Using Ellipsoid as Vertical Using Ellipsoid as Vertical Reference for Seabed MappingReference for Seabed Mapping

Why?Why?How?How?

Hanne Hodnesdal, Herman Iversen, Birgit K. Lynge, Lars K. Nesheim, Arne E. Ofstad, Stig vstedal

Presented by Noralf Slotsvik,Norwegian Hydrographic Service

(NHS)

A A definitiondefinition

A specialist is a person far away from home

Using Ellipsoid as Vertical Using Ellipsoid as Vertical Reference for Seabed MappingReference for Seabed Mapping

GPS antenna

Sea level

Echo sounder

DES

Seabed

Ellipsoide

v

DEllipsoide

H DEllipsoide=H-DES-v

DK0DK0= DMSS - Z0

Chart datum (CD)Z0

Mean sea Surface (MSS)

MSSH

DMSS=MSSH-DEllip.

DMSS

The other alternative: The other alternative: Seabed mapping with water level dataSeabed mapping with water level data

Chart Datum (CD)

Echo Sounder

Seabed

Sea level

draught

DepthCD

DepthES

Water Level (WL)

DepthCD = DepthES + draught - WLCD

Water levelWater level at the time of seabed mapping at the time of seabed mapping must be removed from the depth datamust be removed from the depth data

Methods to obtain water level data: Permanent tide gauge possibly combined with

tide zone Temporary tide gauge associated with a

permanent tide gauge to get the relation to mean sea level

Predicted tide (based on tide models) Use GPS to measure water level with ellipsoid

as reference

Permanent tide gauges Permanent tide gauges operatedoperated by NHSby NHS22 gauges along the coast of Norway og one gauge on Svalbard

Oslo

0

1

2

3

5 10 15 20 26 30

Vard

0

1

2

3

5 10 15 20 26 30

Troms]

0

1

2

3

5 10 15 20 26 30

Rrvik]

0

1

2

3

5 10 15 20 26 30

Stavanger

0

1

23

5 10 15 20 26 30

Bergen

0

1

2

3

5 10 15 20 26 30

OsloOscarsborgVikerHelgeroa

Tregde

Stavanger

Bergen

MlylesundKristiansund

HeimsjTrondheim

Rrvik

Bod

Kabelvg

Andenes

NarvikHarstad

Troms

Hammerfest

HonningsvgVard

Tidal zones: Almost equal tide within a zoneTidal zones: Almost equal tide within a zone

HAMMERFEST

1,0-25

0,9810

0,9715 0,97

25

0,99-15

0,995

0,99-20

0,99-10

0,99-10

Hammerfest

Troms

0,990

0,99-5

Range factor

Time delay

Range factor and time delay from Range factor and time delay from HammerfestHammerfest

8o 25

N

30'

oN

30'

72oN

Hammerfes t

(0.92, -15 min)

(0.92, -5 min)

(0.80, -10 m in)

(0.83, -20 min)

(Hgde- og tidskorreksjon fra Hammerfest)(Range factor, time delay)

What we need for using the ellipsoid as vertical What we need for using the ellipsoid as vertical reference for seabed mapping? (1 of 2)reference for seabed mapping? (1 of 2)

Mandatory: Vertical position of high quality

High quality GPS-receiver Software for post processing of GPS-data (Terrapos) Continuous logging of GPS-data

The position of the GPS-receiver must be known in the coordinate system of the vessel

The motion of the vessel must be known (attitude data: heave, roll, pitch and heading)

What we need for using the ellipsoid as vertical What we need for using the ellipsoid as vertical reference for seabed mapping? (2 of 2)reference for seabed mapping? (2 of 2)

Not mandatory: If we want to convert to mean sea level we must

know the difference between mean sea level and the ellipsoid. MSS (Mean Sea Surface) models can be used.

