Differences in Rate and Direction of Shifts between Phytoplankton Size Structure and Sea Surface Temperature

Abstract

Species distributions are changing with various rates and directions in response to recent global warming. The velocity of sea surface temperature (SST) has been used to predict species migration and persistence as an expectation of how species track their thermal niches; however, several studies have found that evidence for species shifts has deviated from the velocity of SST. This study investigated whether estimation of the velocity of shifts in phytoplankton size structure using remote sensing data could contribute to better prediction of species shifts. A chlorophyll-a (Chla) size distribution (CSD) model was developed by quantifying the relationships between the size structure of the phytoplankton community and the spectral features of the phytoplankton absorption coefficient (a ph (λ)), based on the principal component analysis approach. Model validation demonstrated that the exponent of CSD (hereafter, CSD slope), which can describe the synoptic size structure of a phytoplankton community, was derived successfully with a relative root mean square error of 18.5%. The median velocity of CSD slope across the ocean was 485.2 km·decade −1 , broadly similar to Chla (531.5 km·decade −1 ). These values were twice the velocity of SST, and the directions of shifts in CSD slope and Chla were quite different from that of SST. Because Chla is generally covariant with the size structure of a phytoplankton community, we believe that spatiotemporal changes in Chla can explain the variations of phytoplankton size structure. Obvious differences in both rate and direction of shifts were found between the phytoplankton size structure and SST, implying that shifts of phytoplankton size structure could be a powerful tool for assessing the distributional shifts of marine species. Our results will contribute to generate global and regional maps of expected species shifts in response to environmental forcing.

Impacts of Mesoscale Eddies on Phytoplankton Size Structure

“…These spatial patterns of the CSD slope exhibited an inverse relationship with Chl a , suggesting that areas of the relative dominance of larger‐ and smaller‐sized phytoplankton support high and low productivity, respectively. Although the original version of the CSD model (Waga et al, ) contained large uncertainties for oligotrophic regions, the reconstructed CSD model demonstrated its reliable performance for global monitoring of phytoplankton size structure, because larger and smaller CSD slopes indicating larger proportions of small‐ and large‐sized phytoplankton were found in oligotrophic and eutrophic regions, respectively. Note that the in situ measured Chl a values ranged from 0.04 to 17.04 and 0.07 to 6.80 mg m −3 for reconstruction and match‐up data set, respectively.…”

“…The size structure of phytoplankton communities was detected using a Chl a size distribution (CSD) model (Waga et al, 2017(Waga et al, , 2019 which retrieves the exponent of the CSD (hereafter, the CSD slope) as a proxy of the size structure of phytoplankton communities based on the spectral shape of a ph (λ). A small CSD slope represents a large fraction of larger-sized phytoplankton, whereas a large CSD slope indicates that smaller-sized phytoplankton dominate.…”

Changing Occurrences of Fall Blooms Associated With Variations in Phytoplankton Size Structure in the Pacific Arctic

“…where the coefficients a and b are empirically derived values and R is the base-10 logarithm of the maximum band ratio of R rs (λ) for blue to green wavelengths. In addition, the size structure of phytoplankton communities was detected using a chla size distribution (CSD) model (Waga et al, 2017(Waga et al, , 2019 which retrieves the exponent of the CSD (hereafter, the CSD slope) as an index of the synoptic size structure of phytoplankton communities. A large CSD slope indicates a large proportion of smaller phytoplankton, whereas a small CSD slope suggests that larger phytoplankton dominate.…”

Retrieving Phytoplankton Size Class from the Absorption Coefficient and Chlorophyll A Concentration Based on Support Vector Machine

“…Practically, multiple size class is repetitious and cumbersome for biogeochemical and biological studies. For this reason, several studies tried to represent PSC by using a single index such as PSD slopes [21] or CSD slopes [60]. These methods are possible ways to avoid poor estimation accuracy of nano-plankton.…”