Forecast Uncertainty Research Articles

Carbon price forecasting using leaky integrator echo state networks with the framework of decomposition-reconstruction-integration

Carbon trading is one of the strategies to achieve the goal of dual carbon. However, carbon prices exhibit non-stationary and nonlinear characteristics, posing significant challenges for accurate forecasting. Inspired by the ideas of “division and conquest” and “granularity reconstruction”, a decomposition-reconstruction-integration framework is built for carbon price forecasting. The proposed model leverages a comprehensive ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and sample entropy (SE) to extract features from two perspectives of time and complexity, and thus the one-dimensional time series of carbon price is reconstructed into different items. Subsequently, leaky integrator echo state networks (LiESNs) are utilized to forecast each of these items individually. These individual forecasts are then integrated to generate the final point forecast. To evaluate the uncertainty of the point forecast, a Gaussian process model is applied to interval forecast. The proposed model comprehensively considers the time series features of carbon price including the temporal modal features in the decomposition stage, the entropy probability features in the reconstruction stage, and the essential temporal dependencies via dynamic reservoir of interconnected neurons in the integration forecasting stage. Experimental results demonstrate the model's exceptional forecasting performance, surpassing the effectiveness of alternative approaches.

A causation-based computationally efficient strategy for deploying Lagrangian drifters to improve real-time state estimation

Deploying Lagrangian drifters that facilitate the state estimation of the underlying flow field within a future time interval is practically important. However, the uncertainty in estimating the flow field prevents using standard deterministic approaches for designing strategies and applying trajectory-wise skill scores to evaluate performance. In this paper an information measurement is developed to quantitatively assess the information gain in the estimated flow field by deploying an additional set of drifters. This information measurement is derived by exploiting causal inference. It is characterized by the inferred probability density function of the flow field, which naturally considers the uncertainty. Although the information measurement is an ideal theoretical metric, using it as the direct cost makes the optimization problem computationally expensive. To this end, an effective surrogate cost function is developed. It is highly efficient to compute while capturing the essential features of the information measurement when solving the optimization problem. Based upon these properties, a practical strategy for deploying drifter observations to improve future state estimation is designed. Due to the forecast uncertainty, the approach exploits the expected value of spatial maps of the surrogate cost associated with different forecast realizations to seek the optimal solution. Numerical experiments justify the effectiveness of the surrogate cost. The proposed strategy significantly outperforms the method by randomly deploying the drifters. It is also shown that, under certain conditions, the drifters determined by the expected surrogate cost remain skillful for the state estimation of a single forecast realization of the flow field as in reality.

Enhancing irrigation water productivity using short-range ensemble weather forecasts at basin scale: A novel framework for regions with high hydro-climatic variability

Integrating short-term weather forecasts with crop growth models has emerged as a valuable decision-support tool for enhancing irrigation water productivity in water-intensive crops. Our primary focus lies in formulating a simplified Quasi-Farmer Behavior Routine to utilize ensemble weather forecasts to provide irrigation guidance. The study proposes the concept of Rainfall Confidence Quotient (RCQ) that is optimized based on the forecast uncertainty levels for rainfall categories that affect the irrigation decision. We utilize two publicly available ensemble weather forecasts from NCEP-GEFS and ECM and a regionally fine-tuned high-resolution Weather Research and Forecasting (WRF) ensemble forecast system with a 3-day lead time. We have refined our methodology by conducting 800 simulation experiments across diverse scenarios, each reflecting distinct climate regimes, soil types, cropping seasons, and management practices. The proposed method consists of a sequential 3-checkpoint algorithm incorporating ensemble rainfall forecasts, antecedent soil moisture, and evapotranspiration. During the wet crop season of 2015–16, the enhanced methodology yielded a substantial increase in irrigation water productivity (IWP), averaging around 20–30 % increase across spatial locations, in contrast to the conventional irrigation practices that do not consider weather forecasts. However, during 2016–17, a dry year, the improvement in IWP only ranged between 2–10 %. On average, water savings (at field scale) of 200–500 mm were recorded in the wet year crop season of 2015–16, while 100–300 mm of water savings were achieved during the dry year crop season of 2016–17, without any considerable reductions in the crop yield. The robustness and adaptability of the developed approach have been established through comprehensive evaluation with field observation and remote sensing techniques, suggesting its potential scalability to similar hydro-climatic typologies across the world.

