Search for: All records

Abstract Urbanization affects atmospheric boundary layer dynamics by altering cloud formation and precipitation patterns through the urban heat island (UHI) effect, perturbed wind flows, and urban aerosols, that overall contribute to the urban rainfall effect (URE). This study analyzes an ensemble of numerical simulations with the Weather Research and Forecasting (WRF) model and its version with coupled chemistry and aerosols (WRF-Chem) through a Functional ANalysis Of VAriance (FANOVA) approach to isolate the urban signature from the regional climatology and to investigate the relative contributions of various mechanisms and drivers to the URE. Different metropolitan areas across the United States are analyzed and their urban land cover and anthropogenic emissions are replaced with dominant land-use categories such as grasslands or croplands and biogenic only emissions, as in neighboring regions. Our findings indicate a significant role of the urban land cover in impacting surface temperature and turbulent kinetic energy over the city, and precipitation patterns, both within and downwind of the urban environment. Moreover, simulations of a deep convection event suggest that the aerosols impact dominates the sign and spatial extent of the changes in the simulated precipitation compared to the UHI effect, leading to a significant precipitation enhancement within the urban borders and suppression in downwind regions. 

ABSTRACT Tropospheric ozone (O₃) is a widespread air pollutant that impairs crop physiology and threatens global food security. Most global‐scale assessments have relied on exposure‐based metrics, which overlook plant–environment interactions that control O₃uptake. This study presents a global flux‐based assessment of future O₃risk for wheat (Triticum aestivum) using a dual‐sink dry deposition model driven by Earth System Models from the Coupled Model Intercomparison Project 6 (CMIP6) under three Shared Socioeconomic Pathways (SSP1‐2.6, SSP3‐7.0, and SSP5‐8.5). We quantify phytotoxic O₃dose (POD₆) and production losses from 2000 to 2100, analyze regional trends, and perform multiple simulations to assess the influence of soil water availability and atmospheric CO₂on O₃risk. Finally, we explore the roles of radiative forcing (RF), emission policies on O₃precursors (EP), and their interaction, in determining O₃risk changes. We find a general decline in O₃risk, although regional disparities remain. Under SSP1‐2.6 (strong EP, low RF) POD₆declines throughout the century, leading global mean production losses to decrease from 3.3% to 5.0% at the beginning of the century to less than 1.4% at its end. In contrast, SSP3‐7.0 (weak EP, high RF) shows end‐century losses between 1.3% and 4.9% and may exacerbate risks in several regions (South and East Asia, South America, Sub‐Saharan Africa). SSP5‐8.5 displays intermediate outcomes: O₃risk increases until mid‐century in many regions, and then declines by 2100 (0.5%–2.6%), due to delayed EP adoption. Increasing atmospheric CO₂concentrations will likely hinder future O₃risk due to reduced stomatal conductance, but some hotspots will persist near the Southern and Eastern edges of the Tibetan Plateau. These findings provide a basis for prioritizing region‐specific mitigation strategies to reduce O₃damage to wheat under future climate conditions. 

Abstract Citizen science data for monitoring air pollution have recently emerged as a powerful yet under-explored resource to complement expensive and sparse national air quality monitors. In urban environments, these new data have the potential to allow for high-resolution and high-frequency forecasts, and thereby to provide an assessment of population exposure at neighbourhood level. The complex spatio-temporal structure of these data, however, requires new flexible methods that are also able to provide timely forecasts. In this work, we propose a novel method that first provides forecasts with a reservoir computing approach, an echo-state network, adjusts the forecast with a transformer network with attention mechanism and then merges the echo-state and transformer forecast into a combined network. The stochastic nature of the method allows for a fast and more accurate forecast then individual predictors as well as standard statistical methods. Simulation and application to San Francisco air pollution show how the proposed method is able to produce high-resolution urban maps of air quality. Additionally, we show how these forecasts can be used to provide neighbour-level exposure assessment using population data, a task that would not be achievable with sparse government-sponsored air quality networks. 

An accurate and timely assessment of wind speed and energy output allows an efficient planning and management of this resource on the power grid. Wind energy, especially at high resolution, calls for the development of nonlinear statistical models able to capture complex dependencies in space and time. This work introduces a Convolutional Echo State AutoencodeR (CESAR), a spatio‐temporal, neural network‐based model which first extracts the spatial features with a deep convolutional autoencoder, and then models their dynamics with an echo state network. We also propose a two‐step approach to also allow for computationally affordable inference, while also performing uncertainty quantification. We focus on a high‐resolution simulation in Riyadh (Saudi Arabia), an area where wind farm planning is currently ongoing, and show how CESAR is able to provide improved forecasting of wind speed and power for proposed building sites by up to 17% against the best alternative methods. 

Abstract: In the past decades, clean and renewable energy has gained increasing attention due to a global effort on carbon footprint reduction. In particular, Saudi Arabia is gradually shifting its energy portfolio from an exclusive use of oil to a reliance on renewable energy, and, in particular, wind. Modelling wind for assessing potential energy output in a country as large, geographically diverse and understudied as Saudi Arabia is a challenge which implies highly non-linear dynamic structures in both space and time. To address this, we propose a spatio-temporal model whose spatial information is first reduced via an energy distance-based approach and then its dynamical behaviour is informed by a sparse and stochastic recurrent neural network (Echo State Network). Finally, the full spatial data is reconstructed by means of a non-stationary stochastic partial differential equation-based approach. Our model can capture the fine scale wind structure from a high-resolution simulation and produce more accurate forecasts of both wind speed and energy in lead times of interest for energy grid management.