Hi Laurie.  In my SEM work on many crusts and rhodos- at any one time there seems always to be part of the epithallus that is missing or splitting off.  the amount relative to the total sample could be estimated by taking the thickness of the epithallus:total crust thickness and estimating the amount of carbonate turned over relative to the total crust thickness.

For example, in the P. onkodes, there is ~300 microns thickness of living crust and the epithallus is ~ 20 microns on the top of this.  If the epithallial cells formed at the same rate the perithallial did (I do not know if they do) then as there is only one epithallial cell layer, there would have to be a turnover every time a perithallial cell was formed.  Say the P. onkodes grow 1000 microns a year and the average cell length is 13 microns and 10 microns in the epithallus. This would mean that 77 cells form per year in the perithallus and a matching 77 cells in the epithallus.  But the epithallial cells are not retained.  Therefor of the total 1770 microns of carbonate formed, only 1000 is retained.  Meaning ~43% of production is lost to epithallial turnover each year based on these estimates.   Of course the tricky - and difficult- part is determining exactly how much epithallial growth there is per cell of perithallus so that this proportion could be determined accurately.  Estimates of carbonate production using alkalinity anomaly and weight change would be accurately recording the net accumulation but would be underestimating the total production.

The work we have done on temperature/ Mg change showed a change in the epithallus after 8 weeks (Diaz-Pulido et al. 2014), indicating they are turning over at least each 8 weeks. We have not tested corallines grown experimentally for less time than this but this would be one way to test the rate of growth of epithallus.  Eg, grow crusts in different temperatures and test the epithallus after 1 day, 1 week, 2 weeks etc, to see when the temperature signature shows up.  In our work to date, the epithallus always responds to temperature at a rate of ~0.33 mol% per degree- both experimentally and in the field (Nash et al. 2016 and another paper in review) . Perhaps try offsets of 3-6 degrees to exaggerate the signal. Theoretically this should be quite straight forward to do.  You only need to use XRD to test for the Mg change and it takes about 10 minutes per sample to run the analysis. We usually see the pattern emerge with just 3 samples per temperature, and then typically run 8 for nice statistics. 

Diaz-Pulido, Guillermo, et al. "Greenhouse conditions induce mineralogical changes and dolomite accumulation in coralline algae on tropical reefs." Nature communications 5 (2014).

Nash, Merinda C., Sophie Martin, and Jean-Pierre Gattuso. "Mineralogical response of the Mediterranean crustose coralline alga Lithophyllum cabiochae to near-future ocean acidification and warming." Biogeosciences 13.21 (2016): 5937-5945.

Hi Merinda. Thanks so much or that information. Based on your calculations, that's a pretty incredible amount of production that is lost to turnover, don't you think? Seems like a lot of energy to waste.

Yes, it does seem like a lot lost to turnover.  I do not know if this is a lot of energy to waste in context of other plants, eg, trees that lose their leaves every autumn.  But it is not an issue i have considered in depth. 

Hello everyone, and thanks for the nice discussion. Past studies related the cell sloughing seen in coralline algae to the amount of shading performed from epiphytic organisms. So may be the increase or decrease in this shading could be another variable that must be considered when predicting the amount of CaCO3 that is been slough off the surface? This leads to the hypothesis that may be the cost of slough off this part of the thallus is exceeded by the gain in photosynthesis and energy of a clear surface.