Upload
dangxuyen
View
218
Download
4
Embed Size (px)
Citation preview
Information for decision making: weather 85
In 2003, the Fourteenth Congress of the World MeteorologicalOrganization (WMO) established The Observing System Research andPredictability Experiment (THORPEX). This international programmeseeks to acceler ate improvements in the accuracy and utility of high-impact weather forecasts up to two weeks ahead of an event.
Today, 10 of the world’s leading weather forecast centres regularlycontribute ensemble forecasts to the THORPEX Interactive Grand GlobalEnsemble (TIGGE) project in order to support the development of prob -abilistic forecasting techniques. Ensemble prob abilistic forecasting is anumerical prediction method that uses multiple simulations, sometimes
20 or more, in a given time frame to generate a representative sample ofthe possible future states of weather systems.
Within the realm of numerical weather prediction, ensembleprob abilistic forecasting is a major new tool for improving early warningof such high-impact events. This is particularly important for predictingsevere tropical cyclones, also known as hurricanes and typhoons, whichare the most powerful and destructive weather systems on the planet.The photo above shows damage caused by Typhoon Parma in thePhilippines in September 2009.
Source: J. Van de Keere/Bloggen.be
TYPHOON LUPIT
86 Crafting geoinformation
NCAR
CMC
NCEP
NCDC
UKMO
ECMWF
Météo-
France
CPTEC
CMA
KMA
BoM
JMA
Archive centre
Data provider
LDM
FTP
HTTP
In situ, airborne and satellite observations are used to initialize TIGGEEnsemble Prediction Systems (EPS1, EPS2, etc.). The EPS outputs can inturn be combined to generate weather prediction products. These can thenbe distributed via regional centres to national centres and end users. Tenof the world’s leading weather forecasting centres regularly contribute
ensemble forecasts to the TIGGE project. The map below shows how dataare transferred from the forecast centres to three archiving centres.
Source: WMO.
TIGGE Ensemble Prediction System
Information for decision making: weather 87
In October 2009, Typhoon Lupit approachedthe Philippines from the Pacific Ocean.Forecasters wanted to determine thestorm’s probable path and whether it wouldstrike the Philippines, adding to thedestruction recently brought by TyphoonParma, or veer off in another direction. Thisis an example of how TIGGE can help withforecasts of tropical cyclones.
Source: Hurricane Lupit 17 October 2009 MTSAT-1Rprocessed by Japan’s National Institute of Informatics.
Typhoon Lupit
Forecasting Typhon Lupit's path
The US/Japan Tropical Rainfall MeasuringMission (TRMM) observed that some ofLupit’s towering thunder storms reachedas high as 14 kilometres (more than 8.5miles), indicating very powerful stormswith heavy rainfall threat.
Source: TRMM.
TRMM Precipitation Radar
200km
Light rain Moderate rain Heavy rain
15 20 25 30 35 40 45 50 50 55 dBZ
10/17/2009 1629Z LUPIT West Pacific
15
10
5
0km
88 Crafting geoinformation
Information for decision making: weather 89
Forecast paths of Typhoon Lupit, 18 October 2009, 12:00 UTC, from six of the TIGGE data providers, as displayed on the JapaneseMeteorological Research Institute’s tropical cyclone forecast website.
The colour changes every 24 hours along each forecast track. The blackline is the storm’s actual path (as recorded a posteriori ).
115ºE 120ºE 125ºE 130ºE 135ºE
20ºN
25ºN
10ºN
15ºN
115ºE 120ºE 125ºE 130ºE 135ºE
20ºN
25ºN
10ºN
15ºN
21 members 24 members 15 members
115ºE 120ºE 125ºE 130ºE 135ºE
20ºN
25ºN
10ºN
15ºN
Four-day forecast of Typhoon Lupit’s path, 18 October
11 members 51 members 51 members
115ºE 120ºE 125ºE 130ºE 135ºE
20ºN
25ºN
10ºN
15ºN
115ºE 120ºE 125ºE 130ºE 135ºE
20ºN
25ºN
10ºN
15ºN
115ºE 120ºE 125ºE 130ºE 135ºE
20ºN
25ºN
10ºN
15ºN
Source: JMA.
