278 Gerbrand J. Komen
. .
O
IV
E
..J
,
-"
. .
•
~
,
~
z
,
....
.#'
~
~
.1~
I
..-,
u
>
'"
.. -
o
z
c:
"
y
u
~
Q.
=:
~
O>
U
::J
W
, ,
IV
s:;
' "
"
â
!:!
::J
. .
I
"
:lE
.,
~~
.2.
.9
,
,..
IV
II>
U
E
8l ~
ii.
~
!3
~
ca
ro
ii;
::J
9
E.s
" ro
~~
~ ~
.,
E
J. ... . . . . . . . . . . . . ...
g ~
~
i E
«~ ~
g
« +
~ "'"
'O
iii
>
~ .... ' ' "' '
o
1"
l:
z
~ ', ....
y
§
" "
iii
<>.
i
", .. -:. .,
o
~
Q.
m
:I:
iI~ .. W : ".
:Il
w
a::
>
, 1
O>
...
« •
r ' 11
::J
!
~
I
I
. .
J:
"_-t :
â
lIS
x
~
I
c:
8
-\
.2.
:
+
\
,..
~
,'\
.. . .
E
li
I I
\
ii.
...
l' .. ,
*
i
, ..
ii
I
'i,l.: ....
ii;
J:
il
, E
. .
j
~
,
, -"
;
il
< ~
e
!
,, 'J.o."' "
~
OII
~
o
:!J
"l
."...
'"
..
ci
9
toi
ci
9
9
rl
(W)S'IIB
a
~
Fig. 14.6 Validation statistics ofthe global WAM model as run by ECMWE HW = Hawaii;
NPC = North Pacific; USWC = US West Coast; USES = US East Coast; CANEC =
Canadian East Coast; NEATL = Northeast Atlantic.
example, Lionello et al., 1992; or Voorrips et al., 1997). But the practical applicability of more advanced methods (adjoint, Kalman filter, etc) should be explored.
Finally, there remain some very basic questions. What exactly is the mechanism
of wave generation? How important is turbulence, and the interaction between
wave motion and turbulence in the air? How can one better understand the role of
breaking waves? These questions have been addressed in many studies (see, for
some recent studies: Janssen, 1994; Belcher and Hunt, 1993; Makin et al., 1995),