GIS and Remote Sensing for Environment Statistics

The use of GIS and remote sensing for environment statistics

Jacques-André NDIONE1, Mamadou FAYE2

1- Centre de Suivi Ecologique (CSE), BP 15532, Fann Residence, Dakar, SENEGAL [email protected] 2- Agence Nationale de la Statistique et de la Démographie (ANSD), (ANSD), Dakar, Sénégal, [email protected], [email protected] Outline

• Introduction • Concepts – Environment statistics – Remote sensing – GIS • Examples of using RS and GIS • Conclusion

Climate change should be added… Ongoing dialogue between data demand and data supply… Environment statistics

“Environment statistics are statistics that describe the state and trends of the environment, covering the media of the natural environment (air/climate, water, land/soil), the biota within the media, and human settlements.”

UNSD. 1997. Glossary of Environment Statistics, Studies in Methods, Series F, No. 67 Environment statistics (2) Scope of Environment Stats

Perception of major users

and producers

(UNECA)

Socioeconomic and Specific to Ezigbalike environmental particular

policies conditions

Dozie

: : Courtesy Environment statistics (3) FDES

The challenge of producing environment statistics

(UNECA)

Ezigbalike

Dozie

: : Courtesy

Remote sensing: definition

• Remote sensing is the collection of information about an object without being in direct physical contact with the object.

• Remote Sensing is a technology for sampling electromagnetic radiation to acquire and interpret non-immediate geospatial data from which to extract

information about features, objects, and classes on the Earth's land surface,

oceans, and atmosphere. (UNECA)

- Dr. Nicholas Short

Ezigbalike

Dozie : :

9 Courtesy Elements involved in Remote sensing

1. Energy Source or Illumination (A) 2. Radiation and the Atmosphere (B) 3. Interaction with the Object (C) 4. Recording of Energy by the Sensor (D) 5. Transmission, Reception and Processing (E) 6. Interpretation and Analysis (F) 7. Application (G)

Courtesy: Dozie Ezigbalike (UNECA) Satellites can detect a wide range of reflected or emitted frequencies of electromagnetic radiation. Sources de données Niveau de collecte • Imagerie satellitaire - Météosat, MODIS - NOAA - LANDSAT - SPOT - Corona - QuickBird…

• Prises de vue aériennes - Photographie aérienne - Vols systématiques de reconnaissance - Vidéographie aéroportée

• Etudes de terrain - Inventaires - Mesures GPS - Enquêtes Remotely sensed imagery products.

Sensor Panchro Visible NIR Coverage Cost matic /product1 Landsat-7 ETM+ 15m 30m 30m 1999-2003 free IRS RESOURCESAT-1 LISS-III sensor 23.5m 23.5m 23.5m 2003- $2500 present LISS-IV sensor 5.8m 5.8m 5.8m 2003- $2500 present IRS 1C/1D 5m -- -- 1998- $2500 present IKONOS 1m 1m 4m 1999- approx. $377 present OrbView-3 1m 4m 4m 2003- approx. $1800 present Quickbird .67m .67m 2.44m 2001- $600 present Airborne DMSV 0.3m 0.3m 2m 1999- $20-25k present

Sensor Resolution Coverage Cost Dickinson

: : RADARSAT 8m (Fine Beam) 1995- $3750 present

Courtesy

1 Based on 2004 prices from various online vendors. Benefits from remote sensing data

• Very useful where areas are inaccessible or where the cost of collecting spatial data over extensive areas is prohibitive Disadvantages • Provide good “pictures” for convincing the public of the value of participating in environmental assessment • Needs ground verification • Provide data over large areas that have standard format • Doesn’t offer details • Are available on a repetitive basis •andNot has the been best used tool tofor provide small data for areas over a long time basis areas

• May be used to monitor the progress• Needs of environmental expert system projects to extract data • Faster extraction of GIS-ready data

Application of Remote sensing

• Urbanization & Transportation – Updating road maps – Asphalt conditions – Wetland delineation

