CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Analysis of long-term coupling of the thermospheric density and the tropospheric CO2

Estudiants que han llegit aquest projecte:

Director/a: GIL PONS, PILAR

Departament: FIS

Títol: Analysis of long-term coupling of the thermospheric density and the tropospheric CO2

Data inici oferta: 22-05-2026     Data finalització oferta: 22-01-2027


Estudis d'assignació del projecte:
    DG ENG AERO/SIS TEL
    DG ENG AERO/TELEMÀT
Tipus: Individual
 
Lloc de realització: EETAC
 
Segon director/a (UPC): GUTIÉRREZ CABELLO, JORDI
 
Paraules clau:
Thermosphere, PCMCI, CO2, Atmospheric density, Density, F10.7, Ap, Long-term, Atmosphere, LEO, VLEO
 
Descripció del contingut i pla d'activitats:
Context: Assessing the thermospheric density is essential for accurately determining satellite trajectories and re-entry stages, particularly in Low Earth Orbit (LEO) and Very Low Earth Orbit (VLEO). Recent literature has identified an interesting trend linking the increase in tropospheric CO$_2$ concentrations to a decrease in thermospheric density of approximately 5% per decade. The underlying mechanism is qualitatively understood: CO2 is expected to cool and contract the thermosphere, reducing neutral density at fixed altitudes and altering satellite drag. However, isolating the specific contribution of CO2 from observational data remains challenging, as thermospheric density is simultaneously affected by strong drivers, including solar ultraviolet radiation, geomagnetic activity, and other space-weather phenomena.


Goal: This thesis aims to develop a reproducible data pipeline that harmonises multiple observational datasets of thermospheric density-including those from TU Delft, PANGAEA, and the Swarm mission-and integrates them with a range of atmospheric CO2 records (NOAA, EDGAR, Global Carbon Budget, CarbonTracker CT2022, among others). The analysis will explicitly account for space weather variability to disentangle the CO2-driven signal from other external effects.


Methodology: To assess causal relationships under the influence of strong external drivers, the analysis employs nonlinear, lag-aware causality testing combined with surrogate-based significance assessment. This approach enables detection of directional coupling between tropospheric CO$_2$ and thermospheric density while accouting for other factors such as solar and geomagnetic activity.
Expected results: The study aims to identify the dominant physical mechanisms driving the decline in thermospheric density and to characterise the properties of its potential correlation with tropospheric CO2 concentrations. By quantifying this relationship, the work will contribute to improved modelling of upper atmospheric evolution and its implications for predicting satellite drag and orbital debris.
 
Overview (resum en anglès):
This thesis studies long-term thermospheric density variability and its relationship with solar, geomagnetic and carbon-dioxide-related forcing. The analysis combines three density sources: an orbit-derived global mean density record, the TU Delft satellite density dataset, and a local HASDM subset near Mauna Loa. Time-series inspection, frequency-domain analysis, correlation, activity-binned comparisons and PCMCI+ causal discovery are used with F10.7, Ap, Kp, Mauna Loa CO2, and SABER CO2 cooling-rate diagnostics.
The results show that resolved thermospheric density variability is dominated by solar and geomagnetic forcing. The negative association between CO2 and density is physically consistent with thermospheric cooling and contraction, but it is not isolated conclusively by correlation or causal discovery in the tested configurations. The model-validation results show that the model log-density-ratio error provides a consistent way to compare NRLMSISE-00, NRLMSIS 2.0 and NRLMSIS 2.1 against TU Delft and HASDM reference densities. NRLMSIS 2.0 and NRLMSIS 2.1 reduce part of the middle-thermosphere bias relative to NRLMSISE-00, although the improvement depends on altitude and activity regime. Overall, the work provides a reproducible framework for separating solar-geomagnetic control, CO2-related long-term structure and residual empirical-model error.

© CBLTIC Campus del Baix Llobregat - UPC