What Twenty Years Does to a 2000s Meridian Deck

Meridian’s first big subdivision wave is now old enough to have a second generation of decks coming due. Bear Creek built out between 2001 and 2006. Baldwin Park ran 2003 to 2007. Kelly Creek 2004 to 2009. Lochsa Falls was developed in 2003 with homes going up through 2012. The decks that went on the back of those houses, whether with the original build or a few years after, are now fifteen to twenty-five years into a climate that is hard on exposed timber.

They do not fail all at once. They fail in an order, and the order is useful because only the last one is dangerous.

First: the boards

Surface decking goes first, and it goes cosmetically. Ultraviolet light degrades the surface layer of the wood, the color turns silver-gray, and fine checking opens along the board faces. Boards start cupping, usually crowning up in the middle or curling at the edges depending on how they were laid and how wet they were when they went down.

A long clear high desert summer produces that result quickly on an unmaintained deck. Two summers to gray, three to visible checking.

None of this is structural. A gray checked board still carries load fine. It is unpleasant underfoot, it holds water in the checks, and it looks tired, but a deck at this stage is a maintenance question, not a safety one. The finishing guide covers what bringing it back involves.

Second: the rail and the stairs

Movement shows up next. A rail post that has a bit of give when you lean on it. A stair that has racked slightly so the risers are no longer uniform. A handrail that has worked loose at one end.

These are worth taking seriously, because a guard exists specifically to stop somebody going over the edge, and the way it does that is not intuitive. A guard post is a lever: the push lands at the top and the connection holding it sits down in the framing, so what the fixing at the bottom has to resist is considerably larger than the shove that started it.

Decks from the early 2000s frequently have rail posts simply lag screwed to the outside face of the rim joist, which was ordinary practice at the time. That connection resists the pull only by having its screws drawn straight out of the wood, and withdrawal is the weakest thing you can ask of a fastener. Twenty years of seasonal movement takes what was marginal to begin with and makes it loose. It is repairable, it is not expensive relative to a rebuild, and it is not a job to defer.

Third: the ledger

The one that matters. The ledger is the board bolted to the house carrying one end of the joists, and on an attached deck it is the most safety-critical connection in the structure. Done correctly it is through-bolted or lag screwed into house framing with corrosion resistant flashing over the top so water runs out of the wall rather than into it.

Done incorrectly, it is nailed to the sheathing, or bolted without flashing, or both. Neither condition announces itself. The flashing failure in particular is invisible from outside: water tracks in behind the ledger, the band joist rots slowly from the inside, and the fasteners lose the material they are anchored to. Everything looks fine right up until it does not.

This is why any honest repair assessment on an older deck starts by pulling two or three boards next to the house and looking. A builder who quotes a re-board from photos has not seen the part of the deck that decides whether re-boarding is a good idea.

What that means for repair decisions

The dividing line is the frame.

Frame sound, surface gone. Re-boarding over an existing frame costs a good deal less than a full rebuild, and it is a sensible move. One catch: if the plan is composite, the frame has to suit it. Composite wants its joists closer together than wood does, and closer again under a diagonal layout, so a frame built decades ago for thick wood decking will not take it without new joists threaded in alongside the old ones.

Frame compromised. If the ledger was nailed rather than bolted, if there is no flashing, if the footings are shallow or the posts are set in soil, then money spent on the surface is money spent on top of the fault. At that point a rebuild is usually the honest answer, however much the smaller job would suit everybody’s budget better.

The middle case

Often the frame is mostly fine and one element is not. A ledger re-bolt and re-flash on a typical back yard deck runs $1,800 to $4,500, because siding has to come off and go back properly. That is a real cost and it is far below a rebuild, and it converts a deck with a hidden liability into one worth putting new boards on.

That is the calculation worth making before choosing a decking color: not what the deck looks like, but what the first three boards next to the house reveal when they come up. The repair page covers the scopes, and the footings guide covers the other half of the frame that nobody can see.