Professional Documents
Culture Documents
Mapping Theory
Mapping Theory
Mapping
Introduction to MATLAB mapping Toolbox
Map projection
MATLAB mapping Toolbox and georeferencing
Laboratory exercises:
The simplest (although perhaps not the most general) definition of a map is a
representation of geographic data.
Geospatial data are spatial data (data where things are within a given
coordinate system) that are absolutely or relatively positioned on a planet, or
georeferenced (it has a terrestrial coordinate system that can be shared by
other geospatial data)
Geodata is coded for computer storage and applications in two principal ways:
vector and raster representations. It has been said that “raster is faster but
vector is corrector
Map examples
90° N
45° N
0°
180° W 135° W 90° W 45° W 0° 45° E 90° E 135° E 180° E
45° S
90° S
Example 1
---------
% World map with coarse coastlines
worldmap('World') World
load coast
plotm(lat, long)
Map examples
75 °
N
60 °
N
45 °
N
30 °
N
° E
45
15 °
worldmap('Europe') W °
30 E
0° °
15 E
load coast
plotm(lat, long)
Europe
Map examples
45.0 ° N
42.5 ° N
40.0 ° N
37.5 ° N
35.0 ° N
15.0 ° W1 ° °
° 0.0 E
2.5 W10.0° W 7. ° ° E 5.0° E 7 .5 E 1
5 W 5.0 ° W 2.5° W 0.0 ° 2.5
worldmap('Spain')
load coast
plotm(lat, long)
Spain
Example 2
---------
% Worldmap with land areas, major lakes and rivers, and cities and
% populated places
ax = worldmap('World');
setm(ax, 'Origin', [0 180 0])
land = shaperead('landareas', 'UseGeoCoords', true);
geoshow(ax, land, 'FaceColor', [0.5 0.7 0.5])
lakes = shaperead('worldlakes', 'UseGeoCoords', true);
geoshow(lakes, 'FaceColor', 'blue')
rivers = shaperead('worldrivers', 'UseGeoCoords', true);
geoshow(rivers, 'Color', 'blue')
cities = shaperead('worldcities', 'UseGeoCoords', true);
geoshow(cities, 'Marker', '.', 'Color', 'red')
90° N
45° N
0°
0° 45° E 90° E 135° E 180° E 135° W 90° W 45° W 0°
45° S
90° S
Example 3
---------
% Map of Antarctica
worldmap('antarctica')
antarctica = shaperead('landareas', 'UseGeoCoords', true,...
'Selector',{@(name) strcmp(name,'Antarctica'), 'Name'});
patchm(antarctica.Lat, antarctica.Lon, [0.5 1 0.5])
70° S
80° S
90 °
S
Example 4
---------
% Map of Africa and India with major cities and populated places
worldmap({'Africa','India'})
land = shaperead('landareas.shp', 'UseGeoCoords', true);
geoshow(land, 'FaceColor', [0.15 0.5 0.15])
cities = shaperead('worldcities', 'UseGeoCoords', true);
geoshow(cities, 'Marker', '.', 'Color', 'blue')
30° N
30° S
Example 5 ---------
% Map of the geoid over South America and the central Pacific
worldmap([-50 50],[160 -30])
load geoid
geoshow(geoid, geoidrefvec, 'DisplayType', 'texturemap');
load coast
geoshow(lat, long)
The geoid approximates mean sea level. The shape of the ellipsoid was calculated
based on the hypothetical equipotential gravitational surface. A significant
difference exists between this mathematical model and the real object. However,
even the most mathematically sophisticated geoid can only approximate the real
shape of the earth.
geoid, model of the
figure of the Earth—i.e.,
of the planet’s size and
shape—that coincides
with mean sea level over
the oceans and continues
in continental areas as an
imaginary sea-level
surface defined by spirit
level. It serves as a
reference surface from
which topographic
heights and ocean depths
are measured. The
scientific discipline
concerned with the
precise figure of the Earth
and its determination and
significance is known as
geodesy.
Example 6
---------
% Map of terrain elevations in Korea
load korea
not display the
h = worldmap(map, refvec); outside of the
set(h, 'Visible', 'off') projection
geoshow(h, map, refvec, 'DisplayType', 'texturemap')
colormap(demcmap(map))
GEOSHOW(LAT, LON) or
GEOSHOW(LAT, LON, ..., 'DisplayType', DISPLAYTYPE, ...)
GEOSHOW(LAT,LON,Z, ..., 'DisplayType', DISPLAYTYPE, ...)
GEOSHOW(Z,R, ..., 'DisplayType', DISPLAYTYPE,...)
GEOSHOW(LAT,LON,I)
GEOSHOW(LAT,LON,BW)
GEOSHOW(LAT,LON,X,CMAP)
GEOSHOW(LAT,LON,RGB)
GEOSHOW(..., 'DisplayType', DISPLAYTYPE, ...)
GEOSHOW(I,R)
GEOSHOW(BW,R)
GEOSHOW(RGB,R)
GEOSHOW(A,CMAP,R)
GEOSHOW(..., 'DisplayType', DISPLAYTYPE, ...)
GEOSHOW(S) or GEOSHOW(S, ..., 'SymbolSpec', SYMSPEC, ...)
GEOSHOW(FILENAME)
........
In some cases UTM coords are also used, but these are really just a simple
transformation based on the location of the equator and certain lines of longitude.
On the other hand, there are occasions when a coordinate system based on some
other set of axes is useful. For example, in space physics data is often projected in
coordinates based on the magnetic poles
East longitudes are positive, whereas West longitudes are negative !!!!
