Opening remote datasets not lazy with fsspec

I'm finding that the following code reads all of the remote datasets into memory:

fs = gcsfs.GCSFileSystem(project='ldeo-glaciology', mode='ab')
filelist = fs.ls('gs://ldeo-glaciology/AMPS/wrf_d03_20161222_week-cf')

arrays = []
for filename in filelist:
    with  fsspec.open('gs://' + filename, mode='rb') as openfile:  
        arrays.append(xr.open_dataset(openfile, chunks = {}))

Is this the expected behavior? Is there a way to open these files as a remote dataset without fully reading them into memory? Thanks!

Start by reproducing the provided fsspec.open and xr.open_dataset example with the listed GCS paths, then inspect how remote file handles are passed between them. Determine whether the observed memory loading is expected and, if not, identify the relevant behavior to change; done means establishing lazy access or documenting the limitation.