“Earth observation of vegetation dynamics: global ...

“Earth observation of vegetationdynamics: global observations,

local issues, national implementation ”

Martin Herold

Wageningen University

www.fao.org/gtos/gofc-goldwww.gofc-gold.uni-jena.de Global Observations of Forest Cover and Land Dynamics

What is GOFC -GOLD?• A technical panel of the UN Global Terrestrial Observing

System (GTOS/FAO)

• A coordinated international effort: – to ensure a continuous program of space-based and field forest and

land observations for global monitoring of terrestrial resources

• A network of participants implementing coordinated research, demonstration and operational projects

• A vision to share data, information and knowledge

• GOFC-GOLD operates through:– Executive committee, science and technical board

– Implementation teams and 3 project office (CA, US, Germany?)

– Dedicated working groups (REDD, GEO task, biomass)

– 6 Regional networks (Central/West/East Africa, SE-Asia and Latin america)

Integrated land cover observations

From Herold et al 2008, IEEE Systems

Effort

for f

requen

t updat

e

Thematic detail

Spa

tial d

etai

l

high

highlow

Assuming observation continuity and consistency

IN-SITU (+ IKONOS type)periodically (usually 1-10 yrs)

Detailed physiognomyFloristics and species distributionLand use: i.e. crop type/rotationCalibration and validation

high

LANDSAT/SPOT – typeinter-annual (1-5 yrs)

Vegetation physiognomy

Land change processes

Land type/Phenology

MODIS

/MERIS

(intra

-)ann

ual

patte

rnLo

ng-te

rm

trend

s

In-situperiodically(10-15 yrs)

In-situperiodically(10-15 yrs)

high

high

highlowthematic detail

spat

iald

etai

lup

date

effo

rt

�veg. physionomy �local patterns

�phenology

Integrated concepts for ecosystem assessment

land-cover information

LandsatInter-annual

1-3 yrs

�life form �cover�height�species richness�species abundance�soil properties

MODIS8 days

Vegetation type mapping in the Kalahari

Bottom-up vegetation mapping

422 relevé point data available

Upscaling from point to a coarse MODIS scale (232-m pixel size)

Classification of 16-day MODIS time series using Random Forests

In-situ

Landsat

MODIS

Vegetation type mapping in the Kalahari

(Hüttich et al. 2009)

Effect of aquisition time and temporal composite on mapping accuracy



Effort

for f

requen

t updat

e

Thematic detail

Spa

tial d

etai

l

high

highlow




high




Land type/Phenology

MODIS

/MERIS

(intra

-)ann

ual

patte

rnLo

ng-te

rm

trend

s

Harmonized land cover characterization

Existing global land cover datasets

Deciduous

ArtificialSnow & Ice

BareHerbaceous

Shrubs

Common land cover classifiers (LCCS)

Trees

Cover type/ life form

Evergreen

Leaf longevity

Leaf type

Broadleaved

Needle-leaved Cultivated/managedAquatic/

flooded

“Living” validation database for comparative assessment

SYNMAP – for carbon cycle modeling

SYNMAP – a global synthesis product of existing global land cover maps to provide a targeted and improved land cover map for carbon cycle modelling purposes; here shown as life form assemblages (Source: M. Jung et al. 2006, Remote Sensing of Environment).

GLOBCOVER (2005/6)

Dataset release: September 2008

Thematic standards Reference database (GLC2000)

Comparative validation & assessment

Thematic standards Reference database (GLC2000)

Comparative validation & assessment

Probability

FAO FRA 2010 –remote sensing survey

~ 13,500 monitoring

sites

IPCC 4th assessment report

Source: http://ipcc-wg1.ucar.edu/wg1/Report/AR4WG1_Pub_TS.pdf

Local issues

Martin Herold





Effort

for f

requen

t updat

e

Thematic detail

Spa

tial d

etai

l

high

highlow




high




Land type/Phenology

MODIS

/MERIS

(intra

-)ann

ual

patte

rnLo

ng-te

rm

trend

s

Ground-based LIDAR measuremehts

3/15/2010

Ground-based full waveform lidar. Lidar pulses (1064nm) strike a rotating mirror at an angle of 45°, provide a scan through zenith angles of ±130°in a v ertical circle. As the instrument rotates on its vertical axis, data from all azimuths are acquired.

