Merge PR #165 into e3sm/maint-1.0

Vertical index k=1 on activeTracerVerticalAdvectionTopFlux is the ocean
surface, so is always zero. It was using some uninitialized values when
using k=1, so start loop at k=2.

PR #165 was originally into ocean/develop, I'm just repeating into
e3sm/maint-1.0

src/core_ocean/shared/mpas_ocn_tracer_advection_mono.F

@@ -640,8 +640,10 @@ subroutine ocn_tracer_advection_mono_tend(tracers, adv_coefs, adv_coefs_3rd, nAd
            if (tracerGroupName == 'activeTracers') then
               !$omp do schedule(runtime) private(k)
               do iCell = 1, nCells
-                do k = 1, maxLevelCell(iCell)
+                do k = 2, maxLevelCell(iCell)
                   activeTracerVerticalAdvectionTopFlux(iTracer,k,iCell) = low_order_flux(k,iCell) + high_order_flux(k,iCell)
+                end do
+                do k = 1, maxLevelCell(iCell)
                   activeTracerVerticalAdvectionTendency(iTracer,k,iCell) = work_tend(k,iCell)
                 end do
               end do ! iCell loop