14e édition des Rencontres du Ciel et de l'Espace

09/11/2024 11:00

Changement climatique et énergie : là où l'océan fait la part du géant

11/10/2024 12:30

Cycle de conférences sur l'agriculture et la biodiversité

10/10/2024 17:30

New developments for paleoclimate studies

02/06/2025 13:00

Séminaire de l’UMR METIS-IPSL.

A paleolimnological perspective on avian population dynamics and their effect on lake productivity in a rapidly warming Arctic

26/05/2025 13:00

Séminaire de l’UMR METIS-IPSL.

Retour sur l’école “Arts et sciences face au changement climatique et à la transition écologique”

23/05/2025 11:00

Webinaire TRACCS.

Summertime Nocturnal Low-Level Jet in the Paris Region: Interactions with the Urban Heat and Surface Characteristics

13/11/2024 14:00

Urban areas, covering 1-3% of global land but hosting over half of the world’s population, play a crucial role in the interaction between human activities and the Atmospheric Boundary Layer (ABL). Cities modify the ABL through direct heat emissions, impervious materials, and increased surface roughness, leading to unique microclimates with lower wind speeds and increased turbulence. A prominent effect is the nocturnal Urban Heat Island (UHI), where cities remain warmer than their rural surroundings, intensifying heat-related health risks, particularly at night.

Despite the significance of these phenomena, the interactions between synoptic weather patterns and urban-induced atmospheric changes remain poorly understood, mainly due to a lack of detailed observations at urban sites. This thesis investigates the interaction between the nocturnal mesoscale Low-Level Jet (LLJ) and the urban environment in the Paris region. Over two years, wind and turbulence data were collected using Doppler Wind Lidar (DWL) at two sites: an urban location in central Paris and a suburban site 25 km southwest. An automatic algorithm was implemented to detect and characterize LLJ events, alongside a new shallow Doppler Beam Swinging (DBS) scan to monitor shallow LLJs within the blind zone of the urban DWL.

The study focused on the summer periods of 2022 and 2023, providing a detailed analysis of LLJ characteristics. Results showed that LLJs are frequent in the region, occurring on 70% of nights in 2022 and 40-48% in 2023, with differences driven by synoptic weather conditions. LLJ core heights ranged from 200 to 950 meters, with wind speeds between 3 and 18 m/s. The study also found a link between LLJ characteristics and UHI intensity, where weaker vertical mixing led to more pronounced UHI effects during extreme heat events. These findings highlight the critical role of ABL dynamics in understanding urban environmental risks, such as heat stress.

Contraindre les flux de photosynthèse et de transpiration à grande échelle dans un modèle de biosphère par assimilation de mesures de flux in situ dont l’oxysulfure de carbone

05/11/2024 09:30

La production primaire brute (GPP), absorption photosynthétique du CO₂ atmosphérique par la végétation, joue un rôle crucial dans l’atténuation du changement climatique. La transpiration, émission d’eau par les plantes concomitante à la GPP, renvoie une part significative des précipitations terrestres dans l’atmosphère. Bien que la GPP soit le plus grand flux du cycle du carbone et que la transpiration soit la principale composante de l’évapotranspiration terrestre, leurs estimations globales et leurs réponses au changement climatique demeurent incertaines. Cette recherche vise donc à améliorer la simulation de la GPP et de la transpiration dans un modèle de surfaces continentales, ORCHIDEE. J’ai exploré l’utilisation de mesures d’oxysulfure de carbone (COS), un gaz atmosphérique absorbé par les plantes de manière similaire au CO₂, pour contraindre la GPP et la transpiration dans ORCHIDEE.

J’ai tout d’abord implémenté un modèle des échanges de COS par les sols, complétant le modèle existant d’absorption de COS par la végétation. Ce développement a permis de simuler les flux de COS de l’écosystème à différentes échelles et de fournir de nouvelles estimations des contributions de la végétation et des sols au budget global du COS. Cette étude a montré l’importance de considérer la capacité des sols oxiques à émettre du COS en plus de leur absorption, et que les sols anoxiques peuvent produire des quantités significatives de COS (96 GgS/an), compensant en grande partie l’absorption nette de COS par les sols oxiques (-126 GgS/an), résultant en une absorption nette totale des sols de -30 GgS/an. Avec ces développements, des mesures in situ de flux de COS de l’écosystème ont pu être utilisées pour optimiser les paramètres d’ORCHIDEE via des techniques d’assimilation de données (DA).

J’ai assimilé la plus longue série temporelle de flux de COS de l’écosystème de la forêt boréale de Hyytiälä en Finlande, afin d’optimiser les paramètres impliqués dans la simulation de la GPP et de la transpiration. La comparaison entre une assimilation conjointe ou indépendantes des données de COS et de GPP a montré que l’assimilation des données de COS en plus de celles de GPP améliore à la fois le flux de chaleur latente (LE) et la GPP simulés, ce que ne permet pas l’assimilation de la GPP seule. L’application des paramètres optimisés à l’ensemble de ce biome boréal a augmenté l’absorption de COS dans les hautes latitudes, en accord avec des études indépendantes. Enfin, l’évaluation de la contrainte apportée par le COS sur la GPP et le LE simulés lors d’une sécheresse ayant révélé des erreurs structurelles non corrigées par la DA, la dernière partie de ma thèse s’est concentrée sur l’amélioration de la représentation de la réponse de la végétation à un stress hydrique. J’ai assimilé des données de GPP et de LE sur plus de 40 sites enregistrant les récentes années de sécheresse en Europe.

