Theertha Kariyathan

An Atmospheric Perspective on the Northern Hemisphere Carbon Uptake Period

Terrestrial ecosystems play a crucial role in regulating atmospheric CO2 concentration by absorbing a substantial fraction of anthropogenic emissions. This net sink behaviour emerges from the balance between the two dominant terrestrial carbon fluxes: gross primary production (GPP), the uptake of CO2 through photosynthesis, and total ecosystem respiration (TER), the release of CO2 from autotrophic and heterotrophic respiration; their difference defines the net ecosystem exchange (NEE). When GPP exceeds TER, photosynthetic carbon uptake exceeds respiratory CO2 release, and ecosystems function as a net sink of atmospheric CO2. This period of the year is referred to as the Carbon Uptake Period (CUP). The timing and duration of the CUP are influenced by environmental conditions such as temperature, radiation, and water availability, which regulate the seasonal dynamics of photosynthesis and respiration. In addition to these climatic controls, episodic disturbances and anomalies, such as droughts, heatwaves, or wildfire events, can substantially alter the seasonal balance between uptake and release. By integrating net CO2 exchange between the land biosphere and the atmosphere, the CUP provides a process-relevant indicator of ecosystem carbon balance and a metric for assessing ecosystem responses to climate variability and long-term change. CUP timing can be estimated from long-term atmospheric CO2 records, characterised by the annual rise and decline in CO2 levels driven by ecosystem transitions between net carbon uptake and release. However, extracting reliable CUP timing from atmospheric observations remains challenging, as measurements integrate broad source regions and are influenced by transport, synoptic variability, and methodological uncertainties associated with time-series analysis. This highlights the need to improve CUP estimation methods and better constrain how they reflect changes in surface fluxes.

The overall objective of this thesis is to improve understanding of changes in CUP in the Northern Hemisphere. To achieve this, we primarily use atmospheric CO2 observations, first developing improved methods to estimate CUP from these records, then assessing the influence of atmospheric transport on these estimates, and finally diagnosing recent changes in CUP timing and their ecological drivers. The thesis addresses three central questions: (1) how CUP metrics can be robustly estimated from atmospheric CO2 time series, (2) how atmospheric transport affects the relationship between atmospheric signals and surface carbon fluxes, and (3) what recent changes in CUP timing reveal about regional ecosystem responses to climate variability.

The first chapter of this thesis focuses on improving the estimation of CUP timing from atmospheric CO2 time series, which are discrete, noisy, and often contain gaps. We developed the Ensemble of the First Derivative (EFD) method, which reduces biases associated with the choice of curve-fitting algorithms and decreases the uncertainty in the estimated CUP metrics compared with traditional proxy definitions, such as the zero-crossing point of the detrended seasonal cycle. The method generates an ensemble of plausible fits to the time series and estimates CUP timing from the first derivative of these fits using a threshold defined relative to the minimum of the derivative. The threshold is selected such that it minimizes uncertainty across the ensemble. When evaluated using two different curve-fitting frameworks, the EFD approach shows that threshold-based CUP estimates exhibit substantially lower uncertainty than traditional zero-crossing metrics. Moreover, while earlier derivative-based approaches (Barlow et al., 2015) relied on relatively high thresholds (25% of the seasonal minimum of the first derivative) to identify the onset of the CUP, the reduced uncertainty achieved with the EFD framework allows the use of a lower threshold (15%). This results in CUP estimates that include a larger portion of the early-season carbon uptake period.

The second chapter of the thesis investigates how atmospheric transport affects the relationship between the CUP derived from atmospheric CO2 observations (CUPMR) and that derived from NEE fluxes (CUPNEE). Using atmospheric transport model simulations with prescribed surface flux modifications, we assess how changes in surface carbon exchange propagate through the atmosphere and are reflected in the seasonal CO2 signals observed at monitoring sites. The results show that interannual variability (IAV) in atmospheric transport contributes substantially to variability in the observed CUPMR signals. In addition, atmospheric mixing strongly dampens surface signals: at widely studied sites such as Mauna Loa, Utqiaġvik (formerly Barrow), and Alert, only about 50% of the imposed surface changes are reflected in the CUP derived from atmospheric CO2 observations, due to the spatial integration of asynchronous regional flux variations.

The third chapter of the thesis applies the developed framework to investigate recent changes in CUP timing across the Northern Hemisphere. Combining atmospheric CO2 observations with surface flux estimates from atmospheric inversions and data-driven products, the study analyzes spatial and temporal patterns in CUP timing for 2001–2021. The results show a robust and spatially coherent advancement of CUPMR onset across many high-latitude sites (≈ 60°–90° N), with a mean trend of -2.5 ± 0.27 days per decade. Analysis of regionally aggregated CUPNEE metrics indicates that the observed advancement in CUPMR onset originates primarily from Eurasian boreal ecosystems, while North American regions exhibit weaker or inconsistent trends in CUPNEE onset.

Regions with advancing CUPNEE onset are characterized mainly by earlier GPP, with TER also advancing in similar amounts, suggesting an increase in magnitude of early season GPP. These patterns are consistent with stronger spring warming across Eurasia, suggesting that earlier photosynthetic activation in Eurasian boreal ecosystems likely influences the observed advancement of CUP onset in high-latitude atmospheric CO2 records.

Overall, this thesis shows that atmospheric CO2 observations are a powerful tool for diagnosing large-scale changes in the timing of terrestrial carbon exchange, while also highlighting limitations arising from atmospheric transport and spatial signal integration. The work contributes methodological advances and new insights into the seasonal dynamics of the terrestrial carbon cycle. By improving the reliability of CUP estimates derived from atmospheric observations and clarifying the relationship between atmospheric signals and surface fluxes, it provides a framework for diagnosing seasonal changes in ecosystem carbon exchange at large spatial scales. Together, these advances strengthen the potential of atmospheric CO2 observations to constrain the timing of the terrestrial carbon fluxes.

Lees verder
Publicatiedatum 19 november 2026
Universiteit Wageningen University
Auteur Theertha Kariyathan
DOI nummer 10.18174/681265

Bekijk ook deze projecten

Chris De Witte

Chris De Witte

Wageningen University Lees meer
Martijn Vink

Martijn Vink

Wageningen University Lees meer
Inge Veldhuizen

Inge Veldhuizen

Wageningen University Lees meer