90 Crafting geoinformation
100ºE 110ºE 120ºE 130ºE 140ºE 150ºE 160ºE 100ºE 110ºE 120ºE 130ºE 140ºE 150ºE 160ºE
Ensemble forecast tracks Strike probabilitiesLupit 18 Oct 2009
Colour change each 24 hours in forecast
Probability the storm will pass within 75 miles
5-19% 20-39% 40-59% 60-79% 80-100%
T0-24
T24-48
T48-72
T72-96
T96-120
T120-144
T144-168
T168-192
T192-216
These images show the UK Met Office’s early forecasts for Lupit (left).Based on these tracks, it was forecast (right) that there was a highprobability of the typhoon striking the northern end of the Philippines– but there is also a hint that the hurricane could instead turn towardthe northeast.
Source: UK Met Office.
Six-day forecast of Typhoon Lupit’s path, 18 October
Information for decision making: weather 91
SouthKorea
Japan
Taiwan
Hong Kong
Macau
Philippines
Vietnam
Lupit
21 October 2009 12ZECMWF (51) members)UKMO (51 members)NCEP (21 members)
Laos
Cambodia
Thailand
China
Later ensemble forecasts showed increasing probability that Lupitwould turn north-eastward, as shown on this web site display from theUS National Oceanic and Atmospheric Administration (NOAA). Forecasttracks are shown in colour, with the actual track in black. Lupit is shownclearly turning north, sparing the Philippines this time.
Source: NOAA.
Forecasting Typhoon Lupit’s path: 21 October
Ecosystems provide many valuable products and services, from food,fuel and fibre to purification of water, maintenance of soil fertility andpollination of plants. To sustain these societal benefits, managers needto fully understand the types, spatial patterns, scales and dis tributionsof the ecosystems under their care. Some of the critical tools they relyon are ecosystem classifications and maps at scales ranging from globalto local. The ecosystem mapping activities described in the followingpages are being carried out by the US Geological Survey and its partners. Source: USGS.
ECOSYSTEM SERVICES
92 Crafting geoinformation
Information for decision making: ecosystems 93
EC
OSY
STE
M
Macroclimate
Topoclimate
Biota
Landform
Surface water
Soils
Groundwater
Bedrock
An ecosystem can be viewed as a spatial integration of itslandforms, climate regime, vegetation and other features thatoccur in response to the physical environment.
Ecosystem structure
Source R. Bailey/US Forest Service
94 Crafting geoinformation
Tropical pluvial
Tropical pluvial/seasonal
Tropical xeric
Tropical desertic/hyperdesertic
Mediterranean pluvial/seasonal oceanic
Mediterranean xeric oceanic
Mediterranean desertic/hyperdesertic
Temperate hyperoceanic/oceanic
Temperate xeric
Boreal hyperoceanic
N
0
0
500 1,000 km
250 500 miles
Bioclimate maps
Satellite and in situ observations are the building blocks for creatingthe elements of ecosystem structure. The observations are processedusing a range of techniques; for example, bioclimate is modelled fromin situ and automated weather station observations.
Key environmental data layers are first developed for large regions or continents. They include landforms, geology, bioclimate regions and land cover. These data layers serve as input data into models to map ecosystems.
Source: TNC/NatureServe.
Information for decision making: ecosystems 95
Tree cover, broadleaf evergreen forest
Bamboo dominated forest
Tree cover, broadleaf deciduous forest
Mangrove
Freshwater flooded forest
Permanent swamp forest
Temperate evergreen forest
Temperate mixed forest
Temperate deciduous broadleaf forest
Converted vegetation
Degraded vegetation
Herbaceous cover
Shrub savannah
Periodically flooded savannah
Closed shrubland
Sparse herbaceous
Periodically flooded shrubland
N
Moorland/heathland
Closed steppe grassland
Grassland
Barren
Desert
Salt pan
Water
Montane transitional forest
Montane flooded forest
0
0
500 1,000 km
250 500 miles
Land cover maps
Land cover maps are produced by using statistics to categorize regionsin multispectral satellite images as particular land cover types.
Source: TNC/NatureServe.
96 Crafting geoinformation
N
Forest
Flooded forest
Savannah
Flooded savannah
Shrubland
Flooded shrubland
Grassland
Desert
Barren
Salt
Water
Converted
0
0
500 1,000 km
250 500 miles
Terrestrial ecosystems
Towards a global ecosystems map: South America
The ecosystem structure approach to ecosystem mappinginvolves mapping the major attributes of the physicalenvironment that contribute to the ecosystem’s structure.Satellite imagery, field observations and other data are usedto characterize the biological and physical environment.These data are then integrated and interpreted in order topresent a coherent picture of each ecosystem.
