Tidal mixing modulation of sea-surface temperature and diatom abundance in Southern California

https://doi.org/10.1016/S0278-4343(00)00105-9Get rights and content

Abstract

In the Southern California Bight a clear seasonal cycle in temperature and diatom abundance is observed, with maximum temperatures in summer and maximum diatom abundance in spring, decreasing in summer. Within this seasonal cycle of temperature and diatom abundance, there is a weak fortnightly temperature variability. Here, we show that diatom abundance has lunar as well as seasonal variability, with the highest abundance corresponding to the coldest days within the lunar cycle. This suggests that at least part of the temperature and diatom abundance variability may be due to bottom tidal mixing. To explore the effect of tidal mixing and tidal straining on the control of the thermal structure of the water column in the Southern California region, a one-dimensional, hydrodynamic numerical model is used. The model is successful in explaining the seasonal cycle in temperature and partially explains the fortnightly and monthly variability in temperature. The observed temperature minimum shows a lag of about 2–3 days, when compared with model results. This lag is probably due to the fact that the model does not include the effect of internal waves, which will be an extra source of mixing and may have an advective effect that will modify the water-column structure and the diatom distribution.

Introduction

The fortnightly periodicity in sea-surface temperature (SST) in the nearshore Southern California Bight (SCB, Fig. 1) has been explained in terms of advection, generated by internal tidal bores (Pineda (1991), Pineda (1995)).

The bores occur in two phases in the SCB — the cold phase is characterised by onshore advection of a cold-water bore (Cairns, 1967; Winant, 1974), which displaces warm water offshore, and creates an imbalance in hydrostatic pressure between denser inshore water and lighter offshore water. This phase is followed by the warm phase, where the inshore water recedes offshore due to its greater density, whilst a warm bore returns (Pineda, 1994). Nearshore bores occur in groups of two to nine events in spring and summer, when the water column is well stratified; when these occur, the surface water temperature drops. These events have semi-diurnal or diurnal periodicity. The cold phase can explain the onshore transport of water column planktonic invertebrate and fish larvae (Pineda, 1991; Leichter et al., 1998), whilst the warm phase explains the transport of both surface and water column larval taxa (Pineda (1994), Pineda (1999)). Nutrient-rich waters and phytoplankton can also be transported shoreward. The occurrence of the bores at tidal periodicity (e.g. semi-diurnal), the drop in surface water temperature during groups of events, and the settlement of invertebrates peaking at lunar periodicity has led Pineda (1995) to suggest that the fortnightly periodicity in SST in Southern California could be explained by a fortnightly cycle in internal tidal bores, an idea consistent with a potential fortnightly cycle in the internal tide (e.g. Cairns, 1967).

However, the fact that the temperature reduction between 0.25 and 1°C is maintained for several days actually suggests that this fortnightly variability might be due to mixing, either due to internal solitons or to bottom tidal stirring. Balch (1986) and Pineda (1995) discarded the idea that bottom tidal mixing was directly responsible for the cooling and mixing at the sea surface in their study area, because the most negative anomalies lagged several days after spring tides. Balch (1986) observed a lag of 5–6 days for annual-averaged anomalies, whilst Pineda observed a lag of 8–13 days after a new moon (○) and 6–11 days after a full moon (•), for seasonally averaged anomalies. It was argued that a tidal mixing mechanism would produce colder temperatures closer to spring tides. Pineda (1995) argued also against tidal mixing, based on the supposition that tidal mixing could not explain that at several stations there were two cycles of water temperature anomaly per lunar cycle in summer, but only one in spring.

It has been observed however, that in cases of tidal mixing fronts, there is a lag of approximately 3–4 days between the currents and temperature. The maximum negative anomaly is expected to occur approximately 3 to 4 days after spring tides, with the maximum positive anomaly 3 to 4 days after neap tides (Simpson and Bowers, 1981). Sharples and Simpson (1996) suggest that the reason for this lag is the presence of thermal inertia, as stratification progresses into mixed water towards neap tides, together with the need to re-mix stored buoyancy as the stratified region is eroded towards spring tide. Hence, the lag between the spring tide and mixing is due mainly to more stored potential energy at first requiring more mixing energy. On the other hand, the presence of only one period of a negative temperature anomaly during a lunar cycle could be due to an N2 modulation of the spring–neaps cycle (Sharples and Simpson, 1996). Therefore, it is unclear whether bottom mixing could explain some of the fortnightly phenomena observed by Balch (1986) and Pineda (1995). Furthermore, the fortnightly variability in mixing will bring a fortnightly variability in the availability of nutrients, as nutrients are mixed upwards from the bottom layer to the surface. With an increase in the availability of nutrients, a phytoplankton bloom may develop. Obviously, there would be a lag between the time the nutrients become available and the time the phytoplankton bloom develops; this could vary between 3 days and half a spring–neaps cycle, depending on the phytoplankton species.