Courtesy/Cooperation/Boston University and CSIRO Australia

Hard & Soft Returns in EVI DataHard & Soft Returns in EVI Data

Tree Trunk Foliage


McKinley Sequoia Grove


Laboratory spectro�goniometer system

full BRDF

Angular measurements of:

- ReflectanceASD Fielspec 3 + Solar illuminator

- TemperatureNEC imaging thermal camera

- FluorescenceLaser + camera / fieldspec

Soil spectroscopy for Methane emissions: Data collection

� Soil sampling� Profile descriptions along transects� Plot sampling (dec vs slightly dec)� Spectral measurements� Chemical analysis for model

calibration (pH, C, N, K, P)

� Vegetation descriptions� Species composition� Fractional cover estimates� Spectral measurements

� Hyperion image

Art, Science and Applications of Reflectance Spectroscopy Symposium, February 23�25, 2010, Boulder, CO, USA

Soil spectral analysis

2.511.972.651.171.79RPD

4.6940.257171.64636.8970.4RMSE

0.8410.7470.8560.3010.683R2

99964No of Factors

MCMCMCMCCentre/scale

Med10Med8Med8SG23De-noising

R2LogR2LogR2LogR2LogR2LogData transformations

Total C(%)

Total N(%)

Total K(mg/100g)

Total P(mg/100g)

pH

National issues and implementation

Martin Herold



Prominent international political drivers

1. United Framework Convention on Climate Change:

� Reduce uncertainties in monitoring the global climate system through observing

essential climate variables (ECV)

� Capacity building needs to address stronger role of developing countries in post-

2012 agreement - Reducing emissions from deforestation (REDD)

2. Provide societal benefits through evolving land

observations as part of a Global Earth Observation

System of Systems (GEOSS)

4. Global Monitoring for Environment and Security

(GMES)

5. Effectiveness for global assessments:

� IPCC, UN-FAO forest resources assessment etc.

� Ongoing global land cover mapping activities (GLOBCOVER, MODIS)

Key issues for observing ECV ’s

www.fao.org/gtos/topcECV.html

1. Building upon coordinated observations

2. Standardized and integrated mapping

3. Independent accuracy assessment

4. Synergy among datasets and different ECV products

5. User engagement

Observing land cover as ECV

www.fao.org/gtos/topcECV.html

Current distribution of aboveground biomass

Terrestrial carbon lost through human activities

Variability in capacities for REDD monitoring

Consideration of factors:1.Requirements for monitoring forest carbon on national level (IPCC GPG)2.Existing national capacities for national forest monitoring3.Progress in national GHG inventory and engagement in REDD4.REDD particular characteristics: importance of forest fires, soil carbon, deforestation rate etc.5.Specific technical challenges (remote sensing): cloud cover, seasonality, topography, remote sensing data availability and access procedures

Capacity gap:

Capacity gap

Source: Herold, 2009 http://princes.3cdn.net/8453c17981d0ae3cc8_q0m6vsqxd.pdf

Some technical challenges for remote sensing

Mean annual cloud cover

Seasonality

Topography

Country coverage of Landsat 5 receiving stations

Mean annual cloud free country coverage with SPOT data 2006-08

Average internet download speed

Percent

Variability in cloud cover (%)

Area with steep slopes (%)

Source: Herold, 2009 http://princes.3cdn.net/8453c17981d0ae3cc8_q0m6vsqxd.pdf

Earth observation contribution to UNFCCC-REDD

2005

GOFC-GOLD working group est.

