Land Cover – Land Use - NEESPIneespi.org/web-content/meetings/IIASA/NEESPI_Laxenburg... · Land Cover – Land Use Methods Problems Perspectives ... Scatterometer vs. LPJ-DGVM.

Land Cover – Land Use

Methods Problems Perspectives

Chris Schmullius, FSU Jena

Def

initi

on

-1-

Land Cover / Land Use includes allland surface parameters:

• vegetation characteristics,• disturbances,• water bodies• snow,• anthropogenic impact.

Use

rs-2-

Who is the user?

- Carbon Community: GCMs, DGVMs, flux towers, Kyoto implementation,…

- Water Community: water re-charge, run-off, permafrost,…

- Forest Community: FRA, state inventories, logging industry, NGOs,…

- Human Dimension: infrastructure, env.protection, exploration,…

Met

hods

-3-

Satellite-driven Land Cover Projects

OpticalMODIS LC, GLC 2000, GLOBCOVER, Landsat Nightlights, NOAA VCF, also GlobCarbon, geoland,…

RadarOnly mosaics (Amazon, Africa, GRFM/GBFM) –except SIBERIA 1 mio sqkm forest cover.

BUT upcoming soil moisture product!

IS2

IS7

ASAR WS Biomass map (2004) Siberia-1 Biomass map (1997-8)

Prob

lem

s-4-ContinuityGaps: Landsat, SPOT VGT, MERIS,…

SynergyMulti-scale understanding, product-product complimentarities, optical-radar exploration (VCF&biomass!), dynamic features (water bodies, disturbances)

AccessArchives, expensive processing, global communication

Validation

Classifier ResultsSpatial resolution

MODIS, 500m

MERIS, 300m

GLC2000, 1km

Pers

pect

ives

-5-

Political• GEOSS (GTOS GOFC-GOLD participation)• GMES• GTOS IGOL

ScientificEarth System Science Networks (The Federation, GRID, INSPIRE,…)

Showcasese.g. Sib-ESS-C as part of NEESPI

SIB

ERIA

-IISIBERIA-IIDedicated to spatially explicit approaches.

Spatial patterns from Earth Observation.

Empirical Generalisation from GIS Modelling.

Multi-Sensor - Multi-Model - Multi-Approach… but always spatially exlicit.

Theoretical Generalisation from DVM Modelling.

Spat

ial H

eter

ogen

eity

in E

O d

ata Trend in snow melt

days 1978-2001

Duration ofThaw Period

Greenup dates

Land cover

Correlation Soil MoistureScatterometer vs. LPJ-DGVM

Car

bon

& P

erm

afro

st/V

eg. C

over

LPJ-PIK

LPJ-A

LPJ-P

LPJ-L

IIASA

NPP and VegC 2003

Run

off:

Perm

afro

st &

Veg.

Cov

er

Permafrost / Arctic Runoff

Fire

and

PFT

Dis

trib

utio

nFire Return Intervalls PFT Distribution

MODIS

Standard LPJ

LPJ Adapted

+ Permafrost

+ Veg. Density

SIBERIA II „SIBERIA II „TheThe last last yearyear...“, UWS: ...“, UWS: GregynogGregynog, 22, 22--24 April 200424 April 2004

Current extent of water bodies mask

- pink areas: MODIS water class (all other area classified as tundra)

20 km

mid-July

- No water bodies in the corresponding IIASA GIS layer for this area

Water bodies – accuracy assessment

(1) How much spatial heterogeneity do we need to take into account?

How can we use what we have toproduce the GHG budget of a large area?

(2) What approach do we follow?

a) Process-guided upscaling of measurements?

(accounting-oriented, precision sought)

b) Data-guided downscaling of bgc theories?

(temporal extrapolation-oriented, dynamics sought)GH

G B

udge

ts o

f Lar

ge A

reas

W. Lucht/PIK@SIB-II Final Mtg, July‘05

Is this sufficiently uniform?

Is this sufficiently uniform?

Is this sufficiently uniform?

We still lack a full theory ofecology and ecosystems: e.g. a theory linking the metabolism of cells, individuals and ecosystems.

How much heterogeneity dowe need to take into account?

Is there a justication for3°, 0.5°, 1km, 10m resolution?

Sib - E S S – C [sibessi]SIBERIA-II Earth System Science Cluster 2006-2010

Parameter Years Partner responsible

EO data(main sensors and

other sensors)

Pixel Size / Scale

Land cover 2001-2004 UWS MODISMERIS

300m to 1km

Biomass 2004 (1997/98) CESBIO ASAR( and SIBERIA-I )

75 m

Phenology 98-03 CESBIO VGT 1km to 10km

FAPAR LAI

2002-2004 DLR MODIS 1km to 10km

Snow Depth & Snowmelt date

1988-02 CESBIO SSMI 25km

Freeze/ Thaw 2000-2003 IPF Quikscat 10km

Waterbodies

2003/ 04 IPF ASAR WS 75 m

Disturbances 1990-2002,2003 monthly

CEH MODIS, ATSR-2AVHRR

1km(1:100.000)

ARD 1990vs2000 FSU Landsat, 25 m(1:50000)

Topography 1999 Gamma SRTM 3 arc sec (S)30arc sec(N)

SIBERIA-II Operational EO-Products for GHG Accounting

Prob

lem

s-4-ContinuityGaps: Landsat, SPOT VGT, MERIS,…

SynergyMulti-scale understanding, product-product complimentarities, optical-radar exploration (VCF&biomass!), dynamic features (water bodies, disturbances)

AccessArchives, expensive processing, global communication

Validation