It takes a month to run after setting the number of iterations.

[870627]

When estimating Uncertain parameters by fitting the model, after setting the number of iterations of 5000, it takes me a month to run. I wonder if this is normal.

[michaelspence]

Hi

Are you using the LeMaRns example with only 3 species? This took a day or so to do. However, the Celtic Sea mizer model took about a month.

Cheers Mike

[870627]

Thanks. There is a uncertain parameters in the LeMaRns model: the recruitment parameter b, which is often referred to as the density-dependent part of the hockey-stick recruitment curve. I'd like to ask you how to obtain the recruitment parameter b.

[michaelspence]

This is often done by calibration, although other sources could be used, e.g. from stock-assessments. These references may help:

S2 in https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0227767

Supplementary 1 in https://onlinelibrary.wiley.com/doi/abs/10.1111/faf.12543

We will be publishing this soon that may also help: https://www.frontiersin.org/articles/10.3389/fmars.2021.700534/abstract

[870627]

I read the paper “LeMaRns: A Length-based Multi-species analysis by numerical simulation in R”, and there is a supplementary file S3, titled “Fitting the North Sea LeMans model to survey and landings data”. Before calibration, the recruitment parameter b defined in “Evaluation and management implications of uncertainty in a multispecies size-structured model of population and community responses to fishing” can be found in orig_NS_par. I still don't quite understand how to get the specific value of b in orig_NS_par.

[870627]

Hello

I'm using the LeMaRns model and there is a supplementary file S3, titled “Fitting the North Sea LeMans model to survey and landings data”. Before calibration, the recruitment parameter b can be found in orig_NS_par. But how to get the b value before calibration?

orig_NS_par

species_names Linf W_a W_b k Lmat a

1 Sprat 16.55 0.0059 3.1088 0.681 12.14 885.6667731

2 Npout 23.75 0.0075 3.0244 0.849 17.21 225.0988055

3 Sandeel 23.57 0.0049 2.7826 1.000 12.16 231.6879245

4 Poor cod 23.00 0.0082 3.0865 0.520 15.00 254.2279144

5 Long rough dab 25.00 0.0053 3.1434 0.340 15.00 185.3071561

6 Dab 32.40 0.0159 2.8639 0.536 12.96 69.3184724

7 Herring 33.35 0.0048 3.1984 0.606 23.39 62.1228287

8 Horse mackerel 28.00 0.0316 2.6520 0.380 19.00 120.5695838

9 Lemon sole 37.00 0.0123 2.9713 0.420 27.00 41.8978427

10 Sole 46.41 0.0089 3.0172 0.284 20.96 17.7411663

11 Mackerel 38.00 0.0052 3.1674 0.510 26.00 37.8676686

12 Whiting 52.50 0.0099 2.9433 0.323 21.39 11.1149557

13 Witch 44.00 0.0017 3.3887 0.200 29.00 21.71750603

14 Gurnard 42.97 0.0160 2.8151 0.266 17.67 23.7587503

15 Plaice 65.32 0.0125 2.9425 0.122 22.22 4.8535417

16 Starry ray 66.00 0.0107 2.9400 0.230 46.00 4.6666095

17 Haddock 70.65 0.0092 3.0126 0.271 26.91 3.6046815

18 Cuckoo ray 92.00 0.0026 3.2169 0.110 59.00 1.3242529

19 Monkfish 106.00 0.0297 2.8410 0.180 61.00 0.7738734

20 Cod 149.87 0.0081 3.0502 0.216 54.39 0.2080939

21 Saithe 118.94 0.0085 3.0242 0.175 48.61 0.4999956

b

1 67352.478225

2 20821.054248

3 57632.695589

4 40014.253426

5 1710.014437

6 3837.179493

7 35498.649232

8 4672.577765

9 208.776950

10 320.405464

11 1881.553572

12 1173.452560

13 165.613357

14 793.337934

15 486.130733

16 69.384083

17 378.035208

18 2.068218

19 2.321620

20 3.602605

21 32.769715

[michaelspence]

I believe the parameters b are from the Thorpe et al. 2015 paper: https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/2041-210X.12292

There's no need to know b prior to calibration. Initial screening of the parameters should give reasonable initial values. See https://www.frontiersin.org/articles/10.3389/fmars.2021.700534/full and faf12543-sup-0001-Supinfo.pdf found here: https://onlinelibrary.wiley.com/doi/abs/10.1111/faf.12543 Particularly have a look at the figure at the top of page 7.

[870627]

I want to know whether the parameters b is the inflection point of the hockey-stick spawner–recruit relationship. And is there a unit about the parameters b?

[michaelspence]

In the hockey-stick spawner-recruit relationship b is the density dependent parameter. See the help files for rec_hockey, rec_const, rec_BH, rec_linear, and rec_Ricker as well as the other recruitment functions as well as the vignette.