1,720,974 research outputs found
Transfer learning in environmental data-driven models: A study of ozone forecast in the Alpine region
Full text linkMany environmental variables, in particular, related to air or water quality, are measured in a limited number of points and often for a limited time span. This forbids the development of accurate models for interesting locations with missing or insufficient data and poses the question of whether a model developed for another measurement site can be reliably applied. Such a question is particularly critical when the model is entirely data-driven, such as a neural network. In this context, the paper proposes a procedure to evaluate the expected performance of an existing neural network model applied to a new unmonitored station. This transferability assessment is exemplified by the problem of forecasting ozone concentrations in different environmental settings around the Alpine Arc. Long Short-Term Memory (LSTM) neural network models are applied for predicting hourly concentrations in 20 stations of different types (urban, rural, and mountain). The analysis of the results allows us to determine the expected performance of such models in new cases and reduce the transferability uncertainty when the existing models can be partitioned into clusters. The LSTM models demonstrate the possibility of high accuracy in ozone forecasting at all sites. Given the significant impacts of this gas on human health and the environment, this can contribute to better decision-making and mitigation strategies for air pollution control
Multi-objective planning of building stock renovation
No full textThe energy consumption of residential buildings represents a relevant portion of energy use at the global level. In some regions of old Europe, a large portion of the residential building stock was built in an era of low energy prices, without much attention to energy efficiency, local pollution, and greenhouse gases emissions. Today they can be renewed with energy-saving measures providing consistent benefits to the economy, air pollution, and climate. The paper shows how such energy-saving measures can be optimally planned when dealing with a regional domain, that presents a wide variety of urban conditions, construction styles, and localizations. The study builds upon the extensive dataset of individual building characteristics developed by Lombardy region in Northern Italy and proposes a plan that implements the regional directives already in force. The approach is based on the solution of a mathematical programming problem with two objectives. An efficient compromise solution is then selected considering all economic and environmental impacts. Such a preferred plan determines which actions should be undertaken, on which building types and in which municipality. The results suggest that the regional authorities should complement the current national policy with more specifically directed actions
Local vs regional neural air pollution forecasting models
Full text linkSelecting a suitable dataset to develop a data-based forecasting model is often problematic. This is particularly important in the case of air pollution, where concentration measures are scattered over large areas. On the one hand, the classical approach creates a single-station (local) forecasting model using only the data collected at the same station. This guarantees a training dataset that considers all the site's specific characteristics. On the other hand, these data may be limited and not sufficient to develop a robust predictor. Thus, one may use data from other stations to complement the dataset or develop a unique model considering all the data available within a region/domain. While this approach may be prone to filtering high variations, it may consider information on peculiar episodes that have not occurred in the past to a specific station. This paper discusses the topic of air pollution forecasting using the example of several stations in the Padana Plain, Northern Italy. Local forecasting models are developed using LSTM neural networks for nitrogen dioxide and ozone and hourly data from 2010 to 2023 and then compared with regional models. All these models perform extremely well under various regression-based and classification-based performance indicators, except for a few sites with peculiar characteristics that can be considered at the border of the information domain
Potential of wastewater reuse to alleviate water scarcity under future warming scenarios
Full text linkWater scarcity is a critical issue, expected to worsen with global warming. Tackling water scarcity requires strategies to both decrease water consumption and enhance water availability. One promising solution to mitigate water scarcity is wastewater reuse, which involves collecting, treating, and repurposing used water. By employing a water balance model in conjunction with climate model outputs, we quantified the potential of wastewater reuse to reduce water gaps—situations where water consumption exceeds renewable water availability—under a baseline climate and two warming scenarios. We find that wastewater reuse could reduce the global water gap by 9.1% (from 457.9 to 416.1 km3 per year) under a baseline climate and by 8.3% in a 3 °C warming scenario (from 524.6 to 480.9 km3 per year). Our analysis highlights the potential for wastewater reuse to alleviate water gaps in water scarce countries and metropolitan areas. India, facing the world’s largest water gap, could reduce its baseline water gap by 6.6% (8.2 km3 per year) reusing all available wastewater, compared to a 1.2% (1.5 km3 per year) reduction with reuse of only currently treated wastewater. Specifically, in Delhi, India, the water gap could be reduced by 29.6% with full wastewater reuse and by 16.6% with the reuse of currently treated wastewater. As the global water gap widens with climate change, wastewater reuse is a promising solution to assure sustainable water access. Drawing attention to global disparities in access to wastewater treatment, our findings can guide targeted investments in wastewater treatment and reuse, aiming to alleviate water scarcity, reduce pollution from untreated wastewater, and support circular economies, ultimately ensuring sustainable access to water and sanitation