Sources of seasonal water supply forecast uncertainty during snow drought in the Sierra Nevada

AbstractUncertainty attribution in water supply forecasting is crucial to improve forecast skill and increase confidence in seasonal water management planning. We develop a framework to quantify fractional forecast uncertainty and partition it between (1) snowpack quantification methods, (2) variability in post‐forecast precipitation, and (3) runoff model errors. We demonstrate the uncertainty framework with statistical runoff models in the upper Tuolumne and Merced River basins (California, USA) using snow observations at two endmember spatial resolutions: a simple snow pillow index and full‐catchment snow water equivalent (SWE) maps at 50 m resolution from the Airborne Snow Observatories. Bayesian forecast simulations demonstrate a nonlinear decrease in the skill of statistical water supply forecasts during warm snow droughts, when a low fraction of winter precipitation remains as SWE. Forecast skill similarly decreases during dry snow droughts, when winter precipitation is low. During a shift away from snow‐dominance, the uncertainty of forecasts using snow pillow data increases about 1.9 times faster than analogous forecasts using full‐catchment SWE maps in the study area. Replacing the snow pillow index with full‐catchment SWE data reduces statistical forecast uncertainty by 39% on average across all tested climate conditions. Attributing water supply forecast uncertainty to reducible error sources reveals opportunities to improve forecast reliability in a warmer future climate.

Quantifying and reducing flood forecast uncertainty by the CHUP-BMA method

Abstract. The Bayesian model averaging (BMA), hydrological uncertainty processor (HUP), and HUP-BMA methods have been widely used to quantify flood forecast uncertainty. This study proposes the copula-based hydrological uncertainty processor BMA (CHUP-BMA) method by introducing a copula-based HUP in the framework of BMA to bypass the need for a normal quantile transformation of the HUP-BMA method. The proposed ensemble forecast scheme consists of eight members (two forecast precipitation inputs; two advanced long short-term memory, LSTM, models; and two objective functions used to calibrate parameters) and is applied to the interval basin between the Xiangjiaba and Three Gorges Reservoir (TGR) dam sites. The ensemble forecast performance of the HUP-BMA and CHUP-BMA methods is explored in the 6–168 h forecast horizons. The TGR inflow forecasting results show that the two methods can improve the forecast accuracy over the selected member with the best forecast accuracy and that the CHUP-BMA performs much better than the HUP-BMA. Compared with the HUP-BMA method, the forecast interval width and continuous ranked probability score metrics of the CHUP-BMA method are reduced by a maximum of 28.42 % and 17.86 % within all forecast horizons, respectively. The probability forecast of the CHUP-BMA method has better reliability and sharpness and is more suitable for flood ensemble forecasts, providing reliable risk information for flood control decision-making.

Selecting and weighting dynamical models using data-driven approaches

Abstract. In geosciences, multi-model ensembles are helpful to explore the robustness of a range of results. To obtain a synthetic and improved representation of the studied dynamic system, the models are usually weighted. The simplest method, namely the model democracy, gives equal weights to all models, while more advanced approaches base weights on agreement with available observations. Here, we focus on determining weights for various versions of an idealized model of the Atlantic Meridional Overturning Circulation. This is done by assessing their performance against synthetic observations (generated from one of the model versions) within a data assimilation framework using the ensemble Kalman filter (EnKF). In contrast to traditional data assimilation, we implement data-driven forecasts using the analog method based on catalogs of short-term trajectories. This approach allows us to efficiently emulate the model's dynamics while keeping computational costs low. For each model version, we compute a local performance metric, known as the contextual model evidence, to compare observations and model forecasts. This metric, based on the innovation likelihood, is sensitive to differences in model dynamics and considers forecast and observation uncertainties. Finally, the weights are calculated using both model performance and model co-dependency and then evaluated on averages of long-term simulations. Results show good performance in identifying numerical simulations that best replicate observed short-term variations. Additionally, it outperforms benchmark approaches such as strategies based on model democracy or climatology when reconstructing missing distributions. These findings encourage the application of the proposed methodology to more complex datasets in the future, like climate simulations.

