There are so many reasons why the NCS is not a good idea.

The most important is that Climate is cross-cutting all of NOAA, and NOAA Line Organizations always end up being stove-pipes that interact poorly with other LOs. So setting up a new Climate Service LO will actually reduce the ability of NOAA to address the full range of climate-relevant activities that NOAA needs to be doing to serve our national interests.  

Inevitably the activities of the NCS will come to be seen as the key NOAA-climate activities.

The prelim NAPA report says that matrix management hasn't worked.  This is because of the fundamental lack of commitment by senior NOAA managment, not because matrix management was unworkable.  Matrix management has to be backed by the Adminstrator and performance has to be insisted upon from the LOs.  

Much of NOAA Research (OAR) is fundamentally climate research, yet it's being split up in a completely arbitrary way.  This will be damaging to OAR in ways that cannot be compensated for.

NWS carries out critical climate functions in observations, modeling and services, yet these are being left out.

NMFS needs to become far more climate-connected to do the science of climate and fisheries, yet this is being left out.

Etc. Etc.

In short, to meet it's climate responsibilities to the nation, NOAA is going to have to do matrix management anyway, even if an NCS is set up. 

It would be far better to try to do serious matrix management of climate activities within NOAA, and have NOAA learn to do cross-LO work than to set up an NCS which only does pieces of the job and harms the rest of the organization due to its construction.

At least TRY to do serious matrix management of climate first. 

An NCS can always be set up later.

Setting up a new Line Organization will take a lot of time (already has and will take more), will take resources (human and financial) away from existing programs, and should be done only if it is necessary.  It is not necessary.