Nord latitudes are positive, whereas Sud latitudes are negative!!!
Also note that a decimal degree notation is used, so that a longitude of 120 30'W is
specified as -120.5. .
MATLAB mapping toolbox
>>clear
>> load coast
>> axesm mercator
>> framem
>> plotm(lat,long)
whos
lat 9589x1 76712 double
long 9589x1 76712 double
plotm(lat(1:20),long(1:20),'r')
Plotting raster (grid) data in MATLAB
>> load topo
>> whos
Name Size Bytes Class Attributes
>> indiana=shaperead('usastatehi.shp','UseGeoCoords',true,'Selector',...
{@(name)strcmpi('Indiana',name),'Name'});
>> inLat=indiana.Lat;
>> inLon=indiana.Lon;
>> gridDensity=40; Convert vector
>> [inGrid,inRefVec]=vec2mtx(inLat,inLon,gridDensity); information to
>> whos a raster grid image
Name Size Bytes Class Attributes
Mapping
Introduction to MATLAB mapping Toolbox
Map projections
MATLAB mapping Toolbox and georeferencing
Laboratory exercises:
Usually, maps of the whole world are Mercator, although often the Miller Cylindrical
projection looks better because it doesn't emphasize the polar areas as much.
The Robinson projection is not equal-area or conformal, but was the choice of
National Geographic (for a while, anyway), and also appears in the IPCC reports.
If you are plotting something with a large north/south extent, but not very wide (say,
North and South America, or the North and South Atlantic), then the Sinusoidal or
Mollweide projections will look pretty good.
Another choice is the Transverse Mercator, although that is usually used only for
very large-scale maps.
Map Projections
MATLAB mapping toolbox
Projection Name Map Projection ID
--------------- -----------------
Transverse Mercator tranmerc
Mercator mercator
Lambert Conformal Conic (2SP) lambert
Lambert Conformal Conic (1SP) lambert
Lambert Azimuthal Equal Area eqaazim
Albers Equal-Area Conic eqaconic
Azimuthal Equidistant eqdazim
Equidistant Conic eqdconic
Polar Stereographic ups
Oblique Stereographic stereo
Equirectangular eqdcylin
Cassini-Soldner cassini
Gnomonic gnomonic
Miller Cylindrical miller
Orthographic ortho
Polyconic polycon
Robinson robinson
Sinusoidal sinusoid
Van der Grinten vgrint1
Map Projections
But what projection should I use?
For smaller areas within one hemisphere or other (say, Australia, the United
States, the Mediterranean, the North Atlantic) you might pick a conic projection.
The differences between the two available conic projections are subtle, and if
you don't know much about projections it probably won't make much difference
which one you use.
If you get smaller than that, it doesn't matter a whole lot which projection you
use. One projection useful in many cases is the Oblique Mercator, since you can
align it along a long (but narrow) coastal area.
If map limits along lines of longitude/latitude are OK, use a Transverse Mercator
or Conic Projection. The UTM projection is also useful.
Polar areas are traditionally mapped using a Stereographic projection, since for
some reason it looks nice to have a "bullseye" pattern of latitude lines.
If you want to get a quick idea of what any projection looks like, default
parameters for all functions are set for a "typical" usage, i.e. to get a quick idea
of what any projection looks like, you can do so without having to figure out a lot
of numerical values:
How to specify map projections in MATLAB
mapping Toolbox?
AXESM GUI
AXESM activates a GUI to define a map projection for the current axes.
AXESM(PROPERTYNAME, PROPERTYVALUE,...)
AXESM(MSTRUCT,...)
AXESM(PROJFCN,...)
The MATLAB functions in the table below read files from URLs. Then you can write
the files to a local directory or load them into your MATLAB Workspace. With the
exception of urlread and urlwrite, all the functions can read files on local and
network file systems using path syntax as well.
geotiffread Read a georeferenced image from GeoTIFF file
gunzip Uncompress files in the GNU-Zip format
imread Read image from graphics file
untar Extract the contents of a Tar-file
unzip Extract the contents of a Zip-file
urlread Returns the contents of a URL as a string
urlwriteSave the contents of a URL to a file
Mapping
Mapping
Introduction to MATLAB mapping Toolbox
Data Models
Map projections
MATLAB mapping Toolbox and georeferencing
Laboratory exercises:
Raster geodata consists of georeferenced data grids and images that in MATLAB
are stored as matrices. Raster geodata looks like any other matrix in MATLAB
Referencing vectors
In many instances (when the data grid or image is based on latitude and
longitude and is aligned with the geographic graticule), a referencing matrix has
more degrees of freedom than the data requires. In such cases, you can use a
more compact representation, a three-element referencing vector.
A referencing vector defines the pixel size and northwest origin for a regular,
rectangular data grid:
Note that if the latitude extent of cells differs from their longitude extent you
cannot use a referencing vector, and instead must specify a referencing
matrix.
The second element, north-lat, specifies the northern limit of the data grid
(as a latitude), and the third element, west-lon, specifies the western extent
of the data grid (as a longitude). In other words, north-lat, west-lon is the
northwest corner of the data grid.
Note, however, that cell (1,1) is always in the southwest corner of the
grid. This need not be the case for grids or images described by referencing
matrices, as opposed to referencing vectors.
