we act significantly and soon to reduce greenhouse-gas emissions. Since temperature
affects every chemical and biological reaction, even moderate warming will change,
well, everything. To predict the effects on ocean biota, we must extrapolate from the
longest time-series data available. Apart from fossil proxies, those series are less than
a century long. In general, the expectations from greater warmth begin with more
stable water-column stratification. That means fewer phytoplankton nutrients mixed
upward, less primary production, and shorter rations all the way up the food chain and
all the way down to the benthos. Stratification variation in fact drives a major part of
variation in global ocean productivity (Behrenfeld et al. 2006; Fig. 16.15). That is
because the dominant factor is productivity in the huge equatorial belt. That drops
dramatically when the West Pacific warm pool collapses along the equator during El
Niño (see below), overriding the productive eastern tropical waters. The warm layer
buries nutrient-rich layers more than 100 m deeper, making them inaccessible to
upwelling. Thus, the anomaly in global primary production estimated from
chlorophyll and temperature data closely tracks both an index (MEI) of El Niño
strength and stratification estimates from a data-assimilating model of water-column
structure. The enhanced stratification from global warming will operate differently
and vary in different modes, but the principle is quite general. Warming increases
stratification and reduces upward nutrient mixing, which reduces primary production.
Guesses at the magnitude of the effects can be generated by models that consistently
carry substantial uncertainties. So, we leave them for others. The responses of marine
ecosystems to the longer climate cycles (alternating decades or longer stretches of
similar seasonal cycles of weather) may give us a preliminary sense of what is
coming.
Fig. 16.15 Monthly anomaly of global ocean net primary production (NPP) rate,
estimated from satellite chlorophyll (mostly) and temperature data, for 1997 through
2005, compared to (a) the Multivariate ENSO Index (MEI) and (b) global anomaly of
stratification (mostly determined by El Niño vs. normal conditions in the equatorial
Pacific).
(^) (After Behrenfeld et al. 2006.)