State of the art climate models continue to be plagued by some standard biases especially in the crucial regions such as the tropical Pacific (http://www.usclivar.org/Meeting_Files/SSC 11/Bias_worskhop_summary.pdf) which has a global reach through El Nino-Southern Oscillation (ENSO). One such bias is the so-called ‘cold bias’ where the model sea surface temperatures are typically colder than observed leading to a stronger east-west gradient, stronger winds, and the related cascades into annual cycle errors and thus ENSO and its teleconnections.
It has been known for decades that microscopic algae in the ocean convert light to heat during photosythesis and other suspended solids in the ocean also affect light attenuation in the water column. Satellites like SeaWiFS now provide global maps of pigments and other materials which affect light absorption in the ocean. With this global information, we can now specify the e-folding depth for light-attenuation in ocean general circulation models and represent the conversion of light to heat by the photosynthesizing critters (http://essic.umd.edu/~ragu/biofeedbacks/murtuguddeqpen.pdf).
The eastern equatorial Pacific is a really unique place with the mixed layer variability interacting closely with the location of the maximum chlorophyll concentrations. During the boreal spring months when the mixed layer is shallow due to weak winds and maximum surface heating, the chlorophyll maximum is just below the mixed layer since the thermo/nutricline are just below the mixed layer, i.e., in the euphotic zone. This provides a heat source of about 5-20 W/m2 depending on the strength of the chlorophyll maximum and this heat restratifies the water column leading to a deeper mixed layer and momentum penetration, thus weaker surface divergence and subsurface upwelling. Much of the ‘cold-bias’ is thus alleviated if this bio-climate feedback is represented appropriately (http://essic.umd.edu/~ragu/biofeedbacks/quimfeedback.pdf).
A hybrid coupled model and physical-biological models confirm that improving the annual cycle in this way is a natural solution to improving the annual phase-locking of the ENSO events in the model (http://essic.umd.edu/~ragu/biofeedbacks/coupledfeedbacks.pdf). The assumption that annual biases do not impact ENSO simulations and predictions thus appears not to be very robust since improving the annual cycle of the coupled system also improves the ENSO cycle by improved mixed layer-thermocline or Bjerknes feedbacks (http://essic.umd.edu/~ragu/biofeedbacks/marzeion.pdf).
Models tend to produce ENSO-like behavior that is surface-trapped and quasibienniel (http://www.grida.no/climate/ipcc_tar/wg1/024.htm) whereas improved Bjerknes feedback due to accurate biological feedbacks provides a more accurate representation of the recharge-discharge processes thus improving the ENSO amplitude and frequencies. As we continue to increase the complexity of our coupled climate models, these processes have to be represented appropriately since bio-climate feedbacks also occur in other regions of the World Ocean and also over land (http://www.gfdl.noaa.gov/~jpd/esmdt.html).
Raghu Murtugudde, Assoc. Professor, ESSIC/DAOS, Univ of MD, College Park, MD 20742