Regular data grids
minigrid=[1,2,3,4,5,6;7,8,9,10,11,12;13,14,15,16,17,18]
minivec=[1/60,90,-180]
minigrid = -90,180 SE
-90,-180 SW
1 2 3 4 5 6
7 8 9 10 11 12
13 14 15 16 17 18
90,-180 NW
90,180 NE
Remember!
minigrid=[1,2,3,4,5,6;7,8,9,10,11,12;13,14,15,16,17,18]
minivec=[1/60,90,-180]
>> meshm(minigrid,minivec);
>> colorbar
SW SE
minigrid =
1 2 3 4 5 6
7 8 9 10 11 12
13 14 15 16 17 18 NE
NW
5) Map de data using mesm and display with a color map legend
meshm(minigrid,minivec);
colormap('autumn');colorbar
minigrid =
-90,-180 SW 1 2 3 4 5 6 -90,180 SE
7 8 9 10 11 12
90,-180 NW 13 14 15 16 17 18 90,180 NE
Computing Map limits from reference vectors
load korea
refvec = 12.0000 45.0000 115.0000
[latlimits,longlimits]=limitm(map,refvec)
latlimits =
30 45
longlimits =
115 135
This means that korea region extends from 30oN to 45oN and 115oE to 135oE
>> [rows,cols]=size(map)
rows =
180
cols =
240
>> southlat=refvec(2)-rows/refvec(1)
southlat =
30.0000
>> eastlon=refvec(3)+cols/refvec(1)
eastlon =
135.0000
latitudes decrease southwards, longitudes increase eastwards
Geographic interpretation of matrix elements
>> [r,c]=setpostn(map,refvec,lat,long)
r=
23
c=
79
This means that this lat,long is in the 23,79 row coliumn
BUILDING A GEOREFERENCE MATRIX, R
geoidR =
0 1.0000
1.0000 0
-0.5000 -90.5000
row =
95
col =
79
ans =
-106.9260
Example 3
---------
Create a half-resolution version of a georeferenced TIFF image, using
Image Processing Toolbox functions IND2GRAY and IMRESIZE.
Laboratory exercises:
Plotting images and maps
Google map loader
Getting satellite data from internet
Representing results of a monitoring campaign
150
200
250
300
350
400
450
500
Laboratory exercises:
Plotting images
Google maps and Google Earth
Getting satellite data from internet
Representing results of a monitoring campaign
>> load('mapex.mat')
>> whos
Name Size Bytes Class Attributes
41.94
41.92
41.9
41.88
% scatter(y(i),x(i),(z(i)-min(z)+0.1)*100,[a,b,c],'.')
% scatter(y(i),x(i),(z(i)-min(z)+0.1)*20,'o','MarkerEdgeColor','k','MarkerFaceColor','c')
54
52
50
48
46
44
42
40
38
36
-5 0 5 10 15 20 25
plot_google_map
54
52
50
48
46
44
42
40
38
36
-5 0 5 10 15 20 25
>> axis([0,5,40,44]);
>> plot_google_map
44
43.5
43
42.5
42
41.5
41
40.5
40
0 1 2 3 4 5
>> axis([2,5,38,41]);
>> plot_google_map('maptype','roadmap');
>> plot_google_map('maptype','satellite');
41 41
40.5 40.5
40 40
39.5 39.5
39 39
38.5 38.5
38 38
1.5 2 2.5 3 3.5 4 4.5 5 5.5 1.5 2 2.5 3 3.5 4 4.5 5 5.5
>> load mapex
>> mapgoogle(gpsxy,sampnumbers,z)
42.04
42.02
42
41.98
41.96
41.94
2.24 2.26 2.28 2.3 2.32 2.34 2.36 2.38 2.4 2.42 2.44
Other possibilities???
Simple Google Map Loader in MATLAB.
Nord 42
Est 2.0
Simple Google Map Loader in MATLAB.
λEm 0.35
0.3
0.25
Loadings
0.2
λEx-C
D1 1 0.15
0.1
0.05
0
240 260 280 300 320 340 360 380 400 420 440
Wavelength (nm)
Emission MCR
ALS
λEx-C
D1
Excitation
D2 2
D D2
D3 SEx1-C1 new
SEx1-C1=C1 ⊗ SEx1
D4
λEx-C
D3
3
SVD
1 2 3 4 …… n
1 2 3 4 …… n
Dn Concentration
λEx-C
D4
4
1st component
0.18
0.16
0.14
0.12
……
0.1
……
0.08
0.06
0.04
440
0.02
420
0
0 20 40 60 80 100 120 140
400
Dn n
λEx-C
380 360
W avelength(nm)
320
300
280
260
240 340
0.3
0.25
0.2
0.15
0.1
0.05
0
Figure 2
a
Lo g
din
Excitation
3 components 0.4
MCR-ALS Ex
4 components0.4
MCR-ALS Ex
5 components
0.4
MCR-ALS Ex
6 components0.4
MCR-ALS Ex
0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em
PARAFAC Ex PARAFAC Ex PARAFAC Ex PARAFAC Ex
PARAFAC Em PARAFAC Em PARAFAC Em PARAFAC Em
0.3 0.3 0.3 0.3
Loadings
Loadings
Loadings
Loadings
0.2 0.2 0.2 0.2
0 0 0 0
240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440
