1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
|
\documentclass[a4paper,11pt]{article}
%\usepackage{epsf}
\usepackage[latin1]{inputenc}
%\usepackage{graphicx}
%Check if we are compiling under latex or pdflatex
\ifx\pdftexversion\undefined
\usepackage[dvips]{graphicx}
\else
\usepackage[pdftex]{graphicx}
\fi
\advance\textwidth by 60pt
\advance\oddsidemargin by -25pt
\advance\evensidemargin by -25pt
%
\begin{document}
\newcommand{\etal}{{\it et al.}}
\newcommand{\DegN}{$^{\circ}$N}
\newcommand{\DegW}{$^{\circ}$W}
\newcommand{\DegE}{$^{\circ}$E}
\newcommand{\DegS}{$^{\circ}$S}
\newcommand{\Deg}{$^{\circ}$}
\newcommand{\DegC}{$^{\circ}$C}
\title{cdfovide : user manual}
%\author{R. Dussin \thanks{Laboratoire de Physique des oceans, CNRS-Ifremer-UBO, Brest, France}, J.M. Molines \thanks{Laboratoire des Ecoulements Geophysiques et %Industriels, CNRS UMR 5519, Grenoble, France} }
\author{R. Dussin, J.M. Molines}
\maketitle
\section{Introduction}
cdfovide is part of a package called CDFTOOLS. About the install and common features of CDFTOOLS, please refer to CDFTOOLS documentation.
This document explains briefly how to use it and some details of the code. The usage is :
\begin{verbatim}
cdfovide gridT gridU gridV
\end{verbatim}
\noindent
The grid files \textbf{coordinates.nc} , \textbf{mesh\_hgr.nc} and \textbf{mesh\_zgr.nc} must be in your directory.
\section{Some details of the code}
The Ovide section is approximated by three legs defined such as :
\begin{itemize}
\item leg 1 : ( $43.0$ \DegW , $60.6$ \DegN ) to ( $31.3$ \DegW , $58.9$ \DegN )
\item leg 2 : ( $31.3$ \DegW , $58.9$ \DegN ) to ( $12.65$ \DegW , $40.33$ \DegN )
\item leg 3 : ( $12.65$ \DegW , $40.33$ \DegN ) to ( $8.7$ \DegW , $40.33$ \DegN )
\end{itemize}
\noindent
those values are hardcoded. However it is possible to change them in the code. It corresponds to the following lines :
\begin{verbatim}
!! We define what are the 3 segments of OVIDE section
!! so that the user don't have to worry about it
!! sec1 : (lonsta1,latsta1) -> (lonsta2,latsta2)
!! and so on
lonsta(1)=-43.0
lonsta(2)=-31.3
lonsta(3)=-12.65
lonsta(4)=-8.7
latsta(1)=60.6
latsta(2)=58.9
latsta(3)=40.33
latsta(4)=40.33
\end{verbatim}
The model F gridpoints corresponding to the 4 ends ot the legs are computed using the same code as cdffindij. Then
a broken line is computed using the same code as cdftransportiz. The indices of all the F-points are saved in the arrays isec and jsec.
Their corresponding longitudes and latitudes are stored in nav\_lon and nav\_lat (NB : those are the lon and lat of the F points). If those
arrays' size is $N$ (number of points), the others arrays' size is $N-1$ (number of segments). This is an example of the standard output
with a ORCA025 run :
\begin{verbatim}
------------------------------------------------------------
leg 1 start at -43.00N 60.60W and ends at -31.30N 58.90W
corresponding to F-gridpoints( 986, 796) and (1026, 782)
------------------------------------------------------------
------------------------------------------------------------
leg 2 start at -31.30N 58.90W and ends at -12.65N 40.33W
corresponding to F-gridpoints(1026, 782) and (1098, 675)
------------------------------------------------------------
------------------------------------------------------------
leg 3 start at -12.65N 40.33W and ends at -8.70N 40.33W
corresponding to F-gridpoints(1098, 675) and (1114, 675)
------------------------------------------------------------
\end{verbatim}
\noindent
Once we have these list of F-gridpoints isec and jsec, we have to pick the values of $u$ or $v$ corresponding to the segment.
If the segment is an horizontal one, we will pick a value for $v$ and $u=0$ and vice-versa. Hence the vozocrtx and vomecrty arrays
in the output netcdf files will have empty lines, this is perfectly normal. We loop from F-point $1$ to $N-1$ and we define the current
F-point as f(i,j). Four cases are investigated :
\begin{itemize}
\item horizontal segment, eastward : $v = v(i+1,j)$ and $u = 0$
\item horizontal segment, westward : $v = v(i,j)$ and $u = 0$
\item vertical segment, southward : $v = 0$ and $u = u(i,j)$
\item vertical segment, northward : $v = 0 $ and $u = u(i,j+1)$
\end{itemize}
\noindent
The $e1v$, $e2u$, $e3v$ and $e3u$ arrays are picked at the same points that $u$ and $v$. Also, $u = 0$ leads to $e2u = e3u = 0$ and vice-versa.
The temperature and salinity are interpolated on the $u$ or $v$ point. In the bottom, if one value on the T-point is zero (land), the value is set to
zero.
\end{document}
|