The resulting ecosystem map can be used for a variety of applications, such as climate change assessments, eco -system services evaluations, conservation applications andresource management.
Source: TNC/NatureServe.
Information for decision making: ecosystems 97
0
0
500 1,000 km
250 500 miles
Forest and woodland ecosystems
Herbaceous ecosystems
Shrubland ecosystems
Steppe/savannah ecosystems
Woody wetland ecosystems
Herbaceous wetland ecosystems
Sparsely vegetated ecosystems
Water
Terrestrial ecosystems
N
Towards a global ecosystems map: United States of America
The methodology for producing ecosystems maps using Earthobservation data has been imple mented for several continentalregions and a global ecosystem map is in development.
Source: USGS.
98 Crafting geoinformation
Kalahari camel thorn woodland and savannah
Limpopo mopane
Lower Karoo semi-desert scrub and grassland
Lowveld-Limpopo salt pans
Makarenga swamp forest
Moist Acacia (-Combretum) woodland and savannah
Moist highveld grassland
Namaqualand Hardeveld
Namibia-Angola mopane
North Sahel steppe herbaceous
North Sahel treed steppe and grassland
Pro-Namib semi-desert scrub
Southern Indian Ocean coastal forest
Southern Kalahari dunefield woodland and savannah
Southern Namib Desert
Southern Namibian semi-desert scrub and grassland
Sudano-Sahelain herbaceous savannah
Sudano-Sahelain shrub savannah
Sudano-Sahelain treed savannah
Upper Karoo semi-desert scrub and grassland
Wet miombo
Zambezian Cryptocepalum dry forest
Zambezi mopane
Zululand-Mozambique coastal swamp forest
African temperate dune vegetation
African tropical freshwater marsh (dembos)
Antostema-Alstoneia swamp forest
Baikiaea woodland and savannah
Bushmanland semi-desert scrub and grassland
Central Congo Basin swamp forest
Central Indian Ocean coastal forest
Drakensberg grassland
Dry Acacia woodland and savannah
Dry Acacia-Terminalia-Combretum woodland and savannah
Dry Combretum-mixed woodland and savannah
Dry miombo
Eastern Africa Acacia woodland
Eastern Africa Acacia-Commiphora woodland
Eastern Africa bushland and thicket
Etosha salt pans
Gabono-Congolian mesic woodland and grassland
Gariep desert
Guineo-Congolian evergreen rainforest
Guineo-Congolian littoral rainforest
Guineo-Congolian semi-deciduous rainforest
Guineo-Congolian semi-evergreen rainforest
Indian Ocean mangroves
Sub-Saharan ecosystems
Towards a global ecosystems map: Sub-Saharan Africa
Source: USGS.
Information for decision making: agriculture 99
The GEO Global Agricultural Monitoring Community of Practice is leadingthe effort to develop a global agricultural monitoring system of systems.Based on existing national and international agricultural monitoringsystems, this comprehensive network will improve the coordination ofdata and indicators on crop area, soil moisture, temperature, precipitation,crop condition, yield and other agricultural parameters. The end resultwill be better forecasting of agricultural yields and enhanced food security.
FOOD SECURITY
100 Crafting geoinformation
Multiple spatial and temporal scales of Earth observation data are needed for monitoring agriculture because cropping systems vary widely in terms offield size, crop type, cropping intensity and complexity, soil type, climate, and growing season. This global crop land distribution map (top), based on 250 metre resolution data from the MODIS Earth Observing Satellite sensor,is useful for monitoring global vegetation conditions and identifyinganomalies.
Source: MODIS.
The image of the Indian state of Punjab (right) is based on 30-65 metreresolution images. Even finer resolutions, down to 6 metres, are used formonitoring at the district and village levels.
Source: AWiFS/NASA.