• Agriculture – Crop health analysis – Precision agriculture – Compliance mapping – Yield estimation

15 Courtesy: Bodruddoza Mia Application of Remote sensing (2) • Natural Resource Management – Habitat analysis – Environmental assessment – Pest/disease outbreaks – Impervious surface mapping – Lake monitoring – Hydrology – Landuse-Landcover monitoring – Mineral province – Geomorphology – Geology • National Security - Targeting - Disaster mapping and monitoring - Damage assessment - Weapons monitoring - Homeland security - Navigation - Policy

16 Courtesy: Bodruddoza Mia Applications sur le suivi des feux de brousse

300000 400000 500000 600000 700000 800000 900000 %[ Cas de feux du 1er au 10 Nov. 2008 Podor N $ Guede Village (Incluant les feux précoces) Dagana Gae Fanaye $ Ronkh $ %[ $Z $ $Z Dodele $ $ Aere Lao Ndiayene Pendao Gamadji Sarre $ Cas-Cas

$Z % 1 8

0 Cas de feux 0

0 Ross-Bethio Madina Ndiatbe 0

0 Mbane $ Chef-lieu de région 0

0 $Z %[ 0

0 $Z 0 8 Mboumba %[ Chef-lieu de département 1 $ Salde $Z $Z Chef-lieu d'arrondissement SAINT-LOUIS $ Galoya $ Chef-lieu de communauté rurale $ Syere I (Boki Nedo) Pete $ Agnam Civol Route principale %[ $ Orefonde $Z $ RaoMpal GareKeur Momar Sarr Dabia Chemin de fer $Z $ $Z $ Gandon Sakal Tessekere Forage $ Lim arrondissement $ Labgar Bokidiawe $Z $ Nguer Malal Nabadji Mbal Leona $ $ $ $ Niomre Lo Mboula MATAM $ Pete Ouarack $ %[ $ $ Yang Yang %[ LOUGA $ Gande Ogo $ $Z Dodji Lougre Thioly Thiepe $ Coky Kamb $ $Z Kanel $ $ $Z $ $Z $ $Z $ %[

Thiamene 1 7

0 Linguère $ 0

0 Kébémer Gueoul $ Sinthiou Bamambe 0

0 Ndiagne $ Ouarkhokh %[ 0

0 $ %[ $ $ $Z 0

0 Diokoul $ Ranérou Ouro Sidy 0 7 Loro Orkadiere $ $ Barkedji 1 $Z $Z $ Touba Merina Thiamene Passe %[ $Z $ $ $Z Kab Gaye $ % Ndande Sam Yabal $Z Thiargny I %% $ $ $ %% % % %% % Houdalaye Aoure Mbayene %%% % $ $ $ Sagatta Djolof %%%%%%%%% $ Koul $Z $ $ % %%%%%% %%% BokiladjiMoudery $ Mbadiane Dealy % % % % Taiba Ndiaye $Z $Z % %%%%%% $ $Z $ $ $Z $ Velingara $ Niakhene % $ $Z % Woppa %% $ Missirah % %% $Z %[ Thiel Ouolof I % % Bakel Tivaouane $Z Ndindy $Z $ %%% % % $ $ % % Ndiott %[ Mont Rolland $ $ % Boulone Seno % % $ $ $Z %%%%%%%% %Fete Bowal %% $ $ $ %%% %%%% %% Gawane $ Gassane % % %% % $ $ $ %%%% % %%%%%%%%%% $ THIES $Z $Z Ndame Taif % %%%%% % %% %% %% $ $ $ $ %% %% %%% %%% %%% % % %[ $Z %[ $Z $ %[ $Z %%% %%% %%% %% %%%% Ballou %[ $ Bambey % %%%%% % Sendou Gabou $ $ $ $ $Z $ $ %% %% %% % $ %[ $ $ $ %% % %% %% % % $Z %[ $Z Madina %%% %% % % %% % % Douby Doro %%%%% [% Ndiass $ % %% % % % % $ $ Notto $ %[ Patar % %%% %% %% %%%%% DAKAR $ $ $ $Z Colobane % % Niarouwel Djoloff % %% %%% % $ %% % % %% % $Z Fissel $ % % %%%%% % %% $ % % 1