Similar to the internal bores, tidal straining could be responsible for the semi-diurnal variability in temperature and for plankton and nutrient cross-shore transport. The interaction of the density gradients, with the differential advection resulting from the sheared velocity profile, will generate an alternation between a stratified warm period and a mixed cold period. During the ebb, the faster surface tidal currents will push the warmer coastal waters offshore, generating a warm surface layer. Conversely, during the flood, colder oceanic waters will be advected by the faster surface tidal currents. However, because the colder oceanic waters are denser than the underlying coastal warm waters, the water column would be unstable, resulting in convective mixing which will generate a cooler more homogeneous water column (Simpson et al., 1990). This semi-diurnal differential advection could be responsible also for the onshore and offshore larval transport, especially if combined with vertical larval migration.

A classical theme in marine ecology is how hydrodynamic phenomena impose their temporal and spatial scales on ecological systems (Haury et al., 1978). Knowledge of the temporal and spatial scales of variability in ecological phenomena is important in identifying the underlying forcing mechanisms. Ecological phenomena such as phytoplankton abundance sometimes exhibit lunar periodicity. Proximate explanations, to account for this periodicity, include “extrinsic” hydrodynamic forcing and “intrinsic” individual behaviour. These include differential phytoplankton responses due to changes in stratification in the spring to neap cycle, with associated variability in nutrient input due to tidal mixing (Balch, 1986; Demers et al., 1986). In cases where there is an ecological response to the lunar cycle, the question arises as to whether the response is due to extrinsic hydrodynamic forcing or intrinsic behaviour. If the ecological response is due to hydrodynamics, it is obviously important to identify the forcing mechanism, as different forcing may have different ecological consequences.

In this paper, we show that a simple one-dimensional numerical hydrodynamic model explains some aspects of the fortnightly variability in sea-surface temperature observed on the West Coast of the US. We then discuss previous temperature observations and contrast them with model results, and propose that periodicity in SST is related to bottom mixing. We also present evidence of lunar variability of diatom abundance and discuss the potential effects of tidal mixing.

Section snippets

Temperature observations

Water temperature has been sampled at the end of the Scripps Institution of Oceanography (SIO) Pier, in Southern California, since ∼1920. Fig. 2 shows the 7-day running mean filtered day-of-the-year average surface (1920–1989) and bottom (1927–1989) water temperature. As has been observed before (e.g. Allen, 1941), there is a seasonal cycle, with the lowest temperatures in winter and the highest temperatures in summer. Within this seasonal cycle of temperature, there is a fortnightly/monthly

Modelling mixing in the Southern California Bight: a one-dimensional model

In order to test the hypothesis that the fortnightly cooling is related to bottom tidal mixing, a modified version of the Simpson and Sharples (1992) model has been used. The model uses an explicit scheme to integrate the equations of motion:∂u∂t=−1ρ∂P∂x+fv+∂zNz∂u∂z,∂v∂t=−1ρ∂P∂x−fu+∂zNz∂v∂zwith x and y positive in the east and north directions, respectively, and z increasing positively from the seabed. The second term on the right is the Coriolis forcing and the third term is the effect of

Diatom abundance

The phenomenon that temperature is predictably lowest on some days of the lunar cycle suggests a correlation with the lunar cycle in phytoplankton abundance. To explore this idea, we used a phytoplankton time-series. Allen and his associates sampled phytoplankton from 1920 to 1939 at the end of the SIO pier (depth ca. 6 m). Allen reported both dinoflagellate and diatom abundance, but here we only consider the diatoms. Sampling frequency was daily with the exception of holidays (and Sundays,