Crop mapping
Probability0
100
Crop rotation in Punjab State, 2004-05
Rice / wheatCotton / wheatMaize-basedSugar cane-basedCotton / other cropsRice / other cropsOther crops / wheatTriple cropping
Other rotationsNon-agricultureDistrict boundary
0 30 60 km
N
MuktsarBathinda
Mansa
Sangrur
Moga
Barnala
Faridkot
LudhianaFirozpur
Amritsar
Gurdaspur
Taran Taran Kapurthala
Jalandhar
Hoshiarpur
FatehgarhSahib
Patiala
RupnagarNawan Shahar
Nagar
PAKISTA
N
PAKISTA
N
Rajasthan HaryanaHarya
na
Him
achal Pradesh
J&K
Ch
andig
arh
Information for decision making: agriculture 101
Crop yield models are a critical tool for decision making on agricultureand food security, and daily weather data are a critical input for thesemodels. These data are gathered by weather stations and coordinatedon a global basis. This image depicts zones in Europe that suffered from
high temperatures throughout June and July 2010 and where the cropmodel depicts soil moisture values for spring barley 20 per cent belowthe average.
Crop yield models
Dry and hot regions
Number of days1 - 3
4 - 6
7 - 9
10 - 12
13 - 15
16 - 18
19 - 21
21 +
Crop analysed: spring barley soil moisture/soil moisture 20% below the average
Period of analysis: 11 June 2010 - 20 July 2010
Data source: MARS agrometeorological database
Number of days with temperature over 30ºC in areas with low soil moisture
102 Crafting geoinformation
Another example based on observed meteorological data shows areaswhere crops are under stressing conditions due to consecutive days withhigh temperatures.
Number of heat waves
Number of occurrences
0
> = 1- < 2
> = 2- < 3
> = 3- < 4
> = 4
Source: National Meteorological Services
Processed by Alterra Consortium on behalf
of AGRI4CAST Action - MARS Unit
> = 2 consecutive days where TMAX > 30, cumulated valuesFrom 11 June 2010 to 10 July 2010
13/07/2010
Interpolated grid
of 25x25km
Modelling heat stress
Information for decision making: agriculture 103
Clustering: arable landbased on NDVI actual data
SPOT-Vegetation (P) from 1 October to 30 April 2010
Clusters11%
10%
14%
15%
15%
14%
19%
Masked
No data
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Produced by VITO (BE)
on behalf of the
AGRI4CAST Action
AGRICULTURE Unit
on 02 May 2010
Oct Nov Dec Jan Feb Mar Apr
2009/2010
10-daily NDVI [-]
Besides crop models which are used for qualitative forecasts, low-resolution data are used to monitor agricultural areas. Below is anexample from the JRC MARS Remote Sensing database. The map displaysthe results of a cluster analysis of NDVI values throughout the season fromMarch to June. The NDVI (Normalized Difference Vegetation Index), a
"greenness index", is an indicator of green biomass derived from satelliteobservations and widely used for vegetation monitoring. The diagramdisplays the early start of the season around the Mediterranean Basin andthe winter dormancy of most crops in central Europe.
Monitoring agricultural areas
104 Crafting geoinformation
Daily rainfall is also a key input for crop yield models. It is estimated byintegrating satellite-derived precipitation estimates with weather-station
observations, as presented in the example below from the Famine EarlyWarning System Network (FEWS-NET) system.
Rainfall estimates
8 August 2010
9 August 2010
7 August 20106 August 2010
10 August 2010 11 August 2010
Rainfall estimates 6 - 11 August 2010
Data:
NOAA-RFE 2.0
0 - 0.10.1 - 11 - 22 - 55 - 1010 - 1515 - 2020 - 3030 - 4040 - 5050 - 75> 75No data
Daily totals (mm)
Information for decision making: agriculture 105
Effective early warning of famine is vital for quickly mobilizing foodaid and other support. Areas of maize crop failure due to droughtin the Greater Horn of Africa in August 2009 are here indicated in pink and red, based on the Water Requirement SatisfactionIndex (WRSI).
Source: FEWS-NET.
Crop water requirement
Crop WRSIGrains: 2010-08-1
< 50 failure50 - 60 poor60 - 80 mediocre80 - 95 average95 - 99 good99 - 100 very goodNo start (late)Yet to start
Famine early warning
106 Crafting geoinformation
Calculating NDVI anomalies
Central America - eMODIS 250m NDVI Anomaly Period 2, 1-10 January 20102010 minus average (2001-2008)
NDVI anomaly< -0.3-0.2-0.2-0.05-9.02No difference0.020.050.10.2> 0.3Water
NDVI anomalies can be calculated on aregular basis to identify vegetationstress during critical stages of cropgrowth. An example below shows howdrought effects on crops were trackedduring 2010 over Central America usingvegetation index data. The image fromthe Moderate Resolution Imaging Spec -tro radiometer (MODIS) contrasts theconditions between data collected from2000 to 2009 (average conditions) andthe conditions under the drought of 2010.The brown and red areas on the Mexico–Guatemala border indicate the areasaffected by the drought where thevegetation index is lower than average,meaning that less photosynthesis wasoccurring.
