In places where the ocean is strongly heated from above and away from intense currents, very little of these bottom nutrients are recycled, and these form regions of the ocean that support very little sea life.
Of course, increased warming from above makes surface water lighter; hence increases the stratification and suppresses the upwelling of nutrients. Could this be a negative consequence of global warming?
Possibly so, apparently:
Climate-ecosystem models predict that global warming will exacerbate ocean desert expansion, but not this quickly, Polovina notes. During the past 9 years, gyre deserts expanded 10 to 25 times faster than modeled. The trend feels solid to other scientists. "Everything seems to hold together," says SeaWiFS project scientist Charles McClain of NASA's Goddard Space Flight Center in Greenbelt, Maryland, who was not affiliated with the study. A variety of oceanographic observations and modeling is consistent with warming driving the expansion of the gyres and their low-productivity waters, he says.Emphasis added. Chalk up another one for the "worse than expected" column. Of course in the absence of bias, you would expect literally half of everything to be worse than expected! (The other half would be better.) It's not a field of my expertise but still, a tenfold error!
Such an error in rate estimation looks to this casual onlooker to be a big deal.
That's an odd sort of model and needn't reflect too harshly on the main GCM efforts, but I definitely think this particular effort should go back to the drawing board!
Do the big models use data like this to estimate oceanic biological fixation or is that parameterised? Presumably the deserts are pulling down less CO2 and, if the expected change from C3 to C4 fixation occurs in nutrient-starved areas, are also altering the C12/13/14 numbers.
ReplyDeleteJulian Flood
The carbon cycle is not part of existing climate models at all; CO@ concentration in the atmosphere is specified.
ReplyDeleteThere is a lot of interest in tightly coupling carbon cycle models to climate models. I am not sure that it is a technically sound approach to do so at present; it depends on how it gets done.