How would you do ? - data on trout reproduction in river

#1
Hello to all,
I am currently studying the reproduction of brown trout (salmo trutta fario) on a French river. I have at my disposal a whole series of data that I collected myself from January to the end of May.
Firstly, all the nests from 2022 have been geolocated and associated with the week of spawning, a flow facies, the presence of grass beds, the distance to the source and various other climatic parameters.

id semaine parcours source facies herbiers latitude longitude TempMoy AmpliTherm....
1 1 1 1 4867 3 2 43.91665 5.072260 8.51 6.38
2 2 1 1 4867 3 2 43.91667 5.072263 8.51 6.38
3 3 1 1 4853 3 2 43.91651 5.072444 8.51 6.38
4 4 1 1 4850 3 2 43.91651 5.072484 8.51 6.38
5 5 1 1 4848 3 2 43.91647 5.072510 8.51 6.38
6 6 1 1 4847 3 2 43.91660 5.072510 8.51 6.38

Secondly, I have three years of data (2019, 2021 and 2022, 2020 being incomplete because of covid) of the number of nests per week. This data is associated with the corresponding atmospheric pressure, photoperiod, wind, flow, air temperature....

Annee Semaine Frais TempMoy AmpliTherm Insolation PercentInso Rayonnement Precip....
1 2019 2018_48 4 9.11 10.24 4.05 47.17 531.79 3.71
2 2019 2018_49 5 8.29 8.71 5.31 65.75 635.64 0.14
3 2019 2018_50 22 5.08 7.79 3.01 37.38 394.14 4.24
4 2019 2018_51 12 6.61 11.00 4.76 56.14 576.36 0.14
5 2019 2018_52 13 4.65 9.41 6.15 77.47 694.43 0.07
6 2019 2019_1 24 5.66 5.21 6.09 74.76 708.50 0.01
7 2019 2019_2 22 6.79 7.88 6.33 75.77 760.00 0.04

I ask here how you would analyse this very (too) exhaustive data. I am struggling to get a handle on it and would need some external advice.

So far, I have found with the first set of data that trout choose their flow facies to spawn. Indeed, the flat stream has the highest spawning density even though it represents only 11% of the total surface.
FaciesSélectivité.png
(Sorry, this one is in french)
Furthermore, with the second set of data, I am trying to understand what triggers spawning on my river (the Sorgue). I have so far found that it is the photoperiod.
1658241714004.png
As you can see on this figure (GAM under R), some years are more or less well explained like 2021 where the pressure plays on the activity peak.
1658241961649.png
But for example I can't transcribe this with a stat model hence my post on this forum.

I have spoken well, now it's up to you and don't hesitate if you need more details!