Influences of Space Weather Forecasting Uncertainty on Satellite Conjunction Assessment

AbstractA significant increase in the number of anthropogenic objects in Earth orbit has necessitated the development of satellite conjunction assessment and collision avoidance capabilities for new spacecraft. Neutral mass density variability in the thermosphere, driven by enhanced geomagnetic activity and solar EUV absorption, is a major source of satellite propagation error. This work investigates the impacts of space weather driver forecasting uncertainty on satellite drag and collision avoidance maneuver decision‐making. Since most operational space weather driver forecasts do not offer an uncertainty assessment, the satellite operator community is left to make dangerous assumptions about the trustworthiness of the forecast models they use to perform satellite state propagation. Climatological persistence‐based forecast models are developed for F10.7 and Kp. These models accurately capture the heteroscedastic and, at times, highly non‐Gaussian uncertainty distribution on forecasts of the drivers of interest. A set of realistic satellite conjunction scenarios is simulated to demonstrate the contributions of space weather driver forecast uncertainty on the probability of collision and maneuver decisions. Improved driver forecasts, especially forecasts of F10.7, are demonstrated to be very useful for enabling durable maneuver decisions with additional lead time (up to 24 hr for the period examined), though the improvement depends on the specific conjunction scenario of interest.

Forecast Uncertainties Real-Time Data-Driven Compensation Scheme for Optimal Storage Control

This study introduces a real-time data-driven battery management scheme designed to address uncertainties in load and generation forecasts, which are integral to an optimal energy storage control system. By expanding on an existing algorithm, this study resolves issues discovered during implementation and addresses previously overlooked concerns, resulting in significant enhancements in both performance and reliability. The refined real-time control scheme is integrated with a day-ahead optimization engine and forecast model, which is utilized for illustrative simulations to highlight its potential efficacy on a real site. Furthermore, a comprehensive comparison with the original formulation was conducted to cover all possible scenarios. This analysis validated the operational effectiveness of the scheme and provided a detailed evaluation of the improvements and expected behavior of the control system. Incorrect or improper adjustments to mitigate forecast uncertainties can result in suboptimal energy management, significant financial losses and penalties, and potential contract violations. The revised algorithm optimizes the operation of the battery system in real time and safeguards its state of health by limiting the charging/discharging cycles and enforcing adherence to contractual agreements. These advancements yield a reliable and efficient real-time correction algorithm for optimal site management, designed as an independent white box that can be integrated with any day-ahead optimization control system.

Probabilistic multi-step ahead streamflow forecast based on deep learning

Abstract The use of deep learning methods for fluvial flood forecasting is rapidly gaining traction, offering a promising solution to the challenges associated with accurate yet time-consuming numerical models. This paper presents two physics-inspired deep learning approaches specifically designed for fluvial flood forecasting, each embracing different learning principles: centralized and federated learning. The centralized model utilizes an Encoder-Decoder technique to handle input data of varying types and scales, while the federated model is based on a node-link graph with a seq2seq internal model. Both models are enhanced with a probabilistic forecasting head to account for the inherent uncertainty in streamflow forecasts. The objective of these approaches is to address the limitations of traditional numerical models while leveraging the potential of deep learning to improve the speed, accuracy, and scalability of flood forecasting. To validate their effectiveness, the models were tested across different use cases. The findings from the federated learning approach emphasize the importance of catchment clustering before model utilization and demonstrate the models’ ability to generalize effectively in catchments with similar properties. On the other hand, the results of the centralized method highlight the model’s reliance on the test set falling within the data range of the training set (Average NSE and KGE for the sixth hour ahead of 0.88 and 0.78, respectively). To address this limitation, the paper suggests the development of a method for the future, such as leveraging a numerical model or using Generative Adversarial Networks, to generate highly extreme events, particularly in the context of a changing climate. The models are implemented in a flexible operational framework based on open standards, ensuring their adaptability and usability in various settings.

Prescriptive analytics for low and high water management in German waterways

Abstract The real-time management of multi-purpose storage reservoirs aims at an efficient operation of existing hydraulic infrastructure. This management process can be structured as a prescriptive analytics setup that considers both current and predicted system states to recommend actions and outline potential implications. In application to a reservoir and river system, it combines hydrological modelling components for the system schematization, observations and data assimilation for the identification of the current system state, meteorological and hydrological predictions as well as optimization-based techniques to support decision-making regarding reservoir operations. In this paper, we present the application of such a framework to the short-term management of the Eder and Diemel storage reservoirs. These reservoirs are operated by the German Federal Waterways and Shipping Administration (WSV) with the primary goal to support navigation in the River Weser during low flow periods. In addition, partially conflicting objectives such as flood protection, energy generation and recreation are considered. The implementation includes an explicit consideration of forecast uncertainty and its impact on the decision-making by using probabilistic forecasts in combination with a multi-stage stochastic optimization approach. We demonstrate the applicability of the approach based on low and high water use cases. Special attention is paid on the benefits of the probabilistic forecast in combination with the multi-stage stochastic optimization versus a deterministic setup. It provides an explicit translation of the forecast uncertainty in the decision variables, in this case the reservoir releases helping the operators to better anticipate the range of future release decisions. Furthermore, the stochastic approach is expected to provide more stable decisions in an operational setting, based on more stable forecasts by considering various possible realizations of the future instead of picking a single one, which gets random after 4–5 days.