Ce travail a montré que le paramètre déterminant la vitesse de fermeture stomatique pendant un stress hydrique pouvait être défini en fonction des conditions à long terme de déficit de pression de vapeur. L’implémentation de cette réponse dans ORCHIDEE permet de mieux prendre en compte la diversité de réponse de la végétation aux sécheresses et sa capacité d’acclimatation. Enfin, j’ai réalisé des projections pour évaluer l’impact de cette nouvelle réponse dans un climat futur. En conclusion, cette recherche encourage l’utilisation du COS comme proxy pour la GPP et la transpiration, et préconise davantage de mesures de flux et de concentration de COS pour affiner la paramétrisation des modèles. Elle suggère également des pistes pour améliorer la GPP et la transpiration simulées dans les modèles en tenant compte de la capacité des plantes à s’acclimater face au changement climatique.

 

---

 

Gross primary production (GPP), the photosynthetic absorption of atmospheric CO2 by continental vegetation, plays a critical role in climate mitigation. In parallel, transpiration, the emission of water by plants, returns a significant portion of land precipitation to the atmosphere. Despite GPP being the largest carbon cycle flux and transpiration the largest component of terrestrial evapotranspiration, their global estimates and responses to climate change remain uncertain. Therefore, this research aims to enhance the simulation of GPP and plant transpiration in a land surface model (LSM), ORCHIDEE. First, I investigated the use of carbonyl sulfide (COS) measurements, an atmospheric gas absorbed by plants similarly to CO2, to constrain GPP and plant transpiration in ORCHIDEE. This involved implementing a model for soil COS exchanges, complementing the existing vegetation COS uptake model. This development enabled the simulation of ecosystem COS fluxes from the site to the global scale and provided new estimates of vegetation and soil contributions to the global atmospheric COS budget. Notably, this study highlighted the importance of considering that oxic soils can emit COS in addition to absorbing COS, and that anoxic soils can produce significant amounts of COS (96 GgS y-1), offsetting much of the net oxic soil COS uptake (-126 GgS y-1), and resulting in a total global net soil uptake of -30 GgS y-1. With these developments, in situ measurements of ecosystem COS fluxes have been used to optimize the parameters of ORCHIDEE through data assimilation (DA) techniques.

Therefore, in a second phase, I assimilated the longest time series of ecosystem COS flux from the Hyytiälä boreal evergreen needleleaf forest in Finland to optimize parameters involved in GPP and plant transpiration simulation. Comparing a joint assimilation or independent assimilations of COS and GPP data showed that assimilating COS along with GPP data improves the simulated latent heat flux (LE) as well as GPP, unlike the GPP-only assimilation. Upscaling the optimized parameters across the boreal evergreen needleleaf forest biome increased COS uptake in high latitudes, aligning with independent studies. Finally, as the evaluation of COS constraint on the simulated GPP and LE during a drought event revealed structural errors in the simulated fluxes uncorrected by DA, the final part of my PhD focused on improving the representation of vegetation response to drought events in ORCHIDEE.

To this end, I assimilated GPP and LE data at over 40 sites capturing recent drought years across Europe. This work demonstrated that the parameter determining stomatal closure speed during soil moisture stress, influencing both GPP and transpiration, could be defined as a function of long term vapor pressure deficit conditions. Implementing this response in ORCHIDEE allows better consideration of site-specific vegetation responses to droughts and vegetation acclimation capacity.

Finally, I performed projections to assess the impact of this new response under future climate. Overall, this research supports using COS as a proxy for GPP and transpiration, advocating for more COS flux and concentration measurement campaigns to refine LSM parameterization. It also suggests future avenues to improve GPP and plant transpiration representations in LSMs by accounting for plants’ ability to acclimate in their response to climate change.

Changes in precipitation in the Mediterranean with climate change: between extreme rainfall and dry-days

14/10/2024 14:00

Climate change significantly impacts precipitation at global and regional scale, especially in the Mediterranean region, a climate change hotspot. Existing literature suggests a « water cycle paradox » in this region (decrease of mean precipitation but intensification of heavy precipitations) particularly in the Northern part of the Mediterranean, but this is not yet clear in observations.
This study investigates how the entire precipitation distribution in the Mediterranean has changed over the past and how it might evolve in the future. Using ERA5 reanalysis data (1950-2020) and Euro-CORDEX simulations (1950-2100), the study analyzes temporal trends in precipitation quantiles. We firsy study wet-days (days with more than 1 mm/day) in the past distributions, then look at how changes in dry-day frequency influence the all-days distribution. The study identifies four distinct precipitation regimes: a « U-shape » quantile trend curve (decreasing low to medium quantiles but intensifying extremes), a reversed « U-shape », and two regimes where all quantiles either consistently increase or decrease. Using a Weibull model for the wet-days distribution, the study further determines these four regimes thanks to two parameters. We find regional differences, with a significant intensification in Europe, while in the Mediterranean, changes in precipitation are actually dominated by the increase of dry-days frequency.
Future projections under the RCP5.8 scenario (high emissions) suggest robust trends, with a sharp increase in the frequency of dry-days across the Mediterranean, contrasting with a decline in dry-days and an intensification of wet-days in Northern Europe. The time of emergence of these trends is faster in Northern Europe than in the Mediterranean. This study underlines the essential role of dry-day frequency in the evolution of the precipitation distribution in the Mediterranean.