floating-point warning

[rs028]

Likely not a problem, but I am getting the following message. The model results look normal.
gfortran version 4.9.2

------------------
 Final statistics
------------------
 No. steps = 181   No. f-s = 224   No. J-s = 460   No. LU-s = 45
 No. nonlinear iterations = 223
 No. nonlinear convergence failures = 0
 No. error test failures = 4

 Runtime = 13
 Deallocating memory.
Note: The following floating-point exceptions are signalling: IEEE_UNDERFLOW_FLAG IEEE_DENORMAL

[spco]

These warnings just indicate that within our calculations we're getting such small numbers that there are significant possibilities that we're losing accuracy. This ties directly into why I think some of #265 could be solved with renormalising the equations.