Search

We present an Observing System Simulation Experiment (OSSE) dedicated to the evaluation of the added value of the Sentinel 4 and Sentinel 5P missions for tropospheric nitrogen dioxide (NO2). Sentinel 4 is a geostationary (GEO) mission covering the European continent, providing observations with high temporal resolution (hourly). Sentinel 5P is a low-Earth Orbiting (LEO) mission providing daily observations but with a global coverage. The OSSE experiment has been carefully designed, with separate models for the simulation of observations and for the assimilation experiments, and with conservative estimates of the total observation uncertainties. In the experiment we simulate Sentinel 4 and Sentinel 5P tropospheric NO2 columns and surface ozone concentrations at 7 by 7 km resolution over Europe for two three-month summer and winter periods. The synthetic observations are based on a nature run (NR) from a chemistry transport model (MOCAGE) and error estimates using instrument characteristics. We assimilate the simulated observations into a chemistry transport model (LOTOS-EUROS) independent from the NR to evaluate their impact on modelled NO2 tropospheric columns and surface concentrations. The results are compared to an operational system where only ground-based ozone observations are ingested. Both instruments have an added value on analysed NO2 columns and surface values, reflected in decreased biases, and improved correlations. The Sentinel 4 NO2 observations with hourly temporal resolution benefit modelled NO2 analyses throughout the entire day where the daily Sentinel 5P NO2 observations have a slightly lower impact that lasts up to 3–6 hours after overpass. The evaluated benefits may be even higher in reality as the applied error estimates were shown to be higher than actual errors in the now operational Sentinel 5P NO2 products. We show that an accurate representation of the NO2 profile is crucial for the benefit of the column observations on surface values. The results support the need for having a combination of GEO and LEO missions for NO2 analyses in view of the complementary benefits of hourly temporal resolution (GEO, Sentinel 4) and global coverage (LEO, Sentinel 5P).

We present an Observing System Simulation Experiment (OSSE) dedicated to the evaluation of the added value of the Sentinel-4 and Sentinel-5P missions for tropospheric nitrogen dioxide (NO2). Sentinel-4 is a geostationary (GEO) mission covering the European continent, providing observations with high temporal resolution (hourly). Sentinel-5P is a low Earth orbit (LEO) mission providing daily observations with a global coverage. The OSSE experiment has been carefully designed, with separate models for the simulation of observations and for the assimilation experiments and with conservative estimates of the total observation uncertainties. In the experiment we simulate Sentinel-4 and Sentinel-5P tropospheric NO2 columns and surface ozone concentrations at 7 by 7 km resolution over Europe for two 3-month summer and winter periods. The synthetic observations are based on a nature run (NR) from a chemistry transport model (MOCAGE) and error estimates using instrument characteristics. We assimilate the simulated observations into a chemistry transport model (LOTOS-EUROS) independent of the NR to evaluate their impact on modelled NO2 tropospheric columns and surface concentrations. The results are compared to an operational system where only ground-based ozone observations are ingested. Both instruments have an added value to analysed NO2 columns and surface values, reflected in decreased biases and improved correlations. The Sentinel-4 NO2 observations with hourly temporal resolution benefit modelled NO2 analyses throughout the entire day where the daily Sentinel-5P NO2 observations have a slightly lower impact that lasts up to 3–6 h after overpass. The evaluated benefits may be even higher in reality as the applied error estimates were shown to be higher than actual errors in the now operational Sentinel-5P NO2 products. We show that an accurate representation of the NO2 profile is crucial for the benefit of the column observations on surface values. The results support the need for having a combination of GEO and LEO missions for NO2 analyses in view of the complementary benefits of hourly temporal resolution (GEO, Sentinel-4) and global coverage (LEO, Sentinel-5P).

