Recent trends in the wind-driven California current upwelling system

Long-term changes in the marine ecosystems of the Eastern Boundary Upwelling Systems (EBUS) are predicted due to anthropogenic climate change. In particular, global ocean acidification is having a profound effect on the coastal waters of the EBUS, affecting the entire trophic chain, net primary production (NPP) and related economic activities such as fisheries. Another predicted change related to human activity is that of upwelling dynamics with expected long-term changes in upwelling winds as proposed by Bakun (1990), Bakun et al. (2015) and Rykaczewski et al. (2015). Although these predicted long-term changes may emerge only later in the 21st century, this has fueled many studies using historical data. Long-term increase in upwelling winds has thus been a much debated topic, showing that there is considerable uncertainty depending on the EBUS considered, the effect of natural climate fluctuations, the choice of wind dataset, the time period considered, and the methodologies and significance tests applied. Therefore, there is an immediate interest in being able to monitor upwelling using verified and self-consistent wind data sets. This work focused on a sensitivity study of the estimated trends in upwelling winds in the California Current Upwelling System (CCUS), for the most recent period 1996–2018, using the two state-of-the-art satellite wind analyses and two atmospheric model re-analyses. Embedded into the strong modulation by natural climate fluctuations on interannual and decadal time scales, we do see an increase in upwelling-favorable winds in the core of the CCUS, with a local increase of more than 25% in seasonal upwelling transport for the period considered. In this central upwelling zone, a good agreement on stronger equatorward winds for the winter and spring seasons is found between the different datasets, although with different significance levels. Conversely, conflicting results are found in the southernmost part of the CCUS between the satellite analyses and the model reanalyses. Systematic, time-dependent differences are found between the wind products, highlighting the need to further investigate the poorly documented temporal stability of these widely used wind long-term climatology products. The observed spatial structuring of the estimated wind trends is consistent with the trend analysis of water chlorophyll-a, partial pressure of CO2, and basity (pH) analysis products. This result is consistent with changes being important for modulating the carbonate system within the CCUS