If we want to convert to Chart Datum (CD) we must know the difference between Chart Datum and mean sea level . Along the Norwegian coast the Chart Datum is equal to LAT (lowest astronomical tide) with a few exceptions. Tide models can be used to estimate LAT at high sea.

Accuracy achievable today (95%)Accuracy achievable today (95%)

Vertical position GNSS height: GNSS = 8 cm

Surface models MSS: MSS = 10 cm (high sea) MSS = ? cm (coast) LAT: LAT = 30 cm (high sea) LAT = 10 cm (coast)

Accuracy is improving

In a complete error budget all error sources must be included (echo sounder, sound velocity profile, attitude data, horizontal positioning,...)

Advantages and disadvantagesAdvantages and disadvantages Advantages:

Water level measurements are not needed

Knowledge of the draught is not needed

Do not have to consider different tidal zones

Less possibility of mismatch in overlapping survey areas by using a consistent reference

GPS measured low frequency waves not measured by the heave sensor

(GPS measures heave better than the heave sensor?)

...

Disadvantages High quality vertical position is

needed The position of the GPS-antenna

relative to the echo sounder must be known

Confusing: The ellipsoidal depth will not represent the true ocean depth

MSS an LAT-models must be known to convert to mean sea level and and Chart Datum

MSS- and LAT-models are continuously improving (version control is necessary)

...

Seabed mapping at high seas Seabed mapping at high seas and close to the coastand close to the coast

At high seas it is favourable to use the ellipsoid as reference: 1. Difficult to get good water level data at high seas2. There are good MSS-models covering these areas (except

for high latitude and areas with sea ice)3. It does not matter that the LAT-surface is of less quality at

high seas since the requirements for depth accuracy in navigational charts are weak at high seas.

NHS can still not use the ellipsoid as reference close to the coast:1. MSS-models are still of poor quality2. The requirements for depth accuracy in navigational charts

are stringent

Still it is attractive in the future to use the ellipsoid close to the coast as well

Using ellipsoid as vertical referenceUsing ellipsoid as vertical referenceHow it has been done at NHS so farHow it has been done at NHS so far

Existing production line has been used Water level (based on permanent tide gauge in

Hammerfest) has been delivered continuously to the vessel

GPS has been logged continuously (1 Hz) and post processed with Terrapos HGPS

GPS-height has been used to make a file with water level data. WLGPS = HGPS - heave - MSS ( Chart Datum)

GPS-water level (WLGPS ) has replaced former delivered water level before depth processing and quality control

GPSGPS--water levelwater level Calculating GPS-water level

WLGPS = HGPS - heave - MSS ( Chart Datum)

The calculation is reversible. We store MSS and Chart Datum values

GPS-water level contain rests of heave probably because the heave-sensor is not perfect (we want to carry out a test)

Test 1: GPS-water level has been compared with traditional water level (Svalbard and Mareano project)

Test 2: Depth-data has been processed with both GPS-water level and water level from tide gauge and then compared

Test 3: Accuracy of HGPS has been tested

Seabed mapping on Svalbard July 2005Seabed mapping on Svalbard July 2005 Comparing

traditional method with seabed mapping with ellipsoid as reference

Analysing: 20-21 July-05

Ellipsoid as reference : KMS04 from Danish National

Space Centre

Traditional method : Permanent tide gauge in Ny-

lesund Temporary tide gauge in

Hornsund

00:00 06:00 12:00 18:00 00:0034

34.5

35

35.5

36

36.5

37

Time

[

m

]

Measured GPS height, 20 July 2005

GPS height with heaveGPS height with heave removed

03:07 03:08 03:0934

34.5

35

35.5

36

36.5

37

Time

[

m

]

Measured GPS height, 20 July 2005

GPS height with heaveGPS height with heave removed

GPSGPS--water level and water level from water level and water level from temporary tide gaugetemporary tide gauge

00:00 06:00 12:00 18:00 00:0034

35

36

37

[

m

]

