part of TAR, and the third cluster scattered in the western part of the country. the XGBoost-SHAP analysis efficiently quantifies the contribution of each cluster feature to the glacier melting, revealing the different roles of different clustered features.

The transition from 2D planar displays to immersive holographic 3D environments has brought advancements in visualization technology. However, there remains a lack of effective interactive visualization tools for complex multi-dimensional structured or unstructured datasets in immersive space. To address this gap, we have developed MetIVA, a state-of-the-art multiscale interactive data visualization platform that leverages the Extended Reality (XR) and cloud rendering technology for immersive data exploration. In this paper, we firstly outline the historical development of scientific visualization and the recent shift towards 3D and higher-dimensional visualization, and then basically introduce the conceptual framework and platform structure of MetIVA, and finally present the evaluation results from recruited potential users. The results confirm that MetIVA is a powerful tool to accelerate data exploration and decision-making processes. Its interactive and intuitive features, along with ongoing optimization efforts, make it a valuable tool for researchers and practitioners in the field of Earth science.

We introduce Bayesian Optimization for Anything (BOA), a high-level Bayesian Optimization (BO) framework and model wrapping toolkit, which presents a novel approach to simplifying BO, with the goal of making it more accessible and user-friendly, particularly for those with limited expertise in the field. BOA addresses common barriers in implementing BO, focusing on ease of use, reducing the need for deep domain knowledge, and cutting down on extensive coding requirements. A notable feature of BOA is its language-agnostic architecture, which facilitates broader application in various fields and to a wider audience. We showcase BOA's application through three examples: a high-dimensional optimization with 184 parameters of the SWAT + watershed model, a highly parallelized optimization of this intrinsically non-parallel model, and a multi-objective optimization of the FETCH Tree-Crown Hydrodynamics model. These test cases illustrate BOA's effectiveness in addressing complex optimization challenges in diverse scenarios.