Problems to find the right model structure - rates, level

[Christian Haun]

Dear community,
I am a complete beginner in using Vensim PLE.
I tried to build a foodweb model consisting of the following parts and still have some problems.
Constants: intake rates, Q/B (Consumption/Biomass), P/B (Production/Biomass) for the 3 species Whiting, Sandeel and Bass
“Easy calculated” Auxiliaries: intake
Other Auxiliaries: Q and P for the 3 species
Q is the total consumption of every species minus the needed consumption divided
by Q/B. Therefore it should reduce the Biomass when the total consumption is smaller than the needed consumption and increase the Biomass when this value is positive.
P is the production and easily the result of P/B * Biomass of the species
Level: By now the Biomass is written as a level function. It should start with the initial value and then vary depending on Q and P. But should always be bigger than 0. I`m not sure if I should cancel the dependency on P.
I put the small model which is just a small part out of a big model in the attachments so everyone can see what I`m talking about. There`s a group with a factor C and some intake rates but this group is not the problem. Just the rates with the variables Q, species name (e.g. Whiting) and P are important. In my model the values for Q, species and P get really crazy values (positive and negative) and doesn`t have something to do with my expactations. Maybe it`s because of the structure I`ve chosen. Can someone who is advanced in using Vensim tell me what I`ve done wrong (except using box variables and making every variable dimensionless)? How can I manage the loops in a better way? How can I restrict the impact of Q to the Biomass (I thought about using an IF THEN ELSE function). Why do I receive an floating point overflow error when I try to set the simulation time in the range between 0 and 120 years (step = 1 year? -- by now I set it from 1 to 50 years to avoid that.
In the second model I put in an additional variable called “extern parameter” which is strongly decreasing dependent of time. Why is the affected Q not showing a similar behaviour? If this parameter should decrease the Biomass of one species over the lifetime, how should I link it to the model?
I know there are quite a lot questions on a very small model. But I really need your help. I watched most of the tutorials and looked in the manual many times.
Thanks a lot.
Christian

problem extern parameter.mdl

(8.39 KiB) Downloaded 161 times

problem structure.mdl

(8.07 KiB) Downloaded 157 times

[Administrator]

I'd suggest asking something much simpler. I've read through your post twice and have no idea what you are asking for help with. For example, your model doesn't contain a parameter called "Q".

Two suggestions though.

1. Fix the unit errors.

2. Change the equation for "p ba0" so it's
p ba0 = min ( Bass0*"P/B BA" , Bass0 / time step)

This structure will prevent you removing more than is in the level (which is probably why you are getting crazy values). Do this for all outflows from all levels.