Measured GPS height, 20 July 2005

GPS heightMSS

00:00 06:00 12:00 18:00 00:00-1

0

1

2

Time

[

m

]

GPS height rel. MSSWater level, Hornsund

Mean of difference is 12.5 cm

GPSGPS--water level and water level from water level and water level from temporary tide gaugetemporary tide gauge

00:00 06:00 12:00 18:00 00:0033

34

35

36

[

m

]

Measured GPS height, 21 July 2005

GPS heightMSS

00:00 06:00 12:00 18:00 00:00-1

0

1

2

Time

[

m

]

GPS height rel. MSSWater level, Hornsund

Mean of difference is 15.5 cm

Range factor and time delay from Range factor and time delay from HammerfestHammerfest

8o 25

N

30'

oN

30'

72oN

Hammerfes t

(0.92, -15 min)

(0.92, -5 min)

(0.80, -10 m in)

(0.83, -20 min)

(Hgde- og tidskorreksjon fra Hammerfest)(Range factor, time delay)

Water level relative mean sea levelWater level relative mean sea level

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00

-1.0

0.0

1.0

2.0Hammerfest (no correction)Hammerfest (average secondary harbour corrections)Hammerfest (interpolated secondary harbour corrections)Water level from GPS

Water level relative mean sea levelWater level relative mean sea level

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00

-1.0

0.0

1.0

2.0Hammerfest (no correction)Hammerfest (average secondary harbour corrections)Hammerfest (interpolated secondary harbour corrections)Water level from GPS Filtered water level from GPS

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00Time [s]

-1.0

0.0

1.0

2.0

H

e

i

g

h

t

[

m

]

Hammerfest (no correction)

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00Time [s]

-1.0

0.0

1.0

2.0

H

e

i

g

h

t

[

m

]

Hammerfest (no correction)Hammerfest (average secondary harbour corrections)

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00Time [s]

-1.0

0.0

1.0

2.0

H

e

i

g

h

t

[

m

]

Hammerfest (no correction)Hammerfest (average secondary harbour corrections)Hammerfest (interpolated secondary harbour corrections)

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00Time [s]

-1.0

0.0

1.0

2.0

H

e

i

g

h

t

[

m

]

Hammerfest (no correction)Hammerfest (average secondary harbour corrections)Hammerfest (interpolated secondary harbour corrections)Water level from GPS

05/11/06 0:00 05/11/06 12:00 05/11/07 0:00 05/11/07 12:00 05/11/08 0:00Time [s]

-1.0

0.0

1.0

2.0

H

e

i

g

h

t

[

m

]

Hammerfest (no correction)Hammerfest (average secondary harbour corrections)Hammerfest (interpolated secondary harbour corrections)Water level from GPS Filtered water level from GPS

Tide gauge GPS - heights

5-10 m depth, EM3000D

Ca 50 m

Tide gauge GPS - heights

40 m depth, EM1002

Ca 100 m

Tide gauge GPS - heights

40 m depth, EM1002

Ca 50 m

Further work at NHSFurther work at NHS Document the accuracy of GPS-height

Test has been carried out

Find out how to handle heave Can we use GPS-height to measure heave? Test?

Document error sources in connection with traditional water level measurements

Compare GPS-water level and traditional water level

Consider and possibly implement changes in the production line Should depth data be stored relative to the ellipsoid, mean

sea level or chart datum?

Provide MSS and Chart Datum surfaces as well as a system for updating and version control

ConclusionsConclusions To use the ellipsoid as vertical reference for

seabed mapping is today a relevant method since the GPS-height has become more accurate.

At high seas this method is especially favourable since it is difficult to provide reliable water level data

There are MSS-models covering high seas. These can be used to convert from ellipsoid to mean sea level

Tide models can be used to make Chart Datum surfaces. These can be used to convert from mean sea level to Chart Datum.

GPS-height contain heave. When we subtract heave measured by the heave sensor we still observe rests of heave.