Ticket #1482 (closed assessed: fixed)

Opened 18 months ago

Last modified 6 days ago

Ecotracer computations update

Reported by: willw Owned by:
Priority: normal Milestone: Ecopath 6: release 6.5.1
Component: Contaminant tracing Version: 6.5 beta
Severity: major Keywords:
Cc:

Description (last modified by jeroens) (diff)

For Ecosim Ecotracer calculations: CUpdate (cContaminantTracer)

  • Moved to an adaptive number of sub-timesteps per ecosim timestep (used to be fixed at 3). The number of timesteps is based on how far the concentration is away from the estimated equilibrium, and the fastest loss-rate parameter. The largest timestep is 0.1 times 1/(max loss rate).
  • Changed the time-integration method to explicit Adams-Bashforth multistep. Previously, a semi-analytical solution was assumed over the timestep, but this was not great at conserving tracer mass.

For Ecospace Ecotracer calcuations:

  • Fixed initialization of spatial CCell array for the environment (as a function of RelCin and CZero)
  • Same as Ecosim version, moved to an adaptive number of sub-timesteps per ecosapce timestep (used to be fixed at 1). Same criteria as ecosim to get the timestep.
  • Changed integration method same as Ecosim. Previously this was a single-step backward Euler implicit integration. The old method performed very poorly if the loss rates were large compared to the timestep.
  • Added a !SolveCellC routine that performs all the Ecotracer sections of the SolveCell code since this needs to be done multiple times per ecospace timestep (while the rest of SolveCell does not).
  • Added multi-threading for the SolveCellC calculations.

For *ALL* Ecospace calculations:

  • In setMovementParameters: set the Depth, width, HabCap, Xvel and Yvel variables on the boundary to inherit the values just inside the boundary. Previously, open boundaries would sometimes appear to be closed under advection/diffusion since these values were not being set properly.

As far as interface stuff goes, I think we need one more column for the tracer input: excretion or metabolism rate. This should be the rate that the contaminant is excreted from itself into the environment. Currently there is a CexcretionRate parameter, but the name should probably be changed to CexcretionFraction or something, since it is the fraction of contaminant that is assimilated into tissue, and is not a rate at all. Maybe the CexcretionRate should be used for excretion/metabolism, and a new variable made like CassimFrac or something to represent the fraction assimilated by the organism.

Change History

comment:1 Changed 18 months ago by jeroens

  • Description modified (diff)

comment:2 Changed 18 months ago by jeroens

  • Type changed from not assessed to assessed
  • Milestone changed from Ecopath 6: future version to Ecopath 6: release 6.5.1

comment:3 Changed 6 days ago by jeroens

  • Status changed from new to closed
  • Resolution set to fixed
Note: See TracTickets for help on using tickets.