Wavelength(nm) Wavelength(nm) Wavelength(nm) Wavelength(nm)
0.4 0.4 0.4 0.4
MCR-ALS Ex MCR-ALS Ex MCR-ALS Ex MCR-ALS Ex
0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em
PARAFAC Ex PARAFAC Ex PARAFAC Ex PARAFAC Ex
PARAFAC Em PARAFAC Em PARAFAC Em PARAFAC Em
0.3 0.3 0.3 0.3
Loadings
Loadings
Loadings
0.2 0.2 0.2 0.2
0 0 0 0
240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440
Wavelength(nm) Wavelength(nm) Wavelength(nm) Wavelength(nm)
0.4 0.4 0.4 0.4
MCR-ALS Ex MCR-ALS Ex MCR-ALS Ex MCR-ALS Ex
0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em
PARAFAC Ex PARAFAC Ex PARAFAC Ex PARAFAC Ex
PARAFAC Em PARAFAC Em PARAFAC Em PARAFAC Em
0.3 0.3 0.3 0.3
Loadings
Loadings
Loadings
0.2 0.2 0.2 0.2
0 0 0 0
240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440
Wavelength(nm) Wavelength(nm) Wavelength(nm) Wavelength(nm)
0.4 0.4 0.4
MCR-ALS Ex MCR-ALS Ex MCR-ALS Ex
0.35 MCR-ALS Em 0.35 MCR-ALS Em 0.35 MCR-ALS Em
PARAFAC Ex PARAFAC Ex PARAFAC Ex
PARAFAC Em PARAFAC Em PARAFAC Em
0.3 0.3 0.3
Loadings
Loadings
0.2 0.2 0.2
0 0 0
240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440
Wavelength(nm) Wavelength(nm) Wavelength(nm)
0.4 0.4
MCR-ALS Ex MCR-ALS Ex
0.35 MCR-ALS Em 0.35 MCR-ALS Em
PARAFAC Ex PARAFAC Ex
PARAFAC Em PARAFAC Em
0.3 0.3
Loadings
0.2 0.2
0.15 0.15
0.1 0.1
0.05 0.05
0 0
240 260 280 300 320 340 360 380 400 420 440 240 260 280 300 320 340 360 380 400 420 440
Wavelength(nm) Wavelength(nm)
0.4
MCR-ALS Ex
0.35 MCR-ALS Em
PARAFAC Ex
Figure 3
PARAFAC Em
0.3
0.25
Loadings
0.2
0.15
0.1
0.05
0
240 260 280 300 320 340 360 380 400 420 440
Wavelength(nm)
41.94
9
41.93
8
41.92
7
41.91
a 41.9
6
%Fmax
5
41.89
4
41.88
3
41.87
2
2.22 2.24 2.26 2.28 2.3 2.32 2.34 2.36 2.38 2.4 2.42
41.94
1
9 10 20 30 40 50 60
Stations
41.93
8
41.92
7
41.91
6
%Fmax
41.9
5
41.89
4
41.88
3
41.87
2
2.22 2.24 2.26 2.28 2.3 2.32 2.34 2.36 2.38 2.4 2.42
1
41.94
9 10 20 30 40 50 60
Stations
41.93 8
41.92 7
41.91 6
%Fmax
41.9
5
41.89 4
41.88 3
41.87 2
Figure 4 2.22 2.24 2.26 2.28 2.3 2.32 2.34 2.36 2.38 2.4 2.42 1
10 20 30
Stations
40 50 60
Lat 40.01 N, Lon 2.09 E
Laboratory exercises:
Plotting images
Google map loader
Getting satellite data from internet
Representing results from a monitoring campaign
For example:
in subdirectory nasa from lab_maps
bsst2007.hdf
Use import bizard of MATLAB
Choose bsst
It is a matrix of dimensions 4096 8192
It can be represented directly using image
image(bsst)
>> x=0:8192;
>> y=0:4095;
>> dx=360/8192;
>> dy=180/4096;
>> latitude = (90. - (y * dy)) - (dy/2);
>> longitude = (-180. + (x * dx)) + (dx/2);
Data Characteristics:
Parameter/Variable:
Units of Measurement:
for version 4.1 and below pixel value with slope of 0.15 and y-intercept of -
3.0 to convert to °C.
for version 5.0 pixel value with slope of 0.075 and y-intercept of -3.0 to
convert to °C.
>> temp=bsst.*0.075-3;
>> imagesc(temp)
temp2=temp(1:4:4096,1:4:8192);
dx=360/8192
dy=180/4096
axesm Robinson
temp2=temp(1:4:4096,1:4:8192);
refmat2=makerefmat(-180,90,dx*4,-dy*4)
geoshow(temp2,refmat2,'DisplayType','texturemap')
gridm
Mapping
Mapping
Introduction to MATLAB mapping Toolbox
Data Models
Map projections
MATLAB mapping Toolbox and georeferencing
Introduction to M_map Toolbox
Laboratory exercises:
Plotting images
Google map loader
Getting satellite data from internet
Representing results from a monitoring campaign
15 18c
18a
16 3436a
3739a
39c42a
33a
33c 40 42c
15c
15a
10 14 21
20a 2830a
32
13 22 29
27b
12a
12c 2526b 31
23c
43b
44
11 23a 45
5 10
9c 46a
46c
6b789a 4849b
long
0
5b
-5
4b
longitud
80 ° -10
N 39
36
42
41
4019
18
17
34
35 3b
4330
32
2037
38
16
33
28
15
44
26
29
25