Discussion

The main aim of this paper is to assess the effect of tidal mixing over the southern California shelf; it is not the intention to reproduce the effect that internal tidal bores may have in this region. For this reason, we use a one-dimensional vertical model, which cannot include internal tidal bores but includes other advective effects, such as the interaction of tidal straining with temperature gradients. The model reproduces the seasonal cycle and several aspects of the fortnightly

Conclusion

Whilst the model is one-dimensional, it has achieved its purpose of reproducing some of the aspects of the seasonal and fortnightly variability in temperature; this suggests bottom tidal mixing as an important process controlling the water temperatures of the area. Nevertheless, the fact that there is a lag of 2–3 days between observations and predictions suggests that other processes might be involved. The occurrence of internal tidal bores in this area (Pineda, 1995) makes them a strong

Acknowledgments

The authors thank Ms. P. Habziabdic for proof-reading the latest version of the manuscript. AS thanks Dr. Bethan Jones for facilitating the climatological and tidal data. A portion of this manuscript is based upon JP.'s thesis dissertation, at the Scripps Institution of Oceanography. JP thanks the Committee Members for reading some of the material, the SIO for providing temperature data and the US National Science Foundation for support. This is a Woods Hole Oceanographic Institution

References (28)

  • J. Pineda

    An internal tidal bore regime at nearshore stations along western USApredictable upwelling within the lunar cycle

    Continental Shelf Research

    (1995)
  • J.H. Simpson et al.

    Models of stratification and frontal movement in the shelf seas

    Deep-Sea Research

    (1981)
  • W.E. Allen

    Quantitative studies on inshore marine diatoms and dinoflagellates of southern California in 1921

    Bulletin of the Scripps Institution of Oceanography

    (1927)
  • W.E. Allen

    Methods in quantitative research of marine microplankton

    Bulletin of the Scripps Institution of Oceanography

    (1930)
  • W.E. Allen

    Twenty years’ statistical studies of marine plankton dinoflagellates of Southern California

    The American Midland Naturalist

    (1941)
  • W.M. Balch

    Are red tides correlated to spring–neap tidal mixing?use of a historical record to test mechanisms responsible for dinoflagellate blooms

  • T. Berliand

    Metodika climatologicheskih raschetov radiatsii

    Meteorologia i Hydrologia

    (1960)
  • J.L. Cairns

    Asymmetry of internal tidal waves in shallow coastal waters

    Journal of Geophysical Research

    (1967)
  • J.J. Cullen et al.

    Vertical motion of the thermocline, nitracline, and chlorophyll maximum layers in relation to currents on the Southern California shelf

    Journal of Marine Research

    (1983)
  • S. Demers et al.

    Phytoplankton responses to vertical tidal mixing

  • J.E. Edinger et al.

    The response of water temperatures to meteorological conditions

    Water Resources Research

    (1968)
  • L.R. Haury et al.

    Patterns and processes in the time-space scale of plankton distribution

  • J.J. Leichter et al.

    Breaking internal waves on a Florida coral reefa plankton pump at work?

    Marine Ecology Progress Series

    (1998)
  • G.L. Mellor et al.

    Development of a turbulence closure model for geophysical fluid problems

    Reviews in Geophysics and Space Physics

    (1982)
  • Cited by (14)

    • Modelling explicit tides in the Indonesian seas: An important process for surface sea water properties

      2018, Marine Pollution Bulletin
      Citation Excerpt :

      This study has focused on the impact of mixing on sea surface temperature. The results can be easily transposable for nutrients and biomass as shown in previous studies (e.g. Franks and Chen, 1996; Souza and Pineda, 2001; da Silva et al., 2002; Jan and Chen, 2009) since the vertical mixing would upwell deeper layer richer in nutrients at the surface. Microscale turbulence maintains well-mixed biomass and nutrients within the turbulent surface boundary layer as well as driving nutrient fluxes into the mixed layer (e.g. Lewis et al., 1976).

    • Climate variability and Dinophysis acuta blooms in an upwelling system

      2016, Harmful Algae
      Citation Excerpt :

      Our data support the view that specific lunar phases modulate D. acuta bloom intensity in north-western Iberia. Geographic differences are to be expected because maximum tidal range is not always synchronous with the new or full moon, with lags (called “age of the tide”) from hours to days (Souza and Pineda, 2001). These lags exist because tides are determined by the moon phase, but their mode is site specific and related to bottom topography, basin resonance and geographic location (Pond and Pickard, 1978).

    View all citing articles on Scopus
    View full text