Source: MODIS.
Information for decision making: agriculture 107
Significantly improved
Improved
Normal
Worse
Significantly worse
No data
Crop-growing profile - Shandong
Crop-growing profile - Henan
Crop-growing profile - Anhui
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.005 05 05 05 05 05 05 05
Jan Feb Mar Apr May Jun Jul Aug
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.005 05 05 05 05 05 05 05
Jan Feb Mar Apr May Jun Jul Aug
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.005 05 05 05 05 05 05 05
Jan Feb Mar Apr May Jun Jul Aug
Max 2005-09Average 2005-0920092010
Max 2005-09Average 2005-0920092010
Max 2005-09Average 2005-0920092010
Crop condition in China, April 2009
The timely and accurate assessment of crop condition is a determiningfactor in the process of decision making in response to crop stress. Cropcondition maps and crop growth profile charts of several provinces inChina in mid-April 2009, retrieved from the global Crop Watch System,show the crop condition in drought-affected areas relative to the
previous year. The crop growth profile charts of three selected provincesillustrate how crop growth responds to drought conditions.
Source: China CropWatch System.
Crop assessment
108 Crafting geoinformation
Protected areas are often seen as a yardstick for evaluating conser -vation efforts. While the global value of protected areas is not in dispute,the ability of any given area to protect biodiversity needs to be evaluatedon the basis of rigorous monitoring and quantitative indicators.
The GEO Biodiversity Observation Network (GEO BON) AfricanProtected Areas Assessment demonstrates how field observationscombined with satellite imaging can be combined to assess the value ofprotected areas.
Data for 741 protected areas across 50 African countries have beenassembled from diverse sources to establish the necessary informationsystem. The data cover 280 mammals (including the African wild dogpictured here), 381 bird species and 930 amphibian species as well as alarge number of climatic, environmental and socioeconomic variables.
Source: P. Becker and G. Flacke.
PROTECTED AREAS
Information for decision making: biodiversity 109
Dat
a in
tegr
atio
nD
ata
pres
enta
tion
Mammal
Bir
d
Agriculture Population
Amphibian
Ha
bita
t
Irreplaceabilit
y
Protected areas
Species maps
Assessing pressures on biodiversity
Indicators of Protected Areas Irreplaceability (where the loss of uniqueand highly diverse areas may permanently reduce global biodiversity)and Protected Areas Threats are developed as practical and simplifiedestimates of the highly complex phenomenon of biodiversity. They areestablished using a wide range of geographic, environmental and
species data from the World Database on Protected Areas and othersources. The habitat of each protected area is characterized on the basis of its climate, terrain, land cover and human population. The datalayers are then integrated and the multiple pressures on biodiversityare assessed.
110 Crafting geoinformation
Ia Science
Ib Wilderness protection
II Ecosystem protection and recreation
III Conservation of specific natural features
IV Conservation through management intervention
Convention on wetlands of international importance
UNESCO World Heritage Convention
Other national parks
Categories of protected area management
This map shows the protected areas in Africa.The colour code indicates their protectionstatus. The information is gathered fromnational governments and internationalagencies and is used by the assessment teamas its starting point.
Source: GEO BON.
Protection status
The following three continent-scale mapshave been processed to show the vegetationindex, the percentage of land covered by treesand crops, and the land elevation.
Source: NASA.
Vegetation index
Information for decision making: biodiversity 111
0
0.3
0.6
0.9
Vegetation index
Cropland and tree cover
30 - 40
40 - 60
> 60
Per cent cropland
< 10
10 - 30
30 - 60
> 60
Per cent tree cover
112 Crafting geoinformation
Source: NASA.