31
45 14
21 -15
46
471124
27
13
22
12 2b
49
48 23
-20 11b
910 10 20 30 40 50 60 70 80
78
6 Sample
Decluttered
5
82
39b 42c
1718b 33a
33c 38 4042a
3536b 43b
44
80 15c 20a
15a 21 2526b 2931
27b 46a
46c
4 11
13
12b 22
23a
23c
4849b
9b10
8
78 6b7
3 5b
70 °
N 76
latitud
latd
2 4b
74
3b
1 72
2b
70
30 ° ° E
W 20 11b
20 ° W ° 68
10 E 10 20 30 40 50 60 70 80
10 ° W °
0 Decluttered
Sample
PROYECTO ATOS (Julio 2007)
Deposición atmosférica de carbono y contaminates a los ecosistemas polares
group tasks:
North Pole
POP analysis in samples:
Higher Latd -air and aerosols
and long -sea-water
Ice
39
36 19
-Ice
42
41 18
17
35
40 38
34
37
16
-Fito- and zooplancton
33
28
30
32
43 20 15
44 26 Experiments:
Approximate 31 29
25 14
21
45
ice limit -fotodegradation
2427
-biodegradation (zoo)
46 13 -accumulation
22
12
48 47
49 23 -toxicity
11
Open
Method Implementation
sea -SBSE+TD+GC-MS
10
9
8 Data evaluation and
7
6 Integration
Example: CTD data evaluation and Integration
temp
6b 20b23a
33b
36a 45
20 9b 41 43c 2 28
12c 15c 39b 42a 43b
5a 10 24 33a 42c 48 4a
23b
2a 6c 9c 12b 13 28 42b 46c
15 9a 12a 22 43a 1 4b 18c 36a
2b 8 15b 23c 36b 46a
5b 1b
1 36b
10 44 0 3b
7 31
39c
5 -1 2a 39b
18c 2b 43a 4849b
0
AT THE -2
4142b 43c 46b
4749a
10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80
Decluttered
Sample
MAXIMUM Decluttered
Salt conc
Sample
Conductivity
36 FLUORESCENCE. 35.5
27b
35 22
89a9c 22 26a27b
23c26a
23a
35 6a 23a
16 20b 25
23c 35 37
32 33c
31 37 6b7 15b 31 39a40
34
9b 32 33c
35 4b5b 6c 12a 15a 18a
17 21 24 30a 33b 38
45
9a 12a 15b 16 10 12b
18a 20b 25 39a40 11 13 18b
33 8 9c 12c 20a 33b 45 34.5 15c 36a
7 14 15c 38 11b 26b
10 24 30a 39c
5a6a6c 11 13 17 21 29
32 6b 3a 44
18b 3b 36b
salt
44
cond
26b 34
31 47
2b 28 43a
43c
2a 49a
4a 46b
30 28 39b4142b 46a
46c 49b
4b 36a 33.5 42c
29 11b
36b
3b 18c
28 39c 33 48
39b 18c
27 2b 43a 47 49b
42c43c46a
42a 46c
48
26 32.5
10 20 30 40 50 60 70 80 10 20 30 40 50 60 70 80
Sample Sample
Decluttered Decluttered
420 95
90 2a 4b 49b
400 39b
1b 43b 4a 6b 36b 40
17 27b
28 39a 45 48
2b3b 43c46a 1b 22 27a29
46b 6c 9b 18c 38
85
380 1 2a 36a 46c 9a 3436a
33b
4b 26b 8 33a 35
4a 7 33c
5b 30a 36b 32 42c
21 23c26a 39c 48 80 1 16 4446a
5a 11 23b 15b 46c
49b 15a 18a 25 30a
oxyg
13 24 9c 20b
btra
360 10 18a 49a 26b28 46b
12c18c 32 12a 14 15c 31
18b20b 33b 40 44 75 5a 21 23b 26a
12b 15c 10 12b
12c 39b
7 12a 14 17 25 38 5b 24
340 9c 33c 47
31 34 43a
6c89a 15a16 29 70 23c 41
15b 39a
6b 9b 35 37 23a
45
320 11
22 27a
27b
AT THE 65 42a 43c
43b
300
10 20 30 40
Sample
50 60 MAXIMUM
70 80
60
10 20 30 40
Sample
50 60 70 80
Decluttered Decluttered
50
FLUORESCENCE.
5b 23a 39b 42a43a 100
1112b
13 23c26a 43c
9c 12a
45 14 16 21
20b 24 31 46b 90 Turb. 39b
2526b
40 15c 30a 80
46c 23c
15a 23a
35 18a 35 37 42b
70 11 43c
20a 32 36a 26a 43b
30 23b 42a
15b 34 60 5b
8 44 41
9c 26b
fluo
25 45 12c 15c
turb
28 39a 50 1 5a 13 21 25
18c 33a
33c 12b 28 43a
20b 24
33b 46a 12a
20 9b
7 10 14 20a 31 33c
27a 40 1b 32
7 9b 15a16
22 29 40 8 27b 36a
15 38 15b 18a 33b35 42b 44
17 48 30 27a 37 46c
4b 27b 22 30a 39a
40 42c
36b 4a 6a 18b 45
17 18c
10 1 6a 46a
4a
1b 3b 6c
Int. Fluor. 39c 49b
4749a
20 2a 3b4b 6b
2b
48
36b 39c 49b
5 10
4749a
2a
0 2b
10 20 30 40 50 60 70 80 0
10 20 30 40 50 60 70 80
Sample Sample
Decluttered Decluttered
correlation in dcm (máximo de clorofila
máximo de fluorescencia)
Correlation Map, Variables in Original Order
10
1
2
3
4
5
6
7
8
9
0.8
1 'press'
1 0.6
2 'temp' 2
3 'cond' 3
0.4
4 'salt' 4
0.2
5 'oxyg' 5
0
6 'btra' 6 -0.2
7 'fluo' 7 -0.4
8 'turb' 8 -0.6
9 'long' 9
-0.8
10
10 'latd'
Scale Gives Value of R for Each Variable Pair