Is a more physical representation of aerosol chemistry needed for fog forecasting?

AbstractWith the changing climate, the study of fog formation is essential due to the impact of the complexity of natural and anthropogenic aerosols. The evolution of the droplet size distribution in the presence of different aerosol species remains poorly understood. To make progress towards reducing the uncertainty of fog forecasts, the Eulerian–Lagrangian particle‐based small‐scale model for the diffusional growth of droplets is used to better understand the droplet activation and growth. The small‐scale model simulations are performed using observed data from the Winter Fog Experiment study over Indira Gandhi International Airport, New Delhi. The microphysical properties, such as droplet number concentrations (Nd) and liquid water content (LWC), important for fog simulation, are evaluated to gain more insights. The small‐scale simulations have shown the droplet microphysical properties at different evolutionary stages. The Nd and effective radius change with variations in LWC for different aerosol chemistries (i.e., organics, mix, and inorganic). The calculated visibility at small scale is also shown with the variation of Nd and LWC. This study compared visibility from an existing parametrization with parcel–direct numerical simulation calculation. The hygroscopicity , which is highly related to the activation of aerosols to condensation nuclei, is taken into account to demonstrate the contribution of aerosol chemistry to fog droplet formation. The results highlight that hygroscopicity is essential in the numerical model for fog and visibility prediction as the microphysical properties of fog are regulated by aerosol species.

Constraint optimization of an integrated production model utilizing history matching and production forecast uncertainty through the ensemble Kalman filter

The calibration of reservoir models using production data can enhance the reliability of predictions. However, history matching often leads to only a few matched models, and the original geological interpretation is not always preserved. Therefore, there is a need for stochastic methodologies for history matching. The Ensemble Kalman Filter (EnKF) is a well-known Monte Carlo method that updates reservoir models in real time. When new production data becomes available, the ensemble of models is updated accordingly. The initial ensemble is created using the prior model, and the posterior probability function is sampled through a series of updates. In this study, EnKF was employed to evaluate the uncertainty of production forecasts for a specific development plan and to match historical data to a real field reservoir model. This study represents the first attempt to combine EnKF with an integrated model that includes a genuine oil reservoir, actual production wells, a surface choke, a surface pipeline, a separator, and a PID pressure controller. The research optimized a real integrated production system, considering the constraint that there should be no slug flow at the inlet of the separator. The objective function was to maximize the net present value (NPV). Geological data was used to model uncertainty using Sequential Gaussian Simulation. Porosity scenarios were generated, and conditioning the porosity to well data yielded improved results. Ensembles were employed to balance accuracy and efficiency, demonstrating a reduction in porosity uncertainty due to production data. This study revealed that utilizing a PID pressure controller for the production separator can enhance oil production by 59% over 20 years, resulting in the generation of 2.97 million barrels of surplus oil in the field and significant economic gains.

Improved robust model predictive control for residential building air conditioning and photovoltaic power generation with battery energy storage system under weather forecast uncertainty

The rising demands for comfort alongside energy conservation underscore the importance of intelligent air conditioning control systems. Model Predictive Control (MPC) stands out as an advanced control strategy capable of addressing these demands. However, accurate prediction of all relevant variables remains a challenge in practical scenarios, complicating MPC’s ability to devise effective control actions amid prediction inaccuracies. To counteract this issue, this paper introduces an enhanced Double-Layer Model Predictive Control (DLMPC) algorithm. This innovative approach adjusts for discrepancies between forecasted and actual values without the need for additional variables and models, thereby reducing the adverse effects of prediction errors. Additionally, we develop precise models for room temperature simulation and for calculating air conditioning (AC) load and energy consumption, grounded in empirical data from residential settings and AC performance tests. Validation of these models demonstrates their efficacy in enabling MPC to formulate efficacious control strategies. When juxtaposed with a baseline model, the DLMPC algorithm significantly improves temperature regulation accuracy by up to 15.12% and achieves a 10.50% reduction in energy consumption over the heating season.