114 Crafting geoinformation
African protected areas: Value compared
to pressure
VALUE
PR
ESSU
RE
High
Low
Low High
G200 Ecoregions
30ºN
20ºN
10ºN
0º
10ºS
20ºS
30ºS
20ºW 10ºW 0º 10ºE 20ºE 30ºE 40ºE 50ºE
Based on the previous maps of pro -tection status, vegetation coverage,elevation and other variables, indic -ators have been dev eloped to scoreeach pro tected area for the value of itsbio diversity and the threats that it faces.
Source: GEO BON.
Assessing protected areas
Information for decision making: biodiversity 115
Visual products that can be understood and interpreted by a wide rangeof end users can also be created and used to inform decision making onconservation actions and funding priorities. For example, protected areasin Ghana (left) can be contrasted with all protected areas in Africa (right)to determine their relative status. The coloured sectors of the graphdepict indicators of biodiversity and habitat value (increasing to the right)
and indicators of pressure (increasing to the top). Ghana’s protectedareas are represented by the square symbols. The upper-right portionof the graphic identifies the protected areas – including several in Ghana –that have high biodiversity value and are also under high pressure.
Source: GEO BON.
Informing decision making
Mole NP
Bui NP
Digya NP
Nini-Suhien NP
Kumasi
Accra
Ankasa FRKakum NP
Semi-arid
Dry sub-humid
Moist sub-humid
Humid
Very humid
GHANA
Mole NP
Bui NP
Digya NP
Protected areas in Ghana
Kumasi
Index of value (biodiversity and habitat)
High value/low pressure
Low value/high pressure
High value/high pressure
High value/pressure
Others
All protected areas in Africa
Inde
x of
pre
ssur
e (p
opul
atio
n an
d ag
ricu
ltur
e)
1.00
0.75
0.50
0.25
0.000.00 0.25 0.50 0.75 1.00
Low value/low pressure
Average value/average pressure
116 Crafting geoinformation
CONCLUSION: GLOBAL CHANGE AND TRENDS
The nine stories in this book have described how geoinformation canbe used to support decision making in nine separate societal benefitareas. None of these issues, of course, exists in isolation. They areall interrelated: water supplies affect agriculture, ecosystems affecthealth, climate affects biodiversity, and so forth. Drawing theselinkages together in order to monitor and understand the Earthsystem as an integrated system of systems is essential foraddressing today’s complex global challenges.
The Global Earth Observation System of Systems makes itpossible to do this by assembling a large number of consistent,validated and interoperable data sets of Earth observations. Thesediverse data sets can be used to generate a snapshot of the Earth ata given moment in time. This snapshot can serve as a comprehensivebaseline against which to measure global change over the years anddecades to come. It can provide an essential point of departure forboth retrospective analysis and ongoing monitoring.
The individual baselines presented in the following pagesinclude parameters that do not change substantially over shortperiods of time but are fundamental for understanding globalchange. These relatively static data sets include elevation, soils andgeology. Also featured are data sets for continuously changingvariables that must be gathered at regular intervals. These data sets include surface reflectance, temperature, precipitation andvegetation.
The establishment of a comprehensive 2010 baseline for theEarth and its oceanic, atmospheric and terrestrial componentswould serve as a lasting contribution of the Earth observationcommunity to international efforts to protect and manage the planetfor future generations.
Source: NASA.
Topography
Global digital elevation models (DEMs) are created through the stereo -scopic analysis of multiple satellite images, in this case from theAdvanced Spaceborne Thermal Emission and Reflection Radiometer(ASTER).
Digital elevation models are used to extract terrain parameterssuch as slope, aspect and elevation. They can be used as inputs for floodprediction models, ecosystem classifications, geomorphology studies andwater-flow models.
Source: ASTER NASA.
Conclusion: global change and trends 117
The OneGeology initiative is working to make geological maps morewidely available. It has assembled maps from geological surveys aroundthe world.
Geological maps are used to identify natural resources,understand and predict natural hazards, and identify potential sites forcarbon sequestration. Source: OneGeology .
Geology
118 Crafting geoinformation
Surface reflectance
Surface reflectance images such as the map above provide an estimateof the surface spectral reflectance as it would be measured at groundlevel without the distortion of atmospheric effects. To achieve this, rawsatellite data are corrected for the effects of atmospheric gases andaerosols and the positions of the satellite and the sun.
Surface reflectance data can be used for improving land-surface
type classification, monitoring land change and estimating the Earth’sradiation budget. These data can also serve as building blocks for otherprocessed data such as vegetation indices and land cover classification.