International audience; The amount of ice injected up to the tropical tropopause layer has a strong radiative impact on climate. In the tropics, the Maritime Continent (MariCont) region presents the largest injection of ice by deep convection into the upper troposphere (UT) and tropopause level (TL) (from results presented in the companion paper Part 1). This study focuses on the MariCont region and aims to assess the processes, the areas and the diurnal amount and duration of ice injected by deep convection over islands and over seas using a 2 • × 2 • horizontal resolution during the austral convective season of December, January and February. The model presented in the companion paper is used to estimate the amount of ice injected (∆IWC) up to the TL by combining ice water content (IWC) measured twice a day in tropical UT and TL by the Microwave Limb Sounder (MLS; Version 4.2), from 2004 to 2017, and precipitation (Prec) measurement from the Tropical Rainfall Measurement Mission (TRMM; Version 007) at high temporal resolution (1 hour). The horizontal distribution of ∆IWC estimated from Prec (∆IWC P rec) is presented at 2 • × 2 • horizontal resolution over the MariCont. ∆IWC is also evaluated by using the number of lightnings (Flash) from the TRMM-LIS instrument (Lightning Imaging Sensor, from 2004 to 2015 at 1-h and 0.25 • ×0.25 • resolutions). ∆IWC P rec and ∆IWC estimated from Flash (∆IWC F lash) are compared to ∆IWC estimated from the ERA5 reanalyses (∆IWC ERA5) degrading the vertical resolution to that of MLS observations (∆IW C ERA5). Our study shows that, while the diurnal cycles of Prec and Flash are consistent to each other in timing and phase over lands and different over offshore and coastal areas of the MariCont, the observational ∆IWC range between ∆IWC P rec and ∆IWC F lash is small (to within 4-20% over land and to within 6-50% over ocean) in the UT and TL. The reanalysis ∆IWC range between ∆IWC ERA5 and ∆IW C ERA5 has been also found to be small in the UT (22-32 %) but large in the TL (68-71 %), highlighting the stronger impact of the vertical resolution on the TL than in the UT. Combining observational and reanalysis ∆IWC ranges, the total ∆IWC range is estimated in the UT between 4.17 and 9.97 mg m-3 (20 % of variability per study zone) over land and between 0.35 and 4.37 mg m-3 (30% of variability per study zone) over sea, and, in the TL, between 0.63 and 3.65 mg m-3 (70% of variability per study zone) over land and between 0.04 and 0.74 mg m-3 (80% of variability per study zone) over sea. Finally, from IWC ERA5 , Prec and Flash, this study highlights 1) ∆IWC over land has been found larger than ∆IWC over sea, and 2) the Java Island is the area of the largest ∆IWC in the UT (7.89-8.72 mg m-3 daily mean).

The global context of my PhD thesis work is the optimization of air pollution monitoring networks, but more specifically it concerns the monitoring of accidental releases of radionuclides in air. The optimization problem of air quality measuring networks has been addresses in the literature. However, it has not been addresses in the context of surveillance of accidental atmospheric releases. The first part of my thesis addresses the optimization of a permanent network of monitoring of radioactive aerosols in the air, covering France. The second part concerns the problem of targeting of observations in case of an accidental release of radionuclides from a nuclear plant.; Mon travail de thèse se situe dans le contexte général de l'optimisation de réseaux de mesure de pollution atmosphérique, mais plus spécifiquement centré sur la surveillance des rejets accidentels de radionucléides dans l'air. Le problème d'optimisation de réseaux de mesure de la qualité de l'air a été abordé dans la littérature. En revanche, il n'a pas été traité dans le contexte de la surveillance des rejets accidentels atmosphériques. Au cours de cette thèse nous nous sommes intéressés dans un premier temps à l'optimisation du futur réseau de télésurveillance des aérosols radioactifs dans l'air, le réseau DESCARTES. Ce réseau sera mis en œuvre par l'Institut de Radioprotection et de Sûreté Nucléaire (IRSN), afin de renforcer son dispositif de surveillance de radionucléides en France métropolitaine. Plus précisément, l'objectif assigné à ce réseau est de pouvoir mesurer des rejets atmosphériques de radionucléides, provenant de l'ensemble des installations nucléaires françaises ou étrangères. Notre principal rôle était donc de formuler des recommandations vis-à-vis aux besoins exprimés par l'IRSN, concernant la construction optimale du futur réseau. Á cette fin, l'approche que nous avons considérée pour optimiser le réseau (le futur réseau), vise à maximiser sa capacité à extrapoler les concentrations d'activité mesurées sur les stations du réseau sur tout le domaine d'intérêt. Cette capacité est évaluée quantitativement à travers une fonction de coût, qui mesure les écarts entre les champs de concentrations extrapolés et ceux de références. Ces derniers représentent des scénarios de dispersion accidentels provenant des 20 centrales nucléaires françaises et, calculés sur une année de météorologie. Nos résultats soulignent notamment l'importance du choix de la fonction coût dans la conception optimale du futur réseau de surveillance. Autrement dit, la configuration spatiale du réseau optimal s'avère extrêmement sensible à la forme de la fonction coût utilisée. La deuxième partie de mon travail s'intéresse essentiellement au problème du ciblage d'observations en cas d'un rejet accidentel de radionucléides, provenant d'une centrale nucléaire. En effet, en situation d'urgence, une prévision très précise en temps réel de la dispersion du panache radioactif est vivement exigée par les décideurs afin d'entreprendre des contre-mesures plus appropriées. Cependant, la précision de la prévision du panache est très dépendante de l'estimation du terme source de l'accident. À cet égard, les techniques d'assimilation de données et de modélisation inverse peuvent être appliquées. Toutefois, le nuage radioactif peut être localement très mince et pourrait s'échapper à une partie importante du réseau local, installé autour de la centrale nucléaire. Ainsi, un déploiement de stations de mesure mobiles en suivant l'évolution du nuage pourrait contribuer à améliorer l'estimation du terme source. À cet effet, nous avons exploré la possibilité d'améliorer la qualité de la prévision numérique du panache radioactif, en couplant une stratégie de déploiement optimal de stations mobiles avec un schéma d'assimilation de données pour la reconstruction séquentielle du panache radioactif. Nos résultats montrent que le gain d'information apporté par les observations ciblées est nettement mieux que l'information apportée par les observations fixes.