Allergies to airborne pollen are a significant issue affecting millions of Americans. Consequently, accurately predicting the daily concentration of airborne pollen is of significant public benefit in providing timely alerts. This study presents a method for the robust estimation of the concentration of airborne Ambrosia pollen using a suite of machine learning approaches including deep learning and ensemble learners. Each of these machine learning approaches utilize data from the European Centre for Medium-Range Weather Forecasts (ECMWF) atmospheric weather and land surface reanalysis. The machine learning approaches used for developing a suite of empirical models are deep neural networks, extreme gradient boosting, random forests and Bayesian ridge regression methods for developing our predictive model. The training data included twenty-four years of daily pollen concentration measurements together with ECMWF weather and land surface reanalysis data from 1987 to 2011 is used to develop the machine learning predictive models. The last six years of the dataset from 2012 to 2017 is used to independently test the performance of the machine learning models. The correlation coefficients between the estimated and actual pollen abundance for the independent validation datasets for the deep neural networks, random forest, extreme gradient boosting and Bayesian ridge were 0.82, 0.81, 0.81 and 0.75 respectively, showing that machine learning can be used to effectively forecast the concentrations of airborne pollen.

International audienceThis study aims in filling the gap in understanding the relationship between trend and extreme in diurnal and nocturnal temperatures (Tx and Tn) over the Gulf of Guinea area and the Sahel. Time-evolution and trend of Tx and Tn anomalies, extreme temperatures and heat waves are examined using regional and station-based indices over the 1900–2012 and 1950–2012 periods respectively. In investigating extreme temperature anomalies and heat waves, a percentile method is used. At the regional and local scales, rising trends in Tx and Tn anomalies, which appear more pronounced over the past 60 years, are identified over the two regions. The trends are characterized by an intensification of: i) nocturnal/Tn warming over the second half of the 20th century; and ii) diurnal/Tx warming over the post-1980s. This is the same scheme with extreme warm days and warm nights. Finally annual number of diurnal and nocturnal heat waves has increase over the Gulf of Guinea coastal regions over the second half of the 20th century, and even more substantially over the post-1980s period. Although this trend in extreme warm days and nights is always overestimated in the simulations, from the Coupled Model Intercomparison Project Phase 5 (CMIP5), those models display rising trends whatever the scenario, which are likely to be more and more pronounced over the two regions in the next 50 years

The horn of Africa is susceptible to droughts because of the persistent heat waves and insufficient precipitation. The growth of urban population and built-up urban environments exacerbate the overheating problems due to the urban heat island effects. Understanding the impacts of anthropogenic activities in such dry environments is important to control or mitigate extreme heat leading to droughts. This is required to preserve soil moisture, pothole waters, lakes and rivers that are required for pasture and drinking water. Nonetheless, the intensity and duration of the urban heat island effects have not been investigated in this region resulting in the underestimation of the intensity and severity of the extreme heat events. This study therefore performs the quantitative analyses of the intensity, duration and causality of the tropical surface urban heat islands (TSUHIs) for the first time using earth observation information at a regional to local scale. It also identifies the factors that control TSUHIs, considering background climate, population, vegetation and the impervious urban fractions. Results showed that the TSUHI in the capital cities of tropical east Africa varies from 1 $$^{\circ }$$∘ C in Dodoma to 4 $$^{\circ }$$∘ C in Kampala and reaches up to 8 $$^{\circ }$$∘ C in Khartoum. The mean temperature contribution to regional climate from 2000 to 2020 is 0.64 $$^{\circ }$$∘ C during the day and 0.34 $$^{\circ }$$∘ C during the night, a mean total of around 0.5 $$^{\circ }$$∘ C, a 0.25 $$^{\circ }$$∘ C increase per decade. This is a quarter of the increase in global surface temperature, which is $$\approx$$≈ 1.09 $$^{\circ }$$∘ C from 2011 to 2020 compared to the 1850–1900 level. Most of these capital cities in this region exhibited high TSUHIs from late summer to winter and are dependent on mainly population, vegetation, evapotranspiration and soil moisture in different proportions. This urban induced additional temperature has been intensifying droughts in tropical east Africa. Therefore, urban planners are advised to consider the impacts of TSUHIs to reduce the severity of droughts in the tropical east Africa region.

An increase in global temperature will likely result in more intense and frequent heatwaves that would last longer. Simultaneously, the growth of urban population requires more areas of land incorporated into urbanization, because most people are expected to live in cities, which will increase the intensity and duration of urban heat islands. However, the extent of the link between global warming induced heatwaves and urbanization caused heat islands is barely understood. Understanding the link would give a new information about catastrophic heat mitigation strategies. This paper, therefore, quantifies, at the sub-continental scale of Eastern North America, the effects of background perturbations by the synergies between heatwaves and urban heat islands using simulations from the Weather Research and Forecasting (WRF) model, and focusing on the responses of urban energy balances, boundary layer height and vertical profiles of heat, momentum and moisture. Results showed that urban heat islands exacerbate heatwaves by deepening the turbulent boundary layer height, modifying the urban surface energy and regional winds. The fractional energy shift from latent to sensible heat fluxes and the consequent changes to the urban planetary boundary layer tends to amplify the intensity, extent and duration of extensive heatwaves. The response of ground heat fluxes to urban surfaces lags, while urban canopy humidity dissipates earlier because at the onset of the heatwave the surface water evaporates quickly to the point where there is less water left for evaporation leaving the urbanized regions vulnerable to more heating. During the heatwave event, the mean wind speed dropped by 2.5 m/s, hence less cool air is available for ventilation. The planetary boundary layer deepens by a maximum of 90-m over urban compared to rural and this may prolong urban surface heating. Based on the results, it can be concluded that the best heat-stress management strategies from the perspectives of urban energy balance and planetary boundary layer height is an integral approach that would lower sensible heat fluxes and increase surface albedo, latent heat fluxes and wind flows towards urban centers.