COP11Montreal

GTOSsubmission I

Subm. of view by parties I

1st GOFC-GOLD workshop

ED/GOFCSBSTA

side event

SBSTA 24

1st SBSTA workshop

GTOS/GOFCreport

COP12Nairobi

ESA/GOFCSide event +

booth

Subm. of view by parties II

2nd SBSTA workshop

SBSTA 26

2nd GOFC-GOLD workshop

ESA/GOFC-GOLD

side event

GTOSsubmission II

COP13Bali

1.+ 2. REDD Sourcebook draft

WINROCK/GOFC-GOLDside event

2006 2007 2008 2009

Subm. of view by parties III + IV

GTOSsubmission III

SBSTA 28

GTOS/GOFC-GOLD

side event

3rd GOFC-GOLD workshop

COP14Poznan

3rd SBSTA workshop

SBSTA 30COP15Copenh.

PNG / GOFC-GOLDside event

Informal REDDworkshop

GOFC-GOLDside event

3. REDD Sourcebook version

GEO forest carbon tracking task demonstrator countries & sites

Remote sensing support for carbon estimation

�Direct quantitative biomass mapping from space remains

a challenge for national level estimations

�Existing capabilities:

� Satellite observation may help to map some specific forest

types (i.e. mangroves, plantations etc.)

� Targeted remote surveys to support carbon monitoring:

• Very high resolution satellite or airborne data of air-photo quality to assist field surveys

• Sensitivity of LIDAR and long-wave RADAR observations

• Integration of in-situ and satellite data for large scale biomass mapping

• Direct estimation of emissions from fire radiative power

� Technologies are not operational globally but evolving

Some key issues• Availability of baseline observations on all scales

• In situ data as weakest point?

• Take advantage of full remote sensing potentials• New sensors (Sentinels, ESA-BIOMASS)

• Verification of land (area) change for scientific credibility and compliance

• From land cover to change in land use and dynamic biophysical parameters

• Synergy different ECV observations and products (i.e. fire, biophysical parameters, snow)

• Global monitoring, local issues and cal/val, and national implementation are country specific issues!

WRSLN: Werkgemeenschap voor

remote sensing in landbouw en

natuurbeheer

Members

� vacancy (chair)

� ir. M.C. Bronsveld, ITC

� dr.ir. A. van Dijk, DHV

� dr. R.J.A. Grim (NSO)

� ir. P. Hoefsloot (Hoefsloot Software)

� ir. I. Janssen, RWS�DID

� prof.dr. S.M. de Jong, UU

� dr.ir. R.E.E. Jongschaap, PRI

� ir. A. Klaasse, WaterWatch

� dr.ir. H.J.C. van Leeuwen (Argeops)

� Dr.ir. C.A. Mücher, Alterra

� dr.ir. P.J. van Oevelen, GEWEX

� ir. A. Rosema, EARS

� ir. C. v.d. Sande, NEO

� ir. C. Schotten, PBL

� dr. R. Sluiter (KNMI)

� prof.dr. W. Verhoef, ITC

� dr. R.A. Roebeling, KNMI

� dr.ir. J.G.P.W. Clevers (secretary)

In white the “active” members

Mission

� The WRSLN is an independent platform where experts from government, science and industry participate on the theme remote sensing and land applications.

� The group promotes national cooperation, coordinates research activities, informs (potential) users and advises authorities in organizational, programmatic and research specific issues within the mentioned theme.

Objectives

� Act as intermediate between the field and (inter)national agencies

� Stimulate cooperation and exchange of information between persons and institutions active in the field of land applications with RS

� Develop a vision on the chances and threats of RS for the land domain, supported by members from government, industry and research

� Increase awareness of remote sensing possibilities for land applications

Some key issues

� Political developments poses new requirements on land remote sensing of national and international importance

� Lack of strategic investments in land remote sensing within the Netherlands

� Potential to foster improved services (uses and users national/international), education and more fundamental research (new sensors, technologies and applications)

� NSO perspectives on roles and mandates for WRSNL

� WRSNL should be ready to respond

� WUR remote sensing keen to continue/increase efforts

“Earth observation of vegetation dynamics: global ...

Documents