Search

In this paper, we present an approach that combines data-driven and physical modelling for predicting the runoff occurrence and volume at catchment scale. With that aim, we first estimated the runoff volume from recorded storms aided by the Green-Ampt infiltration model. Then, we used machine learning algorithms, namely LightGBM (LGBM) and Deep Neural Network (DNN), to predict the outputs of the physical model fed on a set of atmospheric variables (relative humidity, temperature, atmospheric pressure, and wind velocity) collected before or immediately after the beginning of the storm. Results for a small urban catchment in Madrid show DNN performed better in predicting the runoff occurrence and volume. Moreover, enriching the input primary atmospheric variables with auxiliary variables (e.g., storm intensity data recorded during the first hour, or rain volume and intensity estimates obtained from auxiliary regression methods) largely increased the model performance. We show in this manuscript data-driven algorithms shaped by physical criteria can be successfully generated by allowing the data-driven algorithm learn from the output of physical models. It represents a novel approach for physics-informed data-driven algorithms shifting from common practices in hydrological modelling through machine learning. • Data-driven and physical modelling are merged to predict runoff. • Data-driven algorithm fed on weather variables before the storm starts. • Light GBM and Deep Neural Networks were coupled with Green-Ampt model. [ABSTRACT FROM AUTHOR]

This paper analyses the probabilistic future behaviour of heat-waves (HWs) in the city of Madrid in the twenty-first century, using maximum daily temperatures from twenty-one climate circulation models under two representative concentration pathways (RCP 8.5 & RCP 4.5). HWs are modelled considering three factors: number per annum, duration and intensity, characterised by three stochastic processes: Poisson, Gamma and truncated Gaussian, respectively. Potential correlations between these processes are also considered. The probabilistic temperature behaviour is combined with an epidemiological model with stochastic mortality risk following a generalized extreme value distribution (gev). The objective of this study is to obtain probability distributions of mortality and risk measures such as the mean value of the 5% of worst cases in the 21st century, in particular from 2025 to 2100. Estimates from stochastic models for characterising HWs and epidemiological impacts on human health can vary from one climate model to another, so relying on a single climate model can be problematic. For this reason, the calculations are carried out for 21 models and the average of the results is obtained. A sensitivity adaptation analysis is also performed. Under RCP 8.5 for 2100 for Madrid city a mean excess of 3.6 °C over the 38 °C temperature threshold is expected as the average of all models, with an expected attributable mortality of 1614 people, but these figures may be substantially exceeded in some cases if the highest-risk cases occur. [ABSTRACT FROM AUTHOR]

While the COVID-19 pandemic is still in progress, being under the fifth COVID-19 wave in Madrid, over more than one year, Spain experienced a four wave pattern. The transmission of SARS-CoV-2 pathogens in Madrid metropolitan region was investigated from an urban context associated with seasonal variability of climate and air pollution drivers. Based on descriptive statistics and regression methods of in-situ and geospatial daily time series data, this study provides a comparative analysis between COVID-19 waves incidence and mortality cases in Madrid under different air quality and climate conditions. During analyzed period 1 January 2020–1 July 2021, for each of the four COVID-19 waves in Madrid were recorded anomalous anticyclonic synoptic meteorological patterns in the mid-troposphere and favorable stability conditions for COVID-19 disease fast spreading. As airborne microbial temporal pattern is most affected by seasonal changes, this paper found: 1) a significant negative correlation of air temperature, Planetary Boundary Layer height, and surface solar irradiance with daily new COVID-19 incidence and deaths; 2) a similar mutual seasonality with climate variables of the first and the fourth COVID-waves from spring seasons of 2020 and 2021 years. Such information may help the health decision makers and public plan for the future. • Climate factors and air quality can trigger SARS-CoV-2 viral infection transmission. • Synoptic atmospheric circulation patterns are related to COVID-19 waves start-up. • 1st and 4th COVID-19 waves have similar correlations with environmental variables. • COVID-19 and climate driving factors have mutual seasonality patterns in Madrid. • Early health strategies are essential to control the COVID-19 pandemic in Madrid. [ABSTRACT FROM AUTHOR]