Integrating Uncertainties in a Multi-Criteria Decision Analysis with the Entscheidungsnavi

The Entscheidungsnavi is an open-source decision support system based on multi-attribute utility theory, that offers various methods for dealing with uncertainties. To model decisions with uncertainties, decision-makers can use two categories: Forecast and Parameter Uncertainties. Forecast Uncertainty is modeled with (combined) influence factors using discrete, user-defined probability distributions or predefined ‘worst-median-best’ distributions. Parameter Uncertainty allows imprecision for utilities, objective weights, and probability distributions. To analyze these uncertainties, the Entscheidungsnavi offers several methods and tools, like a robustness check, based on (Monte Carlo) simulations and a sensitivity analysis. The objective weight analysis provides insights into the effects of different objective weight combinations. Indicator impacts, tornado diagrams, and risk profiles visualize the impact of uncertainties in a decision under risk. Risk profiles also enable a check for stochastic and simulation dominance. This article presents the complete range of methods for dealing with uncertainties in the Entscheidungsnavi using a hypothetical case study.

A generation expansion planning method considering conditional value-at-risk

Abstract The uncertainty of load forecasting can lead to certain risks in generation expansion planning. Different from the traditional large-capacity unit planning in the past, renewable energy has the characteristics of small single-machine capacity and fast construction speed, etc. The uncertainty of load forecasting will have a more significant impact on the results of generation expansion planning including renewable energy. For this reason, this paper proposes a generation expansion planning idea considering load forecast uncertainty based on conditional value-at-risk (CVaR). Firstly, scenario technology is used to simulate the uncertainty, and CVaR is introduced to quantify the risk of generation expansion planning. Then, considering the coordinated development of traditional and renewable energy sources, a generation expansion planning model that considers risk is established to minimize the total costs. Finally, the effectiveness of the model was verified through a case study, and this model has reference significance for balancing the economy and reliability in generation expansion planning.

Adjoint Synthesis for Trans‐Oceanic Tsunami Waveforms and Simultaneous Inversion of Fault Geometry and Slip Distribution

AbstractTsunamis propagate over long distances and can cause widespread damage even after crossing ocean basins. Prediction of tsunamis in distant areas based on observations near their sources is critical to mitigating damage. In recent years, the accuracy of numerical models of trans‐oceanic tsunami propagation has improved significantly due to the incorporation of effects such as the solid earth response to tsunami loading and wave dispersion. However, these models are computationally expensive and have not been fully utilized for real‐time prediction. Here, we derive the adjoint operator for the linear set of equations describing deep‐ocean tsunami propagation and show how a pre‐computed database of adjoint states can achieve rapid synthesis of tsunami waveforms at target sites from nonpoint arbitrary tsunami sources. The adjoint synthesis method allows for an exhaustive parameter search for tsunami source estimation. A method for simultaneous inversion of fault geometry and slip distribution using adjoint synthesis with Sequential Monte Carlo method was proposed and applied to the 2012 Haida Gwaii earthquake tsunami. The influence of model accuracy and the amount of observed data on the estimation of tsunami sources and waveforms was examined. It was found that with a highly accurate propagation model, using only a limited amount of observed data produced source and waveform estimates very similar to the final models obtained with much larger data sets. The final inferred fault model involved megathrust slip distributed between the Haida Gwaii trench and the Queen Charlotte fault. The proposed method can also quantify the uncertainty of the waveform forecasts.

The potential of publicly available weather forecasts for market operations in aggregated photovoltaic plants

Uncertainty in solar generation forecasts affects electricity market operations, forcing producers to bid conservatively and, consequently, reducing the amount of solar energy fed into the grid. This study analyses the potential of aggregated forecasts to reduce the errors of energy generation in medium-sized PV plants. To this objective, publicly available forecasts from the Global Forecasting System were processed with a basic neural network to generate hourly energy forecasts for individual PV plants of 1.1 MW and for the aggregated production of three of them (3.3 MW). The validation was carried out considering typical day-ahead and intraday market horizons, two zones with different atmospheric complexity, and discerning three types of days in terms of the received irradiation. Results show that the aggregation of forecasts would be beneficial as a general strategy in all cases, but especially in the zone of complex atmospheric dynamics and for the day-ahead horizons (36-42 h).

SF-Transformer: A Mutual Information-Enhanced Transformer Model with Spot-Forward Parity for Forecasting Long-Term Chinese Stock Index Futures Prices.

