Improve default for metabolic rate

[gustavdelius]

Currently, if the coefficient ks of the allometric metabolic rate is not given in the species parameters, it is set to 0.2 * h. Given that the coefficient of the allometric growth rate is alpha * feeding_level * h - ks, species will not grow unless alpha * feeding_level > 0.2. Would it perhaps be more appropriate to set ks to 0.5 * alpha * feeding_level * h so that only half of the consumption is lost to metabolism? Any advice @Kenhasteandersen @juliablanchard @astaaudzi ?

[Kenhasteandersen]

I’m not quite sure I understand what you ask. When you write “feeding_level”, is it then “f(w, t)”, or the constant “f_0”? It is important that the metabolism is a constant and not a dynamic function of the feeding level. There are two losses: 1) assimilation, which is a fraction of consumption, represented by alpha; 2) standard metabolism, which is independent of consumption. Ken

[gustavdelius]

@Kenhasteandersen , what I am looking for is a good default value for ks. I believe the philosophy of mizer is that when people have only partial knowledge of their species, i.e., only know some of the parameters, mizer tries to provide defaults for the unknown parameters in such a way that the species is at least viable. The current choice of ks = 0.2 * h does not do that. Assume for example that it is known that the species has a low assimilation efficiency of alpha = 0.1. In that case the current default for ks says that the species has a higher metabolic demand than it could ever cover by feeding. Such a species would never have evolved. So I am looking for a better default value for ks.

Do you know what the motivation for the choice ks = 0.2 * h was?

In my previous post by feeding_level I meant the f0 specified by the user.

[Kenhasteandersen]

Got it. Yes, the correct rule would be: ks = f_c*alpha*h, where f_c approx 0.2 is the critical feeding level. The value “0.2” is rule of thumb. When I wrote the book I tried to make a data analysis to derive it, but it did not work out very well. I’ve attached the pages where I discuss this. Ken

…

--------------------------------------------------------------------------------------------------------------- Ken H. Andersen, http://ken.haste.dk<http://ken.haste.dk/>, twitter: @69kno Professor in theoretical marine ecology, head of section, and deputy director of Centre for Ocean Life http://www.oceanlifecentre.dk<http://www.oceanlifecentre.dk/> New book: Fish Ecology, Evolution, and Exploitation - A New Theoretical Synthesis https://press.princeton.edu/titles/13516.html [cid:50afd584-2091-4184-a2a6-995288d20920@win.dtu.dk] On 17 Sep 2019, at 08.40, Gustav W Delius <notifications@github.com<mailto:notifications@github.com>> wrote: @Kenhasteandersen<https://github.com/Kenhasteandersen> , what I am looking for is a good default value for ks. I believe the philosophy of mizer is that when people have only partial knowledge of their species, i.e., only know some of the parameters, mizer tries to provide defaults for the unknown parameters in such a way that the species is at least viable. The current choice of ks = 0.2 * h does not do that. Assume for example that it is known that the species has a low assimilation efficiency of alpha = 0.1. In that case the current default for ks says that the species has a higher metabolic demand than it could ever cover by feeding. Such a species would never have evolved. So I am looking for a better default value for ks. Do you know what the motivation for the choice ks = 0.2 * h was? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#101?email_source=notifications&email_token=ADFHLANTVPN7IIH6AY36VZ3QKB3W5A5CNFSM4IXHEHA2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD63PFGA#issuecomment-532083352>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ADFHLALYTJURSDFP2DKZYYDQKB3W5ANCNFSM4IXHEHAQ>.

[astaaudzi]

This is a very interesting topic. Assuming that critical feeding level is 0.2 and alpha is 0.6, then ks should be 0.12 of h (the default is 0.2*h, I believe, which is too high). For the Tasmanian model and other papers I have looked at various energy budgets, and parameterised ks to give metabolism at 30-50% of daily intake in equilibrium conditions (not necessarily maximum intake, but realised daily intake).

