LaSyS – workshop Tune 2005 Spatial Indicators for Landscape Development in Denmark - GIS-based Analysis of Long Term Changes and Persistence Bernd Münier,

LaSyS – workshop Tune 2005

Spatial Indicators for

Landscape Development in

Denmark- GIS-based Analysis of Long Term

Changes and Persistence

Bernd Münier, Pia Frederiksen, Gregor Levin

National Environmental Research Institute - Denmark

Department of Policy Analysis

Spatial indicators for landscape development in DenmarkBernd Münier ... LaSyS - Tune 2005

Why spatial indicators ?

• Landscape composition (a-spatial):

– OK for human utilization (census, agricultural

production, ...)

– not satisfying for ecological applications and biodiversity

issues.

• Lack of information on:

– configuration

– persistence

– => need analysis of spatial and temporal landscape

development

• CB 2010 requires spatial indicators !


Mapping local changes – 3 lake areas

0 100 200 300 Meters

1900

0 100 200 300 Meters

1900 on 1886

0 100 200 300 Meters

1886

Topographic map

=> Common boundaries1886 - 1900 - 1950 - 1996

Krag Nat4x (xe)andetmoseengOverdrevhedeStrandengSøløvnålblandet nål og løvråstofgravbyhaveuklassificeret


SignaturerandetmoseengOverdrevhedeStrandengSøløvnålanden skovby

19001886

19961950

Legendnot mappedfenswet meadowsdry grasslandsheathlandssalt marshesopen waterforest - decid.forest - conif.forest mixed.urban

Dallerup sø study area

10 km



19001886

19961950


St. Gribsø study area

10 km



19001886

19961950


Kragsø study area

10 km


Topographic maps ~ 1886 - 1900 - 1950 - 1996

Urban

Mixed forest

Conif. forest

Decid. forest

Open water

Heath

Wet meadows

Fens

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

DallerupSø

DallerupSø

DallerupSø

DallerupSø

St. Gribsø St. Gribsø St. Gribsø St. Gribsø Kragsø Kragsø Kragsø Kragsø

Dallerup sø St.Gribsø Kragsø

Composition: land cover net changes


Configuration: fate of land tracts NA8_1886 NA8_1900 NA8_1950 NA8_1996

0 0 0 0

0 0 0 02 0 0 02 2 0 02 2 2 02 2 2 25 5 5 52 6 6 6

6 6 6 66 6 8 66 6 6 76 6 7 7

0 0 0 95 5 5 9

0 0 9 99 9 9 9

Dallerup sø886 - 1900 - 1950 - 1996

NA8_1886 NA8_1900 NA8_1950 NA8_1996

0 0 0 01 1 1 12 2 2 10 0 0 22 2 2 2

5 5 5 50 0 0 61 6 6 6

6 6 6 67 6 6 68 6 6 67 7 6 68 8 6 66 6 7 67 7 7 66 6 8 60 0 0 71 1 1 7

6 6 6 7

6 6 7 77 6 7 76 7 7 7

7 7 7 78 8 7 76 6 8 78 8 8 7

0 0 0 9

0 0 9 99 9 9 9

NA8_1886 NA8_1900 NA8_1950 NA8_1996

0 0 0 0

1 0 0 0

2 0 0 0

3 0 0 01 1 0 01 2 0 0

2 2 0 0

3 3 0 02 2 2 0

3 3 3 01 1 2 12 2 2 12 2 2 23 0 0 33 3 0 33 0 3 3

3 3 3 33 3 3 63 7 7 60 0 0 73 0 0 7

3 3 0 7

3 3 3 70 0 7 73 0 7 7

3 3 7 70 7 7 7

3 7 7 73 3 9 9

Kragsø1886 - 1900 - 1950 - 1996

St. Gribsø1886 - 1900 - 1950 - 1996


Landscape spacio-temporal dynamics

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

I

Dalle

rup S

ø II

III

I

St. Grib

sø

II

III

I

Kragsø

II

II

ILakes & periods

NET_CHANGE / 2

SWAP / 2

Persistence


• Compare to development

by chance – persistence is:

– Lowest:

every trace of land may

change to every land

cover type = max entropy

– Highest:

persistence = MIN(T1,T2)

Meadows 1886 to 1900GainLossPersistentNewer

Period I: 1886-1900 Meadows [ha]

Observed persistence 633

Persistence by chance 59

Maximum persistence 975

Dallerup sø

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Other Fens Meadows Heath OpenWater

ForestDecid.

ForestConif.

Forstmixed

Build up

Period I: 1886-1900

Period II: 1900-1950

Period III: 1950-1996

St. Gribsø

0

1000

2000

3000

4000

5000

6000

7000

8000

9000


ForestDecid.

ForestConif.

Forstmixed

Build up

Period I: 1886-1900Period II: 1900-1950Period III: 1950-1996

Kragsø

0

1000

2000

3000

4000

5000

6000

7000

8000

9000


ForestDecid.

ForestConif.

Forstmixed

Build up

Period I: 1886-1900Period II: 1900-1950Period III: 1950-1996

Temporal development – how persistent is land cover?

Persistent areas. Hi-Lo bars indicate maximum persistence (Hi) and persistence by chance (Lo)


Kragsø swap 1886 vs. 1900

0%

5%

10%

15%

20%

25%

30%

35%

0 200 400 600 800 1000 1200 1400 1600Gridsize [m x 10]

Are

a [%

]

Other

Fens

Meadows

Heath

Forest

Grain size: The spatial dimension of temporal changes


0%

5%

10%

15%

20%

25%

30%

35%

0 200 400 600 800 1000 1200 1400 1600

Gridsize [m x 10]

Are

a [

ha

]

Other

Fens

Meadows

Heath

Forest


0%

5%

10%

15%

20%

25%

30%

35%

0 200 400 600 800 1000 1200 1400 1600

Gridsize [m x 10]

Are

a [

ha

]

Other

Fens

Meadows

Heath

Forest

Kragsø 1886 - 1900Other Fens Meadows Heath Forest

Rel. Swap [% TotChange] 9.6% 14.4% 83.0% 8.4% 0.0%Swap average / Swap max 0.14 0.27 0.44 0.16 -


Rel. Swap [% TotChange] 27.1% 23.5% 73.6% 9.6% 9.2%Swap average / Swap max 0.21 0.54 0.55 0.17 0.19


Rel. Swap [% TotChange] 90.0% 9.3% 27.9% 34.4% 19.0%Swap average / Swap max 0.33 0.35 0.23 0.29 0.21


Conclusions• The ‘long’ perspective spans only about 120

years.• Topographic maps contribute with extra spatial

information– but some important information is missing

• time gaps of ~ 40 years• dry grasslands not mapped

– => additional information from remote sensing

• Indicators expressing landscape configuration, persistence and changes.

• Spatial and temporal dynamics.

• Outlook: integration with physio-geographical data, agricultural census, EU subsidies, protected areas (Natura 2000,...) , environmental monitoring data etc.


Thank you very much for your attention

Karg_na3_merge.shpH H H HH H H XH H X XH X X XX X X XH H H SH H S SH S S SH X X SH H X SH X S SX X X SX X S SAnden HedeAnden SkovANDET

Kragsø 1996VandBebygget

Background pictures provided by www.naturnet.dk

LaSyS – workshop Tune 2005 Spatial Indicators for Landscape Development in Denmark - GIS-based Analysis of Long Term Changes and Persistence Bernd Münier,

Documents

spatial indicators

denmark bernd mnier

lasys tune

persistence bernd mnier

landscape configuration

landscape composition

lasys workshop tune

need analysis of spatial