Improving the Performance of Multiobjective Genetic Algorithms: An Elitism-Based Approach
Full text linkToday, many complex multiobjective problems are dealt with using genetic algorithms (GAs). They apply the evolution mechanism of a natural population to a “numerical” population of solutions to optimize a fitness function. GA implementations must find a compromise between the breath of the search (to avoid being trapped into local minima) and its depth (to prevent a rough approximation of the optimal solution). Most algorithms use “elitism”, which allows preserving some of the current best solutions in the successive generations. If the initial population is randomly selected, as in many GA packages, the elite may concentrate in a limited part of the Pareto frontier preventing its complete spanning. A full view of the frontier is possible if one, first, solves the single-objective problems that correspond to the extremes of the Pareto boundary, and then uses such solutions as elite members of the initial population. The paper compares this approach with more conventional initializations by using some classical tests with a variable number of objectives and known analytical solutions. Then we show the results of the proposed algorithm in the optimization of a real-world system, contrasting its performances with those of standard packages
NN-based implicit stochastic optimization of multi-reservoir systems management
Full text linkMulti-reservoir systems management is complex because of the uncertainty on future events and the variety of purposes, usually conflicting, of the involved actors. An efficient management of these systems can help improving resource allocation, preventing political crisis and reducing the conflicts between the stakeholders. Bellman stochastic dynamic programming (SDP) is the most famous among the many proposed approaches to solve this optimal control problem. Unfortunately, SDP is affected by the curse of dimensionality: computational effort increases exponentially with the complexity of the considered system (i.e., number of reservoirs), and the problem rapidly becomes intractable. This paper proposes an implicit stochastic optimization approach for the solution of the reservoir management problem. The core idea is using extremely flexible functions, such as artificial neural networks (NN), for designing release rules which approximate the optimal policies obtained by an open-loop approach. These trained NNs can then be used to take decisions in real time. The approach thus requires a sufficiently long series of historical or synthetic inflows, and the definition of a compromise solution to be approximated. This work analyzes with particular emphasis the importance of the information which represents the input of the control laws, investigating the effects of different degrees of completeness. The methodology is applied to the Nile River basin considering the main management objectives (minimization of the irrigation water deficit and maximization of the hydropower production), but can be easily adopted also in other cases
End-to-end Artificial Intelligence to analyze dynamical processes: A linear benchmark test
Full text linkWe envisage AI architectures to analyze complex time series in an end-to-end fashion, meaning that the quantitative metrics of the time series are learned directly from data, without the use of specific human-thought algorithms. That is, we challenge AI to learn those specific algorithms. We present a first step in this direction, a benchmark test on linear dynamical processes. We tackle the archetypical task of learning the eigenvalues of the state-transition matrix of a linear (discrete-time, stable) dynamical system, from output data. We train a scalable LSTM neural network with artificially generated data from random matrices of dimension 2-to-5. With noise-free data, the performance of the trained network is very good (average R2=0.955), especially in estimating the dominant eigenvalues, whereas there is space for improvements on non-dominant real eigenvalues and on the dimension of the generating matrix. Remarkably, the performance is robust to measurement noise and the network outperforms the mean-square identification of the corresponding AR process (the latter giving exact eigenvalues on noise-free data) at noise standard deviation starting from 10−6
Advancing Physical Interpretability of Arctic Sea Ice Dynamics through Automatic Feature Selection
No full textModelling Arctic sea ice dynamics has proven to be a successful application for machine learning, leveraging its ability to generate accurate and computationally efficient forecasts. Nevertheless, prevailing limitations lie in the need for physical interpretability and the inability to unveil the dynamics and interdependencies between relevant ice-related variables and their drivers. In this study, we provide a two-step framework designed to combine the high accuracy and computational efficiency characteristics of machine learning while ensuring high interpretability.