$ Mbar %% % %% %% %%% %% 6 0 $ %

Sindia $ $Z $ $ $ % %% % 0 0 Gossas

$ % %%%% % %% %% % % % Kidira 0

0 $ Darou Miname II %%% %% % %% % 0 0 %

$ $Z $Z %[ Gniby Mbegue Goume % % % % % % $Z 0 0 $ %% % % Bele

$ $ $ %% % % % %%% %% % %% % 0 6 $Z %% Sinthiou Fissa $ %[ $Z %% %% % % % % %% % 1 % % % $% % $ Mboss Ndioum Ngaint%h%% %% % % $ $ %%%% $ %% % %% % % % %% $ Ndiebel $ % %%% %% %% % % Gathiary Sessene $ $ Ndiago % % % % % Leva Diavandel $ %[ $ Gainte Pathe% %% %% %% % % % % $ $ Boulel % % %% % % % %% % $ %% %% % % % %%% % $ Dya % $ % % % %% % % Ngueniene $ Dianke Souf Lour Escale % % %% % %% %%% % FATICK $ $ Mbadakhoune % % % % G%oud%iry % %% %%% %% % $ $ %% %%% %% $Z Kouthiaba Ouolof % % % $Z % Fimela $Z Birkelane $ % %% % %% %% %% % % % % % % Kenieb%a $Z %[ $ $Z $ Kaffrine % %% %% %% % %[ Djilor %%% % %%% % %% $Z $ %[ % %%% %% % %% % %% %% %% $ $Z Maka Yop % %%% %%% % Bala % % %% % %% % %% Palmarin Facao $Z % % % % % % % % % %%%%% %% % $ $ $Z % Koumpe%ntoum %% % % $Z D%ougu%e % Djirnda $Z Ndiognick % % % % % Koulor %% % % $ % % %%% $ Koumbal Maleme Hodar %%%% $ % % % $ % % $ $Z % % % %% %% % BassoulDiossong $ $ $ % %% %% % % % $ $ Mabo %%% $ % %% $ Thiare $ Kathiotte % % % % Kothiary % % %% Niodior $ % Koussanar % %%%% % %% %% %%%%%%%%% %% % % $ Ida Mouride % % $ %% %%% % % %% % % Ndiedieng Nganda $ Malem Niani% % %%% % Medin%a $Foulbe% $Z $ $Z % Bani Isr%ael $ $ % %% $ % %% %% %% % Toubacouta Paoskoto $Z %% $ TAMBACOUNDA % %%Ndiarendi %% $ Kayemor $ % % % %% % $Z $ Kahene %% %% % %%% % $ $Z $ % %[ % % %% % $ $Z %[ $ %%Ndog%a Babacar %% %% % %%% % % %% % $ % % % %% Ngayene % %% % % Sad%atou% Keur Samba Gueye $ $ Kahao Fally %% Goumbayel Wack Ngouna % % $ 1

$ $ Prokhane $ % % %% % % % 5 0 % %

%% %% % % 0

0 $ Neteboulou % % 0 0 Colibantang % % % %

%% $ %% % % 0 0 %

%% Missira % %%% % %%% 0 0 %

$ % 0 5 % % %% % 1 % % % % % % %%% % % $ % % % %% %% % %%% % % % % % % Gal%o% % Dialacoto % $ % $ Medina Yoro Foula % % $Z Sinthiang Koundara %% % % $ % Nemataba %%% $ % % Ndorna % % $ Medina Gounass %[ Vélingara $ $ %%% Kandia $ Diouloulou Djibidione Fafacourou $ % Missirah Sirimana Bounkiling $ Bonconto $Z $ $ % Sindian $Z Diaroume $Z Djinaki Suel Bona $ Kounkane Linkering Kafountine $ $ $Z $ $ Sare Bidji $ %% $ KOLDA $Z % $ DaboMampatim Saraya %[ $ Mlomp $ KartiackBignona SakarDiannah Malary $Z $Z $ $ $ Ba$gadadji Tomboronkoto $ %[ DjibabouyaSansamba Diende $ $ $Z Paroumba $ $ $ $ $ $Z $Z $ Ouassadou %