Eigenvalues and Cross-validation Results for yred11 Variables/Loadings Plot for yred11
3.5 0.5
3
PCA 0.4
temp cond
salt
0.3 long
latd
2.5
Loadings on PC 1 (39.52%)
fluo
0.2 turb
RMSECV, RMSEC
Loadings on PC 2 (31.00%)
0.2 latd
Loadings on PC 2 (31.00%)
0
0
-0.1 -0.1
temp temp
cond
cond -0.2
-0.2 salt
salt
press press
-0.3 -0.3
-0.4 -0.4
42a 39b
43b
4
43c
41
43a Higher fluorescence
3 Higher dissolved O2 and turbidimetry
Scores on PC 2 (31.00%)
42c 11 23c
2 42b
46b 23a
46c 28 5b
26b 13
24 12c 23b
21 26a
1 46a 18c 5a 10 12b
15c Intermediate
36a 12a 31
Last samples 14
9c samples
44 20b
closer to the ice
0 48 30a 20a18a25
36b 15a
7 33a 15b
32 16
-1 8 33b 33c
Initial 49b 9a
39c 6c 18b
38 34 3537
samples 1
49a 29 45 9b
2b2a 3b1b 4a 17 40 39a
-2 3a 22 Higher
47 4b 6b 6a 27a Salt
Higher beam transmission
27b Conduct
-3 Temp
-5 -4 -3 -2 -1 0 1 2 3
Scores on PC 1 (39.52%)
Study of the fluorescence at different depths for multiple samples
100x2 deepths (CTD down and up)
Xfluor(80,200) 10
50
45
20 40
Fluorescence
35
30
at 425 nm excitation
80 stations 40
30
80
60
60 50 100 150 200
50
50
40 40
30 30
Initial Last
20 stations20 stations
10 10
0 0
0 20 40 60 80 100 120 140 160 180 200 0 10 20 30 40 50 60 70 80
MCR-ALS resolution of 4 fluorescence patterns
0.35
0.25 14 26 1813
25
1620 272319 7
26
11 18 24
0.2 29 28 6
20 12
10 31 3841 29 11 5
23 3337 21
0.15 43
9 45 3631
4953 41 32 41
7 54 37
0.1 6 56 45
4 60 49
1 6468
0.05 605652
697377 6864
71 7883
8287919599 96928884807672
0
20 40 60 80 100 120 140 160 180 200
Decluttered Variable
0.25
14
15 max. fluor. 7
16
10
11 17
at low depth. 6
0.2
18
11
10 5
0.15
9 19 12
Row 3
8 1
13 4
7
0.1 6 20
5
3 21 14
1 15
22
0.05 2016
23
27
2419
MCR-ALS 2428 3935 27
474338 3126
42 505559 51
4651
C1 0
20
31
32
33
34
35
36
37
38
40
39
40
4144 53 61
60
6367
68
69
7074
71
72
73
75
76
77
78
79
80
80
81
82
83
84
8589
8790
91
92
93
94
9598
96
97
99
100
100
100
99
98
97
96
95
94
93
92
9188
90
89
87
86
85
84
83
82
81
120
78
80
7773
76
75
74
72
7167635955
70
69
140 160 180 200
Decluttered200 Variable
31
180
1112b
160 5b
14 23c
140 10
12a 15c
120
8
Column 3
5a 15a
9c 12c 16 44
100
13 15b
80
7
60 23b 46b
9b 22 43a
40
18c 23a 27a 42a
20 42b 45
6a
6b 18a 24
28 33a
37 43c 46a
6c
2a3a4a4b 1718b 20a 34
33b35 39a40 42c 46c49a
47 49b
0 11b 20b
21 25
26a
26b 29
30a32 33c 36a 39b39c
41
10 20 30 40 50 60 70 80
Decluttered Sample
0.35
0.3
14
17
0.25
1813
20 max. fluor. 19
0.2 21 At interm.depthts 20 12
Row 1
17
0.15 16 23
21
24 11
15 25
0.1
26 22 10
1427 23
0.05 9
28
MCR-ALS 13 30
3337414549
25 8
12
0 12345678910
11 535761
6266707478
68 808387919599
79 97
100
99
98 9288
9693 878379
78
77 62585450464238
75716763 3733
3532
31
30
29
28
27 7654321
C1 20 40 60 80 100
Variable
120 140 160 180 200
Decluttered
200
180 43b
42c
42a
160 23b
43c
41 43a
140
5a 39b
13 21
120
46c
Column 1
24 46b
12b 16
100 5b
11
10 39c
80 14
15c 20b 36a
60 31
9a 15b 20a 23a
23c 26b 33b
9c
8 9b 33a 45
40 28 4849b
18a 30a
6c 37
20 27b 33c 42b 44
6b
6a 12a 22 46a
3b 29 35
2a 4a 4749a
0 11b2b
3a 4b 7 1718b
18c 25
26a 32 34 36b3839a40
10 20 30 40 50 60 70 80
Decluttered Sample
0.25 26
25
23
28 max. fluor.
0.2 2429 28 22
31
at interm.-high depths
29
23 32 21
0.15
31
Row 2
22 20
34 19
32
0.1 21 33 18
35
35
20 37 17
0.05 38
36
39 38
MCR-ALS 19
18
40
44
39 16
741
47
C3 11
0 12345678910
12
13
1417
15
16 48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
6367
64
656872 7782
7376
94 84
83 89 99 9590 79 7368 62
61
60
59
58
57
56
55
54
5350
51
49
48
47
4642
44
43 15
14
13
129
11
10
20 40 60 80 100 120 140 160 180 200
Decluttered Variable
180
23a
160 14 25
140 26a
11 26b
120 20b
Column 2
0.18 38
3741
max. fluor.