The first step of our framework entails time series clustering to identify subregions that are homogeneous with respect to the spatiotemporal variability in the considered variable and obtain the barycentric time series of each cluster. We then use an advanced feature selection algorithm, the Wrapper for Quasi Equally Informative Subset Selection, that identifies neural predictors, specifically Extreme Learning Machines, to forecast the future evolution of sea ice. It then provides the most relevant set of inputs necessary for accurately describing the evolution for each cluster.
Our investigation focuses on the monthly evolution of sea ice thickness and uses data from the Pan-Arctic Ice-Ocean Modeling and Assimilation System (PIOMAS). Other PIOMAS variables (i.e., sea ice concentration, snow depth, sea surface temperature, and sea surface salinity) as well as observed discharge from five major Arctic rivers (i.e., Ob, Yenisey, and Lena in Asia, Mackenzie and Yukon in North America, provided by the Arctic Great Rivers Observatory discharge dataset) are considered as potential driving factors.
Our results indicate the pivotal role of past sea ice thickness values, since the forthcoming state of sea ice seems to be influenced by both the current situation and historical trends and periodicity. Sea surface salinity in the open Arctic Ocean is highly persistent, and therefore is not used by the algorithm to explain the sea ice evolution. On the other hand, the Arctic rivers’ flows are more representative of the processes occurring in the clusters along the coast. Finally, the interaction between sea surface temperature and snow depth controls the interplay between ice formation and melting, and therefore plays a significant role in shaping the sea ice evolution in the short term.
Our framework aims to advance our comprehension of the complex physical processes governing sea ice thickness evolution in the Arctic region. Moreover, its effectiveness in uncovering sea ice related processes is expected to further improve with the inclusion of additional input variables and, possibly, of a longer data record
Assessing the Effect of Droughts on Complex Multi-sector Water Systems
No full textIn recent years, climate change has significantly intensified the frequency and severity of drought events. Rising temperatures, altered precipitation patterns, and changing weather dynamics have led to more prolonged and intense droughts altering water availability and exacerbating tradeoffs, especially in complex multi-sector water systems.
An archetypal example of this situation is the Lake Como water system, in the north of Italy. Lake Como is operated to provide water downstream to the agricultural sector, control floods on the lake shores, and contrast low water levels that would negatively impact navigation and aquatic ecosystems. The conflict between the interests of these sectors is expected to exacerbate in the years to come due to the evolving hydroclimatic regimes. Among different adaptation options considered by the regional authority, we investigate the potential expansion of the lake's active storage capacity enabled by the recent construction of flood mobile dykes.
Here, we contribute a framework for evaluating the impact of droughts on multiple water users. Specifically, we adopt a synthetic weather generator to create multiple streamflow ensembles (scenarios) controlling the drought’s frequency, duration, and intensity. Drought features are then linked to impacts (e.g., agricultural deficit) using a simulation model of the lake. Failure thresholds are defined for each impact indicator to set the minimum level of performance acceptable to each sector. Finally, a logistic classifier is used to identify the combination of drought features leading to a system failure.
Our results show that system failures can be accurately estimated using a linear combination of drought frequency, duration, and intensity. The combined effect of these three characteristics, rather than the extreme values of one of them, is responsible for system failure. Our analysis also proves that storage expansion is fundamental to reduce the downstream deficit, as well as to prevent most of the flood events
- …
