Can We Use Evolutionary Algorithms to Outdo Evolution? A Computational Approach to Guiding Increased Crop Photosynthetic Productivity through Genetic Transformation.

Global atmospheric change, an increase in population to 8 billion and increasing land use pressure require an accelerated rate of improvement in the yield potential (Y_p) of our crops, especially the major grains.  Large increases in Y_p over the past 50 years are from increased partitioning of biomass to grain and production of canopies that intercept most available radiation during the growing season.  These strategies have been so successful that there both now are close to their theoretical maxima, with little room for further improvement.  By contrast photosynthesis been improved little and realized efficiencies of solar energy conversion are far short of the theoretical potential.  Photosynthesis is perhaps the best understood plant process, from genes to crop stands.  Mechanistic dynamic models describing the >100 discrete biophysical and metabolic steps through linked differential equations have been developed.  These successfully simulate the behavior of crop leaves in vivo.   Evolutionary algorithms were applied to this system to determine: a) if investment of resources within the photosynthetic apparatus is optimal for productivity; b) how investment should be re-optimized for rising CO₂.   Results suggest that altered investment, could increase photosynthetic efficiency by 60%.  Results are consistent with recent transgenic manipulations.  Why has this re-optimization not occurred as a result of natural evolution?  The answer, at least in part, may be increased vulnerability to extreme events and failure to adapt to the increase in CO₂ that has already occurred.