Human activities, primarily greenhouse gas emissions, are unequivocally responsible for the observed global warming since preindustrial times, with accelerating and discernible impacts (IPCC, 2023). Anticipating the rate of global warming until the end of the century however remains delicate since it does not only depend on future emissions but also on complex and model-dependent climate feedbacks. While global warming has been shown to occur at near-constant relative humidity on a global scale (Allan 2022; Douville 2022), thus leading to a strong positive water vapor feedback, regional departures from this equilibrium may represent a significant source of uncertainty for projections of land surface temperatures. Similarly, increasing water vapor may intensify policy-relevant extreme events like heavy precipitation, flooding, and hurricanes, whose projections are also highly model-dependent at the regional scale (Byrne & O’Gorman 2021; Douville 2021, John 2022).

Understanding and projecting climate change relies both on observations (including reanalyses) and coupled Earth System Models. Quality-checked observations and reanalyses are fundamental not only for the tuning and evaluation of the model’s ability to capture the key features of present-day climate, but also for the detection and attribution of observed climate change. The latter is particularly challenging for water cycle changes given their weaker signal-to-noise ratio (greater internal climate variability) than for surface temperatures and the competing effects of other greenhouse gases and external radiative forcings such as natural and anthropogenic aerosols (Douville 2021). Yet, the growing emergence of the anthropogenic signals at the regional scale offers the opportunity not only to better understand the recent climate but also to constrain the future climate projections (Douville 2022; Qasmi & Ribes 2022).

Global atmospheric reanalyses have well-known limitations in representing water and energy cycles (Wild & Bosilovich 2024). In particular, multiple and sometimes abrupt changes in the global observation system, such as the availability of new satellite data, may induce artificial trends at both global and regional scales. Bias correction with independent observations is therefore crucial for identifying missing or poorly simulated processes in historical climate simulations, but also improving trend accuracy in global reanalyses (Douville, 2021). The GNSS (Global Navigation Satellite System) provides precise global positioning, supporting geodesy and meteorology. While weather centers use near real-time GNSS tropospheric data, it is not yet integrated into global reanalyses. GNSS offers hourly Integrated Water Vapor (IWV) data with up to 30 years of global coverage, serving as an independent resource for evaluating both reanalyses and climate models (Bock & Parracho 2019; Bock 2024).

This PhD project will use GNSS IWV data to investigate regional water cycle dynamics in climate simulations for the next IPCC assessment report. The work will involve collaboration with the SPOTGINS community, Data Terra, CLIMERI-France, and PEPR TRACCS. The objectives are:

1) Establish a reference global GNSS IWV dataset by leveraging the expertise of the IPGP-IGN geodesy team and the processing capacity of the SPOTGINS consortium using CNES’s GINS software. The dataset will include hundreds of reference stations and thousands of auxiliary stations, with Data Terra’s infrastructure supporting post-processing and homogenization (Nguyen 2024), and AI-based spatialization for high-density networks (Bochow 2025). A particular attention will be paid over specific regions such as western Europe and the South Asian and West African monsoon regions.

2) Investigate uncertainties in the water cycle of recent global reanalyses and climate models using the enhanced GNSS IWV dataset, alongside radiosonde and satellite products. The analysis will span global and regional scales, using CMIP7 and CORDEX simulations, with a focus on French climate models (CNRM, IPSL) to assess the impact of parameterizations on water and energy cycles. An attempt will be also made to constrain both available (CMIP5 and CMIP6) and future (CMIP7) climate projections over the regions of interest.

References:

- Allan 2022 https://doi.org/10.1029/2022JD036728

- Bock & Parracho 2019 https://doi.org/10.5194/acp-19-9453-2019

- Bock 2024 https://doi.org/10.1175/BAMS-D-24-0116.1

- Bochow 2025 https://doi.org/10.1126/sciadv.adp0558

- Byrne & O’Gorman 2018 https://doi.org/10.1073/pnas.1722312115

- Douville 2021 https://doi.org/10.1017/9781009157896.010

- Douville 2022 https://doi.org/10.1038/s43247-022-00561-z

- IPCC 2023 https://doi.org/10.59327/IPCC/AR6-9789291691647

- John 2022  https://doi.org/10.1016/j.wace.2022.100435

- Nguyen 2024 https://hal.science/hal-04014145

- Qasmi & Ribes 2022 https://doi.org/10.1126/sciadv.abo6872

- Wild & Bosilovich 2024 https://doi.org/10.1007/s10712-024-09861-9

=================

For more Information about the topics and the co-financial partner (found by the lab!); contact Directeur de thèse - bock@ipgp.fr

Then, prepare a resume, a recent transcript and a reference letter from your M2 supervisor/ engineering school director and you will be ready to apply online before March 13th, 2026 Midnight Paris time!