Karantaba $ $ $Z % 1 4

0 $ Ouonck $ Salikegne Coumbacara 0

0 $Z $ %[ $ $ $ 0

0 $ Sédhiou Tanaff Niangha 0

0 Niamone $ Salemata 0

0 Tendouck $ $Z $ % % 0 4 $ $ Kédougou Adeane $ % 1 %[ Bandafassi %[ Loudia Ouolof $ $Z $ $Z Madina Baffe Enampor Niaguis Bambali $Z $ $ Dakateli Dimboly $Z %% $ $%[ $ $ 30 0 30 60 Kilometers %% $ $Z Nyassia BoutoupaDjibanar $ Kabrousse $Z % $Z $ Sources: Images MODIS, UMD, CSE

300000 400000 500000 600000 700000 800000 900000

Le traitement des données journalières de l’imagerie MODIS a permis de faire des synthèses mensuelles et annuelles des statistiques concernant les feux de brousse. Ces synthèses sont utilisées pour une meilleure compréhension des évolutions temporelle et spatiale du phénomène des feux de brousse. Image Processing

• Image Pre-Processing - Image Restoration - Sensor Calibrations - Atmospheric Corrections - Solar Illumination Corrections - Topographic Corrections - Geometric Corrections

• Image processing - Spatial enhancement - Spectral enhancement - Classification - Feature Extraction

Courtesy: Bodruddoza Mia Image Processing Software

• ERDAS Imagine • ENVI • ILWIS • ArcGIS • PCI Geomatica

Courtesy: Bodruddoza Mia 1986

Monitoring vegetation degradation in Mau Forest on the Mau escarpment, Kenya. – 1986 (top), 2001 bottom)

Threat to the further deterioration 2001 through logging in the forest in 2001 was exposed by using remotely sensed images GIS  GIS for Data combination GIS (Geographic Information System) is used to input, store , retrieve, manipulate, analyze and output geographically referenced data or geospatial data, in order to support decision making for planning and management of land use, natural resources, environment, transportation, urban facilities, and other administrative records

 GIS for Visual Presentation Basic Functions of GIS

• Data Acquisition and prepossessing

• Database Management and Retrieval

• Spatial Measurement and Analysis

• Graphic output and Visualization

Courtesy: Bodruddoza Mia Benefits of GIS

• Geospatial data are better maintained in a standard format.

• Revision and updating are easier.

• Geospatial data and information are easier to search, analysis and represent.

• More value added product.

• Geospatial data can be shared and exchanged freely.

• Productivity of the staff improved and more efficient.

• Time and money are saved.

• Better decision can be made.

Courtesy: Bodruddoza Mia GIS Use

Locations - What is at….?

v Objects - Where is…?

Patterns - Which things are related…?

v Models - What if…?

Trends - What has changed since…?

Courtesy: Bodruddoza Mia The basic elements of a GIS

• A GIS is a 5-part system: Six Functions of a GIS: – People  Capture data  – Data Store data  Query data – Hardware  Analyze data – Software  Display data – Procedures  Produce output

Courtesy: Bodruddoza Mia Application of GIS

Area GIS Application

Facilities Management Locating underground pipes & cables, planning facility maintenance, telecommunication network services

Environmental and Natural Environmental impact analysis, disaster Resources Management management and mitigation Locating houses and streets, car Street Network navigation, transportation planning Planning and Engineering Urban planning, regional planning, development of public facilities Land Information Taxation, zoning of land use, land acquisition

Courtesy: Bodruddoza Mia Application of GIS (2) GIS can be found in most any field … but generally can be grouped into four basic categories:

• NATURAL RESOURCE MANAGEMENT » Forest & Wildlife » Hydrological » Minerals • URBAN & REGSIONAL MANAGEMENT » Land Use Planning/Environmental Impact » Public Works » Emergency Response » Legal Records

Courtesy: Bodruddoza Mia Application of GIS (3) • COMMERCIAL » Market Area Analysis » Site Selection » Routing • AGRICULTURAL MANAGEMENT » Field Records » Animal Management » Climate Change / Human Impact

Courtesy: Bodruddoza Mia Figure 3 29

Dates des premiers semis Comparaison entre les dates des premiers semis en 2002 et en 2001

Progression des infestations de criquets au Sénégal

VCI (Vegetation Condition Index) en mi-Août 2004

Standardized Precipitation Index SPI en fin juillet Appui au suivi de la campagne agricole Changes of the agrocultural sector evolution from 1900 to 2000

CSE - EROS

2009 19871973 % Classe Superficie (ha) Evolution des superficies

1973 1987 2009 1987-1973 2009-1987 2009-1973 2009-1973

Savane boisée 6275,02 4162,57 397,36 -2112,45 -3765,21 -5877,66 -2,90 Savane arborée 44300,65 11606,49 13092,73 -32694,17 1486,24 -31207,92 -15,42 Savane arbustive à arborée 95550,42 85888,83 28762,20 -9661,58 -57126,63 -66788,22 -33,00 Savane arbustive 8686,51 11061,41 24227,33 2374,90 13165,92 15540,82 7,68 Steppe arbustive à arborée 19781,01 28296,15 63991,98 8515,15 35695,83 44210,98 21,84 Steppe arbustive 10293,18 14583,71 19184,04 4290,53 4600,33 8890,87 4,39 Culture/Jachè re 15380,78 44676,35 50466,79 29295,57 5790,44 35086,01 17,34 Mare 613,18 579,81 541,37 -33,37 -38,44 -71,80 -0,04 Cuirasse 242,55 554,33 561,92 311,78 7,59 319,37 0,16 Sol nu 1146,63 834,25 1006,77 -312,38 172,52 -139,86 -0,07

Habitat 127,13 153,15 164,56 26,01 11,41 37,42 0,02

LU/LC in Barkedji region (BA, 2010; not to distributed!) Détection des surfaces en eau et robustesse des index

NDPI < Threshold 1

No Yes NDPIMIRVert Détection des mares en utilisant la MIRVert classification hiérarchique par No pond MIR < Threshold 2 arbre décisionnel basé sur le NDPI No Yes

No pond Pond

‘Objet’ # 2: Vegetation à l’intérieur des mares

‘Objet’ # 1: Vegetation en dehors des mares Comportement similaire Comportement différent

Aucune différentiation n’est possible Les 2 objets sont pourtant entre les 2 objets bien différents Composition colorée NDVI NDPI 08/26/2003 08/262003 08/26/2003

Bibliographie : Lacaux et al (2007), Tourre et al (2008), Ndione et al (2009), Vignolles et al (2010) SPOT 5 10m SPOT-VEGETATION dekadal synthesis 1km 26/08/2003 21/08/2003

1703 ha water bodies identified 100 ha (1 pixel) water bodies identified Correct pond cartography Breeding sites Conclusion  Remote sensing can be useful for applications reagrding to environment statistics

 Remote sensing allows the collection of detailed data about bio-physical characteristics that cannot be

collected by questionnaires

(UNECA)

Ezigbalike

Dozie

: : Courtesy Conclusion (2)  GIS allows the data, even questionnaire data to be visualized graphically

 GIS also provides for interpolation of values from spatial samples, and complex combination of data

based on their location for analyses

(UNECA)

Ezigbalike

Dozie

: : Courtesy Aknowledgements

Questions ?

Contact: [email protected], [email protected], [email protected] MERCI DE VOTRE AIMABLE ATTENTION

CENTRE DE SUIVI ECOLOGIQUE