The NSSDC/SPDF people have lots of data on-line.  Some is WIND/WAVES data.
Here:

https://cdaweb.gsfc.nasa.gov/pub/data/wind/waves/

So in trying to scoop up directory trees - in this case from the SPDF people - I tried a CLI app called wget.  I installed it with MacPorts.
Here we go:

The first time I did:

I made a new target directory and put myself there.

% pwd
/Volumes/mySpace/GGS/WIND

% wget https://cdaweb.gsfc.nasa.gov/pub/data/wind/waves/ --mirror

I control-c'ed that after it seemed to get what I wanted but then started making repeated passes.  It seemed to be getting the whole f'n tree.  Filling my whole f'n disk.

The second time I did:

I made another new and empty target directory and put myself there.

% pwd
/Volumes/mySpace/GGS/WIND

% wget https://cdaweb.gsfc.nasa.gov/pub/data/wind/waves/ --mirror -np

The -np switch means "no parent" and that stopped it from going around and around.
Which left me with just the tree I wanted.

At the end of a few hours:

It did not seem to run wild.  Good.
It did seem to give me what I wanted with a lot of "supporting" files and directories.

FINISHED --2025-03-01 16:20:57--
Total wall clock time: 6h 1m 35s
Downloaded: 307103 files, 229G in 1h 41m 51s (38.3 MB/s)

So that's good. 

For fun, I did the exact same thing again.  A test of mirroring.
It seems to do nothing but it took a long long time to do it.
So I control-c'ed the thing.
I started it again later.

FINISHED --2025-03-02 03:01:03--
Total wall clock time: 2h 41m 59s
Downloaded: 6280 files, 323M in 1m 51s (2.91 MB/s)
 
The app says 229GB.  Finder says 246GB.
The app says 307,103 files.  Finder says 308,005 files.
But.
There are a ton of junk files.

Essentially every directory or folder has a file called index.html which is some actual html made to show all of the files in the directory.
So I could point a web browser here.  If I wanted to.

And then there are sometimes files like index.html?C=M;O=A which are maybe intended to add value.  

find . -name "index.html"   -type f -delete
find . -name "index.html?*" -type f -delete

Essentially every directory or folder has a file called SHA1SUM which gives a checksum sort of thing for every file in the folder.

% find . -name "SHA1SUM" -type f -delete

In two old folders with some old files, we sometimes see attribute folders.
For example:

pub/data/wind/waves/spectra_3min_gif_old/1996/attrib/1996006_wind_wav_3min_v01.att 

We could delete the attr junk or we could just delete these old folders.

Three directories contain compressed ASCII files.

pub/data/wind/waves/rad1_ascii_compressed/2000/wind_waves_rad1a_20000101.R1.Z 
pub/data/wind/waves/rad2_ascii_compressed/2001/wind_waves_rad2a_20010101.R2.Z 
pub/data/wind/waves/tnr_ascii_compressed/2001/wind_waves_tnra_20010101.tnr.Z 

These files could be uncompressed (not unzipped) into new trees.
The following lines expand the .Z files and then delete them after.

find . -name "*R1.Z"  -type f -exec uncompress -f {} \;
find . -name "*R2.Z"  -type f -exec uncompress -f {} \;
find . -name "*tnr.Z" -type f -exec uncompress -f {} \;

These files could have their filenames changed to have a filetype of .txt
but I didn't do this.

Three directories would seem to contain IDL savesets.

pub/data/wind/waves/rad1_idl_binary/2001/wind_waves_rad1_20010101.R1 
pub/data/wind/waves/rad2_idl_binary/2001/wind_waves_rad2_20010101.R2 
pub/data/wind/waves/tnr_idl_binary/2001/wind_waves_tnr_20010101.tnr 

These files could have their filenames changed to have a filetype of .sav
but I didn't do this.

Remove the unsightly junk:
...index.html
...index.html?*
...SHA1SUM
...attrib/*.att
...attrib/*.gif_att
...attrib/00readme.txt
...attrib/
...robots.txt