0.16 33 42
43 at high depths
0.14 45 36
0.12 32 4853 4339 32
52 34
54
Row 4
0.1 31 45
56 47
60 29
0.08 30 59 49 28
29 62
64 5854
0.06 68 605652 27
15
16
10 28 697377 63 14
0.04 3 6 13 6864 11
8 27 7175 71 17
8186
848892 100 8884807672
MCR-ALS 0.02
15 26
96 989490
2521 10
18 9
0 16
17
18
19
20
2125
22
23 22
23 8765321
C4 20 40 60 80 100 120 140 160 180 200
Decluttered Variable
250
16
200
150
Column 4
15a
11 15c
100 10 15b 44
34
26b
32
35 39b
50 18c 25 27a 30a 39a
14 17 37 40 46b
89a 18a 26a27b
4b5b6a 22 33a
33c
1 20a 33b 38 47
1b 5a 6b6c7 9b 12a 21 24 29
28 36b 43a 46c
46a
4a 12c 43c 4849a
0 2a3a
2b3b 9c 13 23a
23b
23c 31 36a 41 42c 45
42a 49b
10 20 30 40 50 60 70 80
Decluttered Sample
SIMPLS Variance Captured and Statistics for yred11
Fluorescence PLS prediction from others 8
7.8
7.6
Linear regression model using
Partial Least Squares calculated with SIMPLS 7.4
RMSECV, RMSEC
Cross validation: random samples w/ 8 splits 7.2
0.2
0.2 temp cond latd
Loadings on LV 1 (29.00%)
long
salt oxyg
-0.4
-0.2
btra
-0.6 -0.3
13
RMSECV, RMSEC
Partial Least Squares calculated with the SIMPLS
12
Cross validation: random samples w/ 8 splits
Percent Variance Captured by Regression Model 11.5
-----X-Block----- -----Y-Block----- 11
Comp This Total This Total
10.5
---- ------- ------- ------- -------
10
1 34.94 34.94 33.23 33.23
2 39.55 74.49 10.53 43.76 9.5
1 2 3 4 5 6 7
3 11.72 86.21 12.44 56.20 Latent Variable Number
0.3
0.1
0 1
-0.1
press press
-0.2 0.5
temp cond
% worldmap of the desired lat and long, with the coast lines
worldmap(latlim,lonlim)
load coast
geoshow('landareas.shp', 'FaceColor', [0.15 0.5 0.15])
plotm(latdr,longr,'r--')
textm(latdr,longr+0.4,ident)
10 CTD measured variables:
scatterm(latdr,longr,y,y,'filled');
1 Depth, press
gridm on
2 Temperature, temp
3 Conductivity, cond
linem(latdr,longr,y);
4 Salt concentration, salt
5 Oxygent dissolved, oxyg
6 beam light transmission, btrm
7 fluorescence, fluor
8 turbidimetry, turb
9 latitude, latd
10 longitude, long
Fluoresc in dcm
81 ° N
90
42b
42c39a
39b
39c
42a36a
36b
41
40 35
34
38
37
43a 17
18a
18c
18b
4443c
43b 33c
33b
33a 16
30a
28
80 ° N 2526b
26a
32
20b
20a 15b
15c
15a
45 31 29 21
14 85
46a
46c
46b
24
27b
27a
49a
49b
48 47 13
22 80
12c
12b
12a
23a
23b
23c
79 ° N 11
10 75
9a9b
9c
8
7
78 ° N 6c
6b
6a
70
65
77 ° N°
2 E3 ° E ° ° °
° E 6° E17° E18 E19
4 E5 ° E 6 ° E
° ° ° ° ° ° ° °
E 7 E 8 E 9 E10 E11 E12 E13 E14 E15 1
Salt conc in dcm
81 ° N
35
39a
39b
39c
42a
42b
42c 36a
36b 34.8
41
40 35
34
38
37
43a 17
18a
18c
18b
4443c
43b 33c
33b
33a 16
30a
28
80 ° N 2526b
26a
32
20b
20a 15b
15c
15a
34.6
45 31 29 21
14
34.4
46a
46c
46b
24
27b
27a
49a
49b
48 47 13
22 34.2
12c
12b
12a
23a
23b
23c
79 ° N 11
34
33.8
10
9a9b
9c
8 33.6
7
78 ° N 6c
6b
6a
33.4
33.2
33
77 ° N
2 ° E3 ° E ° ° °
° E 6° E17° E18 E19
4 E5 ° E 6 ° E
° ° ° ° ° ° ° °
E 7 E 8 E 9 E10 E11 E12 E13 E14 E15 1
conduct in dcm
81 ° N
35
42b
42c39a
39b
39c
42a36a
36b 34
41
40 35
34
38
37
43a 17
18a
18c
18b
4443c
43b 33c
33b
33a 16
30a
28
80 ° N 2526b
26a
32
20b
20a 15b
15c
15a
45 31 29 21
14 33
46a
46c
46b
24
27b
27a 32
49a
49b
48 47 13
22
12c
12b
12a
23a
23b
23c
79 ° N 11
31
10 30
9a9b
9c
8
7
78 ° N 6c
6b
6a 29
28
27
77 ° N
2 ° E3 ° E ° ° °
° E 6° E17° E18 E19
4 E5 ° E 6 ° E
° ° ° ° ° ° ° °
E 7 E 8 E 9 E10 E11 E12 E13 E14 E15 1
diss oxyg in dcm
81 ° N
41
40
42b
42c39a
39b
39c
42a36a
36b
41
40 35
34
38
37
43a 17
18a
18c
18b
4443c
43b 33c
33b
33a 16 39
30a
28
80 ° N 2526b
26a
32
20b