Source: NASA/MODIS.
Conclusion: global change and trends 119
120 Crafting geoinformation
Vegetation index
Vegetation indices are created from surface reflectance data. Bycombining spectral bands that are sensitive to chlorophyll absorptionand cellular structure, it is possible to highlight variations in the typeand density of forests, fields and crops.
Vegetation index data are used for a wide variety of applications,including agricultural assessment, land management, forest-fire
danger assessment and drought monitoring. The data are also used askey inputs for land cover mapping, phenological characterization andmany other applications.
Source: ESA/MERIS.
Cultivated and managed areas/rainfed cropland
Post-flooding or irrigated croplands
Mosaic cropland (50-70%)/vegetation (grassland/shrubland/forest) (20-50%)
Mosaic vegetation (grassland/shrubland/forest) (50-70%)/cropland (20-50%)
Closed to open (>15%) broadleaved evergreen and/or semi-deciduous forest (>5m)
Closed (>40%) broadleaved deciduous forest (>5m)
Open (15-40%) broadleaved deciduous forest/woodland (>5m)
Closed (>40%) needle-leaved evergreen forest (>5m)
Closed (>40%) needle-leaved deciduous forest (>5m)
ESA GlobCover Version 2 - 300mDecember 2004/June 2006 [ENVISAT MERIS]
Open (15-40%) needle-leaved deciduous or evergreen forest (>5m)
Closed to open (>15%) mixed broadleaved and needle-leaved forest
Mosaic forest or shrubland (50-70%) and grassland (20-50%)
Mosaic grassland (50-70%) and forest or shrubland
Closed to open (>15%) shrubland (<5m)
Closed to open (>15%) grassland
Sparse (<15%) vegetation
Closed (>40%) broadleaved semi-deciduous and/or evergreen forest
regularly flooded, saline water
Closed (>40%) broadleaved forest regularly flooded, fresh water
Closed to open (>15%) grassland or shrubland or woody vegetation on
regularly flooded or waterlogged soil, fresh, brackish or saline water
Artificial surfaces and associated areas (urban areas >50%)
Bare areas
Water bodies
Permanent snow and ice
No data
Land cover data are produced from relevant data sets such as surfacereflectance, temperature, vegetation indices, and other satelliteproducts. Land cover data are created by statistically clusteringtogether pixels with similar spectral and/or temporal patterns and thenlabelling them accordingly. Large-area land cover data are used for
many applications, including change detection studies, agricultural andforest monitoring, and input to global circulation models and carbonsequestration models.
Source: ESA.
Land cover
Conclusion: global change and trends 121
122 Crafting geoinformation
Tropical rainfall
The Tropical Rainfall Measuring Mission (TRMM) is a research satellitedesigned to increase our understanding of the water cycle. Althoughrainfall has been measured for more than 2,000 years, it is still notknown how much rain falls in many remote areas of the globe, inparticular over the oceans. With the TRMM it is now possible to directlymeasure such rainfall rates. The TRMM satellite carries a passivemicrowave detector and an active spaceborne weather radar called thePrecipitation Radar (PR).
TRMM data enhance the understanding of interactions betweenthe sea, air and land. These interactions produce changes in globalrainfall and climate. TRMM observations also help to improve themodelling of tropical rainfall processes and their influence on globalcirculation. This leads to better predictions of rainfall and its variabilityat various time scales.
Source: TRMM.
Conclusion: global change and trends 123
Forest height
Many data serve as building blocks for more highly processed data sets.These “derived” data sets tend to require a substantial period of time todevelop at a satisfactory level of quality.
Scientists have used a combination of satellite data sets to
produce a map that details the height of the world’s forests. Datacollected by multiple satellites are also being used to build an inventoryof how much carbon the world’s forests store and how fast carbon cyclesthrough ecosystems and back into the atmosphere.
124 Crafting geoinformation
Sea surface temperature
This sea surface temperature (SST) map was created from datacollected by the Advanced Along Track Scanning Radiometer. The imageis an average of all data available for one year. The colours representthe sea surface temperature, from dark blue (cold) to dark red (warm).
SST measures are used to monitor and predict the El Niño and La Niña phenomena. They are extensively used in hurricane and cycloneprediction and numerical weather and ocean forecasts.
Source: AASTR/ESA.