The complexity in stock index futures markets, influenced by the intricate interplay of human behavior, is characterized as nonlinearity and dynamism, contributing to significant uncertainty in long-term price forecasting. While machine learning models have demonstrated their efficacy in stock price forecasting, they rely solely on historical price data, which, given the inherent volatility and dynamic nature of financial markets, are insufficient to address the complexity and uncertainty in long-term forecasting due to the limited connection between historical and forecasting prices. This paper introduces a pioneering approach that integrates financial theory with advanced deep learning methods to enhance predictive accuracy and risk management in China's stock index futures market. The SF-Transformer model, combining spot-forward parity and the Transformer model, is proposed to improve forecasting accuracy across short and long-term horizons. Formulated upon the arbitrage-free futures pricing model, the spot-forward parity model offers variables such as stock index price, risk-free rate, and stock index dividend yield for forecasting. Our insight is that the mutual information generated by these variables has the potential to significantly reduce uncertainty in long-term forecasting. A case study on predicting major stock index futures prices in China demonstrates the superiority of the SF-Transformer model over models based on LSTM, MLP, and the stock index futures arbitrage-free pricing model, covering both short and long-term forecasting up to 28 days. Unlike existing machine learning models, the Transformer processes entire time series concurrently, leveraging its attention mechanism to discern intricate dependencies and capture long-range relationships, thereby offering a holistic understanding of time series data. An enhancement of mutual information is observed after introducing spot-forward parity in the forecasting. The variation of mutual information and ablation study results highlights the significant contributions of spot-forward parity, particularly to the long-term forecasting. Overall, these findings highlight the SF-Transformer model's efficacy in leveraging spot-forward parity for reducing uncertainty and advancing robust and comprehensive approaches in long-term stock index futures price forecasting.

Technical note: Exploring parameter and meteorological uncertainty via emulation in volcanic ash atmospheric dispersion modelling

Abstract. ​​​​​​​Consideration of uncertainty in volcanic ash cloud forecasts is increasingly of interest, with an industry goal to provide probabilistic forecasts alongside deterministic forecasts. Simulations of volcanic clouds via dispersion modelling are subject to a number of uncertainties relating to the eruption itself (mass of ash emitted and when), parameterisations of physical processes, and the meteorological conditions. To fully explore these uncertainties through atmospheric dispersion model simulations alone may be expensive, and instead, an emulator can be used to increase understanding of uncertainties in the model inputs and outputs, going beyond combinations of source, physical, and meteorological inputs that were simulated by the dispersion model. We emulate the NAME (Numerical Atmospheric-dispersion Modelling Environment) dispersion model for simulations of the Raikoke 2019 eruption and use these emulators to compare simulated ash clouds to observations derived from satellites, constraining NAME source and internal parameters via history matching. We demonstrate that the effect of varying both meteorological scenarios and model parameters can be captured in this way with accurate emulation and using only a small number of runs per meteorological scenario. We show that accounting for meteorological uncertainty simultaneously with other uncertainties may lead to the identification of different sensitive model parameters and may lead to less constrained source and internal NAME parameters; however, through idealised experiments, we argue that this is a reasonable result and is properly accounting for all sources of uncertainty in the model inputs.

Urban influence on convective precipitation in the Paris region: Hectometric ensemble simulations in a case study

AbstractThe purpose of this study is to investigate the influence of the urban environment of the Paris region on an isolated convective cell that formed downwind of the city on May 7, 2022, using the Meso‐NH research atmospheric model at a horizontal scale of 300 m. To account for all sources of forecast uncertainty, the initial and lateral boundary conditions of the simulations are provided by an ensemble prediction system. A multi‐layer urban scheme is used to represent the influence of buildings on the airflow accurately. Two sets of ensemble simulations are performed: the first set (URB) uses a fine‐scale surface description of the city, while the second set (NOURB) replaces urban surfaces with vegetation. This sensitivity test shows that, despite the high variability of simulated precipitation within the ensemble, the city of Paris plays a statistically significant role in the initiation of convection in this case. Convective cells are initiated over the city for several members of the URB ensemble, while almost no precipitation is simulated for the same members of the NOURB ensemble. The mean 6‐h rainfall accumulation of the URB ensemble is increased by 70% over Paris (compared with the NOURB ensemble) and no statistically significant trend is found around the city. The analysis reveals that the capital experiences a higher sensible heat flux due to drier and warmer air, resulting in enhanced vertical velocities and an increase in boundary‐layer height in the URB ensemble. Additionally, the total water content and cloud fraction over Paris are intensified, leading to more precipitation. These findings suggest that urbanisation has a notable impact on convection and precipitation processes in this case.