Figure 11. Principal component analysis of the seasonalized data-set including winter (left panel) and summer (right panel) for all studied gases. The panels represent the biplots that displays both the scores of the samples and the loadings of the variables on the principal components in the same plot. Numbers on the axis represent the factor loadings of each variable and each principal component i.e. PC.
Factor 2 elucidated ~ 25% of the total variance, associated with salinity and temperature, albeit exerting a limited impact on CO2 and N2O emissions. During the summer season, Factor 1 played a pivotal role, elucidating approximately 50% of the total variance concerning CO2and N2O emissions. Notably, Δp CO2, F CO2, ΔN2O, F N2O, Chla, and oxygen exhibited a negative correlation with water temperature, particularly within the Lüderitz region.
We know that local upwelling plays a crucial role in transporting cold, nutrient-rich, and gas-enriched deep waters to the surface layers in Lüderitz. This process leads to outgassing while simultaneously fostering biological activity. Additionally, Factor 2 exhibited an association with wind speed, although it had a relatively minor impact on the emissions of CO2 and N2O. Previous studies have also reported a negative relationship between F CO2 and SST/SSS in upwelling regions, depending on the positioning of hydrodynamic features (e.g. Lefèvre and Taylor, 2002).
In the context of CH4, Factor 1 elucidates over 40% of the total variance in winter, with oxygen displaying a negative correlation with temperature, salinity, and wind speed. This observation may be indicative of upwelling events linked to the influx of saltier, colder, and oxygen-depleted waters in the Kunene and Walvis Bay regions. Therefore, factor 1 is likely related to oceanographic conditions. Factor 2, accounting for 33% of the total variance, was largely dominated by ΔCH4 and F CH4, representing significant and persistent disequilibria in CH4 concentrations at the ocean-atmosphere interface in both seasons. This factor emerged as a critical driver of CH4 emissions, particularly in the coastal area of Walvis Bay. A similar pattern was also evident in summer, underscoring the paramount role of the extensive organic-rich diatomaceous mud-belt in driving CH4 production and emissions in Walvis Bay (Mollenhauer et al., 2007; van der Plas et al., 2007; Sabbaghzadeh et al., 2021).
5 Conclusions
We showed that the nBUS is a perennial, yet, seasonally-varying hotspot of CO2, CH4 and N2O production and emissions to the atmosphere. The seasonal variability in water column and sea surface concentrations of CO2, CH4 and N2O could be observed across the major upwelling cells in the region, albeit a meridional gradient with the Lüderitz cell having a smaller annual contribution.
The primary drivers influencing CO2, CH4, and N2O fluctuations include upwelling events, temperature changes, and pronounced biological effects, particularly during upwelling. Non-thermal effects (likely biological effects) consistently control p CO2fluctuations across all regions, with coastal areas displaying more biological influences than offshore locations. Also, the lack of correlation between CT and salinity in the nBUS suggests that factors beyond dilution govern CT fluctuations, emphasizing the complex interplay of physical and biological processes in shaping carbon dynamics. Despite the shared drivers, each gas displays unique seasonal variations, with CO2 and N2O emissions peaking during intensified upwelling in winter, while CH4 dynamics are influenced by organic-rich mud-belt at Walvis Bay and other sedimentary sources at Kunene and Lüderitz, leading to elevated concentrations and sea-air fluxes throughout the year. In addition, analysis of CO2-e emissions indicates a winter prevalence of CO2 emissions, followed by N2O and CH4. Conversely, summer exhibits a notable shift, with N2O taking the lead, followed by CO2 and CH4 in contributing to the total GHG emissions.
Overall, we show the nBUS to be an important region in terms of its contribution to atmospheric CO2, CH4 and N2O. The intrinsic spatial and temporal variability of production and sea-air fluxes of these gases in the region underscores the need for an improved understanding of their seasonal dynamics and its relationship with oceanographic variability. Furthermore, ongoing environmental changes including global warming and associated changes in oceanic stratification and ice coverage, ocean acidification, expansion of OMZs, and eutrophication might significantly alter the oceanic production and consumption of these gases, their distribution patterns, as well as their sea-air fluxes. An improved representation of how these processes affect the seasonal dynamics and driving factors of these gases, might in turn benefit future model projections of climate change scenarios. To this end, we contend that including CH4and N2O along with CO2 is key when assessing the impacts of coastal upwelling systems on climate dynamics, in particular in terms of the quantification of the net radiative balance of these highly productive ecosystems.