The union(RDD) transformation returns a new RDD that is the union of the source and argument RDDs. Look at the following code snippet:

# Create `a` RDD of Washington airports
a = (
    airports
    .zipWithIndex()
    .filter(lambda (row, idx): idx > 0)
    .map(lambda (row, idx): row)
    .filter(lambda c: c[1] == "WA")
)

# Create `b` RDD of British Columbia airports
b = (
    airports
    .zipWithIndex()
    .filter(lambda (row, idx): idx > 0)
    .map(lambda (row, idx): row)
    .filter(lambda c: c[1] == "BC")
)

# Union WA and BC airports
a.union(b).collect()

This will generate the following output:

# Output
[[u'Bellingham', u'WA', u'USA', u'BLI'],
 [u'Moses Lake', u'WA', u'USA', u'MWH'],
 [u'Pasco', u'WA', u'USA', u'PSC'],
 [u'Pullman', u'WA', u'USA', u'PUW'],
 [u'Seattle', u'WA', u'USA', u'SEA'],
...
 [u'Vancouver', u'BC', u'Canada', u'YVR'],
 [u'Victoria', u'BC', u'Canada', u'YYJ'], 
 [u'Williams Lake', u'BC', u'Canada', u'YWL']]