20a 15b
15a
15c
45 31 29 21
14
38
46a
46c
46b
24
27b
27a
49a
49b
48 47 13
22 37
12c
12b
12a
23a
23b
79 ° N 11 23c
36
10 35
9a9b
9c
8
7 34
78 ° N 6c
6b
6a
33
32
77 ° N
2 ° E3 ° E ° °
° 18° E19 E
31
4 E5 ° E 6 ° °
E 7 ° E 8 ° E 9 ° E10° E11° E12° E13° E14° E15° E16 E17 E
beam transm in dcm
81 ° N
90
42b
42c39a
39b
39c
42a36a
36b
41
40 35
34
38
37
43a 17
18a
18c
18b
4443c
43b 33c
33b
33a 16
30a
28
80 ° N 2526b
26a
32
20b
20a 15b
15c
15a
45 31 29 21
14 85
46a
46c
46b
24
27b
27a
49a
49b
48 47 13
22 80
12c
12b
12a
23a
23b
23c
79 ° N 11
10 75
9a9b
9c
8
7
78 ° N 6c
6b
6a
70
65
77 ° N°
2 E3 ° E ° ° °
° E 6° E17° E18 E19
4 E5 ° E 6 ° E
° ° ° ° ° ° ° °
E 7 E 8 E 9 E10 E11 E12 E13 E14 E15 1
turbidimertry in dcm
81 ° N
90
39a
39b
39c
42a
42b
42c 36a
36b 80
41
40 35
34
38
37
43a 17
18a
18c
18b
4443c
43b 33c
33b
33a 16
30a
28
80 ° N 2526b
26a
32
20b
20a 15b
15c
15a
45 31 29 21
14 70
46a
46c
46b
24
27b
27a 60
49a
49b
48 47 13
22
12c
12b
12a
23a
23b
79 ° N 11 23c
50
10
9a9b
9c 40
8
7
78 ° N 6c
6b
6a 30
20
77 ° N 10
2 ° E3 ° E ° ° °
° 17° E18 E19 E
4 E5 ° E 6 ° ° ° ° ° °
E 7 E 8 E 9 ° E10° E11° E12° E13 E14 E15 E16 E
From vectors to raster (grids)
load atosmaps
[Z, refvec] = geoloc2grid(latdr2(9:80,:),longr2(9:80,:),yred(9:80,7),0.1);
worldmap([77,81],[2,16])
From vectors to raster (grids)
function [Z,refvec]=mapgridatos(latlim,longlim,latw,longw,y,cellsize)
% function [Z,refvec]=mapgridatos(latlim,longlim,lat,long,y,cellsize)
close
[Z, refvec] = geoloc2grid(latw,longw,y,cellsize);
worldmap(latlim,longlim)
meshm(Z,refvec)
load coast
geoshow('landareas.shp', 'FaceColor', [0.15 0.5 0.15])
plotm(latw,longw,'k+','LineWidth',1)
linem([79.5,80.5],[2,18],'w--','LineWidth',3)
colorbar
From vectors to raster (grids)
meshm(Z,refvec)
From vectors to raster (grids)
>> load coast
>> geoshow('landareas.shp', 'FaceColor', [0.15 0.5 0.15])
From vectors to raster (grids)
plotm(latdr2(9:80,:),longr2(9:80,:),'k-','LineWidth',5)
colorbar
From vectors to raster (grids)
[Z,refvec]=mapgridatos([77,81],[2,16],latdr(9:80,:),longr(9:80,:),ydeep(9:80,2),0.1);
Investigation of Arctic and Antarctica spatial and depth patterns of sea
water in CTD profiles using chemometric data analysis.
Ewelina Kotwa, Silvia Lacorte, Carlos Duarte, Roma Tauler
Polar Science, in press
Data pre-processing together with outliers and missing values detection and
elimination strategies were applied before the actual exploratory data analysis
was performed. Classical and robust versions of common chemometric tools
such as PCA, PARAFAC, PLS and nPLS were applied for data exploration and
calibration, and MATLAB's mapping toolbox was used for results geo-
referencing and visualization.
Measured variables.
Control plots for predicted Fluorescence
values;
No ”full data” ”5,10,25,50,100m data”
of ROBPCA Classical PCA ROBPCA Classical PCA
PC
This Total This Total This Total This Total
PC PC PC PC
1 61,26 61,26 62,92 62,92 59.78 59.78 55.67 55.67
2 17,02 78,28 14,26 77,18 19.67 79.46 20.78 76.45
3 8,33 86,61 6,56 83,74 7.04 86.49 8.37 84.81
2 23,44 85,09 30,96 83,16 34.33 85.31 32,30 82,34 34.43 84.51 33,03 83,13
3 7,11 92,19 8,29 91,46 6.29 91.60 8,79 91,13 5.95 90.46 8,80 91,93
-2
-4
-2
-6
-4
-8
-6
-10
-4 -2 0 2 4 6 8 10 12
-8
-5 0 5 10
Pc1
10
PC2
PC3
8
-2
-4
-6
-8
0 20 40 60 80 100 120 140 160
0.5 Comp 1
Comp 2
0.4
0.3
0.2
0.1
-0.1
-0.2
-0.3
-0.4
1 2 3 4 5 6 7
Temp Cond. Salin. Oxy. Beamtr Fluor. Seaturb
Variance coverage by 3
components PLS