I think the above rule of thumb (ks = f_calphah) makes lots of sense for a start. Although if you look at literature you find that standard metabolism level (ks) is often higher in small bodied species, whereas h could be expected to be larger in large bodied species. In which case h and ks would be negatively correlated (which is the case in the Tasmanian model). But this could perhaps be left for further explorations :)

[gustavdelius]

Good. I'll change the default in the new version to ks = 0.2 * alpha * h then. I will also add a warning when a user has specified a value for f0 that is less than the critical feeding level of 0.2.

[gustavdelius]

Or how about setting the default to ks = 0.2 * alpha * h * w_mat^(n-p) so that the critical feeding level is equal to 0.2 at w_mat also when n != p?

[Kenhasteandersen]

Hmm. Setting p != n is a little tricky because it introduces some extra effect. In that case it might be better not to set a default, but force the user to reflect upon what it means in terms of the value of ks. If you think that we should have a default, then your suggestion is ok. Ken

…

--------------------------------------------------------------------------------------------------------------- Ken H. Andersen, http://ken.haste.dk<http://ken.haste.dk/>, twitter: @69kno Professor in theoretical marine ecology, head of section, and deputy director of Centre for Ocean Life http://www.oceanlifecentre.dk<http://www.oceanlifecentre.dk/> New book: Fish Ecology, Evolution, and Exploitation - A New Theoretical Synthesis https://press.princeton.edu/titles/13516.html [cid:image001.jpg@01D09493.A377DFB0] On 17 Sep 2019, at 10.07, Gustav W Delius <notifications@github.com<mailto:notifications@github.com>> wrote: Or how about setting the default to ks = 0.2 * alpha * h * w_mat^(n-p) so that the critical feeding level is equal to 0.2 at w_mat also when n != p? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#101?email_source=notifications&email_token=ADFHLANTMU6GAKDTBU654H3QKCF3LA5CNFSM4IXHEHA2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD63V4HQ#issuecomment-532110878>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ADFHLAP4SLWGNP5HQBYUZDLQKCF3LANCNFSM4IXHEHAQ>.

[gustavdelius]

@Kenhasteandersen , I agree with you that it is tricky to give good defaults when p != n. For example the default for h that leads to the right maturity age will become difficult. We should definitely issue a stark warning when we provide any defaults for this case. However I would like mizer to provide some default so that the user can then use shiny gadgets to start investigating the system and interactively explore better parameter values. All I require of the defaults for that purpose is that they allow every species to reach maturity size in order to have some reproduction.

[Kenhasteandersen]

Seems good to me. Ken

…

--------------------------------------------------------------------------------------------------------------- Ken H. Andersen, http://ken.haste.dk<http://ken.haste.dk/>, twitter: @69kno Professor in theoretical marine ecology, head of section, and deputy director of Centre for Ocean Life http://www.oceanlifecentre.dk<http://www.oceanlifecentre.dk/> New book: Fish Ecology, Evolution, and Exploitation - A New Theoretical Synthesis https://press.princeton.edu/titles/13516.html [cid:image001.jpg@01D09493.A377DFB0] On 17 Sep 2019, at 11.19, Gustav W Delius <notifications@github.com<mailto:notifications@github.com>> wrote: @Kenhasteandersen<https://github.com/Kenhasteandersen> , I agree with you that it is tricky to give good defaults when p != n. For example the default for h that leads to the right maturity age will become difficult. We should definitely issue a stark warning when we provide any defaults for this case. However I would like mizer to provide some default so that the user can then use shiny gadgets to start investigating the system and interactively explore better parameter values. All I require of the defaults for that purpose is that they allow every species to reach maturity size in order to have some reproduction. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#101?email_source=notifications&email_token=ADFHLAOWEHIL4SUJFM7WYKTQKCOKPA5CNFSM4IXHEHA2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD634LFI#issuecomment-532137365>, or mute the thread<https://github.com/notifications/unsubscribe-auth/ADFHLAKFSGB2JIGISQKI4PLQKCOKPANCNFSM4IXHEHAQ>.