Wooden floorboards on existing terrace slabs

When I look at it like this, I suspect that the substructure of the slabs is wrong.

Water is rising here and bringing lime with it.

Lift one of the slabs and have a look at the gravel and the mineral concrete.

I've heard that quite often, but I can't find any mistake. Under the slabs there is definitely coarse gravel; I lifted one last year. But I was also there during the installation. The terrace is on the south side and is now even covered. The white efflorescences appear this winter mainly right at the house, where it hardly ever rains.

The substructure consists of:
1. An indefinite amount of gravel. Directly at the house, where it is particularly blooming at the moment, it is at least 60 cm - less at the edge.
2. About 10-12 cm of gravel again.
3. About 3-5 cm of gravel.
4. Terrace slabs.
No idea whether it should maybe be 15 or 20 cm, but this is not about settling, rather about rising moisture, and this layer should definitely be sufficiently capillary-breaking. According to a very detailed article, the bedding is a possible cause for efflorescence, but it rather talks about water standing next to the stones at the bottom. That is definitely not the case with me.

By the way, the rest of the paving on the east and north sides shows hardly any efflorescence, although the substructure here is similar and the location is actually climatically less favorable.

Attached are a few photos from the installation.
What else could be a cause if it’s not the slabs themselves?
Should I re-sand the joints of the slabs?

Our landscape gardener gladly uses WPC beams as a substructure, they do not rot. Wooden beams are the weak point after several years. Leave the panels as they are, lay WPC beams 3 or 4 cm thick on top, level them with rubber pads, clamp the decking boards on, you can dowel the beams to the panels (we do it that way), or not.

Hello " ".
That is indeed an interesting problem you are reporting!
Already mentioned: decking boards on a gravel bed is not recommended.
Also not a "floating (loose) installation." The elements will "float" away from you under mechanical stress. That means: they shift uncontrollably from their original position.
Installation on battens: recommended. What speaks against it (seemingly) is the low build-up height of 60mm.
Ventilation is essential for wood and wood-based materials.
WPC decking boards on system-compliant substructure would be a good idea.
But be careful: I have experienced quite a few surprises (professionally) with hollow chamber systems. Solid profiles, on the other hand, have proven to be very stable.
Pedestal supports can be omitted since the existing slabs were probably leveled and aligned in the slope, and are therefore suitable as a substructure.
Self-constructed support strips should be avoided because system-compliant solutions also include holding or mounting clips, which are supposed to hold the boards in their original position.

Such metal substructures are basically suitable (build-up height <20mm) but should be laid strip-wise on rubber granulate mats which are waterproof-glued to the back of the profile.
The whole thing is not so simple because the slabs must remain and we have only a limited build-up height available.
In my opinion, wood as a support is not ideal because it would then "stand with its feet" in water. However, in the case of extensive roofing, this thought should be reconsidered!
----------------------
Good luck: KlaRa

and Thank you very much for your answers. We have decided to give the plates one last chance. If we still have the same problem at the end of this year, we will revisit the project and take your tips into account.